The Phy and Bio of Pro

Free ebooks ==> www.ebook777.
com
The
Physiology and
Biochemistry
of Prokaryotes
FOURTH EDITION
David White
Indiana University
James Drummond
Indiana University
Clay Fuqua
Indiana University
New York Oxford

OXFORD UNIVERSITY PRESS
www.ebook777.com
Free ebooks ==> www.ebook777.com
Oxford University Press, Inc., publishes works that further Oxford University’s
objective of excellence in research, scholarship, and education.
Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Copyright © 2012, 2007, 2000, 1995 by Oxford University Press, Inc.
For titles covered by Section 112 of the US Higher Education Opportunity Act,
please visit www.oup.com/us/he for the latest information about pricing and
alternate formats.
Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
http://www.oup.com
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data
White, David, 1936-

The physiology and biochemistry of prokaryotes / by David White,
James T. Drummond, Clay Fuqua. — 4th ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-19-539304-0 (alk. paper)
I. Drummond, James T. II. Fuqua, Clay. III. Title.
[DNLM: 1. Bacteria—metabolism. 2. Archaea—physiology.
3. Prokaryotic Cells—physiology. QW 52]
571.2’93—dc23
2011037790
Printing number: 9 8 7 6 5 4 3 2 1
Printed in the United States of America

on acid-free paper
PREFACE
Introduction physiology of specific groups of prokaryotes is

The fourth edition of The Physiology and emphasized. This pattern of organization lends
Biochemistry of Prokaryotes, designed for use itself to the elucidation of general principles of
in advanced undergraduate and beginning physiology, metabolism, responses to environ-
graduate-level biology courses, provides the mental challenges, and cellular/multicellular
most current, authoritative, and relevant presen- development.
tation of prokaryotic physiology and biochem- Topics include cellular structure and func-
istry. It presents microbial metabolism in the tion, growth and cell division, chromosome
context of the chemical and physical problems replication and partitioning of chromosomes,
that cells must solve in order to grow. The text membrane bioenergetics and the proton poten-
is organized by topic rather than by organism, tial, electron transport , photosynthesis, the reg-
therefore helping students understand the gen- ulation of metabolic pathways, bioenergetics in
eral principles of physiology and metabolism. the cytosol, central metabolic pathways, RNA
This new edition builds in comprehensive cov- and protein synthesis, cell wall and capsule
erage of energetics. It also adds broad coverage biosynthesis, inorganic metabolism, C1 metab-
of molecular machinery, applied throughout the olism, fermentations, responses to environmen-
text to help create a unifying narrative across tal stress, solute transport, protein transport
biological principles. Also added is broader cov- and secretion, responses to environmental cues,
erage of chromosomes, macromolecular synthe- chemotaxis, photoresponses, aerotaxis, micro-
sis, biofilms, and cell–cell communications. bial biofilms, cell–cell communication mecha-
The prokaryotes are a diverse assemblage of nisms, and bacterial development.
organisms that consists of the Bacteria (also called
eubacteria) and the Archaea (also called archae- Distinctive Features
bacteria). This text provides an updated descrip- • Topical organization fosters understanding
tion of the major aspects of the prokaryotes, such of the general principles and concepts of the
as cell structure, biochemistry, bacterial devel- biochemistry and physiology of prokaryotes,
opment, adaptation to environmental changes, in addition to providing detailed data and con-
and signaling interactions between the cells that clusions about specific groups of prokaryotes.
occur in bacterial populations such as those living • End-of-chapter summaries and study ques-
in biofilms. The text highlights signaling mecha- tions help students synthesize material and
nisms that allow individual bacterial cells to sense prepare for examinations.
and respond to the environment, and also to sig- • An extensive references and notes section
nal each other so that they can respond as a coop- is available in the chapters to aid further
erating population of organisms. research and to provide access to the data that
supports the conclusions made in the text.
Organization • Boxed sections call out topics of special
The organization of the text is according to interest, adding historical information about
topics rather than organisms, although the earlier discoveries, covering experiments
xv
www.ebook777.com
xvi preface
that established the central tenets of micro- Acknowledgments

biology and biochemistry, or exploring in
We would like to express gratitude to the many
detail especially engaging or important top-
individuals named below, and those who remain
ics discussed in the text.
anonymous, who have read sample chapters and
have made helpful suggestions for the fourth
New to the Fourth Edition edition.
• New coauthors: David White is joined by Jennifer Anthony, University of the Sciences in
two colleagues from Indiana University: Jim Philadelphia
Drummond adds expertise on biological Theodore C. Crusberg, Worcester Polytechnic
macromolecules, while Clay Fuqua contrib- Institute
utes authority on microbial interaction. Both John E. Gustafson, New Mexico State University
are well-known authorities in their respec- Shannon Hinsa-Leasure, Grinnell College
tive fields, as well as experienced educators: Adam J. Houlihan, Wagner College
• New chapters: The fourth edition adds a Carol R. Lauzon, California State University–
more detailed account of chromosome rep- East Bay
lication, protein and RNA synthesis, and a Paul W. Lepp, Minot State University
more complete description of the biology of Robert J. C. McLean, Texas State University–
biofilms and of intercellular communication San Marcos
between bacteria: Tina Salmassi, California State University–Los
Chapter 11, RNA and Protein Synthesis Angeles
Chapter 21, Microbial Biofilms—Structured Kathleen Scott, University of South Florida
Multicellular Assemblies Timothy Secott, Minnesota State University
Chapter 22, Cell–Cell Communication Louis Sherman, Purdue University
Mechanisms Teri Shors, University of Wisconsin–Oshkosh
All chapters from the previous edition have been Om V. Singh, University of Pittsburgh
thoroughly reviewed and revised to incorporate John G. Steiert, Missouri State University
the most recent research. Ann M. Stevens, Virginia Tech
Sonia M. Tiquia, University of Michigan–
• New themes: Two new themes have been
Dearborn
incorporated across the text. A comprehen-
Thomas M Walter, Purdue University
sive coverage of energetics adds another per-
Hwan Youn, California State University–Fresno
spective to the physiological and biochemical
topics covered. A thorough incorporation of We also thank the reviewers of the third edition:
molecular machinery helps create a unifying Carl Bauer, Yves Brun, Jim Drummond, Martin
narrative across biological principles. Dworkin, Pat Foster, Heidi Kaplan, Larry
Shimkets, and Ashley Williams, as well as the
A note on chemical notation of acidic and basic
many individuals who helped by reviewing
groups: Most of the carboxyl groups are drawn
the first two editions. Thanks also to the team
as nonionized and the primary amino groups as
at Oxford University Press, including Jason
nonprotonated. However, at physiological pH
Noe, senior editor for the life sciences; Katie
these groups are ionized and protonated, respec-
Naughton and Caitlin Kleinschmidt, editorial
tively. The names of the organic acids indicate
assistants; Jason Kramer, marketing manager;
that they are ionized (e.g., acetate rather than
Frank Mortimer; director of marketing; Patrick
acetic acid).
Lynch, editorial director; and John Challice, vice
president and publisher. The excellent efforts of
Ancillary Materials the Oxford University Press production team
• Instructor’s Resource Companion Website: are gratefully acknowledged: David Bradley,
www.oup.com/us/white. All images are production editor; Steven Cestaro, production
available to instructors for classroom director; Lisa Grzan, production team leader;
use on a password-protected instructor’s Betty Lew, art director; and Brenda Griffing,
website. Please contact your Oxford sales copy editor. Much of this edition, like the first
representative for access. three editions, was illustrated by Eric J. White.
BRIEF CONTENTS
Boxed Material xiii

Preface xv
Symbols xvii
Conversion Factors, Equations, and Units of Energy xix
Definitions xxi
Chapter 1. Structure and Function 1
Chapter 2. Growth and Cell Division 55
Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77
Chapter 4. Membrane Bioenergetics: The Proton Potential 111
Chapter 5. Electron Transport 146
Chapter 6. Photosynthesis 175
Chapter 7. The Regulation of Metabolic Pathways 199
Chapter 8. Bioenergetics in the Cytosol 207
Chapter 9. Central Metabolic Pathways 222
Chapter 10. Metabolism of Lipids, Nucleotides, Amino Acids, and Hydrocarbons 255
Chapter 11. RNA and Protein Synthesis 281
Chapter 12. Cell Wall and Capsule Biosynthesis 316
Chapter 13. Inorganic Metabolism 335
Chapter 14. C1 Metabolism 358
Chapter 15. Fermentations 383
Chapter 16. Responses to Environmental Stress 403
Chapter 17. Solute Transport 432

v
www.ebook777.com
vi brief contents
Chapter 18. Protein Transport and Secretion 452
Chapter 19. Responses to Environmental Cues 482
Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534
Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551
Chapter 22. Cell–Cell Communication Mechanisms 566
Chapter 23. Bacterial Development 587
Index 613
CONTENTS
Boxed Material xiii

Preface xv
Symbols xvii
Conversion Factors, Equations, and Units of Energy xix
Definitions xxi
Chapter 1. Structure and Function 1

1.1 Phylogeny 3
1.2 Cell Structure 6
1.3 Summary 44
Study Questions 46
Reference and Notes 47
Chapter 2. Growth and Cell Division 55

2.1 Measurement of Growth 55
2.2 Growth Physiology 57
2.3 Growth Yields 65
2.4 Growth Kinetics 66
2.5 Steady State Growth and Continuous Growth 68
2.6 Cell Division 69
2.7 Summary 71
Study Questions 72
References and Notes 72
Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77

3.1 DNA Replication, Chromosome Separation, and Chromosome Partitioning 77
3.2 Summary 103
Study Questions 104
Chapter 4. Membrane Bioenergetics: The Proton Potential 111

4.1 The Chemiosmotic Theory 111
4.2 Electrochemical Energy 112
4.3 The Contributions of the DΨ and the DpH to the Overall Dp in Neutrophiles,
Acidophiles, and Alkaliphiles 118
4.4 Ionophores 119
4.5 Measurement of the Dp 120
4.6 Use of the Dp to Do Work 122
4.7 Exergonic Reactions That Generate a Dp 124
4.8 Other Mechanisms for Creating a DΨ or a Dp 129
vii
www.ebook777.com
viii contents
4.9 Halorhodopsin, a Light-Driven Chloride Pump 137

4.10 The Dp and ATP Synthesis in Alkaliphiles 138
4.11 Summary 139
Study Questions 140
Chapter 5. Electron Transport 146

5.1 Aerobic and Anaerobic Respiration 147
5.2 The Electron Carriers 147
5.3 Organization of the Electron Carriers in Mitochondria 152
5.4 Organization of the Electron Carriers in Bacteria 152
5.5 Coupling Sites 156
5.6 How a Proton Potential Might Be Created at the Coupling Sites:
Q Loops, Q Cycles, and Proton Pumps 159
5.7 Patterns of Electron Flow in Individual Bacterial Species 163
5.8 Summary 170
Study Questions 171
Chapter 6. Photosynthesis 175

6.1 The Phototrophic Prokaryotes 175
6.2 The Purple Photosynthetic Bacteria 178
6.3 The Green Sulfur Bacteria (Chlorobiaceae) 183
6.4 Cyanobacteria and Chloroplasts 184
6.5 Efficiency of Photosynthesis 186
6.6 Photosynthetic Pigments 187
6.7 The Transfer of Energy from the Light-Harvesting Pigments
to the Reaction Center 193
6.8 The Structure of Photosynthetic Membranes in Bacteria 194
6.9 Summary 194
Study Questions 196
Chapter 7. The Regulation of Metabolic Pathways 199

7.1 Patterns of Regulation of Metabolic Pathways 199
7.2 Kinetics of Regulatory and Nonregulatory Enzymes 201
7.3 Conformational Changes in Regulatory Enzymes 204
7.4 Regulation by Covalent Modification 204
7.5 Summary 205
Study Questions 205
Chapter 8. Bioenergetics in the Cytosol 207

8.1 High-Energy Molecules and Group Transfer Potential 207
8.2 The Central Role of Group Transfer Reactions in Biosynthesis 213
8.3 ATP Synthesis by Substrate-Level Phosphorylation 215
8.4 Summary 220
Study Questions 221
Chapter 9. Central Metabolic Pathways 222

9.1 Glycolysis 224
9.2 The Fate of NADH 230
9.3 Why Write NAD+ instead of NAD, and NADH instead of NADH2? 230
9.4 A Modified EMP Pathway in the Hyperthermophilic Archaeon
Pyrococcus furiosus 231
9.5 The Pentose Phosphate Pathway 231
contents ix
9.6 The Entner–Doudoroff Pathway 236

9.7 The Oxidation of Pyruvate to Acetyl–CoA:
The Pyruvate Dehydrogenase Reaction 238
9.8 The Citric Acid Cycle 241
9.9 Carboxylations That Replenish Oxaloacetate: The Pyruvate and
Phosphoenolpyruvate Carboxylases 245
9.10 Modification of the Citric Acid Cycle into a Reductive (Incomplete)
Cycle during Fermentative Growth 246
9.11 Chemistry of Some of the Reactions in the Citric Acid Cycle 247
9.12 The Glyoxylate Cycle 248
9.13 Formation of Phosphoenolpyruvate 249
9.14 Formation of Pyruvate from Malate 251
9.15 Summary of the Relationships between the Pathways 251
9.16 Summary 252
Study Questions 253
Chapter 10. Metabolism of Lipids, Nucleotides, Amino Acids, and Hydrocarbons 255
10.1 Lipids 255
10.2 Nucleotides 264
10.3 Amino Acids 267
10.4 Aliphatic Hydrocarbons 273
10.5 Summary 275
Study Questions 278
Chapter 11. RNA and Protein Synthesis 281

11.1 RNA Synthesis 282
11.2 Protein Synthesis 296
11.3 Summary 310
Study Questions 311
Chapter 12. Cell Wall and Capsule Biosynthesis 316

12.1 Peptidoglycan 316
12.2 Lipopolysaccharide 321
12.3 Extracellular Polysaccharide Synthesis and
Export in Gram-Negative Bacteria 326
12.4 Levan and Dextran Synthesis 331
12.5 Glycogen Synthesis 332
12.6 Summary 332
Study Questions 332
Chapter 13. Inorganic Metabolism 335

13.1 Assimilation of Nitrate and Sulfate 335
13.2 Dissimilation of Nitrate and Sulfate 337
13.3 Nitrogen Fixation 339
13.4 Lithotrophy 344
13.5 Summary 353
Study Questions 354
Reference and Notes 355
Chapter 14. C1 Metabolism 358

14.1 Carbon Dioxide Fixation Systems 358
14.2 Growth on C1 Compounds Other than CO2: The Methylotrophs 374
14.3 Summary 378
www.ebook777.com
x contents
Study Questions 380

Chapter 15. Fermentations 383

15.1 Oxygen Toxicity 383
15.2 Energy Conservation by Anaerobic Bacteria 384
15.3 Electron Sinks 385
15.4 The Anaerobic Food Chain 386
15.5 How to Balance a Fermentation 387
15.6 Propionate Fermentation via the Acrylate Pathway 388
15.7 Propionate Fermentation via the Succinate–Propionate Pathway 389
15.8 Acetate Fermentation ( Acetogenesis) 391
15.9 Lactate Fermentation 392
15.10 Mixed-Acid and Butanediol Fermentation 394
15.11 Butyrate Fermentation 397
15.12 Ruminococcus albus 400
15.13 Summary 400
Study Questions 401
Chapter 16. Responses to Environmental Stress 403

16.1 Maintaining a DpH 403
16.2 Osmotic Pressure and Osmotic Potential 406
16.3 Heat-Shock Response (HSR) 412
16.4 Repairing Damaged DNA 415
16.5 The SOS Response 421
16.6 Oxidative Stress 423
16.7 Summary 425
Study Questions 427
Chapter 17. Solute Transport 432

17.1 The Use of Proteoliposomes to Study Solute Transport 432
17.2 Kinetics of Solute Uptake 433
17.3 Energy-Dependent Transport 434
17.4 How to Determine the Source of Energy for Transport 444
17.5 Drug-Export Systems 445
17.6 Bacterial Transport Systems in Summary 446
17.7 Summary 446
Study Questions 448
Chapter 18. Protein Transport and Secretion 452

18.1 The Sec System 453
18.2 The Translocation of Membrane-Bound Proteins 457
18.3 The E. coli SRP 459
18.4 Protein Translocation of Folded Proteins: The Tat System 459
18.5 Extracellular Protein Secretion 461
18.6 Folding of Periplasmic Proteins 473
18.7 Summary 474
Study Questions 474
Chapter 19. Responses to Environmental Cues 482

19.1 Introduction to Two-Component Signaling Systems 483
contents xi
19.2 Responses by Facultative Anaerobes to Anaerobiosis 488

19.3 Response to Nitrate and Nitrite: The Nar Regulatory System 494
19.4 Response to Nitrogen Supply: The Ntr Regulon 498
19.5 Response to Inorganic Phosphate Supply: The PHO Regulon 503
19.6 Effect of Oxygen and Light on the Expression of Photosynthetic Genes
in the Purple Photosynthetic Bacterium Rhodobacter capsulatus 504
19.7 Response to Osmotic Pressure and Temperature:
Regulation of Porin Synthesis 506
19.8 Response to Potassium Ion and External Osmolarity:
Stimulation of Transcription of the kdpABC Operon by a Two-Component
Regulatory System 507
19.9 Acetyl Phosphate Is a Possible Global Signal in Certain
Two-component Systems 508
19.10 Response to Carbon Sources: Catabolite Repression, Inducer Expulsion,
and Permease Synthesis 510
19.11 Virulence Factors: Synthesis in Response to Temperature, pH,
Nutrient Osmolarity, and Quorum Sensors 515
19.12 Summary 522
Study Questions 524
Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534

20.1 Bacteria Measure Changes in Concentration over Time 534
20.2 Tumbling 535
20.3 Adaptation 536
20.4 Proteins Required for Chemotaxis 536
20.5 A Model for Chemotaxis 537
20.6 Mechanism of Repellent Action 541
20.7 Chemotaxis That Does Not Use MCPs: The Phosphotransferase System
Is Involved in Chemotaxis toward PTS Sugars 541
20.8 Chemotaxis That Is Not Identical with the Model Proposed for
the Enteric Bacteria 541
20.9 Photoresponses 543
20.10 Halobacteria 544
20.11 Photosynthetic Bacteria 544
20.12 Aerotaxis 546
20.13 Summary 546
Study Questions 546
Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551

21.1 Bacterial Multicellular Structures 551
21.2 Prevalence and Importance of Biofilms 552
21.3 Properties of Biofilms 554
21.4 Progression of Biofilm Formation and Dissolution 557
21.5 Regulation of Biofilm Formation 559
21.6 Inhibition of Biofilm Formation 560
21.7 Evolutionary Processes in Biofilms 561
21.8 Summary 562
Study Questions 563
Chapter 22. Cell–Cell Communication Mechanisms 566

22.1 Diversity of Diffusible Signal Molecules Produced by Bacteria 566
22.2 Specific Signaling Systems 566
22.3 Cell–Cell Signaling That Requires Contact 581
www.ebook777.com
xii contents
22.4 Summary 583

Study Questions 583
Chapter 23. Bacterial Development 587

23.1 Myxobacteria 587
23.2 Caulobacter Development: Control of DNA Replication and Cell Cycle Genes 598
23.3 Sporulation in Bacillus subtilis 601
23.4 Summary 610
Study Questions 610
References and Note 611
Index 613
1
Structure and Function
The prokaryote (procaryote) domains of life are movement on solid surfaces via a form of motil-
the Bacteria and the Archaea. Prokaryotes are ity called twitching, for gliding motility among
defined as organisms that have no membrane- the myxobaceria, and for mating. Flagella (sing.
bound nucleus. (The prefix pro, borrowed from flagellum) are used by single cells to swim in liq-
Greek pro, means earlier than or before; kary- uid; they are also used in swarming, a form of
ote, borrowed from Greek káryon, means kernel group swimming on moist solid surfaces.
or nut.) Besides lacking a nucleus, prokaryotes In addition, it turns out that the poles of
are devoid of organelles such as mitochondria, nonspherical bacteria cells are physiologically
chloroplasts, and Golgi vesicles. However, as and structurally different from the rest of the
you will learn from reading this chapter, their cell, and sometimes the cell poles in the same
cell structure is far from simple and reflects the cell differ from each other. This is obvious, for
evolution of prokaryotes into quite sophisti- example, when flagella or pili protrude from
cated organisms that are remarkably successful one or both cell poles, or when a cell pole is dis-
in inhabiting diverse ecological niches. tinguished by having a stalk, as is the case for
Despite the absence of organelles comparable Caulobacter, discussed in Chapter 23. In other
to those in eukaryotic cells, the activities within cases, the differences between poles are less
prokaryotic cells are compartmentalized. For obvious, as shown, for example, in the discus-
example, as this chapter points out, compart- sion of the polar localization of the chemotaxis
mentalization occurs within multienzyme gran- proteins in Chapter 20.
ules that house enzymes for specific metabolic It has become clear that prokaryotes even
pathways, in intracellular membranes within have an internal protein cytoskeleton, a prop-
the cytosol, within the cell membrane, within erty previously thought to be restricted to
a special compartment called the periplasm eukaryotic cells. As described in this chapter
in gram-negative bacteria, within the cell wall and in Chapter 2, the cytoskeleton is critical
itself, and within various inclusion bodies that for determining and maintaining cell shape as
house specific enzymes, storage products, or well as for cell division. For a general review of
photosynthetic pigments. As this chapter also subjects covered in this chapter, the reader is
points out, the prokaryotes also have special- referred to ref. 1.
ized external structures called appendages. Thus, as has been pointed out many times in
The most notable appendages are pili and fla- recent years, the prokaryotic cell is not simply
gella. Different types of pili (sing., pilus) serve a “bag of enzymes,” in accordance with ear-
different functions. Depending upon the type, lier descriptions, but rather, a sophisticated
pili are used for adhesion to other cells when entity, that is dynamic both structurally and
that becomes necessary for colonization, for physiologically. Furthermore, as Chapter 22
1
www.ebook777.com
2 the physiology and biochemistry of prokaryotes
Table 1.1 Major subdivisions of prokaryotes
Bacteria and their subdivisions B. Flavobacterium group

Purple bacteria (now referred to as the division or Flavobacterium, Cytophaga, Saprospira,
phylum Proteobacteria) Flexibacter
α subdivision
Purple nonsulfur bacteria (Rhodobacter, Planctomyces and relatives
Rhodopseudomonas), rhizobacteria, agrobacteria, A. Planctomyces group
rickettsiae, Nitrobacter, Thiobacillus (some), Planctomyces, Pasteuria
Azospirillum, Caulobacter B. Thermophiles
β subdivision Isocystis pallida
Rhodocyclus (some), Thiobacillus (some), Chlamydiae
Alcaligenes, Bordetella, Spirillum, Nitrosovibrio, Chlamydia psittaci, C. trachomatis
Neisseria
γ subdivision Radio-resistant micrococci and relatives
Enterics (Acinetobacter, Erwinia, Escherichia, A. Deinococcus group
Klebsiella, Salmonella, Serratia, Shigella, Deinococcus radiodurans
Yersinia), vibrios, fluorescent pseudomonads, B. Thermophiles
purple sulfur bacteria, Legionella (some), Thermus aquaticus
Azotobacter, Beggiatoa, Thiobacillus (some),
Photobacterium, Xanthomonas Green nonsulfur bacteria and relatives
δ subdivision A. Chloroflexus group
Sulfur and sulfate reducers (Desulfovibrio), Chloroflexus, Herpetosiphon
myxobacteria, bdellovibrios B. Thermomicrobium group
Thermomicrobium roseum
Gram-positive eubacteria
A. High (G + C) species Archaea subdivisions
Actinomyces, Streptomyces, Actinoplanes, Extreme halophiles
Arthrobacter, Micrococcus, Bifidobacterium, Halobacterium, Halococcus morrhuae
Frankia, Mycobacterium, Corynebacterium
B. Low (G + C) species Methanobacter group
Clostridium, Bacillus, Staphylococcus, Methanobacterium, Methanobrevibacter,
Streptococcus, mycoplasmas, lactic acid bacteria Methanosphaera stadtmaniae,
C. Photosynthetic species Methanothermus fervidus
Heliobacterium Methanococcus group
D. Species with gram-negative walls Methanococcus
Megasphaera, Sporomusa
“Methanosarcina” group
Cyanobacteria and chloroplasts Methanosarcina barkeri, Methanococcoides
Oscillatoria, Nostoc, Synechococcus, Prochloron, methylutens, Methanothrix soehngenii
Anabaena, Anacystis, Calothrix
Methospirillum group
Spirochaetes and relatives Methanospirillum hungatei, Methanomicrobium,
A. Spirochaetes Methanogenium, Methanoplanus limicola
Spirochaeta, Treponema, Borrelia
B. Leptospiras Thermoplasma group
Leptospira, Leptonema Thermoplasma acidophilum
Green sulfur bacteria Thermococcus group

Chlorobium, Chloroherpeton Thermococcus celer
Bacteroides; flavobacteria and relatives Extreme thermophiles
A. Bacteroides group Sulfolobus, Thermoproteus tenax,
Bacteroides, Fusobacterium Desulfurococcus mobilis, Pyrodictium occultum
Source: Hodgson, D. A. 1989. Bacterial diversity: the range of interesting things that bacteria do, pp. 4–22. In: Genetics of
Bacterial Diversity. D. A. Hopwood and K. F. Chater (Eds.). Academic Press, London.
explains, depending upon the growth condi- the investigator to experimentally learn how
tions, prokaryotes are capable of living either the organisms are temporally and spatially
as single cells, when suspended in liquid, or as organized so that myriad activities can take
interacting cells in multicellular populations at place within the cells and between the cells to
physical interfaces, such as solid surfaces. The enable survival in the face of environmental
study of prokaryotes presents a challenge to challenges.
structure and function 3
1.1 Phylogeny Korarchaeota, has been found in hot environ-

Figure 1.1 shows a current phylogeny of life- ments such as hot springs, and much less is
forms based upon a comparison of ribosomal known about it, or about archaea belonging to
RNA (rRNA) nucleotide sequences. For a more the newly discovered phylum Nanoarchaeota,
complete explanation of Fig. 1.1 and how it which has been cultivated from a submarine hot
was derived, see Box 1.1. Notice that there are vent.
three lines (domains) of evolutionary descent—
Bacteria, Eucarya, and Archaea—that diverged Phenotypes
in the distant past from a common ancestor.2-4 For a comprehensive description of the Archaea,
(The term archaeon may be used to describe the reader is referred to ref. 5. For a review of the
particular Archaea.) Archaea differ from bac- cell surface structures of the Archaea, see ref. 6
teria in ribosomal RNA nucleotide sequences, and also Box 1.6. Archaea are found not only
in cell chemistry, and in certain physiological in extreme environments such as hot springs,
aspects, described in Sections 1.1.1 and 1.2. and alkaline and acid waters, but also in non-
Table 1.1 lists examples of prokaryotes in extreme environments that exist in the oceans,
the different subdivisions within the domains lakes, soil, sewage, swamps, and the animal
Bacteria and Archaea. Notice that the gram- intestinal tract. They are thus widespread in the
positive bacteria are a tight grouping. Although environment.
there is no single grouping of gram-negative From a morphological point of view (size
bacteria, most of the well-known gram-negative and shape), Archaea are similar to the typical
bacteria are in the Proteobacteria division. Bacteria lineage. However, they form a group
of organisms phylogenetically distinct from
1.1.1 Archaea both bacteria and eukaryotes. The Archaea that
Phylogenetic lineages have been best studied commonly manifest one
The two major archaeal phyla are the Euryar- of three phenotypes: methanogenic, extremely
chaeota and the Crenarchaeota. A third phylum, halophilic, and extremely thermophilic.
Fig. 1.1 Phylogenetic relationships among life-forms based upon rRNA sequences. The line lengths are pro-
portional to the evolutionary differences. The position of the root in the tree is approximate. The “purple
bacteria” are now referred to as the phylum Proteobacteria and, as summarized in Table 1.1, comprise a wide
variety of gram-negative organisms including phototrophic, chemoheterotrophic, and chemolithotrophic
bacteria. Source: Woese, C. R., and N. R. Pace. 1993. Probing RNA structure, function, and history by com-
parative analysis. The RNA World. Cold Spring Harbor Press, Cold Spring Harbor, NY.
www.ebook777.com
The methanogenic archaea (Euryarchaeota), Their highly unusual metabolism is explained

also referred to as methanogens, produce meth- in Chapter 14. Methanogens are obligate anaer-
ane. This ability is important for the organisms’ obes that grow in environments such as anaero-
survival, because they derive energy from the bic groundwaters, swamps, and sewage, as well
process. The methanogens produce methane by as part of the digestive tract (rumen) of animals
reducing carbon dioxide to methane or by con- such as cattle and sheep, where they produce
verting acetate to carbon dioxide and methane. methane.
BOX 1.1 PHYLOGENY

The evolutionary relationships among all can explain the differences in nucleotide
living organisms have been deduced by sequences. The simplest tree is the one that
comparing the ribosomal RNA sequences requires the smallest number of nucleotide
of modern organisms. If the structures of changes to evolve the collection of extant
the ribosomal RNA molecules from differ- sequences from a postulated ancestral
ent living organisms are sufficiently consequence.
served in certain segments of the molecule, It has been pointed out that it is diffi-
conserved sequences and secondary struc- cult to assess the validity of a phylogenetic
tures can be aligned to permit comparison tree. There are several reasons for this. For
of the differences in base sequences between example, none of the methods accounts for
RNA molecules. Researchers analyzed the the possibility that multiple changes may
number of positions that differ between have taken place at any single position.
pairs of sequences, as well as other features This shortcoming is likely to result in an
(e.g., which positions vary, the number of underestimate of the distances.
changes that have been made in going from Underestimation of the distances, in
one sequence to another). The number of turn, might bring two distantly related lin-
nucleotide differences between homolo- eages much closer, and in fact might make
gous sequences is used to calculate the evo- them appear to be specifically related to
lutionary distance between the organisms each other when they are not. Other ambi-
and to construct a phylogenetic tree. Most guities in phylogenetic trees might arise if
recently published phylogenetic trees are different positions in the sequence align-
based upon 16S rRNA sequences. ments have changed at very different rates.
There are two major ways in which phy- Furthermore, the branching pattern itself
logenetic trees are constructed from the can differ if a different algorithm is used,
nucleotide differences. In the evolutionary even with the same sequences. The out-
distance method, the number of nucleotide come is that there exist a variety of phyloge-
differences is used as a measure of the evo- netic trees, and no single published tree has
lutionary distance between the organisms. been accepted as perfectly representing the
The second method, the maximum par- 4 billion years or so of bacterial evolution.
simony method, is more complicated. It Despite these problems, there is consistency
takes into account not only the nucleotide in the trees that are based upon 16S rRNA
differences but also the positions at which sequences, and it is believed that most of the
the differences occur and the nature of the phylogenetic trees derived from sequenc-
differences. Parsimony means “less is bet- ing bacterial 16S rRNA are at least plau-
ter” or “stinginess,” going to extreme, for sible. However, the assumption remains
the sake of economy; the method attempts unproven that the relationships between
to find the simplest evolutionary tree that the 16S rRNAs represent the phylogenies
of the organisms rather than simply the M. Dworkin, W. Harder, and K.-H. Schleifer
phylogeny of a given 16S rRNA. Because of (Eds.). Springer-Verlag, Berlin.
this, it is imperative that the relationships 2. Stackebrandt, E. 1991. Unifying phylogeny
between the different organisms be tested and phenotypic diversity, pp. 19–47. In: The
by means of other characters (e.g., other Prokaryotes, Vol. I. A, Balows, H. G. Trüper,
M. Dworkin, W. Harder, and K.-H. Schleifer
appropriate molecules besides the 16S (Eds.). Springer-Verlag, Berlin.
rRNA, and various phenotypic character-
3. Felsenstein, J. 1982. Numerical methods
istics of the organisms). See refs 1–4.
for inferring evolutionary trees. Q. Rev. Biol.
57:379–404.
REFERENCES 4. Boucher, Y., C. J. Douady, R. T. Papke, D.
A. Walsh, M. E. R. Boudreau, C. I. Nesbo, R.
1. Woese, C. R. 1991. Prokaryotic systemat- J. Case, and W. F. Doolittle. 2003. Lateral gene
ics: the evolution of a science, pp. 3–11. In: The transfer and the origins of prokaryotic groups.
Prokaryotes, Vol. I. A, Balows, H. G. Trüper, Annu. Rev. Gen. 37:283–328.
The extremely halophilic archaea (also in acceptor to oxidize hydrogen gas. These include
the kingdom Euryarchaeota) require very Thermoproteus, Pyrobaculum, Pyrodictium,
high sodium chloride concentrations (at least and Archaeoglobus. [Pyrobaculum and
3–5 M) for growth. They grow in salt lakes Pyrodictium use elemental sulfur (S°) as an elec-
and solar evaporation ponds. The halophilic tron acceptor during autotrophic growth, i.e.,
archaea have light-driven proton and chloride growth on CO2 as the carbon source, whereas
pumps called bacteriorhodopsin and halorho- Archaeglobus uses S2O32−.] Archaea belong-
dopsin, respectively, Bacteriorhodopsin creates ing to the genus Pyrodictium have the highest
an electrochemical proton gradient that is used growth temperature known, with the ability to
to drive ATP synthesis, and halorhodopsin is grow at 110 °C. A few of the sulfur-oxidizing
used to accumulate chloride intracellularly to archaea are acidophiles, growing in hot sulfuric
maintain osmotic stability. These pumps are acid at pH values as low as 1.0. They are called
described in more detail in Chapter 4 (Sections thermoacidophiles, indicating that they grow
4.8.4 and 4.9). optimally in hot acid. For example, Sulfolobus
The extremely thermophilic archaea grow grows at pH values of 1 to 5 and at tempera-
in thermophilic environments (generally tures up to 90 °C in hot sulfur springs, where it
55–100 °C).7 Some of these have an optimal oxidizes H2S (hydrogen sulfide) or So to H2SO4
growth temperature near the boiling point (sulfuric acid). Although most of the extreme
of water! Many use inorganic sulfur either thermophiles are obligately sulfur dependent,
as an electron donor or as an electron accep- some are facultative. For example, Sulfolobus
tor in energy-yielding oxidation–reduction can be grown heterotrophically on organic
(redox) reactions. [The pathways for sulfate carbon and O2 as well as autotrophically on
reduction and sulfur oxidation are described H2S or S°, O2, and CO2. Interestingly, some of
in Sections 13.2.2 and 13.4.1 (see subsection the sulfur-dependent archaea have metabolic
entitled Sulfur-oxidizing prokaryotes), respec- pathways for sugar degradation not found
tively.] Some of these archaea, which are also among the Bacteria. These are mentioned in
called sulfur dependent, oxidize inorganic sul- Section 9.4.
fur compounds such as elemental sulfur and Recently, a new genus of thermoacido-
sulfide, by using oxygen as the electron accep- philic archaea was discovered.8 Picrophilus, a
tor, deriving ATP from the process. Examples member of the kingdom Euryarchaeota, order
include Sulfolobus and Acidianus. Other Thermoplasmales, is an obligately aerobic,
extreme thermophiles are anaerobes that use heterotrophic archaeon that grows at tempera-
elemental sulfur or thiosulfate as the electron tures between 45 and 65 °C at a pH of 0 with
www.ebook777.com
an optimum pH of 0.7. It was isolated from tor, EF-2, that can be ADP-ribosylated by
hot geothermally heated acidic solfataras9 in diphtheria toxin. In contrast, bacteria use
Japan. Despite the sulfur content of the habitat, formylmethionine to initiate protein synthe-
Picrophilus is not sulfur dependent. sis, and their EF-2 is not sensitive to diphthe-
ria toxin. However, the archaeal ribosomes
Comparison of domains Archaea, are similar to bacterial ribosomes in having
Bacteria, and Eucarya an overall size described as 70S; that is, they
Bacteria can be distinguished from Archaea on sediment in a centrifugal field of a velocity of
the basis of the following structural and physi- 70 svedberg units.
ological differences. 6. The methanogenic archaea have several
coenzymes that are unique to Archaea. The
1. Whereas the lipids in the membranes of bac- coenzymes are used in the pathway for the
teria and eukaryotes are fatty acids, which reduction of carbon dioxide to methane and
are esterlinked to glycerol (see later: Figs. in the synthesis of acetyl–CoA from H2 and
1.16 and 10.5), the archaeal lipids are meth- CO2. These coenzymes and their biochemi-
yl-branched, isoprenoid alcohols, ether-
cal roles are described in Section 14.1.5.
linked to glycerol (Fig. 1.18).10 Archaeal
membranes are discussed in Section 1.2.5, In addition to the differences just listed, there
and the biosynthesis of archaeal lipids is dis- are several similarities between archaeal and
cussed in Section 10.1.3. eukaryotic RNA polymerase, and archaeal and
2. Archaea lack peptidoglycan, a universal eukaryotic ribosomes (Table 1.2).
component of bacterial cell walls. The cell
walls of some archaea contain pseudopepti- 1.2 Cell Structure
doglycan, a component absent from bacte-
There are well-known differences between
rial cell walls (Section 1.2.3).
Archaea and Bacteria with respect to cell walls,
3. Archaea contain histones that resemble
cell membranes, and flagella, and these will be
eukaryal histones and bind archaeal DNA
into compact structures resembling eukaryal pointed out. The structure and function of
nucleosomes.11,12 the major cell components will be described,
beginning with cellular appendages (pili and
4. The archaeal RNA polymerase differs from
flagella) and working our way into the interior
bacterial RNA polymerase by having 8 to
of the cell.
10 subunits, rather than 4 subunits, and by
not being sensitive to the antibiotic rifampi-
cin. The difference in sensitivity to rifampi- 1.2.1 Appendages
cin reflects differences in the proteins of the Numerous appendages, each designed for a spe-
RNA polymerase. In fact, the archaeal RNA cific task, can extend from bacterial surfaces.
polymerase resembles the eukaryotic RNA We shall describe two classes: flagella and pili.
polymerase, which also has many subunits The flagella are used by single cells for swim-
(10–12) and is not sensitive to rifampicin. ming in liquid and swarming, a type of group
5. Some protein components of the archaeal swimming on moist solid surfaces. Depending
protein synthesis machinery differ from upon the type, the pili (sometimes called fim-
those found in the bacteria. Archaeal ribo- briae) are used for adhesion to surfaces, includ-
somes are not sensitive to certain inhibitors ing adhesion to the surfaces of animal cells; in
of bacterial ribosomes (e.g., erythromycin, certain cases, described in Box 1.2, they serve
streptomycin, chloramphenicol, and tetracy- for movement on solid surfaces (type IV pili),
cline). These differences in sensitivity to anti- and for conjugal transfer or mating (sex pili).
biotics reflect differences in the ribosomal
proteins. In this respect, archaeal ribosomes Flagella
resemble cytosolic ribosomes from eukary- For a review of all known prokaryotic motil-
otic cells. Other resemblances to eukaryotic ity structures, see ref. 13. For a review of bac-
ribosomes are the use of methionine rather terial flagella, see refs. 14 and 15. Swimming
than formylmethionine to initiate protein bacteria have one or more flagella, which are
synthesis, and a translation elongation fac- organelles of locomotion that protrude from
Table 1.2 Comparison between Bacteria, Archaea, and Eucarya
Characteristic Bacteria Archaea Eucarya
Peptidoglycan Yes No No
Lipids Ester linked Ether linked Ester linked
Ribosomes 70S 70S 80S
Initiator tRNA Formylmethionine Methionine Methionine
Introns in tRNA No Yes Yes
Ribosomes sensitive to No Yes Yes
diphtheria toxin
RNA polymerase One (5 subunits) Several (8–12 Three (12–14
subunits each) subunits each)
Ribosomes sensitive to
chloramphenicol, streptomycin,
kanamycin Yes No No
the cell surface. Flagella allow single cells to coli and S. typhimurium. Archaeal flagella are
swim in liquid and can also be used for swarm- somewhat different, as will be described.
ing on a solid surface. Swarming, a means of
group swimming by which bacterial colonies 2. General structure
can spread in limiting liquid, is discussed later The proteins are named after genes found in
in this section. Escherichia coli and Salmonella typhimurium.17
Mutations in the mot (motility) genes result in
1. An overview paralyzed flagella, and as we shall see later, the
The bacterial flagellum is a stiff, helical fila- Mot proteins provide the torque that causes
ment approximately 20 nm in diameter that the flagellum to rotate. The flagellum consists
rotates like a screw-type propeller; it can be of a basal body, a hook, a filament, a motor, a
either a left-handed helix or a right-handed switch, an export apparatus, capping proteins,
helix, depending upon the species. The bac- and junction proteins (typically referred to as
terial flagellum is unrelated to the eukary- fla, fli, flg, flh, and flb). We discuss some of these
otic flagellum in composition, structure, and now.
mechanism of action. (For a description of
eukaryotic flagella and cilia, see note 16 in The basal body. Examine Fig. 1.2. At the base
the section References and Notes, at the end of the flagellum there is a basal body embedded
of the chapter.) The word flagellum, which in the membrane. In gram-negative bacteria, the
means whip in Latin, was first used to describe basal body consists of three stacked rings (C,
the bacterial filament in 1852. The bacterial M, and S rings) and a central rod. The M and S
organelle, however, is more like a stiff propel- rings are actually joined as a single ring called
ler than like a whip, which is a flexible rod. the MS ring, made from different domains of
When the bacterial flagellum rotates, a helical the FliF protein. This scheme is supported by
wave travels from the proximal to the distal electron microscopic evidence. The term “MS”
end (outward from the cell), and as a con- indicates the ring’s location: membrane and
sequence the cell is pushed forward as illus- supramembranous. The P ring (FlgI protein)
trated later (see Fig. 20.2). The flagellum is a and the L ring (FlgH protein) are also named
very complex machine driven by a tiny rotat- according to their location: peptidoglycan and
ing motor embedded in the membrane. Both lipopolysaccharide. The L ring in S. typhimu-
its structure and the mechanism of its motility rium has been shown to be a lipoprotein.18
will be described. The flagella that are studied Presumably the lipid portion helps anchor the
in most detail are those of Escherichia coli and protein in the lipid regions of the outer enve-
Salmonella typhimurium, and we will begin lope. A central rod, made from the FlgB, FliC,
with a discussion of their flagella. The flagella and FlgF proteins, passes through the rings and
of other bacteria, except for the spirochaetes, leads to the hook portion (H) of the flagellum
are similar in general structure to those of E. on the outside of the cell. The outermost two
www.ebook777.com
BOX 1.2 NONFLAGELLAR MOTILITY

Many bacteria do not have flagella and yet short, intermittent jerks (hence the name
are capable of motility. Depending upon the “twitching”) of the cells.
bacterium, the motility may take place on a
solid surface (twitching or gliding) or in liq-
uid (swimming). The mechanistic bases for Pore complexes, slime extrusion,
these movements differ and are related to and gliding motility
specific cell structure features present in the
particular cell. See refs. 1–4. Gliding motility by filamentous cyanobac-
teria is apparently due to the secretion of
slime through pores near the septa that
Type IV pili, twitching motility, separate the cells in the filament The model
and gliding motility proposes that as a consequence of the slime
secretion the filament of cells is pushed for-
Type IV pili are fibrillar protein append- ward through the so-called junction pores.
ages located at either pole of certain
gram-negative pathogenic bacteria, such
as Pseudomonas aeruginosa, Bacteroides Gliding in the green fluorescent
ureolyticus, Legionella pneumophila, bacteria (GFP) group
Neisseria meningitidis, Ralstonia solan-
acearum, and Vibrio cholerae, that infect Bacteria in the GFP group glide rapidly on
animals, plants, and fungi. They are also solid surfaces. Some species actually rotate
found in many nonpathogenic gram- as they glide. The cells suspended in liquid
negative bacteria such as Myxococcus can propel adsorbed latex beads in multiple
xanthus, which is a social bacterium that paths around the cell, indicating movement
constructs multicellular fruiting bodies by cell surface molecules, perhaps poly-
(discussed in Chapter 23), and in the uni- mers or fibers, to which the beads attach.
cellular cyanobacterium Synechocystis. Perhaps the same molecules are part of the
Type IV pili comprise the mechanosystem machinery that propels the cells when they
for twitching motility, which is a form of are on a solid surface.
gliding that takes place on moist surfaces.
Twitching motility in M. xanthus is called REFERENCES
social motility (S-motility). It is a form of
smooth motility of cells gliding as groups. 1. McBride, M. J. 2001. Bacterial gliding motil-
ity: multiple mechanisms for cell movement over
Surfaces upon which twitching motility surfaces. Annu. Rev. Microbiol. 55:49–75.
can occur include agar, epithelial tissue,
plant tissue, and various other surfaces 2. Mattick, J. S. 2002. Type IV and twitching.
Annu. Rev. Microbiol. 56:289–314.
such as glass, plastics, and metal. Motility
on such surfaces is important for rapid 3. Semmler, A. B., C. B. Whitchurch, and J. S.
Mattick. 1999. A re-examination of twitch-
colonization, for the formation of bio- ing in Pseudomonas aeruginosa. Microbiology
films (discussed in Chapter 21), and for the 145:2863–2873.
building of fruiting bodies by myxobacte-
4. McBride, M. J., T. F. Braun, and J. L. Brust.
ria (Chapter 23). 2003. Flavobacterium johnsoniae GldH is a
Twitching motility in P. aeruginosa and lipoprotein that is required for gliding and chitin
several other bacteria is characterized by utilization. J. Bacteriol. 185:6648
C ring/switch
Fig. 1.2 Bacterial flagellum in a gram-negative envelope (based on Salmonella). The basal body itself, to
which the hook–flagellum assembly is attached, is 22.5 nm × 24 nm in size and is composed of four rings, L, P,
S, and M, connected by a central rod. The M ring is embedded in the cell membrane and the S ring appears to
lie on the surface of the membrane. The S and M rings are actually one ring, the MS ring. The P ring may be in
the peptidoglycan layer, and the L ring seems to be in the outer membrane. The P and L rings may act as bush-
ings that allow the central rod to turn. Gram-positive bacteria have similar flagella but lack the L and P rings.
The MotA and MotB proteins form complexes (Mot) that couple the influx of protons to the rotation of the
rotor. The rotor consists of the MS ring with the FliG protein attached to its cytoplasmic surface and the C ring
attached to the cytoplasmic surface of the MS ring. The switch complex consists of three peripheral membrane
proteins, FliG (also part of the rotor), FliM, and FliN, which probably are closely apposed to the cytoplasmic
face of the M ring. Not shown are hook accessory or adaptor proteins (HAP1 and HAP3) between the hook
and filament, and a protein cap (HAP2) on the end of the filament. The flagellum is assembled from the proxi-
mal to the distal end, with the filament being assembled last. It appears that the HAP1 and HAP3 proteins are
required for the proper assembly of the filament onto the hook. Abbreviations: OM, outer membrane; pg,
peptidoglycan; CM, cell membrane; R, central rod; MOT, MotA and MotB; H, hook; F, filament; L, L-ring;
P, P-ring.
rings (P and L) may act as bushings, allowing and surround the MS ring. The number of such
the central rod to rotate in the peptidoglycan particles in various bacteria has been reported to
and outer membrane. Gram-positive bacteria be 12 to 16. Such arrays of particles surround-
do not have P and L rings. The basal body trans- ing the MS ring were seen in electron micro-
mits torque to the hook and filament, causing graphs of freeze-fractured cell membranes, but
them to rotate. 14 not when either MotA or MotB was missing.20
It has been suggested that the large periplasmic
The motor. One of the fascinating components domain of MotB is attached noncovalently to
of the flagellar apparatus is the tiny motor, the peptidoglycan, explaining why the MotA/
approximately 50 nm in diameter, that lies at MotB complex does not rotate when the motor
the base of the flagellum and causes it to rotate. turns.
For a review of the flagellar motor, see ref. 19. In agreement with the conclusion that MotA
The motor consists of two parts: a nonrotating and MotB are part of the motor, mutations in
part called the stator (made of the Mot pro- the motA and motB genes result in paralyzed
−
teins) and a rotating part called the rotor, which flagella (Mot phenotype). The MotA/MotB
includes the FliG proteins that transmit torque complexes are believed to conduct protons from
to the MS ring. outside the cell to inside, across the membrane,
The stator consists of two different proteins, and to use this proton movement to provide the
MotA and MotB, indicated as Mot in Fig. 1.2. torque to rotate the rotor. Extreme alkaliphiles
These exist as particles of multiple complexes, and some marine bacteria use a sodium ion cur-
(MotA)4(MotB)2, that span the cell membrane rent. This is reviewed in ref. 21. (As discussed in
www.ebook777.com
Chapter 4, bacteria use a proton current across chemotaxis (Chapter. 20), binds to FliM. (For
the cell membrane to do other kinds of work, more information about these proteins, includ-
e.g., ATP synthesis, in addition to rotating fla- ing their function, see note 28.)
gella.) For an explanation of the conclusion that
The hook and the HAP proteins. The central
MotA and MotB form a complex, see note 22.
flagellar rod is attached to an external curved
Membrane vesicles prepared from strains syn-
flexible hook made of multiple copies of a spe-
thesizing wild-type MotA were more permeable
cial protein called the hook protein, FlgE, the
to protons than were vesicles prepared from
product of the flgE gene. There are also two
strains synthesizing mutant MotA, indicating
hook-associated proteins: HAP1, also called
that MotA is likely to be part of a proton chan-
FlgK (the flgK gene), and HAP3, also called
nel. 23 Some of the evidence in support of the
FlgL (the flgL gene). These proteins are nec-
conclusion that the Mot proteins form a com-
essary to form the junction between the hook
plex that is a transmembrane proton channel is
and the filament. Mutants that lack these HAPs
in ref. 24.
secrete flagellin into the medium. According to
How does passage of protons through the
the model, the hook subunits, consisting of the
Mot complex generate torque? One model pro-
FlgE protein, fill the C ring and then are trans-
poses that as protons pass through the MotA/
ferred en masse to the growing hook through
MotB complexes, conformational changes
the export apparatus, described later, which is
occur in the complexes that drive stepwise rota-
in the middle of the C ring. (See later subsec-
tion of the attached rotor. For a model of how
tion entitled The export apparatus for flagellar
this might occur, consult ref. 25.
components.)
What is the rotor? Sometimes the C ring is
referred to as the rotor, but others consider the The filament and the capping proteins. Attached
rotor to be the MS and C rings together. (See to the hook is a rigid, hollow, helical filament
note 26 for references.) An essential rotor com- that, along with the hook, protrudes from the
ponent is FliG. FliG interacts with the Mot pro- cell. When it rotates, it acts as a propeller and
teins and transmits torque generated by the Mot pushes the cell forward. The protein in the fila-
complex to rotate the rotor. The FliG proteins ment is called flagellin (which in Escherichia and
are part of the C ring and attach the C ring to the Salmonella is known as FliC) and is present in
MS ring. As explained next, the C ring functions thousands of copies. Flagellin is not identical in
as a switch that reverses the direction of rota- all bacteria. For example, the protein can vary
tion of the rotor. in size from 20 kDa to 65 kDa depending upon
the bacterial species. Furthermore, although
The switch. In certain bacteria, such as E. coli
there is homology between the C-terminal and
and S. typhimurium, the motor spontaneously
N-terminal ends of most flagellins, the central
changes its direction of rotation periodically.
part can vary considerably and is distinguished
The frequency of switching is influenced by
immunologically in different bacteria. In some
attractants and repellents that the bacterium
cases, there is no homology at all. For example,
might encounter, and as such, is important for
nucleotide-derived amino acid sequences for
chemotaxis, described in Chapter 20. Mutations
the flagellins from Sinorhizobium meliloti show
that cause failure to change flagellar rotation
almost no relationship to flagellins from E. coli,
map in three genes, fliG, fliM, and fliN, which
S. typhimurium, or Bacillus subtilis, but are
code for the complex of switch proteins (FliG,
60% similar to the N and C termini of flagellin
FliM, and FliN).27 It appears that the three
from Caulobacter crescentus.29 Finally, a third
switch proteins form a complex of peripheral
HAP, HAP2, also called FliD (the fliD gene),
membrane proteins closely associated with the
caps the flagellar filament.
cytoplasmic side of the MS ring. In particular,
FliG seems to be bound to the MS ring itself; it 3. Brief summary of assembly
is also bound to FliM and FliN. The latter two of the flagellum
proteins form the cup-shaped C ring, located The flagellum and its associated components
directly beneath the basal body attached to the are assembled in a precise order, beginning with
MS ring. The regulator protein CheY, which is the components closest to the cell membrane.
important in switching flagellar rotation during The first components assembled, probably in a
coordinated fashion, are the MS and C rings. How can growth at the tip of the filament
This step is followed by construction of the be demonstrated? Growth at the tip has been
transport apparatus for export of the remain- visualized by the use of fluorophenylanine,
der of the flagellar components through a chan- a phenylalanine analogue, or radioisotopes.
nel in the center of the MS ring. Then the rod Incorporation of fluorophenylalanine by
is assembled, followed by the hook. The hook Salmonella resulted in curly flagella that had
is not completed until the P and L rings have only half the normal wavelength. When the
been assembled. When the hook is complete, analogue was introduced to bacteria that had
the flagellin monomers are exported and the partially synthesized flagella, the completed fla-
filament is assembled. MotA and MotB are gella were normal at the ends next to the cell
assembled late in the assembly process, to coin- and curly at the distal ends, implying that the
cide with the appearance of the filament. As fluorophenylalanine was incorporated at the
described next, the timing of assembly of the tips during flagellar growth.32 When Bacillus
components is at least partially controlled at the flagella were sheared off the cells and allowed
level of the transcription for the genes encoding to regenerate for 40 min before the addition of
these components. radioactive amino acid ([3H]leucine), radioau-
tography showed that all the radioactivity was
4. Brief summary of gene expression control at the distal region of the completed flagella.33
assembly regulating the timing and assembly The flagellin monomers are presumed to be
of flagellum and its associated components transported through the central channel in the
As we have seen, the flagellum and its com- filament to the tip, where they are assembled.
ponents are assembled in a precise order. For more information about the biosynthesis of
Interestingly, the genes for the flagellum and flagella, consult the review by Macnab.14
its associated components are expressed in the
order that the gene products are used. How is The export apparatus for flagellar components.
this done? In E. coli and Salmonella the genes are In the center of the C ring on the cytoplasmic
in three groups that are sequentially expressed. side is a knob composed of several Flh and Fli
The groups are denoted as belonging to class 1, proteins (FlhA, FlhB, FliH, FliI, FliO, FliP, FliQ,
2, or 3. The genes in class 1 (flhDC) comprise FliR). The ATPase FliI is negatively regulated
an operon that encodes transcriptional activa- by FliH, and its activity is required for flagellar
tors of the class 2 genes. Class 2 gene products assembly. (See note 34.) FliI and FliH are solu-
comprise the basal body and hook, as well as a ble proteins, as is a chaperone protein, FliJ, and
sigma factor (FliA) required for transcription of the others are integral membrane proteins. The
class 3 genes. Class 3 genes are required for the knob is the transport apparatus used for export
synthesis of the flagellin monomers (FliC) and of all the flagellar components (basal body, rod,
the torque-generating unit (MotA and MotB). hook, filament) through a central channel in the
MS ring, to be assembled beyond the cytoplas-
5. Growth of the flagellum mic membrane. One of the exported proteins
Although one might expect new flagellin sub- is a muramidase called FlgJ, and it presumably
units to be added to the growing filament at the functions to allow the nascent rod to penetrate
base next to the cell surface, this is not the case. the peptidoglycan. The transport system for the
The flagellin subunits actually travel through a bulk of flagellar materials is a flagellum-specific
hollow core in the basal body, hook, and fila- type III export system. (See the description of
ment and are added at the distal tip. The cap- type III export systems in Chapter 18) The P and
ping protein (HAP2) is important in this regard L rings are assembled using the Sec system of
because it prevents the flagellin monomers from transport, described in Section 18.1.
leaking out into the medium. Somehow, the cap-
ping protein must be able to move out from the 6. Differences in flagellar structure
growing filament to allow extension of the fila- Although the basic structure of flagella is simi-
ment, neither becoming detached nor allowing lar in all bacteria thus far studied, there are
flagellin subunits to leak out into the medium. important species-dependent differences. For
Models by which this may be done have been example, some bacteria have sheathed flagella,
suggested.30, 31 whereas others do not. In some species (e.g.,
www.ebook777.com
Vibrio cholerae) the sheath contains lipopoly- brane, and closely surrounded by an outer lipid
saccharide and appears to be an extension of the bilayer membrane. The spirochaete flagella are
outer membrane.35 In spirochaetes the sheath is called axial filaments. For a review, see ref. 40.
made of protein. (See later.) For a list of diseases caused by spirochaetes, see
Bacteria also vary with respect to the number note 41.
of different flagellins in the filaments. Depending Most spirochaete cells are helically coiled and
upon the species, there may be only one type have two or more flagella (some have 30 or 40 or
of flagellin, or two or more different flagellins more) inserted subterminally near each cell pole
in the same filament. For example, E. coli has and extending along the cell body. Some species
one flagellin; Bacillus pumilis has two differ- have a flat meandering waveform, rather than
ent flagellins, and C. crescentus has three and being helically coiled. The number of flagella
S. meliloti has four.36–38 The data for B. pumilis inserted at opposite poles is the same. The fla-
and C. crescentus support the conclusion that gella are usually more than half the length of the
the different flagellins reside in the same fila- cell and overlap in the middle. The spirochaete
ment. It has been proposed that the filaments of flagellum is often surrounded by a proteina-
S. meliloti are composed of heterodimers of the ceous sheath. Borrelia burgdorferi, the spiro-
two different flagellins. The flagella of many bac- chaete that causes Lyme disease, is somewhat
teria that have been studied (e.g., E. coli) show different. Its axial filaments are not surrounded
a smooth surface under electron microscopy by proteinaceous sheaths. Furthermore, this
and are called plain filaments. However certain organism has a planar waveform shape rather
bacteria (e.g., Rhizobium lupini, S. meliloti) than the corkscrew type.42 The rotation of the
have “complex” filaments with obvious helical rigid periplasmic flagella between the outer
patterns of ridges and grooves on the surface. membrane sheath and the cell cylinder is thought
Flagella with plain filaments rotate either clock- to move the cell by propagating a helical wave
wise or counterclockwise, whereas flagella with backward down the length of the highly flexible
complex filaments rotate only clockwise, with cell cylinder, propelling the cell forward, which
intermittent stops or slowing of rotation. It is allows the cells to “corkscrew” through viscous
thought that because complex filaments are media, such as mud, sediments, and connective
brittle, they are more rigid than plain filaments, tissue in animals. (For a more detailed model,
hence are better suited for propelling bacteria see note 43.) There are five antigenically related
in viscous media such as the gelatinous matrix flagellins in the axial filaments, but it is not
through which R. lupini and S. meliloti must known whether individual filaments contain
swim to infect root hairs of leguminous plants.39 more than one type of flagellin.44 Despite these
A discussion of the role of bacterial flagella in differences, the spirochaete flagella and those
pathogenicity can be found in the review by found in other bacteria are structurally similar
Moens and Vanderleyden.39 insofar as they have a basal body composed of a
Although E. coli and S. typhimurium have series of rings surrounding a central rod, plus a
three rings (MS, P, and L) through which the cen- hook and a filament.
tral rod passes, other bacteria may have fewer,
or more. For example, gram-positive bacteria 7. Site of insertion of flagella and the
lack the outer two rings (P and L). Additional number of flagella
structural elements of unknown function (e.g., The site of insertion of the flagellum and the
additional rings or arrays of particles surround- number of flagella vary with the bacterium.
ing the basal body) have been observed in cer- Some rod-shaped or curved cells have flagella
tain bacteria. that protrude from one or both of the cell poles.
A bacterium with a single, polar flagellum is said
Spirochaete flagella. A major difference between to be monotrichous. Bacteria with a bundle of
spirochaete flagella and those found in other flagella at a single pole are lophotrichous, from
bacteria is that the flagella in spirochaetes do the Greek words lophos (crest or tuft) and trichos
not protrude from the cell; rather, they are in (hair). Bacteria with flagella at both poles are
the periplasm, wrapped around the length of said to be bipolar. They may have either single
the protoplasmic cylinder next to the cell mem- or bundled flagella at the poles. Amphitrichous
refers to flagella at both poles. The prefix amphi, more than 100 years ago (1885). Now it is rec-
in Greek, means on both sides.) ognized to be a widespread phenomenon among
Some bacteria (e.g., spirochaetes) have sub- many bacterial genera, both gram-positive and
polar flagella, which are inserted near but not gram-negative.
exactly at the cell poles, and some curved bacte- The cells in swarming populations are called
ria (e.g., Vibrio) have a single, medial flagellum. swarmer cells, and they are often morphologi-
If the flagella are arranged laterally all around cally different from swimmer cells grown in
the cell (e.g., as in Escherichia and Salmonella) liquid. The morphological changes that occur
they are said to be peritrichous. (The prefix when swimming cells from a liquid culture are
peri, in Greek, means around.) Peritrichous inoculated onto an agar plate and convert to
flagella coalesce into a trailing bundle during swarmer cells can be more or less pronounced
swimming. depending upon the bacterium and the concen-
tration of the swarming agar. In general, cells
8. Role of flagella in tactic responses and
that convert from swimming cells to swarming
in virulence
cells become nonseptate filaments, multinucle-
Many swimming bacteria are capable of tactic
oid, and hyperflagellated with lateral flagella.
responses: that is, they swim toward environ-
(Bacillus subtilis swarmer cells differ less dra-
ments more favorable with respect to nutrient,
matically from the nonswarmer cells in being
light, and electron acceptors, and away from
only slightly larger and having only two nuclei.)
toxic or less favorable environments. These
The lateral flagella are critical because to migrate
swimming responses occur because some bac-
as populations of cells, the cells physically inter-
teria can sense environmental signals, transfer
act with each other via the lateral flagella.
these signals to the flagellum motor, and modify
Swarming can be facilitated by the produc-
the rotation of their flagella to swim in a particu-
tion of a surfactant that reduces surface tension
lar direction. How this occurs for bacteria of dif-
at the aqueous periphery of the colony on hydro-
ferent types is discussed in detail in Chapter 20,
philic surfaces such as agar, allowing the colony
which describes chemotaxis. Flagella can also
to expand. The surfactant is one of the compo-
make important contributions to the virulence
nents in the extracellular slime produced by the
of pathogenic bacteria.45 For example, it has
bacteria. For example, Serratia marcescens pro-
been suggested that the ability of spirochaetes
duces a cyclic lipopeptide (3-hydroxydecanoic
(spiral-shaped bacteria) such as Treponema pal-
acid attached to five amino acids), and B. sub-
lidum, the causative agent of syphilis, to swim in
tilis produces a lipopeptide surfactant called
a corkscrew fashion through viscous liquid aids
surfactin.48,49 Isolated swarmer cells rarely
in their dissemination (e.g., through connec-
move. One of the differences between isolated
tive tissue or the junctions between endothelial
cells and cells in a group is that the cells in the
cells).
group are encased in much more slime than is
9. Flagella and swarming found around single cells. It could be that wet-
The swarming of flagella is a type of social ting agents in the slime hydrate the external
swimming in which cells move on solid surfaces medium sufficiently for the flagella to rotate,
in groups (called rafts) of physically interact- allowing swarming to take place on the agar.
ing cells. (For a review of the different motility To demonstrate swarming on agar, one
systems that allow bacteria to move on solid inoculates petri plates containing nutrient agar
surfaces, read ref. 46; for a review of swarming at a concentration that is wet enough to allow
behavior, and to learn how swarmer cells dif- swarming to take place (e.g., 0.5–1%, or some-
fer from cells grown in liquid, consult ref. 47.) times 2%), but not so dilute that cells can swim
Swarming allows bacterial populations to rap- as individuals. For example, if the agar concen-
idly spread as multicellular populations, rather tration is between 0.2 and 0.3%, individual bac-
than as single cells, over a wet, solid surface, for teria can swim through water-filled pores in the
example in biofilms discussed in Chapter 21. (As agar. Such very dilute agar plates are referred to
discussed in Box 1.2, gliding represents another as “swim plates.” Swarming is easily recognized
means for bacteria to spread on a solid surface.) because a population of swarming cells spreads
Swarming was first described in Proteus species over the agar plate, whereas nonswarmers grow
www.ebook777.com
as discrete colonies when inoculated in the cen- are assembled by having the subunits pushed
ter of the plate. As reported for B. subtilis, some through the central channel and assembled at
bacteria may lose the ability to swarm when the distal tip. (See earlier subsection entitled 5.
they are maintained as laboratory cultures.49 Growth of the flagellum.) It has been suggested
that archaeal flagella are assembled by adding
A different flagellar system is sometimes pro- subunits at the base.
duced for swarming. Although many bacteria,
such as Escherichia, Proteus, Salmonella, and Fimbriae, pili, filaments, and fibrils
Serratia, use the same lateral flagella system Protein fibrils extending from the cell surface
both for swimming in liquid and for swarming have been called various names, including fim-
on a solid surface (albeit they make many more briae, pili, filaments, and simply fibrils. They
flagella for swarming), other bacteria, such extend from the surface of most gram-negative
as species of Aeromonas, and Azospirillum, bacteria (Fig. 1.3).54-57 Similar fibrils are only
Rhodospirillum centenum, Vibrio alginolyti- occasionally observed in gram-positive bac-
cus, and V. parahaemolyticus, make a spe- teria, such as for Corynebacterium renale and
cial lateral flagella system for swarming and Actinomyces viscosus.58,59 (C. renale is the caus-
growth in viscous environments. (Reviewed in ative agent of bovine pyelonephritis and cystitis,
ref. 50.) These bacteria are polarly flagellated and A. viscosus is part of the normal flora of the
when grown in liquid but become both polar human mouth, where it adheres to other bacte-
and peritrichously flagellated when grown on ria in plaque and to teeth.) All such appendages
surfaces. For example, V. alginolyticus and V. are not alike, nor are their functions known
parahaemolyticus produce many unsheathed in all instances. The size varies considerably.
lateral flagella that aid in swarming when these They can be quite short (0.2 μm) or very long
microorganisms are grown on a solid surface. (20 μm), and they differ in width from 3 nm to
For some bacteria, such as Vibrio species, the 14 nm or greater. Some originate from anchor-
polar and lateral flagella are determined by ing structures in the cell membrane. However,
different genes. The increase in production of most seem not to originate in the cell membrane
lateral flagella may be necessary to overcome at all, and how they are attached to the cell is
surface friction and/or the viscosity of the slime
produced by a swarming population. Perhaps
some sort of mechanosensory system, activated
by surface friction and/or slime viscosity, sig-
nals the expression of the genes for lateral fla-
gella. It has been suggested that the inhibition
of flagellar rotation on the solid surface might
be part of the mechanosensory signaling system
in Vibrio.
10. Archaeal flagella

Archaeal flagella are different from bacterial
flagella.51 The primary sequences of the flagel-
lins from several archaea show no homology
at all to bacterial flagellins, although they do
show sequence homology to each other at the
N-terminal end.52 Also, the basal body struc-
ture in the archaeon Methanospirillum hun-
gatei appears to be simpler than for bacterial
flagella in that a simple knob is present rather
than rings.53 Also, as opposed to bacterial fla- Fig. 1.3 Electron micrograph of a metal-shadowed
gella, there is no central channel in archaeal preparation of a dividing Salmonella typhimurium
flagella. This means that the assembly mecha- showing flagella and fimbriae (or pili). The cell is
nism is different from that of bacterial flagella. about 0.9 μm in diameter. (Reprinted with permis-
The external portions of the bacterial flagellum sion of J. P. Duguid.)
not known. Furthermore, they are not always glycoside residues on the cell surface receptors.
present. Although freshly isolated strains fre- They are called common pili, or mannose-sensi-
quently have fibrils, filaments are often lost dur- tive pili, or (more frequently) type 1 pili. Other
ing extended growth in the laboratory. They are pili recognize different cell surface receptors and
apparently useful in the natural habitat but dis- are sometimes called mannose-resistant pili.
pensable in laboratory cultures. What do they
do? Medical significance. Pilus proteins (and other
adhesins) can have great medical significance.
1. Pili (fimbriae) An example of attachment to host cells via
Many of the fibrils can be observed to mediate bacterial pili is the attachment of Neisseria
attachment of the bacteria to other cells (e.g., gonorrhoeae, the causative agent of the sexu-
other bacteria, or animal, plant, or fungal cells) ally transmitted disease (STD) gonorrhea, to
and to abiotic surfaces. Although adhesive epithelial cells of the urogenital tract. Another
fibrils are sometimes referred to as fimbriae to example is E. coli. Most clinical isolates of E.
distinguish them from the sex pili that are used coli from the urinary and gastrointestinal tracts
in bacterial conjugation, in this discussion, all possess type 1 pili. Many also (or instead) have
such fibrils will be referred to as pili. Pili are galactose-sensitive pili, called P-type pili.60,61
important for colonization because they help the P-type pili bind the α-D-galactopyranosyl-(1–
bacteria stick to surfaces. In nature most bacte- 4)-β-D-galactopyranoside in the glycolipids on
ria grow while attached to surfaces, where the cells lining the upper urinary tract.62
concentration of nutrients is frequently highest. Other known pili include the type 4 pili
Attachment can be via adhesive pili and/or non- (type IV pili) produced by N. gonorrhoeae,
fibrillar material that may be part of the cell wall Pseudomonas aeruginosa, Bacteroides nodo-
or glycocalyx (Section 1.2.2). Adhesive pili have sus, and Moraxella bovis (see note 63). These
adhesins, that is, molecules that cause bacteria pili mediate the form of surface motility known
to stick to surfaces. The adhesins are commonly as twitching. A different type of pili are the Tcp
proteins in the pili, often minor proteins at the (toxin-coregulated pili) pili produced by Vibrio
tip, that recognize and bind to specific receptors cholerae. (See note 64 for the diseases caused by
on the surfaces of cells. Adhesion is studied in these bacteria.) The Tcp pili are necessary for
the laboratory under experimental situations of V. cholerae to colonize the intestinal mucosa of
two types: (1) attachment of bacteria to eryth- mammals, and mutants that do not make these
rocytes, causing them to clump (hemagglutina- pili do not cause disease. Thus it is evident that
tion), and (2) attachment of bacteria in vitro to pili of different types are specialized for attach-
host cells to which their attachment normally ment to specific receptors and can account for
occurs in vivo. Often researchers who study specificity of bacterial attachment to hosts and
such attachments have found that when spe- tissues. Pili can be distinguished by inhibition
cific monosaccharides and oligosaccharides are of binding by mono- and oligosaccharides, as
added to the suspension, they inhibit attach- well as by a variety of other methods, including
ment or hemagglutination. The implications are morphology, antigenicity, molecular weight of
that at least some, if not all, pili bind to oligosac- the protein subunit, isoelectric point of the pro-
charides in cell surface receptors, and that the tein subunit, and amino acid composition and
added monosaccharides and oligosaccharides sequence. Since gram-positive bacteria gener-
are inhibitory because they compete with the ally do not have pili, their adhesins are part of
receptor for the adhesin. Receptors on animal other cell surface components (e.g., the glycoca-
cell surfaces include glycolipids and glycopro- lyx, to be described next, in Section 1.2.2).
teins, which are embedded in animal cell mem-
branes via the lipid and protein portions, and 2. Sex pili
present their oligosaccharide moieties to the Bacteria are capable of attaching to each other
outer surface. It has been reported that many for the purpose of transmitting DNA from a
strains of E. coli, Salmonella, and Shigella carry donor cell to a recipient cell. This is called con-
pili whose hemagglutinin activity is prevented jugation or mating. (See Fig. 1.4.) Some bacteria
by D-mannose and methyl-α-mannoside. These (not all) use pili for mating attachments. The pili
pili, therefore, are believed to attach to mannose that mediate attachment between mating cells
www.ebook777.com
1.2.2 The glycocalyx

1
The term “glycocalyx” is often used to describe
all extracellular material that is external to
donor recipient
the cell wall.68–71 (The word calyx is from the
2 Greek kályx, meaning husk or outer covering.)
Such material includes the extracellular matrix
(ECM) in biofilms, described in Chapter 21. The
3 polymers in glycocalyces are predominantly
polysaccharides and/or proteins. All bacteria
are probably surrounded by glycocalyces as
they grow in their natural habitat, although they
Fig. 1.4 F-pilus-mediated conjugation. Transfer often lose these external layers when cultivated
of a sex plasmid. The donor cell has a plasmid and in the laboratory. The extracellular polymers
an F pilus encoded by plasmid genes. 1, The F pilus may be in the form of S layers, capsules, slime,
binds to the recipient cell. 2, In this model a depo- or a loose network of fibrils.
lymerization of the pilus subunits causes the pilus to The S layers, which are found on the cell wall
retract, bringing the two cells together. 3, The plas-
surfaces of a wide range of gram-positive and
mid is transferred as it replicates so that when the
gram-negative eubacteria, are arrays of protein
cells separate, each has a copy of the plasmid and is
a potential donor. An alternative model [explained or glycoprotein subunits on the cell wall. They
in Babic, A., A. B. Lindner, M. Vulic, E. J. Stewart, are also present in the Archaea, where the S layer
and M. Radman. Direct visualization of horizon- sometimes tightly covers the cell membrane and
tal gene transfer. 2008. Science 319:1533–1536] is serves as the cell wall itself. If the S layer is the
that DNA transfer can occur between separated E. wall itself, it is not considered a glycocalyx.
coli cells through the F-pilus from F+ donor cells to Capsules are composed of fibrous material at
F− recipient cells. the cell surface and generally ensheath the cell
(Figs. 1.5 and 1.6). They may be rigid, flexible,
are different from the other pili just discussed integral (i.e., very closely associated with the
and are called sex pili. The requirement of sex cell surface), or peripheral (i.e., loose) material
pili for mating is found for a variety of gram- that is sometimes shed into the medium.
negative bacteria such as E. coli, but gram-pos- Material that loosely adheres to the cell wall
itive bacteria do not have sex pili. The sex pilus is sometimes called slime or slime capsule. If the
grows on “male” strains that donate DNA to loosely adhering or shed material is polysac-
recipient (“female”) strains. In E. coli, the sex charide, it is often referred to as extracellular
pilus is encoded by a conjugative transmissible polysaccharide (EPS), such as exists in biofilm
plasmid, the F plasmid, that resides in the donor matrices. (In discussing bioflims, “EPS” also
strains and can be delivered to recipients. (See serves as a general term to refer to extracellular
note 65 for an explanation of plasmids.) polymeric substance, not restrictive to polysac-
The gram-positive Enterococcus faecalis charide.) The capsular polysaccharide is usually
forms efficient mating aggregates in liquid sus- covalently bound to phospholipid or phospho-
pension and does not have sex pili. In other lipid A, which is embedded in the surface of the
words, the adhesins are located on the cell sur- cell. This is not the case for extracellular poly-
face rather than on pili. Mating in E. faecalis is saccharide. The synthesis of extracellular poly-
of additional interest because it is induced by saccharides is discussed later, in Section 12.3.
a sex pheromone secreted into the medium by For a review of bacterial capsules and their
recipient cells. The sex pheromone signals the medical significance, see ref. 72.
donor cells to synthesize cell surface adhesins
that promote cell aggregate formation (clump- Chemical composition
ing) and subsequent DNA transfer.66 Mating The glycocalyces from several bacteria have
interactions mediated by surface adhesins is been isolated and characterized. Although
widespread in the microbial world. Other well- many are polysaccharides, some are proteins.
studied examples include Chlamydomonas For example, some Bacillus species form a
mating and yeast mating.67 glycocalyx that is a polypeptide capsule. Also,
Fig. 1.5 Electron micrograph of a thin section of E. Fig. 1.6 Electron micrograph of a thin section of bac-
coli stained with a ruthenium red dye. The cells are teria adhering to a rock surface in a subalpine stream.
adhering to neonatal calf ileum. Source: Costerton, J. The sample, showing fibrous glycocalyx, was stained
W., T. J. Marrie, and K.-J. Cheng. 1985. Phenomena of with ruthenium red dye. Source: Costerton, J. W.,
bacterial adhesion, pp. 3–43. In: Bacterial Adhesion. T. J. Marrie, and K.-J. Cheng. 1985. Phenomena of
D. C. Savage and M. Fletcher (Eds.). Plenum Press, bacterial adhesion, pp. 3–43. In: Bacterial Adhesion.
New York and London. With kind permission from D. C. Savage and M. Fletcher (Eds.). Plenum Press,
Springer Science + Business Media B.V. New York and London. With kind permission from
Springer Science + Business Media B.V.
pathogenic Streptococcus species have a fibril- species. For example, the K1 polysaccharide of
lar (hairlike) protein layer, the M protein, on E. coli is identical to the group B capsular poly-
the external face of the cell wall. (See note 73 saccharide of Neisseria meningitidis.
for a description of the role of M protein in S.
pyogenes pathogenesis.) Role
The capsular polysaccharides are extremely An important role for the glycocalyx can be
diverse in their chemical composition and adhesion to the surfaces of other cells or to inan-
structure.74 Some consist of one type of mono- imate objects to form a biofilm. Such adhesion
saccharide (homopolymers), whereas some are is necessary for colonization of solid surfaces or
composed of more than one type of monosac- growth in biofilms (Figs. 1.5 and 1.6). The bac-
charide (heteropolymers). The monosaccha- teria can adhere to a nonbacterial surface or to
rides are linked together via glycosidic linkages each other via these attachments. An example
to form straight-chain or branched molecules. is the complex ecosystem of several different
They can be substituted with organic or inor- bacteria maintained by intercellular adherence
ganic molecules, further increasing their diver- in human oral plaque.75 Advantages to such
sity. For example, E. coli strains make more adherence include the higher concentrations of
than 80 different capsular polysaccharides, nutrients that may be found on the surfaces.
called K antigens. Sometimes the same capsular Another role of the glycocalyx is protection
polysaccharide is made by different bacterial from phagocytosis. For example, mutants of
www.ebook777.com
pathogenic strains of bacteria that no longer and gram-negative walls, responsible for the
synthesize a capsule, such as unencapsulated strength of bacterial cell walls. Then we con-
strains of Streptococcus pneumoniae, are more sider, in turn, the gram-positive and gram-neg-
easily phagocytized by white blood cells, mak- ative cell walls.
ing the pathogens less virulent.
The glycocalyx can also prevent dehydration The Gram stain
of the bacterial cell, an important role in the The Gram stain was invented in 1884 by a
soil. This is because polyanionic polysaccha- Danish physician, Christian Gram, to allow the
rides, including polysaccharide capsules, are use of ordinary bright-field microscopy in the
heavily hydrated. visualization of bacteria in tissues. For more
about Christian Gram, see Box 1.3.
1.2.3 Cell walls When appropriately stained, bacteria with
Most bacteria are surrounded by a cell wall that cell walls can usually be divided into two
lies over the external face of the cell membrane groups, depending upon whether they retain
and protects the cell from bursting due to the a crystal violet–iodine stain complex (gram-
cell’s internal turgor pressure.76,77 The turgor positive) or do not (gram-negative) (Fig. 1.7).
pressure exists because bacteria generally live Gram-negative bacteria are visualized with a
in environments that are more dilute than the pink counterstain called safranin. During the
cytoplasm. As a consequence, there is a net staining procedure the complex of crystal violet
influx of water that results in a large hydro- and iodine can be readily removed with alco-
static pressure (turgor) of several atmospheres hol or acetone from gram-negative cells, but
directed out against the cell membrane. Two less readily from gram-positive cells. This result
major types of wall exist among the bacteria. is thought to be related to the thickness of the
One type of wall can be stained by using the peptidoglycan layer in the gram-positive wall in
Gram stain procedure, and relatively restricted comparison to the much thinner peptidoglycan
groups of bacteria having walls of this type are layer in the gram-negative wall. During the alco-
called gram-positive. Many more bacteria pos- hol/acetone wash, the outer lipopolysaccharide
sess the second wall type, do not stain, and are layer of the gram-negative bacteria is disrupted
called gram-negative. and the dye leaks out through the thin, porous
We introduce the Gram stain next, follow- peptidoglycan layer.
ing up with a description of peptidoglycan, a The difference in Gram staining is also related
cell wall polymer found in both gram-positive to the outer wall layer of gram-negative bacteria
BOX 1.3 HISTORICAL PERSPECTIVE: CHRISTIAN GRAM

Christian Gram was a Danish physician different bacteria, which later would be
who worked at the morgue of the City characterized as gram-positive and gram-
Hospital of Berlin. In 1894 he published a negative, but in 1886 Carl Flügge pub-
procedure he had devised for staining bac- lished a textbook in which he pointed out
teria. The method allowed the staining of that Gram’s staining procedure is useful for
all bacteria in tissue preparations to dif- the differential staining of bacteria in tis-
ferentiate them from nuclei. Some bacteria sues and for diagnostic purposes. Christian
did not retain the stain, however, making Gram died in 1935.
the procedure less effective than Gram had
desired. Source: Lechevalier, H. A., and M. Solotorovsky.
It is not clear who realized that the Gram 1965. Three Centuries of Microbiology.
stain could be used to distinguish between McGraw-Hill Book Company, New York.
(also called the outer envelope or outer mem- Although Fig. 1.8 depicts the glycan chains
brane), which is rich in phospholipids and thus running parallel to the cell membrane, this is a
made leaky by the lipid solvents, alcohol, and controversial point. Another model, the scaf-
acetone. The Archaea can stain either gram-pos- fold model, proposes that the glycan chains run
itive or gram-negative, but their cell wall com- perpendicular to the cell membrane.78 Others
position is different from that of the Bacteria, as have argued that the parallel model is more
will be described later. consistent with experimental data than the scaf-
fold model.79 The glycan consists of alternating
Peptidoglycan residues of N-acetylglucosamine (GlcNAc or
The strength and rigidity of bacterial cell walls G) and N-acetylmuramic acid (MurNAc or M)
is due to molecules called peptidoglycan or attached to each other via β-1,4 sugar linkages.
murein, which consist of glycan chains cross- Attached to the MurNAc is a tetrapeptide that
linked by peptides (see later: Figs. 1.18, 12.3). cross-links the glycan chains via peptide bonds
(black dotted stems on M residues, Fig. 1.8).
The tetrapeptide usually consists of the follow-
ing amino acids in the order that they occur
outer membrane
from MurNac: L-alanine, D-glutamate, L-R3
peptidoglycan (the R indicates that the amino acid can vary
cell between species: see note 80), and D-alanine.
membrane The structural diversity of the peptidoglycan, as
well its synthesis, is discussed in more detail in
Section 12.1. The peptidoglycan forms a three-
dimensional network surrounding the cell mem-
brane, covalently bonded throughout by the
glycosidic and peptide linkages. It is the cova-
lent bonding that gives the peptidoglycan its
strength. The shape of the peptidoglycan helps
maintain the shape of the cell as well as shape
changes that occur during cellular morphogen-
esis (e.g., during the formation of round cysts
gram- gram-
negative positive
from rod-shaped vegetative cells). In addition
to peptidoglycan, certain cytoskeletal compo-
Fig. 1.7 Schematic illustration of a gram-negative nents such as MreB and crescentin are involved
and a gram-positive bacterial cell wall. Note the pres- in shape determination. This is discussed later
ence of an outer membrane (also called outer enve- in this chapter, in Section 1.2.7. Destruction
lope) in the gram-negative wall and the much thicker of the peptidoglycan by the enzyme lysozyme
peptidoglycan layer in the gram-positive wall. (which hydrolyzes the glycosidic linkages) or
G M G M G M G M
M G M G M G M G
G M G M G M G M
Fig. 1.8 Schematic drawing of a peptidoglycan layer. The peptidoglycan surrounds the cell membrane and
consists of glycan chains (–G–M–) cross-linked by tetrapeptides (solid circles). The peptidoglycan is approxi-
mately one monomolecular layer thick in gram-negative bacteria and several layers thick in gram-positive
bacteria. The direction in which the glycan chains are depicted to be running is not to be taken literally. It has
been suggested to run in a helical pattern in rod-shaped cells. Abbreviations: M, N-acetylmuramic acid; G,
N-acetylglucosamine.
www.ebook777.com
interference in its synthesis by antibiotics (such organic solvents, the cell membrane adheres to
as penicillin, vancomycin, or bacitracin) results the outer envelope in numerous places, while
in the inability of the cell wall to restrain the generally shrinking away in other locations. It is
turgor pressures. Under these circumstances, reasonable to suggest that these zones of adhe-
the influx of water in dilute media causes the sion may be areas in which the peptidoglycan
cell to swell and burst. Since the strength of the bonds the cell membrane to the outer membrane
peptidoglycan is due to its covalent bonding because the peptidoglycan lies between the two
throughout, changes in its shape, or expansion membranes.
during growth of the cell, must be accompanied
by hydrolysis of some of the covalent bonds and Gram-positive walls
synthesis of new bonds. However, because of 1. Chemical composition
the high intracellular turgor pressure, this must The gram-positive cell wall in bacteria is a thick
be under tight control; otherwise a lethal weak- structure approximately 15 to 30 nm wide and
ness in the peptidoglycan structure will be pro- consists of several polymers, the major one
duced when its bonds are cleaved. This problem being peptidoglycan (Figs. 1.8 and 1.9). The
is discussed by Koch.81 kinds and amounts of other polymers in the
There are some important differences wall vary according to the bacterial taxa. The
between the peptidoglycans in gram-positive nonpeptidoglycan polymers can comprise up to
and gram-negative bacteria. The peptidogly- 60% of the dry weight of the wall. Most of these
can of gram-negative bacteria can be isolated are commonly found covalently linked to the
as a sac of pure peptidoglycan that surrounds glycan chain of the peptidoglycan. These poly-
the cell membrane in the living cell. This recep- mers include teichoic acids, teichuronic acids,
tacle, called the murein sacculus, is elastic and neutral polysaccharides, lipoteichoic acids, and
is believed to be under stress in vivo because of glycolipids of different types (Fig. 1.10).
the expansion due to turgor pressure against the
cell membrane.82 In contrast, the peptidogly- Teichoic acids. For a review, see ref. 83. Teichoic
can from gram-positive bacteria is covalently acids are polyanionic polymers of either ribi-
bonded to various polysaccharides and teichoic tol phosphate or glycerol phosphate joined by
acids and cannot be isolated as a pure murein anionic phosphodiester bonds (Fig. 1.10). They
sacculus (See the discussion of the gram-posi- can comprise 30 to 60% of the dry weight of the
tive wall in the next subsection.) wall and may have multiple roles, as described
The peptidoglycan from gram-negative in ref. 84. Teichoic acids vary structurally with
bacteria differs from the peptidoglycan from respect to the extent and type of molecules
gram-positive bacteria in two other ways: (1) covalently bonded to the hydroxyl groups of
diaminopimelic acid is generally the diamino the glycerol phosphate or ribitol phosphate
acid in gram-negative bacteria, whereas this in the backbone. Frequently, the amino acid
D-alanine is attached, as well as the sugar glu-
position is much more variable among gram-
positive bacteria; and (2) the cross-linking is cose or N-acetylglucosamine. The teichoic acids
generally direct in gram-negative bacteria, are attached to the peptidoglycan via covalent
whereas there is usually an additional peptide bonds between the phosphate of glycerol phos-
bridge between the tetrapeptide stems in gram- phate or ribitol phosphate to the C6 hydroxyl of
positive bacteria (see later: Fig. 12.3). N-acetylmuramic acid (MurNac).
As described in Chapter 12, the monolayer Teichuronic acids. Teichuronic acids are polyan-
of peptidoglycan in gram-negative bacteria is ionic acidic polysaccharides containing uronic
attached noncovalently to the outer envelope acids (e.g., some have N-acetylgalactosamine
via lipoprotein. Whether it is also attached to and D-glucuronic acid). When Bacillus subtilis is
the cell membrane is a matter of controversy. grown in phosphate-limited media, the teichoic
The evidence for attachment to the cell mem- acid is replaced by teichuronic acid (reviewed
brane is that when E. coli is plasmolyzed (placed in ref. 84).
in hypertonic solutions so that water exits the
cells), and prepared for electron microscopy by Neutral polysaccharides. Neutral polysaccharides
chemical fixation followed by dehydration with are particularly important for the classification of
Fig. 1.9 Schematic drawing of the gram-positive wall. Components, starting from the bottom, are as follows:
PM, plasma cell membrane consisting of protein (Pr), phospholipid (Pl), and glycolipid (Gl). Overlying the
cell membrane is highly cross-linked peptidoglycan (PG) to which are covalently bound teichoic acids, teichu-
ronic acids, and other polysaccharides (SP). Acylated lipoteichoic acid (aLTA) is bound to the cell membrane
and extends into the peptidoglycan. Some of the LTA is in the process of being secreted (LTAt). LTA that
already has been secreted into the glycocalyx is symbolized as aLTAx. Some of the LTA in the glycocalyx is
symbolized as dLTAx because it has been deacylated (i.e., the fatty acids have been removed from the lipid
moiety). Within the glycocalyx can also be found lipids (L), proteins (Pe), pieces of cell wall (W), and polymers
that are part of the glycocalyx proper (G). Also shown is the basal body and hook structures of a flagellum (B).
Source: Wicken, A. 1985. Bacterial cell walls and surfaces, pp. 45–70. In: Bacterial Adhesion. D. C. Savage
and M. Fletcher (Eds.). Plenum Press, New York and London. With kind permission from Springer Science +
Business Media B.V.
streptococci and lactobacilli, where they are used hydrophobic lipid, the molecule is amphipathic
to divide the bacteria into serological groups (e.g., (i.e., it has both a polar and a nonpolar end).
groups A, B, and C streptococci). The lipid portion is bound hydrophobically to
the cell membrane, whereas the polyglycerol
Lipoteichoic acids. A polyanionic teichoic acid phosphate portion extends into the cell wall.
found in most gram-positive walls is lipote- Unlike the other polymers thus far discussed,
ichoic acid (LTA), which is a linear polymer LTA is not covalently bound to the peptido-
of phosphodiester-linked glycerol phosphate glycan. Its biological role at this location is not
covalently bound to lipid. The C2 position of understood, but in some bacteria it is secreted
the glycerol phosphate is usually glycosylated and can be found at the cell surface, where it is
and/or D-alanylated. Because of the negatively thought to act as an adhesin. For example, LTA
charged backbone of glycerol phosphate and the is secreted by S. pyogenes, where it binds with
www.ebook777.com
Fig. 1.10 Some teichoic and teichuronic acids found in different gram-positive bacteria. (A) Glycerol
phosphate teichoic acid with D-alanine esterified to the C2 of glycerol. (B) Glycerol phosphate teichoic
acid with D-alanine esterified to the C3 of glycerol. (C) Glycerol phosphate teichoic acid with glucose and
N-acetylglucosamine in the backbone subunit. D-Alanine is esterified to the C6 of N-acetylglucosamine.
(D) Ribitol phosphate teichoic acid, with D-glucose attached in a glycosidic linkage to the C4 of ribitol. (E)
Teichuronic acid with N-acetylmannuronic acid and D-glucose. (F) Teichuronic acid with glucuronic acid
and N-acetylgalactosamine. It is believed that teichoic and teichuronic acids are covalently bound to the
peptidoglycan through a phosphodiester bond between the polymer and a C6 hydroxyl of muramic acid in
the peptidoglycan.
the M protein and acts as a bridge to receptors Bacteria belonging to the genus Mycobacterium
on host tissues. (See note 85 for a further expla- include the causative agents of tuberculosis (M.
nation.) Under these circumstances, one should tuberculosis) and leprosy (M. leprae). M. tuber-
consider the secreted LTA to be part of the gly- culosis infects approximately one-third of the
cocalyx along with the M protein. world’s population and causes about 3 million
deaths annually. Between 12 million and 13
Other glycolipids. There is a growing list of million people worldwide are infected with M.
gram-positive bacteria that do not contain leprae. For more complete descriptions of tuber-
LTA but have instead other amphiphilic gly- culosis and leprosy, see Boxes 1.4 and 1.5, respec-
colipids that might substitute for some of the tively. The cells walls of mycobacteria consist of
LTA functions, whatever they might be.86 waxy lipids that can comprise up to 40% of the
Bacteria having these cell surface glycolip- dry weight of the cell wall. The waxy lipids are
ids (also called macroamphiphiles or lipo- responsible for resistance to dehydration, acids,
glycans) belong to various genera, including and alkalies. Indeed, treatment of sputum with
Micrococcus, Streptococcus, Mycobacterium, dilute sulfuric acid or dilute sodium hydroxide
Corynebacterium, Propionibacterium, Actino- enriches for mycobacteria, as opposed to other
myces, and Bifidobacterium. bacteria in the respiratory flora.
BOX 1.4 TUBERCULOSIS

Tuberculosis is a lung disease caused by verb “to caseate” comes from the Latin
Mycobacterium tuberculosis, but the bac- caseus, which means cheese.) The caseated
teria can spread to other tissues. The lung lesions may heal, becoming infiltrated with
tissue is slowly destroyed by a complex pro- fibrous tissue and, often, acquiring depos-
cess involving macrophages activated by its of calcium, which form nodules (large
the immune system. The symptoms are the tubercles) that can be seen in X-ray images.
coughing up of mucus, which may be bloody, The walled-off tubercles remain in the lung
low-grade fever, night sweats, and weight for the rest of the individual’s life and con-
loss (wasting). Tuberculosis used to be called tain viable bacteria. Most active cases of
consumption and also the white plague. tuberculosis are due to the reactivation of
tubercles. This may happen in individuals
with weakened immune systems, such as
Pathogenesis people with AIDS, or in the elderly.
During an early stage in the progress of
The bacteria enter the respiratory tract the disease—that is, during the primary
via inhalation of nasopharyngeal secre- lesion stage—the bacteria may spread into
tions from individuals with active disease. regional lymph nodes, and from there into
Once bacteria have entered the lungs, small the bloodstream. The bacteria may also
lesions, frequently called primary lesions, spread into the bloodstream from case-
are produced, where the bacteria reproduce. ated lung tissue. The circulatory system can
At this stage of the infection, the bacteria bring the bacteria to various organs and tis-
are engulfed by nonactivated alveolar mac- sues, including the bone marrow, spleen,
rophages. A certain fraction of the bacteria liver, meninges, and kidneys. The bacteria
survive and grow inside the macrophages can even return from these organs, to the
at the site of the lesions. When they grow lungs via the blood circulatory system,
inside the macrophages, the macrophages and this can contribute to the reactivation
are killed and release the bacteria. Since, of the disease in the lungs years after the
however, the cellular immune system reacts initial infection. An asymptomatic period
against the bacteria, the alveolar mac- of years frequently precedes the appear-
rophages are activated, becoming better ance of symptoms in various body organs.
killers of the mycobacteria. This stops the Symptoms, when they do occur, are caused
spread of the infection. The lesions heal and by malfunction of the organs or tissues
form granulomas, also called tubercles, con- due to necrotizing lesions, or to damage of
taining large activated macrophages called blood vessels that feed the tissues.
epithelioid cells because they resemble epi-
thelial cells, and multinucleated giant cells
derived from the fusion of epitheloid cells The role of the cellular immune
(Langhans giant cells). The tubercles also response in damaging infected tissue
contain lymphocytes and fibroblasts, the
latter forming fibrous tissue in the tubercles. The macrophages are activated by lympho-
There are relatively few bacteria present in cytes that are activated during the immune
the tubercles, and these are mostly outside response to the bacteria. These are CD4+
the macrophages. However, the activated (T-helper cells) and CD8+ (T-cytolytic)
macrophages eventually destroy the lung lymphocytes. The T-helper (TH) cells acti-
tissue in the center of the tubercles, caus- vate the other lymphocytes, namely, B
ing it to become semisolid or “cheesy.” The lymphocytes, which mature into antibody-
“cheesy” tissue is said to be caseated. (The secreting cells, TC cells, and macrophages.
www.ebook777.com
The B lymphocytes are part of the humoral A more rapid test is to use the polymerase
response, and they mature to make antibod- chain reaction (PCR) to amplify the bacte-
ies. The T-cytolytic lymphocytes (TC cells) rial DNA. The PCR assay can be applied to
are part of the cellular immune response, sputum material, and specific DNA probes
and their major role is to attach to and can be used to detect M. tuberculosis DNA
kill virus-infected cells and foreign tissues. sequences. The reagents can be purchased
Both the TH and TC cells produce interferon in a kit.
gamma (IFN-γ), which is a major activa-
tor of macrophages. The macrophages are
said to be active when they become capable Who gets sick
of new activities, such as the engulfment
and killing of bacteria. Hence, activated Tuberculosis is a major worldwide killer,
macrophages are also part of the cellular certainly one of the most important of
immune response and limit the spread of the lethal infectious diseases. The World
the infection. Tissue necrosis (destruction) Health Organization (WHO) estimates
is not due to toxins produced by the myco- that about 10 million new active cases of
bacteria. Rather, it results when hydrolytic tuberculosis occur each year, and 3 mil-
enzymes and toxic forms of oxygen are lion people die. In the twentieth century
released by the activated macrophages in alone there were about 1 billion deaths
the granulomas. These factors are made due to tuberculosis. It has been estimated
by macrophages to kill the pathogens they that about half the world’s population of
have engulfed; but when leaked into the approximately 6 billion people is infected,
interstitial fluid, they kill surrounding tis- but that only 30 million of infected people
sue. Activated macrophages also release have active cases. This means that about
factors that cause blood clotting (proco- 99% of people who become infected
agulant factors), and when the clots form with Mycobacterium tuberculosis do
in local blood vessels, blood supply to the not become ill and are healthy carri-
tissues is depressed, resulting in death of the ers, although as noted, the disease can
tissues. become activated later in life. Estimates
are that approximately 15 million people
in the United States are infected. A skin
Laboratory diagnosis test called the tuberculin testt can detect
whether a person has been infected with
The identification of acid-fast rods in M. tuberculosis.
smears of sputum is a tentative diagnosis of
tuberculosis. To identify acid-fast rods, the
Ziehl–Neelson or Kinyoun method is used How they get sick
to stain the smears. The cells in the smears
can also be stained with a fluorescent rho- The infection is transmitted from some-
damine–auramine dye and viewed with a one with an active case of tuberculosis
fluorescence microscope. In part because who coughs or sneezes, producing respira-
there may be only very few organisms tory droplets that the next victim inhales.
present in the smears, culturing the speci- Some people who have been infected are
mens is more definitive. The cultures can free of symptoms for many years and may
be grown on medium containing egg yolk die without developing active tuberculo-
or oleic acid and albumin. However, the sis. Activation of the disease is associated
bacteria grow very slowly (the cells divide with malnutrition, old age, and a sup-
appproximately once a day), and it may pressed immune system, as in people with
be 3 to 6 weeks before growth is visible. AIDS.
Treatment from a live, attenuated bovine strain of

tubercle bacillus (M. bovis). “Attenuated”
Antituberculosis drugs must be taken for at means that the strain has been grown in cul-
least 6 months. The drugs that can be taken ture for a long time and has lost so much
are isoniazid (Nydrazid), rifampin (Rifadin virulence that it cannot cause tuberculosis.
and Rimactane), ethambutol (Myambutol), BCG is widely used worldwide; in develop-
streptomycin, and pyrazinamide. Usually ing countries it is routinely given to chil-
more than one medication is taken. dren at birth. The WHO has estimated that
85% of children born in 1990 received the
BCG vaccination during the first year of
Tuberculin skin test life. However, there is a great deal of uncer-
tainty about BCG’s efficiency, which has
Tuberculin (called old tuberculin) refers to reportedly ranged from 0% (one study in
material isolated from M. tuberculosis by India) to 90%. The results of studies con-
boiling the cells. A purified protein deriva- ducted in the United Kingdom indicate
tive (PPD) of old tuberculin is used for the protection against infection of approxi-
skin test. It is a mixure of cell wall proteins mately 70% of vaccinated individuals. All
and polysaccharides prepared by frac- tuberculin-negative children in the United
tionation with trichloroacetic acid (TCA), Kingdom receive the BCG vaccination at
ammonium sulfate, and alcohol. The PPD is 10 to 13 years of age.
injected under the skin. If the person being The BCG vaccine is not routinely used in
tested has not been exposed to M. tubercu- the United States. The reasons for this in the
losis, there is no immediate reaction, and a past have been the relatively low incidence
positive skin reaction occurs 3 to 5 weeks of the disease (<1% of children and young
after inoculation. If the individual has been adults give a positive tuberculin skin test)
infected, a delayed-type hypersensitivity and the fact that any person who has been
reaction occurs at the site of injection vaccinated will test positive in the tubercu-
within a couple of days. This is manifested lin test. Under the latter circumstances, it
as an indurated (raised, hard) area that is would not be possible to use the tubercu-
often red (erythematous). The diameter lin test to ascertain the spread of the infec-
of the induration, measured within 2 to 3 tion. Nor would it be possible to ascertain
days, is interpreted according to the follow- whether an individual has become recently
ing list of ranges: infected and therefore should receive pre-
ventive therapy with isoniazid.
0–4 mm = negative
> 10 mm = positive
History
5–10 mm = uncertain; must be retested
For an immunocompromised person (e.g., The people who have died from tuberculo-
an AIDS patient), a 5 mm diameter is con- sis between 1805 and 1967 include many
sidered to be a positive test. who have been very important in music,
writing, and the theater: the English poet
John Keats (died in 1821), the English
BCG vaccine author D. H. Lawrence (1930), the English
actress Vivien Leigh (1967), the English
There exists a vaccine for tuberculosis: the novelist and essayist George Orwell (1950),
bacille Calmette–Guérin (BCG) vaccine, the American playwright Eugene O’Neill
named after the French microbiologists who (1953), the Polish pianist and composer
developed the drug. The vaccine is made Frédéric Chopin (1849), the Scottish novelist
www.ebook777.com
and poet Walter Scott (1832), the Italian Russian playwright and novelist Anton
violinist and composer Nicolò Paganini Pavlovich Chekhov (1904), and the German
(1840), the German poet and playwright poet and playwright Johann Christoph
Johann Wolfgang von Goethe (1832), the Friedrich von Schiller (1805).
BOX 1.5 LEPROSY

Leprosy, a chronic disease caused by the lungs in response to M. tuberculosis
Mycobacterium leprae, can manifest itself infection and are partly attributable to the
in cutaneous lesions, nerve impairment ability of the bacteria to grow inside the
resulting in sensory loss, and tissue destruc- macrophages. The cell-mediated immune
tion. The bacteria infect the skin, periph- response against the bacteria results in
eral nerves, eyes, and mucous membranes, nerve damage due to inflammation. As a
and cartilage, muscle, and bone may be consequence, there is loss of sensation in
destroyed. Humans are the only known the area of the lesions.
source of M. leprae. The bacteria are obli-
Lepromatous leprosy. The skin lesions are
gate intracellular parasites and grow well
anesthetic, extensive, large, nodular, and
in the peripheral nerves, specifically in the
disfiguring. The face becomes disfigured
Schwann cells, which make up the myelin
(lion’s face) owing to thickening, nodules,
sheaths.
and plaques. Eyebrows and eyelashes are
lost. Lepromatous leprosy results when the
immune system fails to mount an effective
Symptoms
cell-mediated defense. It is a progressive
disease that is fatal if not treated. The
There are two forms of leprosy: leproma-
bacteria disseminate throughout the body
tous and tuberculoid, but these are two
and can be found in all the body organs.
extremes, and infected people can display
The bacteria also grow in the mucous
symptoms intermediate between them.
membranes of the nose. The cartilaginous
Tuberculoid leprosy. This is a self-limiting septum is destroyed, resulting in severe nasal
disease that may regress. Skin lesions are deformities, including collapse of the nose.
present but contain very few organisms. There is also damage to underlying cartilage
The lesions, which can occur anywhere and bone. Fingers or toes may be lost.
on the body, are usually one or two anes-
thetic (without sensation), hypopigmented
macules (flat spots) with raised reddish Laboratory diagnosis
or purple edges. The lesions vary in size
from a few millimeters to larger ones that Diagnosis depends upon detecting the
that may cover the entire trunk. There is bacteria in skin smears and skin biopsies.
palpable thickening of peripheral nerves Generally such tests are positive only for
due to the growth of bacteria in nerve lepromatous leprosy. To make a skin smear,
sheaths. Skin samples subjected to biopsy the dermis is cut and tissue fluid is scraped
show tuberculoid granulomas consisting of from the slit and smeared on a microscope
activated macrophages and lymphocytes. slide. The smears are stained for acid-
These resemble the granulomas formed in fast bacilli by using a modification of the
Ziehl–Neelson stain. For the skin biopsy, as contagious as once thought. Spread by
material is taken from the lesion and fixed skin contact would probably require bro-
in 10% formalin and stained for acid-fast ken skin on the recipient for the bacteria to
bacilli. The bacteria cannot be cultured in penetrate.
cell-free cultures (i.e., in vitro). They can
be grown in mouse footpads, however, by
injecting a suspension of bacteria derived Treatment
from a skin biopsy sample. It takes about
6 months for maximum growth to occur. The drug Dapsone, which is related to
Growing the bacteria in mouse footpads the sulfonamides, will successfully treat
allows diagnosticians to test the sensitiv- leprosy. For strains that show some resis-
ity of the infecting organism to therapeutic tance to Dapsone, a combination of drugs,
drugs such as Dapsone and rifampin. M. including Dapsone, rifampin, and clofaz-
leprae can also be similarly cultivated by imine may be used.
inoculation into armadillos.
History
Who gets sick
Leprosy is an ancient and infamous disease
Between 12 million and 13 million people often referred to as Hansen’s disease, after
are infected worldwide, mostly in develop- the Norwegian scientist Gerhard Hansen,
ing countries, especially those in tropical who discovered the causative agent in
or semitropical areas. In the past, how- 1874. The disease apparently started in the
ever, leprosy has not been confined to the Far East prior to 600 BCE and spread to the
warmer climates. It is most prevalent in Near East, Africa, and finally to Europe,
India, where in 1994 the WHO put the where incidence peaked during the Middle
number of estimated cases at 1,167,900. Ages. It was probably introduced into the
In the United States about 100 to 200 new Americas with the early Spanish explora-
cases a year are reported each year to the tions and the slave trade. The first reference
Centers for Disease Control and Prevention to leprosy in the United States was in the
(CDC) (mostly among immigrants). Floridas in 1758. A hospital for the treat-
ment of leprosy was established in New
Orleans in 1785. The Louisiana Leper
How they get sick Home was established in 1894, near the
village of Carville, and in 1921 this facility
Leprosy is thought to be transmitted via became the National Leprosarium, a fed-
nasal secretions and contact with skin eral facility. It now exists as the Gillis W.
lesions, although the disease is not nearly Long Hansen’s Disease Center at Carville.
For the following discussion, refer to galactose, or to trehalose, which is a disac-

Fig. 1.11. The main waxy lipids are branched- charide of D-glucose. If the layer is bound to
chain hydroxy fatty acids called mycolic acids, trehalose, the compound, which is described
also found in Nocardia, Corynebacterium, next, is called cord factor. The arabinogalactan
and Rhodococcus. A major class of lipid, the itself is covalently bonded to acetyl or glycolyl
mycolic acids form a hydrophobic layer on the groups in the peptidoglycan via phosphodiester
external face of the cell wall. This layer can be bonds. There are variations in the size of indi-
esterified to a polysaccharide called arabinoga- vidual mycolic acids among the different spe-
lactan, which is a copolymer of arabinose and cies of Mycobacterium. In M. tuberculosis the
www.ebook777.com
OH
OH O C–CH–CH–C60H120(OH) mycolic acid
O C24H49
CH–CH–COOH
CH2 OH
R2 R1 O
OH trehalose
OH
O O OH
mycolic acid HO
OH O OH
O C–CH–CH–C60H120(OH) mycolic acid
C24H49
cord factor
Fig. 1.11 Chemical structures of mycolic acid and cord factor. Mycolic acid is a β-hydroxy fatty acid with
alkyl groups R1 attached to the α carbon and R2 attached to the β carbon. In Mycobacterium tuberculosis R1
is C24H49 and R2 is C60H120(OH). Cord factor is a glycolipid in the cell walls of M. tuberculosis. It is a disac-
charide of D-glucose [Glc(α1–α1)Glc], called trehalose, to which mycolic acid is attached via ester linkage to
the C6 hydroxyl of the sugar.
mycolic acid is a 3-hydroxy fatty acid with two take the Gram stain unless the wall lipids have
alkyl branches. The first branch (R1) is at C2 been removed with alkaline ethanol.
and is a hydrocarbon side chain of 24 carbons In summary, gram-positive cell walls have
(–C24H49). The second branch (R2) is at C3 and diverse types of neutral and acidic polysaccha-
is a hydroxylated hydrocarbon with 60 carbons rides, glycolipids, lipids, and other compounds
in the chain (–C60H120–OH). either free in the wall or covalently bound to the
In addition to mycolic acid–arabinogalactan, peptidoglycan. The functions of most of these
the M. tuberculosis cell wall has a mycolic acid– polymers are largely unknown, although some
containing cell surface glycolipid called cord may act as adhesins and/or presumably affect
factor (Fig. 1.11). Cord factor consists of the the permeability characteristics of the cell wall,
glucose disaccharide trehalose, to which are whereas others, such as mycolic acid deriva-
covalently bonded two molecules of mycolic tives in M. tuberculosis, contribute to resistance
acid. The characteristic M. tuberculosis colony properties and virulence.
appearance of long, intertwining cords, due
to the side-by-side interactions of long chains Gram-negative wall
of cells forming serpentine, ropelike rods, is The gram-negative cell wall is structurally and
generally attributed to cord factor—hence the chemically complex. It consists of an outer
name. As reviewed in ref. 87, cord factor is also membrane composed of lipopolysaccharide,
found in noncording mycobacteria. It has been phospholipid, and protein, and an underlying
suggested that cord factor is responsible for the peptidoglycan layer (Fig. 1.12). Between the
wasting, fever, and lung damage symptomatic outer and inner membrane (the cell membrane)
of tuberculosis. Cord factor is required for viru- is a compartment called the periplasm, wherein
lence, and it is lacking in avirulent mycobacte- the peptidoglycan lies.
ria. When injected intravenously into rabbits,
cord factor causes weight loss and granulomas 1. Lipopolysaccharide structure
in the liver and lungs.88 and function
When mycobacteria are appropriately Lipopolysaccharide (LPS) consists of three
stained, the dye (basic fuschin) is not removed regions: lipid A, core, and a repeating oligosac-
by dilute hydrochloric acid in ethanol (acid charide, sometimes called o-antigen or somatic
alcohol) because of the presence of the waxy antigen. The chemical structure and synthe-
lipids. Therefore these microorganisms are sis of LPS are described in Section 12.2. In the
called acid-fast bacteria. Acid-fast staining is an Enterobacteriaceae (e.g., E. coli), the lipopoly-
important diagnostic feature for the identifica- saccharide is confined to the outer leaflet of the
tion of Mycobacterium in clinical specimens, outer membrane and is arranged so that the
such as sputum. Acid-fast bacteria also do not lipid A portion is embedded in the membrane as
Fig. 1.12 Schematic drawing of the gram-negative envelope. The outer membrane consists of lipopolysac-
charide, phospholipid, and proteins (most of which are porins). Underneath the outer membrane is the pep-
tidoglycan layer, which is noncovalently bonded to the outer membrane via murein lipoproteins, the lipid
portion of which is integrated into the inner leaflet of the outer membrane, and the protein portion is cova-
lently attached to the peptidoglycan. The cell membrane is composed of phospholipid and protein. The area
between the outer membrane and the cell membrane is called the periplasm. The wavy lines are fatty acid
residues, which anchor the phospholipids and lipid A into the membrane. Abbreviations: LPS, lipopolysac-
charide; O, oligosaccharide; C, core; A, lipid A; P, porin; PL, phospholipid; MLP, murein lipoprotein; pg,
peptidoglycan; Pr, protein; om, outer membrane; cm, cell membrane.
part of the lipid layer, and the core and oligosac- the enteric bacteria to have such a permeability
charide extend into the medium (Fig. 1.12). barrier because they live in the presence of bile
Mutants of E. coli that lack the oligosac- salts in the intestine. In fact, a common basis
charide experience loss of virulence, and it is for selective media for gram-negative bacteria
believed that the LPS can increase pathoge- is the resistance of these bacteria to bile salts
nicity. The lipid A portion of the LPS, which and/or hydrophobic dyes. The bile salts and
can have toxic effects when released from dyes inhibit the growth of gram-positive bac-
bacterial cells, is described as an endotoxin as teria but, because of the LPS, not gram-nega-
explained in note 89. Loss of most of the core tive bacteria. An example of such a selective
and oligosaccharide in E. coli and related bac- medium is eosin–methylene blue (EMB) agar,
teria is associated with increased sensitivity which is used for the isolation of gram-nega-
to hydrophobic compounds (e.g., antibiotics, tive bacteria because it inhibits the growth of
bile salts, and hydrophobic dyes such as eosin gram-positive bacteria. As explained in note
and methylene blue). This is because the LPS 90 and summarized in Section 17.5, the pres-
provides a permeability barrier to hydropho- ence of drug efflux pumps is another impor-
bic compounds. (A model for how this might tant reason that bacteria are resistant to many
occur is described later.) It is advantageous for antibiotics.
www.ebook777.com
Apparently the outer membrane has low membrane is protein, of which there are several
permeability to hydrophobic compounds: the different kinds. One of the proteins is called the
phospholipids are confined primarily to the murein lipoprotein. This is a small protein with
inner leaflet of the outer envelope, whereas lipid attached to the amino-terminal end (Figs.
the LPS layer, which is a permeability bar- 1.13 and 1.14). The lipid end of the molecule
rier to hydrophobic substances, is in the outer extends into and binds hydrophobically with
leaflet. Since lipid A contains only saturated the lipids in the outer envelope. The protein end
fatty acids, the LPS presents a somewhat rigid of some of the molecules is covalently bound
matrix, and it has been suggested that this to the peptidoglycan, thus anchoring the outer
property, plus the tendency of the large LPS envelope to the peptidoglycan. In E. coli, about
molecules to engage in lateral noncovalent one-third of the murein lipoprotein is bound to
interactions, makes it difficult for hydropho- the peptidoglycan. Mutants unable to synthe-
bic molecules to penetrate between the LPS size the murein lipoprotein have unstable outer
molecules to the phospholipid layer. Mutants envelopes that bleb off into the medium at the
that lack a major region of the oligosaccharide cell poles and septation sites. Therefore, the
and core are more permeable to hydrophobic murein lipoprotein may play a structural role in
compounds because there is more phospho- keeping the outer membrane attached to the cell
lipid in the outer leaflet of these mutants; it is surface.
likely, as well, that there is less lateral interac- There are a small number of other outer mem-
tion among these incomplete LPS molecules. brane or cell membrane lipoproteins.91 These
The asymmetric distribution of phospholipid were discovered by chemically cross-linking the
to the inner leaflet of the outer envelope seems peptidoglycan in whole cells to closely associ-
to be an adaptive evolutionary response to ated proteins with a bifunctional cross-linking
hydrophobic toxic substances in the environ- reagent. (Note 92 explains how cross-linking
ment, such as the intestine of animals. Not all reagents work.) There is no evidence that these
gram-negative bacteria have an outer envelope additional lipoproteins are covalently bonded
with an asymmetric distribution of lipopoly- to the peptidoglycan, and in most cases their
saccharide and phospholipids, but many non- functions remain to be elucidated.
enterics as well as enterics do.
3. Porins and other proteins
2. Lipoproteins The major proteins in the outer envelope are
In addition to lipopolysaccharide and phos- called porins. The porins form small nonspe-
pholipid, a major component of the outer cific hydrophilic channels through the outer
Fig. 1.13 Murein lipoprotein. Attached to the amino-terminal cysteine is a diacylglyceride in thioether link-
age and a fatty acid in amide linkage. The lipid portion extends into the outer envelope and binds hydrophobi-
cally with the fatty acids in the phospholipids and lipopolysaccharide. The carboxy-terminal amino acid is
lysine, which can be attached via an amide bond to the carboxyl group of diaminopimelic acid (DAP) in the
peptidoglycan. The murein lipoprotein therefore holds the outer envelope to the peptidoglycan.
Fig. 1.14 Pseudopeptidoglycan as found in Archaea. Pseudopeptidoglycan resembles peptidoglycan in being

a cross-linked glycopeptide. It differs from peptidoglycan in the following ways: (1) N-acetyltalosaminuronic
acid replaces N-acetylmuramic acid. (2) The glycosidic linkage is β-1,3 instead of β-1,4. (3) There are no
D-amino acids. Abbreviations: G, N-acetylglucosamine; T, N-acetyltalosaminuronic acid. The peptide sub-
unit is enclosed in dashed lines.
envelope, allowing the diffusion of low molecu- bacteria, although they are not all identical. As
lar weight (<600 Da) neutral and charged sol- mentioned earlier, E. coli regulates the amounts
utes, such as sugars and ions. The channels are of the various porins according to growth con-
necessary to allow passage of small molecules ditions. This is discussed in Chapter 18.
into and out of the cell. E. coli has three major Since the porins exclude molecules with
porins: OmpF, OmpC, and PhoE. Each porin molecular weights larger than 600 Da, one
makes a separate channel. Thus, there are would expect to find other proteins in the outer
OmpF, OmpC, and PhoE channels. membrane that facilitate the translocation of
The OmpC channel is approximately 7% larger solutes across the outer membrane. This
smaller than the OmpF channel and is expected is the case. E. coli has an outer membrane pro-
to make the outer envelope less permeable to tein called the LamB protein that forms channels
larger molecules. Both OmpF and OmpC are for the disaccharide maltose and for maltodex-
present under all growth conditions, although trins. Other examples in the outer membrane
the ratio of the smaller OmpC to the larger of E. coli facilitate the transport of vitamin B12
OmpF increases in high-osmolarity media and (BtuB), nucleosides (Tsx), and several other
at high temperature. The increased amounts of solutes.
OmpC relative to OmpF presumably also occur
in the intestine, where osmolarity and tempera- Archaeal cell walls
ture are higher than in lakes and streams that Archaeal cell walls are not all alike and are very
also harbor E. coli. This may confer an advan- different from bacterial cell walls. For example,
tage to the enterics because the smaller OmpC no archaeal cell wall contains peptidoglycan.
channel should present a diffusion barrier to Archaeal cell walls may be either pseudopeptido-
toxic substances in the intestine, whereas the glycan, polysaccharide, or protein (the S layer).
larger OmpF channel should be advantageous Pseudopeptidoglycan (also called pseudo-
in more dilute environments outside the body. murein) resembles peptidoglycan in consist-
The protein PhoE is produced only under ing of glycan chains cross-linked by peptides
conditions of inorganic phosphate limitation. (Fig. 1.14). However, the resemblance stops
This is because PhoE is a channel for phosphate here. In pseudopeptidoglycan, although one of
(and other anions) whose synthesis under lim- the sugars is N-acetylglucosamine as in pepti-
iting phosphate conditions reflects the need doglycan, the other is N-acetyltalosaminuronic
to bring more phosphate into the cell. Porins acid instead of N-acetylmuramic acid.
appear to be widespread among gram-negative Furthermore, the sugars are linked by a β-1,3
www.ebook777.com
glycosidic linkage rather than a β-1,4, and the The literature discusses whether the periplasm
amino acids in the peptides are L-amino acids has zones of adhesion (called Bayer’s patches)
rather than D-amino acids. (The latter are pres- between the inner and outer membranes.94,95
ent in peptidoglycan; compare Figs. 1.14 and Whether zones of adhesion are seen depends
12.2.) Thus, pseudopeptidoglycan is very dif- upon how the cells are prepared for electron
ferent from peptidoglycan. microscopy. One school of thought holds that
the zones are artifacts of fixation; the counter-
1.2.4 Periplasm argument is that the zones are seen only if the
Lying between the cell membrane and the proper techniques are employed to preserve the
outer membrane of gram-negative bacteria is rather fragile adhesion sites. This is an impor-
a separate compartment called the periplasm93 tant issue because the zones of adhesion have
(Fig. 1.15). It appears as a space in electron been postulated to be sites of the translocation
micrographs of thin sections of cells but should apparatuses that export lipopolysacharide,
be considered to be an aqueous compartment polysaccharide (capsule), and protein through
containing protein, oligosaccharides, salts, the inner and outer membranes.
and the peptidoglycan. It seems that the pep- The periplasm should be considered to be a
tidoglycan and oligosaccharides may exist in a cellular compartment with specialized activities.
hydrated state, forming a periplasmic gel.94 These activities include oxidation–reduction
Fig. 1.15 A model of the periplasm in E. coli. The outer region of the periplasm is thought to consist of
cross-linked peptidoglycan attached to the outer envelope via lipoprotein (LP) covalently bound to the pep-
tidoglycan. The inner region of the periplasm (approaching the cell membrane) is believed to consist of less
cross-linked peptidoglycan chains and oligosaccharides that are hydrated and form a gel. The gel phase is
thought to contain periplasmic proteins (e.g., A and B). Thus A might be a periplasmic enzyme and B a solute-
binding protein that interacts with a membrane transporter (C); D and E are integral membrane proteins, per-
haps enzymes. The outer envelope is depicted as consisting of lipopolysaccharide, porins, and specific solute
transporters (receptors), which require a second protein (TonB) for uptake. Source: Ferguson, S. J. 1991. The
periplasm, pp. 311–339. In: Prokaryotic Structure and Function, A New Perspective. S. Mohan, C. Dow, and
J. A. Coles (Eds.). Cambridge University Press, Cambridge.
reactions (Chapters 5 and 13), osmotic regu- carried into the cell via specific inorganic phos-
lation (Chapter 16), solute transport (Chapter phate transporters.
17), protein secretion (Chapter 18), and hydro-
5. Detoxifying agents
lytic activities such as those mediated by phos-
Some periplasmic enzymes are detoxifying
phatases and nucleases. The phosphatases
agents. For example, the enzyme to degrade pen-
and nucleases degrade organophosphates and
icillin (β-lactamase) is a periplasmic protein.
nucleic acids that might enter the periplasm
from the medium and transport the hydrolytic 6. TonB protein96
products into the cell. An interesting periplasmic protein anchored to
the cell membrane in E. coli is the TonB pro-
Periplasmic components tein. (See Fig. 1.15.) It is known that the pro-
The periplasm is chemically complex and car- tein is required for the uptake of several solutes
ries out diverse functions. The following list of that do not diffuse through the porins; rather,
components and their functions, which reflects they require specific transport systems (also
the importance of the periplasm, is simply a called receptors or TonB-dependent trans-
partial inventory emphasizing the periplasmic porters, or TBDTs) in the outer envelope. All
functions about which most is known. these solutes have molecular weights larger
than 600 Da, which is the upper limit for mol-
1. Oligosaccharides
ecules that enter via the porins. Examples of
The oligosaccharides in the periplasm are
solutes with specific outer membrane receptors
thought by some to be involved in osmotic reg-
that require TonB for uptake are iron sidero-
ulation of the periplasm because their amounts
phores and vitamin B12 (cobalamin). (For an
decrease when the cells are grown in media of
explanation of iron siderophores, see note 97.)
high osmolarity. This complex subject is dis-
Interestingly, these solutes are brought into
cussed more fully in Section 16.2.
the periplasm against a large concentration
2. Solute-binding proteins gradient, sometimes 103 times higher than the
Solute-binding proteins in the periplasm assist concentration outside the cell. In a way that
in solute transport by binding to solutes (e.g., is not understood, the TonB protein couples
sugars and amino acids that have entered the the electrochemical energy (the proton motive
periplasm through the outer envelope) and force, Δp) in the cell membrane to the uptake
delivering the solutes to specific transporters of certain solutes through the outer envelope
(carriers) in the cell membrane. and into the periplasm.98-101 TonB is thought
to be an energy transducer. One suggestion
3. Cytochromes c
is that TonB is energized by the electrochemi-
Some of the enzymes in the periplasm are cyto-
cal potential that exists across bacterial cell
chromes c that oxidize carbon compounds or
membranes and that in the energized state,
inorganic compounds and deliver the electrons
TonB causes a conformational change in the
to the electron transport chain in the cell mem-
outer membrane receptor protein that results
brane. These oxidations are called periplasmic
in translocation of the solute (e.g., vitamin B12)
oxidations. There are other oxidoreductases
or ligand-bound solute (e.g., iron–siderophore
in the periplasm as well, but the various cyto-
complexes) through the receptor channel into
chromes c are very common.
the periplasm.101 Accessory proteins in the cell
4. Hydrolytic enzymes membrane, which in E. coli are called ExbB
Hydrolytic enzymes in the periplasm degrade and ExbD, interact with TonB and may use
nutrients to smaller molecules that can be trans- the proton motive force Δp (i.e., uptake of
ported across the cell membrane by specific H+ through ExbB/D) to convert TonB into an
transporters. For example, the enzyme amylase energized conformation that somehow drives
is a periplasmic enzyme that degrades oligo- the uptake of material through transport-
saccharides to simple sugars. Another exam- ers in the outer membrane. This concept was
ple is alkaline phosphatase, which removes reviewed in 2003, in a paper that also discusses
phosphate from simple organic phosphate two experimentally based models designed to
monoesters. The inorganic phosphate is then explain the mechanism by which TonB acts as
www.ebook777.com
an energy transducer.101 One model proposes 1.2.5 Cell membrane

that TonB in an energized conformation leaves We now come to what is certainly the most
the ExbB/D complex in the cell membrane and functionally complex of the cell structures, the
moves to the transporter in the outer mem- cell membrane. The cell membrane is responsi-
brane. It then returns to the ExbB/D complex ble for a broad range of physiological activities
to be reenergized. This has been called the including solute transport, electron transport,
“shuttle” model. An alternative model pro- photosynthetic electron transport, the estab-
poses that one part of TonB remains associated lishment of electrochemical gradients, ATP
with the ExbB/D complex in the cell membrane synthesis, biosynthesis of lipids, biosynthesis
and one part reaches across to the transporter of cell wall polymers, secretion of proteins, the
in the outer membrane. secretion and uptake of intercellular signals,
and responses to environmental signals. To
Is there a periplasm in gram-positive refer to the cell membrane simply as a lipopro-
bacteria? tein bilayer does not do justice to the machinery
In the past it has been assumed that gram- embedded in the lipid matrix, a complex mosaic
positive bacteria lack a space between the cell of parts whose structure and interactions at the
membrane and the cell wall equivalent to the molecular level are not well understood.
periplasm of gram-negative bacteria. This is As expected, the protein composition of cell
because thin sections of gram-positive cells do membranes is complex. There can be more than
not indicate that such a space exists, and of 100 different proteins. Many of the proteins
course gram-positive bacteria do not have an are clustered in functional aggregates (e.g.,
outer envelope. Evidence suggests, however, the proton translocating ATPase, the flagella
that perhaps gram-positive bacteria do have a motor, electron transport complexes, certain
compartment analogous to the periplasm found of the solute transporters). At the molecular
in gram-negative bacteria.102,103 level, the membrane is certainly a complex and
The evidence in favor of the existence of busy place. What follows is a general descrip-
a periplasm in Bacillus subtilis includes the tion of the membrane, without reference to its
release of putative periplasmic proteins after microheterogeneity.
protoplasts of the cells have been made with
lysozyme. (For a definition of protoplasts and Bacterial cell membranes
how they are stabilized, read note 104.) Proteins Bacterial cell membranes consist primarily of
solubilized by removing the cell wall during pro- phospholipids and protein in a fluid mosaic
toplast formation may include proteins resident structure in which the phosphlipids form a
in the area between the cell membrane and the bilayer (Fig. 1.16). The structure is said to be
cell wall. The proteins released by protoplast fluid because there is extensive lateral mobility of
formation include nucleases, which are distinct bulk proteins and phospholipids. Nevertheless,
from cytoplasmic nucleases. certain protein aggregates (e.g., complex solute
Additional evidence has been obtained by transporters and electron transport aggregates)
using ultrarapid freezing and electron micros- remain as aggregates within which the proteins
copy (cryo–transmission electron microscopy, interact to catalyze sequential reactions.
or cryo-TEM) of frozen–hydrated thin sections
of B. subtilis. When this was done, the area out- 1. The lipids
side the cell membrane was seen to be bipartite, The phospholipids are fatty acids esterified to
consisting of a low-density 22 nm region sur- two of the hydroxyl groups of phosphoglycer-
rounded by a 33 mm high-density outer wall ides (Fig. 1.17). The structure and synthesis of
zone, which is thought to consist of peptido- phospholipids is described in detail in Section
glycan, teichoic acid, and cell wall proteins.105 10.1.2. The third hydroxyl group in the glyc-
Further studies must be undertaken to elucidate erol backbone of the phospholipid is covalently
the exact nature of this putative periplasm, bound to a substituted phosphate group, which
including its contents and the relationships of makes one end of the molecule very polar owing
the proteins found therein with the cell mem- to a negative charge on the ionized phosphate
brane and cell wall, as well as the similarities to group. Because the phospholipids are polar at
the gram-negative periplasm. one end and nonpolar at the other end (the end
Fig. 1.17 Phospholipids have both a polar and a non-

polar end. (A) Phospholipid with two fatty acids (R)
esterified to glycerol. The phosphate is conjugated
to X, which determines the type of phospholipid. In
Fig. 1.16 Model of the cell membrane showing bacteria, X is usually serine, ethanolamine, a deriva-
bimolecular lipid leaflets and embedded proteins; tive of glycerol, or a carbohydrate derivative. See
the phospholipid molecules are interacting with one Section 10.1.3 for a more complete description of
another via their hydrophobic (apolar) “tails.” The bacterial phospholipids. (B) Schematic drawing of a
hydrophilic (polar) “heads” of the phospholipids face phospholipid showing the polar (circle) and nonpo-
the outside of the membrane, where they interact with lar (straight lines) regions.
proteins and ions. Proteins can span the membrane
or be partially embedded. Source: Singer, S. J., and
G. L. Nicolson. 1972. The fluid mosaic model of the
structure of cell membranes. Science 175:720–731. across the membrane through or on special pro-
Copyright 1972 by the Association of Academies of tein transporters that bridge the phospholipid
Science. Reprinted with permission from AAAS. bilayer. These modes are discussed in the context
of solute transport in Chapter 17. (However,
the lipid bilayer is permeable to water mole-
cules, gases, and small hydrophobic molecules.)
with the fatty acids), they are said to be amphi-
An important consequence of the lipid matrix is
pathic, able to spontaneously aggregate while
that ions do not freely diffuse across the mem-
their nonpolar fatty acid regions interact with
brane unless they are carried on or through
each other by hydrophobic bonding; their polar
protein transporters. Because of this, the mem-
phosphorylated regions, on the other hand, face
brane is capable of holding a charge that is due
the aqueous phase, where ionic interactions
to the unequal transmembrane distribution of
occur with cations, water, and polar groups on
ions. This is discussed in Chapter 5.
proteins. Phospholipids accomplish all this by
spontaneously forming lipid bilayers in water 4. Aquaporins (water channels)
solutions or in cell membranes. Although the lipid bilayer allows rapid equili-
bration of water, there do exist in E. coli and
2. The proteins
other bacteria water channels that are similar
There are two classes of proteins in membranes,
to the aquaporins found in eukaryotes; these
integral and peripheral. Integral proteins are
water channels, called aquaporins, enhance
embedded in the membrane and bound to the
the rapid equilibration of water across the cell
fatty acids of the phospholipids via hydrophobic
membrane. (Reviewed in ref. 105.) The gene
bonding. They can be removed only with deter-
coding for the water channel protein in E.
gents or solvents. Peripheral proteins, attached
coli is aqpZ. The expression of the aqpZ gene
at membrane surfaces to the phospholipids by
under different extracellular osmolarity con-
ionic interactions, can be removed by washing
ditions and its requirement for viability have
the membrane with salt solutions. The insertion
been investigated.106 Null mutants of aqpZ
of the proteins into the membrane during mem-
are viable, although the colonies are smaller
brane synthesis is discussed in Section 18.2.
than the wild-type strain. Interestingly, when
3. Permeability E. coli is grown in media of high osmolarity,
The phospholipid bilayer acts as a permeabil- the synthesis of the aquaporin channels is
ity barrier to virtually all water-soluble mol- repressed. This may help protect the cell from
ecules. Thus most solutes diffuse or are carried hypo-osmotic stress in the event of a sharp
www.ebook777.com
decrease in the osmolarity of the external facilitating the uptake of organic osmolytes such
medium. as proline and betaine. Recently it was reported
that ProP localizes to the membrane at the poles
5. Mechanosensitive (MS) channels
of cells because of the presence there of a lipid
For reviews of mechanosensitive channels,
called cardiolipin, which is more highly concen-
read refs. 105 and 107 through 109. As will be
trated in the cytoplasmic membrane near the
explained in Section 16.2, bacteria adjust the
cell poles.111 This is very interesting because it
internal osmotic pressure so that it is always
points to an important role of certain lipids in
higher than the external osmotic pressure. This
localizing membrane proteins to specific mem-
keeps water flowing into the cell via osmosis and
brane locations.
maintains the high internal turgor pressure that
is important for growth. The internal osmotic
Archaeal cell membranes
pressure is kept high by the accumulation of cer-
1. The lipids
tain solutes such as K+, glutamate, glutamine,
Archaeal membrane lipids differ from those
proline, trehalose, and betaine.
found in bacterial membranes.112-114 The
One consequence of maintaining a high inter-
archaeal lipids consist of isoprenoid alcohols
nal osmotic pressure is that a sudden decrease
(either 20 or 40 carbons long), linked via ether
in the external osmotic pressure can endanger
bonds, in contrast to the ester linkages in eubac-
the cell by promoting a sudden increase in the
teria, to one glycerol to form monoglycerol
influx of water, leading to overexpansion of the
diethers or to two glycerols to form diglycerol
cell wall and subsequent lysis of the cell. This
tetraethers. These are illustrated in Fig. 1.18.
is sometimes referred to as hypo-osmotic stress
The synthesis of these lipids is described later,
or hypo-osmotic shock. The cells are protected
in Section 10.1.3. (Recall that bacterial glyc-
from this form of destruction by mechanosensi-
erides are fatty acids esterified to glycerol.
tive (MS) channels that open under conditions
Refer to Fig. 1.16 for a comparison.) The C20
of hypo-osmotic stress and provide a means for
alcohol is a fully saturated hydrocarbon called
internal solutes to rapidly exit the cell, thus low-
phytanol. The C40 molecules are two phytanols
ering the internal osmotic pressure. (See note
linked together head to head in the diglycerol
110 for an explanation.) Mutants that do not
tetraether lipids. Thus, the lipids are either
have MS channels lyse when subjected to hypo-
phytanylglycerol diethers or diphytanyl diglyc-
osmotic stress.
erol tetraethers. The diethers and tetraethers
Mechanosensitive channels are present in
occur in varying ratios depending upon the bac-
most bacteria as well as archaea. E. coli has
terium. For example, there may be from 5 to 25
three such channels: a large channel called
different lipids in any one cell. This is really quite
MscL, a small channel called MscS, and a
a diverse mix, and it can be contrasted to the
“mini” channel called MscM (L refers to large,
lipid complement of a typical bacterium, which
S to small, and M to mini).
has only four or five different phospholipids.
6. Osmosensory transporters The diversity of the archaeal lipids is due to the
As explained in Chapter 16, bacteria adapt different polar head groups that exist, as well
to high-osmolarity media by increasing the as to the mix of core lipids to which the head
intracellular concentrations of certain solutes groups are attached (Fig. 1.18). Although the
called osmolytes. This raises the cytoplasmic polar head group is responsible for the polarity
osmolarity so that water does not rush out of of most phospholipids, there is some polarity at
the cell into the more concentrated solution. As one end of the archaeal lipids without a polar
stated in Section 16.2.3 (Turgor pressure and head group because of the free hydroxyl group
its importance for growth), it is very important on the glycerol. (Recall that hydroxyl groups are
for bacteria to maintain an internal osmolar- capable of forming hydrogen bonds with water
ity higher than the external osmolarity so that and proteins.) It is usually stated that the ether
turgor pressure within the cell is maintained. linkages, which are more stable to hydrolytic
E. coli has an osmosensory transporter called cleavage, are an advantage over ester linkages
ProP; this membrane protein senses increas- in the acidic and thermophilic environments in
ing osmolarity in the medium and responds by which many archaea live.
Fig. 1.18 Major lipids of Methanobacterium thermoautotrophicum: (A) glycerol diether (archaeol),
(B) diglycerol tetraether (caldarchaeol), (C) a glycolipid (gentiobiosyl archaeol), (E) a phospholipid (archaeti-
dyl–X), where X can be inositol, serine, or ethanolamine, (F) a phospholipid (caldarchaetidyl–X), (G) a
phosphoglycolipid (gentiobiosyl caldarchatidyl–X). Source: Nishihara, M., H. Morii, and Y. Koga. 1989.
Heptads of polar ether lipids of an archaebacterium, Methanobacterium thermoautotrophicum: structure
and biosynthetic relationship. Biochemistry 28:95–102. Reprinted with permission from Nishihara, M., H.
Morii, and Y. Koga. Copyright 1989 American Chemical Society.
www.ebook777.com
2. The proteins generally believed to be derived from invagi-

There is little information regarding archaeal nations of chemically modified areas of the
membrane proteins. It is known, however, that cell membrane. Connections to the cell mem-
in bacteriorhodopsin and halorhodopsin in brane are not always seen, however, and it is
Halobacterium, the conformational array in the unknown whether the intracytoplasmic mem-
cell membrane is dependent upon interaction branes are derived from an invagination of the
with polar membrane lipids.113 The functions cell membrane or are synthesized independently
of these two proteins are discussed in Sections of the cell membrane (e.g., the thylakoids of
4.8.4 and 4.9. cyanobacteria). A few prokaryotes with intra-
cytoplasmic membranes and their physiological
3. The membrane
roles are listed.
The thermoacidophilic archaea and some
methanogens have tetraether glycerolipids in 1. Methanotrophs
the cell membrane. These lipids have a polar Bacteria that grow on methane as their sole
head group at both ends and span the mem- source of carbon (methanotrophs) possess
brane, forming a lipid monolayer (Fig. 1.19). intracytoplasmic membranes that are suggested
This is the only known example of a membrane to function in methane oxidation. Methane oxi-
having no midplane region. Since there is no dation is discussed later, in Section 14.2.1.
midplane region, the lipid monolayer is more
2. Nitrogen fixers
resistant to levels of heat that would disrupt the
Bacteria for which nitrogen gas serves as a source
hydrophobic bonds holding the two lipids in
of nitrogen use an oxygen-sensitive enzyme
the lipid bilayer together. The increased resis-
called nitrogenase to reduce the nitrogen to
tance to heat of the lipid monolayer may confer
ammonia, which is subsequently incorporated
an advantage to organisms living at high tem-
into cell material. Many of these organisms have
peratures. However, it cannot be claimed that
extensive intracytoplasmic membranes. One
diether lipids or tetraether lipids are a specific
such nitrogen-fixing bacterium is Azotobacter
adaptation to high temperatures, although they
vinelandii, whose intracytoplasmic membranes
may be advantageous in these environments.
increase with the degree of aeration of the
This is because some mesophilic methanogens
have tetraether lipids, whereas two extremely
thermophilic archaea, Methanopyrus kandleri
and Thermococcus celer, do not have tetraether
lipids.
1.2.6 Cytoplasm
The cytoplasm is defined as everything enclosed
by the cell membrane. Cytoplasm is a viscous
material containing a heavy concentration of
protein (100–300 mg/mL),115 salts, and metab-
olites. In addition, there are large aggregates of
protein complexes designed for specific meta-
bolic functions, various inclusions, and highly
condensed DNA. Intracytoplasmic membranes
are also present in many prokaryotes. The
soluble part of the cytoplasm is called the cyto-
sol. We will begin with the intracytoplasmic Fig. 1.19 Lipid layer with membrane proteins (shaded
areas) in archaebacteria membranes. The glycerol
membranes.
diethers form a lipid bilayer, and the tetraethers
form a monolayer. Some archaea (e.g., the extreme
Intracytoplasmic membranes halophiles) contain only the diethers. Most of the
Many prokaryotes have intracytoplasmic sulfur-dependent thermophiles have primarily the
membranes that have specialized physiological tetraethers, with only trace amounts of the diethers.
functions.116 Intracytoplasmic membranes are Many methanogens have significant amounts of both
often connected to the cell membrane and are the di- and tetraethers.
culture. Since respiratory activity is localized in organisms to float in lakes and ponds at depths
the membranes, it is probable that an important that support growth because of favorable light,
role for Azotobacter intracytoplasmic mem- temperature, or nutrients. For example, the
branes is to increase the cellular respiratory green sulfur bacterium Pelodictyon phaeoclath-
activity, to provide more ATP for nitrogen fixa- ratiforme forms gas vesicles only at low light
tion and to remove oxygen from the vicinity of intensities.117 Perhaps this allows the bacteria
the nitrogenase. Nitrogen fixation is discussed to float at depths where the light is optimal for
in Section 13.3. photosynthesis. Many bacteria and cyanobac-
teria with gas vesicles are plentiful in stratified
3. Nitrifiers
freshwater lakes, but they are not as abundant
Intracellular membranes are also found in nitri-
in isothermally mixed waters. Other prokary-
fying bacteria (i.e., bacteria that oxidize ammo-
otes containing gas vesicles (e.g., the halophilic
nia and nitrite as the sole source of electrons:
archaeon Halobacterium) live in hypersaline
Nitrosomonas, Nitrobacter, Nitrococcus).
waters, and a few marine species of cyanobacte-
Several of the enzymes that catalyze ammonia
ria belonging to the genus Trichodesmium have
and nitrite oxidation are in the membranes.
gas vesicles. When gas vesicles are collapsed by
This is discussed in Section 13.4.
experimentally subjecting cells to high hydro-
4. Phototrophs static pressure or turgor pressure, the cells are
In bacteria that use light as a source of energy no longer buoyant and sink. Collapsed vesicles
for growth (phototrophs), the intracytoplasmic do not recover, and the cells acquire gas-filled
membranes are the sites of the photosynthetic vesicles only by de novo synthesis of new vesi-
apparatus. The membrane structure varies: flat cles. Thus, during synthesis of the vesicles, water
membranes, vesicles, flat sacs (thylakoids in is somehow excluded, presumably because of
cyanobacteria), and tubular invaginations of the hydrophobic nature of the inner protein
the cell membrane (photosynthetic bacteria). surface.
See the discussion of photosynthesis and pho-
tosynthetic membranes in Chapter 6, especially 2. Carboxysomes
Fig. 6.16. Bacteria that obligately grow on CO2 as their
sole or major source of carbon (strict autotro-
Inclusion bodies, multienzyme aggregates, phs) sometimes have large (100 nm) polyhedral
and granules protein-walled microcompartments called car-
boxysomes.118, 119 These inclusions have been
Certain bacteria contain specialized compart-
observed in nitrifying bacteria, sulfur oxidizers,
ments in the cytoplasm. Some researchers
and cyanobacteria. The distribution appears to
refer to these entities as inclusion bodies. They
be species specific (e.g., not all sulfur oxidizers
are not surrounded by a lipid bilayer–protein
have carboxysomes). However, it should be
membrane as are organelles in eukaryotic cells,
pointed out that most oceanic microorganisms
although they can have a membrane or coat. In
addition, there are numerous large aggregates that fix carbon dioxide, including all cyanobac-
teria, do so with carboxysomes. In these organ-
and multienzyme complexes in all bacteria.
isms ribulose-1,5-bisphosphate carboxylase
1. Gas vesicles (RuBP carboxylase), the enzyme in the Calvin
Aquatic bacteria such as cyanobacteria, certain cycle that incorporates CO2 into organic car-
photosynthetic bacteria, some nonphotosyn- bon, is stored in carboxysomes. The enzyme
thetic bacteria, and certain archaea have gas is discussed in Section 14.1.1. Although many
vesicles surrounded by a simple protein coat autotrophs lack carboxysomes, it has been
consisting primarily of gas vesicle protein A suggested that an advantage to having them is
(GsvA), a small hydrophobic protein that is that the RuBP carboxylase is sequestered there,
highly conserved among the diverse groups of where the concentration of CO2 is kept high.
organisms. Gas vesicles are hollow, spindle- This may be due to the carbonic anhydrase
shaped structures about 100 nm long, filled associated with the shell of the carboxysome.
with gas in equilibrium with the gases dissolved The carbonic anhydrase catalyzes the conver-
in the cytoplasm. The gas vesicles allow the sion of HCO–3 to CO2.120
www.ebook777.com
3. Chlorosomes Magnetosomes should be thought of as a

Green sulfur photosynthetic bacteria (e.g., navigational device or a magnetic compass that
Chlorobium) have ellipsoid inclusions called orients the bacteria with the Earth’s magnetic
chlorosomes (formerly called chlorobium field so that they swim in a particular direction,
vesicles) that lie immediately underneath the a behavior called magnetotaxis. In the Northern
cytoplasmic membrane. The chlorosomes are Hemisphere the geomagnetic north points
surrounded by a nonunit membrane of galac- down at an angle, and magnetotactic bacteria in
tolipid, with perhaps some protein. At one time the Northern Hemisphere that swim toward the
it was believed that such vesicles were found geomagnetic north swim deeper into the water.
only in the green sulfur photosynthetic bacteria. In the Southern Hemisphere the geomagnetic
However, similar vesicles have been found in the north points up at an angle, and magnetotactic
green photosynthetic bacterium Chloroflexus. bacteria in the Southern Hemisphere are “south
The major light-harvesting photopigments are seeking.” (At the equator, north-seeking and
located in the chlorosomes, whereas the photo- south-seeking magnetotactic bacteria coexist.)
synthetic reaction centers are in the cell mem- One way to demonstrate this in the laboratory
brane. This means that during photosynthesis in is to place a small drop of water on a microscope
these organisms, light is absorbed by pigments coverslip and place the south pole of a small bar
in the chlorosomes and energy is transmitted to magnet 1–2 cm away from the drop. The south
the reaction centers in the cell membrane, where pole of the magnet corresponds to geomagnetic
photosynthesis takes place. In photosynthetic north. It is clear that swimming is important for
bacteria that do not have chlorosomes, the this to occur, because dead cells are not pulled by
light-harvesting pigments surround the reac- a magnetic field. A generally accepted model is
tion centers in the cell membrane. The structure that the earth’s magnetic field is such that when
and function of chlorosomes are discussed in the bacteria are swimming, the magnetosomes
Section 6.6. allow orientation in the magnetic field and guide
them to swim downward in their natural aque-
4. Magnetosomes
ous habitat. All the magnetotactic bacteria are
For a review of magnetosomes in bacteria, see
microaerophilic or anaerobic, and it is thought
refs. 121 through 125.
that magnetotaxis to lower levels is beneficial
Magnetosomes and magnetotactic bacteria. because there is less oxygen at greater depths.
Certain marine and freshwater bacteria have However, although this is the case for most MB
chains of membrane-bound organelles called that have been observed, exceptions do occur.
magnetosomes, which originate as invagina- There has been a report of a population of mag-
tions of the cell membrane. The bacteria are netotactic bacteria in the Northern Hemisphere
referred to as magnetotactic bacteria (MB) or that responds to high oxygen levels by swim-
(MTB) because the magnetosomes influence ming toward geomagnetic south in the drop
the direction of swimming with respect to the assay just described, rather than in the direction
earth’s magnetic field. of geomagnetic north. Bacteria displaying south
Magnetosomes are strings of crystals of iron polarity co-occurred in water samples from the
oxide, magnetite (Fe3O4), or in some cases iron same aquatic redox environment with bacteria
sulfide, greigite (Fe3S4), each one of which is sur- displaying north polarity. Thus, more has to be
rounded by a specialized, complex membrane learned about magnetotactic behavior.
containing a lipid bilayer of phospholipids, and The magnetotactic bacteria are a very diverse
proteins. For a review of the proteins associ- group of motile organisms and belong to a
ated with magnetosomes and their proposed wide range of phylogenetic groups, including
functions, refer to ref. 119. The magnetosomes α-Proteobacteria, β-Proteobacteria, δ-Proteo-
are positioned in the cell by cytoskeletal fila- bacteria, and Nitrospira. At this time, magne-
ments composed of a protein called MamK, tosomes have not been found in gram-positive
which flanks the magnetosomes. MamK is a bacteria or archaea.
homologue of the bacterial actinlike cytoskel-
etal protein called MreB. (See the discussion 5. Granules and globules
of cytoskeletal proteins and their functions in Bacteria often contain cytoplasmic granules
Section 1.2.7.) whose content varies with the bacterium. Many
bacteria store a lipoidal substance called poly- specific sequences and also bends DNA. The
β-hydroxybutyric acid (PHB) in the granules DNA-binding proteins, HU, H-NS, and Fis,
as a carbon and energy reserve. Other bacteria are called histonelike because they resemble
may store glycogen for the same purpose. Other histone in their electrostatic charge, binding
granules found in bacterial cells can include to DNA, and because they have a low molecu-
polyphosphate and, in some sulfur-oxidizing lar weight and a high copy number. However,
bacteria, elemental sulfur globules. unlike HU itself, the other histonelike proteins
are not similar to histone in their amino acid
6. Ribosomes composition or structure. For example, the
Ribosomes are the sites of protein synthesis. histonelike protein H-NS binds to curved or
They are ribonucleoprotein particles, approxi- bent DNA. The structure of H-NS, its role in
mately 22 nm by about 30 nm, or about the DNA superstructure, and its role in gene regu-
size of the smallest viruses. Ribosomes consist lation is reviewed in ref. 127. It is believed that
of over 50 different proteins and three differ- the DNA-binding structural proteins facilitate
ent types of RNA (23S, 16S, and 5S). Bacterial bending of DNA and/or have a high affinity for
ribosomes sediment in a centrifugal field at a bent DNA, or restraining of DNA supercoils in
characteristic velocity of 70S, as opposed to the nucleoid, and for the compact structure of
eukaryotic cytosolic ribosomes, which are 80S. the nucleoid.128
Bacterial ribosomes are very similar regardless The DNA-binding proteins are important
of the bacterium. However, there are some dif- not only for nucleoid structure but also for the
ferences from archaeal ribosomes, which are regulation of expression of certain genes. See
also 70S. These differences were described in the discussion of the H-NS and Fis proteins in
Section 1.1.1. Section 2.2.2 (ribosomal RNA genes), the dis-
7. The nucleoid cussion of H-NS and IHF in Section 19.11.5
The site of DNA and RNA synthesis is the nucle- (virulence genes), and the discussion of the
oid, an amorphous mass of DNA unbounded by regulatory role of IHF in Section 19.3.5, as well
a membrane, lying approximately in the center as Section 20.4 (Ntr regulon) and Section 19.7
of the cell. Faster-growing bacteria may contain (porin biosynthesis).
more than one nucleoid, but each nucleoid has Another protein important for DNA struc-
but one chromosome, and all the chromosomes ture is DNA gyrase (topoisomerase II), which
are identical. The DNA is very tightly coiled. is responsible for negative supercoiling, that is,
Indeed, if the DNA from E. coli were stretched the twisting of the double helix about its axis in
out, it would be 500 times longer than the cell! the direction opposite to the right-handed dou-
ble helix. Cellular DNA is mostly in a negative
DNA-binding proteins influence nucleoid supercoil and otherwise would be much more
structure and gene regulation. The DNA in the difficult to unwind to obtain single strands for
nucleoids is bound to several proteins as well replication and transcription. There is also an
as to nascent chains of RNA. Of course, among enzyme, topoisomerase I, that removes negative
the proteins bound to the DNA are RNA poly- supercoils.
merase molecules engaged in transcription; but
several other proteins can be present as well. 8. Multienzyme complexes
When nucleoids are isolated from E. coli It should not be thought that the enzymes in
under low salt conditions (thus preserving the the cytoplasm are a random mixture of pro-
attachment of proteins to the DNA), membrane teins. There are many examples of enzymes in
proteins are present as well as the DNA-binding the same pathway forming stable multienzyme
proteins HU, IHF, H-NS, and Fis. HU, often complexes, reflecting strong intermolecular
referred to as a histonelike protein because it has bonding.129 For example, pyruvate dehydroge-
physical properties and an amino acid composi- nase from E. coli is a complex of three different
tion similar to eukaryotic histones, is the major enzymes, each present in multiple copies (50
DNA-binding protein present in E. coli.126 It proteins total), that oxidizes pyruvic acid to
binds to DNA without any apparent sequence acetyl–CoA and CO2. The size of the pyruvate
specificity, wrapping the DNA and causing it dehydrogenase complex is (4.6–4.8) × 106 Da.
to bend. Integration host factor (IHF) binds to Contrast this with the size of a 70S ribosome
www.ebook777.com
(another multienzyme complex), which is of glucose to pyruvic acid or lactic acid) cannot
about 2.7 × 106 Da. Other enzyme complexes be isolated as complexes. It has generally been
that catalyze a consecutive series of biochemi- assumed that these enzymes exist either as inde-
cal reactions include the α-ketoglutarate dehy- pendent proteins or in loose associations that
drogenase complex, which consists of three are easily disrupted during cell breakage and the
different enzymes present in multiple copies accompanying dilution of the proteins. (Bear in
(i.e., 48 proteins), 2.5 × 106 Da (in E. coli), it mind that because of dilution of the extracts by
oxidizes α-ketoglutarate to succinyl–coenzyme the addition of buffers during the washing and
A (acetyl–SCoA) and CO2. Another example is suspension of the cells prior to breakage, the
fatty acid synthase in yeast. It consists of seven concentration of proteins in broken cell extracts
different enzymes (2.4 × 106 Da) and synthe- is orders of magnitude lower than in the aque-
sizes fatty acids from acetyl–SCoA. (For com- ous portion of the unbroken cell.)
parison to bacteria, see note 130.) One of the
advantages of a multienzyme complex is that 1.2.7 Cell shape
it facilitates the channeling of metabolites, This section is devoted to a discussion of pro-
thus increasing the efficiency of catalysis. For teins that influence the final shape that bacteria
example, in these stable enzyme complexes the take. A central point is the role of components
biochemical intermediates in the pathway are in determining the sites of synthesis and shape
transferred directly from one enzyme to the of the peptidoglycan sacculus that holds the
next without entering the bulk phase. Thus, cell in its particular shape. The argument will
there is no dilution of intermediates, nor is be made that bacteria have several cytoskel-
there reliance on random diffusion to reach a etal components that provide a scaffolding to
second enzyme. which peptidoglycan-synthesizing machinery is
attached, and the placement of the peptidogly-
Cytosol can machinery along the scaffolding determines
The liquid portion of the cytoplasm, the cytosol, the ultimate shape of the sacculus. We will
can be isolated in diluted form as the superna- begin with background information that briefly
tant fraction obtained after broken cell extracts describes cytoskeletal components found in
have been centrifuged at 105,000 × g for 1 to eukaryotic cells and compare these to cytoskel-
2 h, which should result in sedimentation of etal components present in prokaryotic cells.
the membranes, the DNA, the ribosomes, very
large protein aggregates, and other intracellular Eukaryotic cytoskeletal proteins
inclusions. In the cytosol are found the enzymes Eukaryotic cells have a network of protein fila-
that catalyze a major portion of the biochemi- ments (fibers), collectively called a cytoskeleton,
cal reactions in the cell, such as the enzymes of extending throughout the cytoplasm. Three
the central pathways for the metabolism of car- types of protein filament are present in the
bohydrates (glycolysis, the pentose phosphate eukaryotic cytoskeleton, and they are distin-
pathway, and the Entner–Doudoroff pathway) guished in part by their size. In increasing diam-
as well as the central pathways for organic acid eter, they are actin filaments (7 nm diameter),
metabolism (citric acid cycle and glyoxylate intermediate filaments (10–12 nm diameter),
pathways) and enzymes for other pathways, and microtubules (25 nm diameter). The fila-
such as the biosynthesis and degradation of ments are made from different protein subunits:
amino acids, lipids, and nucleotides. actin (actin filaments), tubulin (microtubules),
The concentration of proteins in the cytosol vimentin, lamin (intermediate filaments), and
is very high, making the material viscous, and so on. The eukaryotic cytoskeleton, along with
it is expected that extensive protein–protein motor proteins such as myosin, dynein, and
interactions occur among the enzymes, even kinesin, has many functions, briefly reviewed
those that do not exist in tight complexes. If in note 131. It is responsible for cell shape, the
such interactions exist, however, they must be movements of cilia and flagella, muscle contrac-
weak, since most of the enzymes that catalyze tion, endocytosis, and the movements of com-
metabolic pathways in the cytosol (e.g., the gly- ponents such as vesicles from one part of the cell
colytic enzymes that catalyze the degradation to another.
Prokaryotic cytoskeleton proteins the synthesis of peptidoglycan in the septum,

Prokaryotes possess proteins that show some necessary for cell division. The Z ring is also
similarity to the eukaryotic cytoskeletal compo- important for cell separation because it recruits
nents actin (bacterial MreB), tubulin (bacterial peptidoglycan hydrolases that are required for
FtsZ), and intermediate filament components the septal peptidoglycan to split into two to
(bacterial IFs, crescentin), suggesting a prokary- allow the daughter cells to separate.
otic origin to these cytoskeletal components.. The Z ring also is required for the relocal-
These bacterial cytoskeletal components play ization of enzymes that are necessary for lat-
important roles in cell shape, cell division, chro- eral peptidoglycan expansion. These enzymes
mosome segregation, plasmid segregation, and include a transglycosylase called PBP1B. It
the maintenance of cell polarity. For reviews, should be mentioned that at least two other pro-
see refs. 132 through 136. teins, RodA and PBP2, are also required for lat-
eral peptidoglycan synthesis and are important
for cells to elongate as rods. See note 137 for an
How the prokaryotic cytoskeletal proteins
explanation of PBPs and their roles.
might determine cell shape
A description of the individual prokaryotic
Mre proteins. These are cytoskeletal proteins
cytoskeletal proteins is usefully preceded by a
that are required for cells to elongate. One
review of some ideas about how these proteine
of them is a protein called MreB. MreB has
determine cell shape. In this regard it is impor-
sequence patterns similar to that of actin and
tant that localized synthesis of peptidoglycan
is a member of the actin superfamily. It can be
reflects whether cells grow as rods or cocci and
found in many rod-shaped, filamentous, and
that the cytoskeletal proteins have been sug-
helical bacteria. It has been reported to form
gested to act as a scaffolding that determines
a cytoskeleton in some bacteria, encircling the
the location of the peptidoglycan biosynthetic
cell as spirals (also referred to as helical cables)
enzymes. Much of the experimental evidence
under the cell membrane along the longitudinal
showing the sites of peptidoglycan synthesis
axis from cell pole to cell pole, and it contributes
in rod-shaped cells and coccoid cells is due to
toward determining the shapes of nonspheri-
the use in gram-positive bacteria of fluorescent
cal bacteria.138 Mutations in the gene encod-
antibiotics that label peptidoglycan precursors
ing MreB cause E. coli to lose its rod shape and
and, in gram-negative bacteria, whose outer
become spherical. Mutants of MreB cause rod-
membrane is not permeable to fluorescent anti-
shaped B. subtilis cells to round up and lyse.
biotics, procedures the use of D-cysteine label-
Examination of genome sequences indicates
ing. What has been learned is that in E. coli and
that mreB is not present in coccoid bacteria.
B. subtilis the insertion of new peptidoglycan
One hypothesis is that the Mre proteins orga-
components occurs along the lateral cell wall in
nize the peptidoglycan biosynthetic machinery
a helical manner, but not at the cell poles. On
so that the cells grow as rods. Another way of
the other hand, new peptidoglycan is inserted
stating this is that the Mre proteins may be a
only in the septal region of cocci, as well as in E.
scaffolding that might recruit peptidoglycan-
coli round mutants (pbpA, rodA) that have lost
synthesizing enzymes such as transglycosylases
their ability to grow as rods. As mentioned ear-
for elongation of the glycan chains, cross-link-
lier, the sites of peptidoglycan synthesis appear
ing enzymes that form peptide bonds between
to be due to certain cytoskeletal protein compo-
the glycan chains, and hydrolases that introduce
nents that act as a scaffolding to which the pep-
gaps in the glycan chains so that additional com-
tidoglycan-synthesizing machinery is attached.
ponents can be added for elongation of the pep-
FtsZ. FtsZ is a cytoskeletal component related tidoglycan. It is the placement of these enzymes
to tubulin. At the site of cell division, it assem- that might determine cell shape. MreB is present
bles as a ring, called the Z ring, and recruits in both E. coli and Caulobacter crescentus, but
other proteins to form a contractile septal ring Bacillus subtilis has three paralogues (MreB,
that constricts the cell during division. In addi- Mbl, and MreBH). Anchored in the cell mem-
tion, two of the proteins recruited to the Z brane and having a major helical portion in the
ring are FtsI and FtsW, which are required for periplasm, MreC might serve in the periplasm
www.ebook777.com
as a scaffolding for the spatial localization of that of a bundle of 10 to 12 flagella. Swimming

peptidoglycan-synthesizing machinery. is by means of helical contractions and exten-
sions of the cell, and rotation about the helical
Crescentin. In addition to MreB, the loss of
axis. For a discussion of this type of motility,
which leads to a loss of the rod shape, resulting in
including the forces and energetics, the student
lemon-shaped cells, Caulobacter crescentus has
is referred to refs. 142 through 144. Movement
a cytoskeleton protein called crescentin, which
is more rapid in viscous media.
is responsible for its vibrioid shape.136,139,140
What causes movements in Spiroplasma?
Crescentin has certain characteristics similar to
Inside each cell is a single flat, cytoskeletal rib-
those of intermediate filaments. The screening of
bon made of seven paired parallel protein fibers,
transposon-insertion libraries revealed mutants
attached to the inner surface of the cell mem-
that grow as rod-shaped bacteria rather than
brane along its innermost (and shortest) helical
vibrios. This led to the discovery of crescentin.
side. The ribbon is continuous and spans the
Immunofluorescence microscopy revealed that
entire length of the cell. Nonmotile mutants do
crescentin exists as a helical filament along the
not have the cytoskeletal ribbon. The cytoskel-
concave side of the cell membrane.
etal ribbon with its associated proteins is a lin-
ear motor, whereas flagella have rotary motors.
Cytoskeletal components unrelated to What this means is that the cytoskeletal ribbon
eukaryotic cytoskeletal components are undergoes linear contractions and extensions;
present in Mollicutes and because it is attached to the cell membrane,
The Mollicutes (Spiroplasma, Mycoplasma, the cell moves forward in the fluid. It has been
Acholeplasma) are a class of very small, bacte- proposed that the length changes in the ribbon
ria without cell walls. They may be free living, are due to conformational changes in the mono-
but many are parasitic on animals (including meric cytoskeletal proteins.
humans) and plants, and can cause disease.
Mollicutes can be found as part of the natural
Mycoplasma
flora in the mouth, throat, and genitourinary
Mycoplasma cells are very small, somewhat
tract of mammals and birds. (See note 141 for
spherical cells. One end is extended as a nar-
more information about diseases caused by
row neck, giving the cell a flasklike appear-
Mollicutes.) Several strains of Mycoplasma can
ance. Many are nonmotile. However, some
glide, and Spiroplasma can swim. These bacte-
are capable of movement on a solid surface.
ria are also chemotactic. However, they do not
Mycoplasma has an internal cytoskeleton com-
have analogues to genes responsible for chemot-
posed of protein fibers that originate at the tip
axis or in other bacteria. All Mollicutes stain
of the “neck” region. These fibers are present
gram-negative; however, 16S rRNA analysis
in both motile and nonmotile Mycoplasma.
indicates that they are closely related to gram-
Mycoplasma adheres to host tissues at the tip
positive bacteria, most notably Clostridium,
of the “neck” region, and it has been suggested
from which they are believed to have evolved.
that the cytoskeletal fibers play a role in adhe-
1. Cytoskeleton sion. How the fibers might be related to motility
Mollicutes have an internal protein cytoskel- is not clear.
eton that determines the shape of the cell and in
some cases (e.g., Spiroplasma) can function in 1.3 Summary
motility. They do not have homologues to any
of the eukaryotic genes for cytoskeletal pro- There are two evolutionary lines of prokary-
teins or motor proteins, such as actin, micro- otes, the Bacteria and the Archaea. Archaea are
tubule proteins, myosin, or dynein. Thus, their similar to bacteria, but they differ in certain fun-
cytoskeleton is made from protein unrelated to damental aspects of structure and biochemistry.
cytoskeletal proteins found in eukaryotes. These include differences in ribosomes, cell wall
chemistry, membrane lipids, and coenzymes.
2. Spiroplasma In addition, certain archaea have metabolic
Spiroplasma cells are very thin, round, helical, pathways (e.g., methanogenesis) not found in
tubelike cells. The cell diameter is approximately bacteria.
Surrounding many bacteria are pili and extra- certain Mollicutes, driven by a cytoskeletal lin-
cellular material called a glycocalyx. The pili ear motor.
are protein fibrils. The glycocalyx encompasses There are two types of bacterial cell wall. One
all extracellular polymers beyond the cell wall, kind has a thick peptidoglycan layer to which
polysaccharide or protein, including capsules there are covalently bonded polysaccharides,
and slime layers. The pili and glycocalyx anchor teichoic acids, and teichuronic acids. Bacteria
the cell to specific animal and plant cell surfaces, with such walls stain gram-positive. A second
as well as to inanimate objects, and in several type of wall has a thin peptidoglycan layer and
cases to each other. an outer envelope consisting of lipopolysac-
Some bacteria possess a fibril called a sex charide, phospholipid, and protein. Cells with
pilus that attaches the cell to a mating part- such walls do not stain gram-positive and are
ner and retracts to draw the cells into intimate called gram-negative. There is also a separate
contact. When the cells are in contact, DNA is compartment in gram-negative bacteria called
transferred unilaterally from the donor cell (i.e., a periplasm, between the outer envelope and the
the one with the sex pilus) to the recipient. cell membrane. The periplasm is the location of
Protruding from many bacteria are flagella numerous proteins and enzymes, including pro-
filaments that aid the bacterium in swimming. teins required for solute transport and enzymes
At the base of the flagellum is a basal body that function in the oxidation and degradation
that is embedded in the cell membrane. At the of nutrients in the periplasm. Archaea have very
base of the flagellum is a rotary motor that in different cell walls. No archaeon has peptido-
most bacteria runs on a current of protons. glycan, although some have a similar polymer
Extreme halophiles and some marine bacteria called pseudopeptidoglycan or pseudomurein.
use a sodium ion current instead of a proton The cell membranes of the bacteria are all
current. A membranous sheath covers the fla- similar. They consist primarily of phosphoglyc-
gella of some bacteria (e.g., Vibrio cholerae, V. erides in a bilayer and protein. The phospho-
parahaemolyticus, Helicobacter, Bdellovibrio glycerides are generally fatty acids esterified to
bacteriovorus). There is a proteinaceous glycerol phosphate. Archaeal cell membranes
sheath covering the flagella (axial filaments) have lipids that are long-chain alcohols, ether-
of most spirochaetes (an exception is Borrelia linked to glycerol. Archaeal membranes can be
burgdorferi, the causative agent of Lyme dis- a bilayer or a monolayer, or perhaps a mixture
ease) which is actually located in the periplasm of the two.
rather than outside the cell. Spirochaetes are The cytoplasm of prokaryotes is a viscous
shaped like a coil, and when the flagellum solution of protein. Salts, sugars, amino acids,
turns, the cell moves in a corkscrewlike fash- and other metabolites are dissolved in the pro-
ion, which apparently is suited for movement teinaceous cytoplasm. In many bacteria intra-
through high-viscosity media. (Again, B. burg- cytoplasmic membranes are present. In many
dorferi is an exception and does not move in cases these membranes are invaginations of
a corkscrewlike fashion.144) Bacterial flagella the cell membrane. They are sites for electron
are important for the pathogenicity of certain transport and specialized biochemical activi-
bacteria and this is discussed in the review by ties. Numerous inclusion bodies are also pres-
Moens and Vanderleyden.39 Many genera of ent. These include ribosomes, which are the
bacteria contain species that swim as interact- sites of protein synthesis, as well as large aggre-
ing populations of cells on moist solid surfaces gates of enzymes that catalyze short metabolic
such as dilute agar, and other surfaces where pathways, gas vesicles, and carbon and energy
biofilms form. The movement is called swarm- reserves (e.g., glycogen particles). Some bacteria
ing and is due to lateral flagella. have magnetic particles called magnetosomes
There exist several other mechanisms of in the cytoplasm. Also present in the cytoplasm
besides flagella-driven motility. These include is DNA, very tightly packed, and referred to
“twitching,” a form of gliding motility pow- as a nucleoid. There is evidence that a protein
ered by type IV pili; gliding motility, which is cytoskeleton in the cytoplasm and periplasm
driven by extracellular slime secretion through plays a role in determining cell shape, as well as
pores in the cell surface; and swimming in serving other functions. It is clear that although
www.ebook777.com
BOX 1.6 CANNULAE AND HAMI:

CELL SURFACE STRUCTURES OF ARCHAEA
The microorganisms in the domain Archaea the exchange of nutrients. Other cell sur-
have cell surface structures called cannulae face appendages found on certain archaea
and hami not present in the Bacteria. (See are the hami (sing. hamus). These have been
Ng, S. Y. M., et al. 2008. Cell surface struc- described as a “halo” of filamentous protein
tures of Archaea. J. Bacteriol. 190:6039– appendages that surround each cell. The
6047). Cannulae are found in members of hami are morphologically complex struc-
the genus Pyrodictium that live in hydro- tures used for adhesion of the archaeal cells
thermal marine environments as anaero- to surfaces as well as to each other. They
bic chemoautotrophs. They derive energy have been called “grappling hooks” (hamus
from oxidizing hydrogen gas and reduc- is Latin for hook or barb) and have been
ing sulfur: H2 + S° → H2S. The cannulae found on the surface of SM1 cocci grow-
are hollow tubules made of glycoprotein, ing closely with Thiothrix, a filamentous
which connect the cells with one another. sulfur-oxidizing γ-proteobacterium grow-
When the cells divide, the daughter cells ing in cold sulfurous marshes. The coccoid
stay connected with the growing cannulae archaeal cells bearing the hami are found in
so that the end result is a dense network of “macroscopically visible string-of pearls-
cells that interconnect with one an other via like arrangement” with the Thiothrix. The
the cannulae. Some cannulae can enter the cell arrangement is very interesting, with the
periplasmic spaces of the cells but not the filamentous Thiothrix forming the “whit-
cytoplasm, whereas other cannulae simply ish” exterior of each “pearl” (SM1 coccus)
touch the outer cell wall layers (S layers) of plus the white interconnecting thread. Each
the cells. It may be that the cannulae allow “pearl” consists of up to 107 euryarchaeal
for cell–cell communication or perhaps even SM1 cells.
the cytoplasm of prokaryotes is not compart- (ribosomes, enzyme aggregates, etc.), and
mentalized into membrane-bound organelles membrane.
like the eukaryotic cytoplasm, it nonetheless
represents a very complex pattern.
Study Questions
The central metabolic pathways that are
described in the ensuing chapters (viz., glycoly-
1. What chemical and structural differences
sis, the Entner–Doudoroff pathway, the pentose
distinguish the archaea from the bacteria?
phosphate pathway, and the citric acid cycle) all
take place in the cytosol, the liquid part of the 2. What are the differences between gram-
cytoplasm. Also found in the cytosol are most of positive and gram-negative cell walls?
the other pathways, including the enzymatic reac-
3. Which cell wall polymers use D-alanine (as
tions for the synthesis and degradation of amino
opposed to L-alanine) as part of their struc-
acids, fatty acids, purines, and pyrimidines.
ture? (Hint: Polymers of one type are found
The cell membrane is the site of numerous
in both gram-positive and gram-negative
other metabolic pathways, including phos-
walls; those of the other are present only in
pholipid biosynthesis, protein secretion, solute
gram-positive walls.)
transport, electron transport, cell wall biosyn-
thesis, the generation of electrochemical ion 4. What structures enable bacteria to adhere
gradients, and ATP synthesis. The prokaryotic to surfaces? What is known about the
cell can therefore be considered to have three chemistry, location, and receptors of the
major metabolic domains: cytosol, particulate molecules that mediate the adhesion?
5. Contrast the functions and cellular loca- 3. Woese, C. R. 1987. Bacterial evolution. Microbiol.
tion of peptidoglycan, phospholipid, and Rev. 51:221–271.
lipopolysaccharide. What is it about the 4. Pace, N. R., D. A. Stahl, D. J. Lane, and G. J.
chemical structure of these three classes of Olsen. 1985. Analyzing natural microbial popula-
compounds that is suitable for their functions by rRNA sequences. ASM News 51:4–12.
tions and/or cellular location? 5. Cavicchioli, R. (Ed.). Archaea, Molecular and
Cellular Biology. 2007. ASM Press, Washington,
6. What are porins, and what do they do? Are DC.
they necessary for gram-positive bacteria?
6. Ng, S. Y. M., B. Zolghadr, A. J. M. Driessen, S.-V.
Explain. Albers, and K. F. Jarrell. 2008. Cell surface structures
7. What are the physiological and enzymatic of Archaea. J. Bacteriol. 190:6039–6047.
functions associated with the periplasm? 7. Stetter, K. O. 1989. Extremely thermophilic
chemolithoautotrophic archaebacteria, pp. 167–
8. The protein TonB is an energy transducer 176. In: Autotrophic Bacteria. H. G. Schlegel and B.
in gram-negative bacteria that transfers Bowien (Eds.). Springer-Verlag, Berlin.
energy from the cell membrane to spe-
8. Schleper, C., G. Puehler, I. Holtz, A. Gambacorta,
cific transporters in the outer membrane. D. Janekovic, U. Santarius, H.-P., Klenk, and W.
Speculate about how this might occur. The Zillig. 1995. Picrophilus gen. nov., fam. nov.: a novel
subject is discussed in ref. 101. aerobic, heterotrophic, thermoacidophilic genus and
family compromising archaea capable of growth
9. Summarize the cell compartmentalization around pH 0. J. Bacteriol. 177:7050–7059.
of metabolic activities in prokaryotes.
9. Solfataras, which are acidic, geothermally heated
10. What are some multienzyme complexes? habitats containing inorganic sulfur compounds,
What advantage do they confer? including sulfuric acid, may be volcanic fissures,
springs, basins, or dried soil that previously con-
11. Which metabolic pathways or activities tained solfataric water. They have an acidic pH in the
are found in the cytosol and which in the aerobic portions because sulfuric acid arises from the
oxidation of hydrogen sulfide, which occurs sponta-
membranes?
neously and also as a result of the growth of sulfur-
12. What distinguishes twitching from swim- oxidizing bacteria.
ming? Compare and contrast the mecha- 10. Koga, Y., M. Nishihara, H. Morii, and M.
nisms and structures. Akagawa-Matsushita. 1993. Ether polar lipids
of methanogenic bacteria: structures, compara-
13. What is the evidence that bacteria possess tive aspects, and biosynthesis. Microbiol. Rev. 57:
a cytoskeleton? Is there more than one type 164–182.
of cytoskeleton? What is the role of the 11. Grayling, R. A., K. Sandman, and J. N. Reeve.
cytoskeleton? Explain. 1996. Histones and chromatin structure in hyper-
thermophilic archaea. FEMS Microbiol. Rev.
14. What is the experimental basis for the con-
18:203–213.
clusion that MotA and MotB form a com-
plex? What is the experimental basis for the 12. Pereira, S. L., R. A. Grayling, R. Lurz, and J.
N. Reeve. 1997. Archaeal nucleosomes. Proc. Natl.
conclusion that MotA conducts protons? Acad. Sci. USA 94:12633–12637.
Why has it been concluded that FliG is the
rotor portion of the flagellar motor? Why 13. Bardy, S. L., S. Y. M. Ng, and K. F. Jarrell.
2003. Prokaryotic motility structures. Microbiology
are the MS and C rings sometimes included 149:295–304.
as part of the rotor?
14. Macnab, R. M. 2003. How bacteria assemble
flagella. Annu. Rev. Microbiol. 57:77–100.
REFERENCES AND NOTES 15. Chevance, F., and K. T. Hughes. 2008.
Coordinating assembly of a bacterial macromolecu-
1. Morris, D. M., and G. J. Jensen. 2008. Toward lar machine. Nat. Rev. Microbiol. 6:455–465.
a biomechanical understanding of whole bacterial
16. Eukaryotic flagella differ from bacterial flagella
cells. Annu. Rev. Biochem. 77:583–613.
in many ways. Rather than being stiff, rotating fila-
2. Woese, C. R., O. Kandler, and M. L. Wheelis. ments, they are flexible and have an undulating wave
1990. Towards a natural system of organisms: motion from base to tip that pushes the cell forward.
proposal for the domains Archaea, Bacteria, and The movement is driven by ATP rather than by a
Eucarya. Proc. Natl. Acad. Sci. USA 87:4576–4579. proton current. The structure is very different from
www.ebook777.com
that of prokaryotic flagella. The core of the filament A. Stoebner, and M. D. Manson. 1995. Protein inter-
is called an axoneme. The axoneme, which has an actions in the bacterial flagellar motor. Proc. Natl.
internal system of rods called microtubules, com- Acad. Sci. USA 92:1970–1974.
posed of tubulin subunits and associated proteins,
is covered by a sheath that is an extension of the 23. Blair, D. F., and H. C. Berg. 1990. The MotA
cell membrane. Each axoneme has nine fused pairs protein of E. coli is a proton-conducting component
of microtubules that are attached to each other as a of the flagellar motor. Cell 60:439–449.
bundle, surrounding two central microtubules, all of 24. Garza, A. G., L. W. Harris-Haller, R. A.
which extend the length of the filament. This is called Stoebner, and M. D. Manson. 1995. Motility protein
the 9 + 2 arrangement. Numerous proteins project interactions in the bacterial flagellar motor. Proc.
along the length of the microtubule doublets. Some Natl. Acad. Sci. USA 92:1970–1974.
of these are cross-links that attach the doublets to
each other. Others, called dynein, cause the bending. 25. Kojima, S., and D. F. Blair. 2001. Conformational
Dynein hydrolyzes ATP as a source of energy and change in the stator of the bacterial flagellar motor.
moves along the length of one microtubule doublet, Biochemistry 40:13041–13050.
pulling the attached doublet so that it bends. Other 26. Some researchers refer to the FliG proteins,
attached proteins constitute a relay system that con- which are bound to the peripheral inner surface of
trols the bending so that the appropriate waveform the MS ring, as well as to the C ring, as the rotor.
is produced. Some researchers refer to the C ring as the rotor, and
Cilia are constructed in a similar fashion, and some researchers refer to both the MS and C rings
the molecular basis for their movement is the same, as the rotor. See Bardy, S. L., S. Y. M. Ng, and K.
although their movement is more whiplike than the F. Jarrell. 2003. Prokaryotic motility structures.
movement of flagella. Microbiology 149:295–304. Berg, H. C. 2003. The
17. Iino, T., Y. Komeda, K. Kutsukake, R. M. rotary motor of the bacterial flagella. Annu. Rev.
Macnab, P. Matsumura, J. S. Parkinson, M. I. Simon, Biochem. 72:19–54. Grünenfelder, B., S. Gehrig, and
and S. Yamaguchi. 1998. New unified nomencla- U. Jenal. 2003. Role of the cytoplasmic C terminus of
ture for the flagellar genes of Escherichia coli and the FliF motor protein in flagellar assembly and rota-
Salmonella typhimurium. Microbiol. Rev. 52:533– tion. J. Bacteriol. 185:1624–1633.
535.
27. Francis, N. R., G. E. Sosinsky, D. Thomas, and
18. Schoenhals, G. J., and R. M. Macnab. 1996. D. J. DeRosier. 1994. Isolation, characterization and
Physiological and biochemical analyses of FlgH, a structure of bacterial flagellar motors containing the
lipoprotein forming the outer membrane L ring of switch complex. J. Mol. Biol. 235:1261–1270.
the flagellar basal body of Salmonella typhimurium.
J. Bacteriol. 178:4200–4207. 28. FliG, FliM, and FliN are thought to be involved
in the turning of the flagellar motor (generating
19. Berg, H. C. 2003. The rotary motor of the bacte- torque) because certain mutations result in paraly-
rial flagella. Annu. Rev. Biochem. 72:19–54. sis. However, the evidence suggests that only FliG
20. Khan, S., M. Dapice, and T. S. Reese. 1988. is directly involved in generating torque. It has been
Effects of mot gene expression on the structure of the concluded that FliG, FliM, and FliN interact with one
flagellar motor. J. Mol. Biol. 202:575–584. another in a complex because certain mutations in
fliG, fliM, and fliN can be suppressed by other muta-
21. Ivey, D. M., M. Ito, R. Gilmour, J. Zemsky, tions in any of the fli genes. CheY-P (phosphorylated
A. A. Guffanti, M. G. Sturr, D. B. Hicks, and T. A. CheY), the molecule that causes the flagellar motor
Krulwich. 1998. Alkaliphile bioenergetics, pp. 181– to reverse its direction of rotation, binds to FliM
210. In: Extremophiles: Microbial Life in Extreme (Sections 20.4–20.7). FliG, FliM, and FliN are prob-
Environments. K. Horikoshi and W. D. Grant (Eds.). ably also involved in flagellar assembly, since null
John Wiley & Sons, New York. mutants do not have flagella. This may be because
22. Suppressor mutations can indicate whether pro- the switch complex must be made prior to more dis-
teins interact in a complex. With respect to MotA and tal portions of the flagellum.
MotB, suppressor mutations in E. coli were isolated 29. Pleier, E., and R. Schmitt. 1989. Identification
that suppressed motB missense mutations that oth- and sequence analysis of two related flagellin genes
erwise would have resulted in paralyzed or partially in Rhizobium meliloti. J. Bacteriol. 171:1467–1475.
paralyzed flagella. DNA sequencing of the suppres-
sor mutants showed that they caused single amino 30. Yonekura, K., S. Maki, D. G. Morgan, D. J.
acid changes in MotA. The conclusion is that MotA DeRosier, F. Vonderviszt, K. Imada, and K. Namba.
and MotB interact as a complex. Electron micro- 2000. The bacterial flagellar cap as a rotary promoter
graphic evidence also indicates that MotA and MotB of flagellin self-assembly. Science 290:2148–2152.
form a complex. Electron micrographs show 10 to 12 31. Hughes, K. T., and P. D. Aldridge. 2001. Putting
particles surrounding the M ring in the cytoplasmic a lid on it. Nat. Struct. Biol. 8:96–97.
membrane. The particles are not present when either
MotA or MotB is absent in mutants. For more infor- 32. Iino, T. 1969. Polarity of flagellar growth in
mation, read Garza, A. G., L. W. Harris-Haller, R. Salmonella. J. Gen. Microbiol. 56:227–239.
33. Emerson, S. U., K. Tokuyasu, and M. I. Simon. consider Borrelia burgdorferi., which has 7 to 11 fla-
1970. Bacterial flagella: polarity of elongation. gella attached to each end of the cell; the flagella over-
Science 169:190–192. lap at the cell center to form a continuous bundle. The
cells themselves have a flat-wave shape, rather than
34. It has been proposed that ATP hydrolysis by
being “corkscrew” helical coils. Mutant cells that lack
FliI may drive the export of proteins through the
flagella owing to an insertion in flaB, which is the gene
translocation apparatus, as well as being necessary
that encodes the major flagellar filament protein, no
in the process of bringing proteins to the transloca-
longer resemble a flat wave but are rod shaped. Thus,
tion site. Since FliH binds to FliI and is a negative
the flagella influence the shape of the cell. Bearing this
regulator of its activity, the ATPase activity functions
in mind, recall that the flagellum rotates between the
only when necessary. See Minamino, T., and R. M.
outer membrane sheath and the flexible protoplasmic
Macnab. 2000. Interactions among components of
cell cylinder. One model is based on the configuration
the Salmonella flagellar export apparatus and its sub-
of the flagella: side by side within the flexible proto-
strates. Mol. Microbiol. 35:1052–1064. Minamino,
plasmic cylinder and held there owing to containment
T., and R. M. Macnab. 2000. FliH, a soluble com-
by the outer membrane sheath. The model proposes
ponent of the type III flagellar export apparatus of
the following: (1) the flagella have a left-handed heli-
Salmonella, forms a complex with FliI and inhibits its
cal shape (going away from the observer) wrapped
ATPase activity. Mol. Microbiol. 37:1494–1503.
around the protoplasmic cylinder; (2) the flagella
35. Fuerst, J. A., and J. W. Perry. 1988. Demonstration rotate counterclockwise as viewed from the back of
of lipopolysaccharide on sheathed flagella of Vibrio the cell; (3) as a result of flagella rotation, backward-
cholerae 0:1 by protein A–gold immunoelectron moving waves are propagated down the length of the
microscopy. J. Bacteriol. 170:1488–1494. flexible protoplasmic cylinder, and the cell is pushed
forward in the viscous medium. For a more detailed
36. Weissborn, A., H. M. Steinman, and L. Shapiro.
discussion of the model, see ref. 40.
1982. Characterization of the proteins of the
Caulobacter crescentus flagellar filament. J. Biol. 44. Parales, J., Jr., and E. P. Greenberg. 1991.
Chem. 257:2066–2074. N-Terminal amino acid sequences and amino acid
37. Pleier, E., and R. Schmitt. 1989. Identification compositions of the Spirochaeta aurantia flagellar
and sequence analysis of two related flagellin genes filament polypeptides. J. Bacteriol. 173:1357–1359.
in Rhizobium meliloti. J. Bacteriol. 171:1467–1475. 45. Moens, S., and J. Vanderleyden. 1996. Functions
38. Driks, A., R. Bryan, L. Shapiro, and D. J. of bacterial flagella. Crit. Rev. Microbiol. 22:67–
DeRosier. 1989. The organization of the Caulobacter 100.
crescentus flagellar filament. J. Mol. Biol. 206:627– 46. Harshey, R. M. 2003. Bacterial motility on a
636. surface: many ways to a common goal. Annu. Rev.
39. Moens, S., and J. Vanderleyden. 1996. Functions Microbiol. 57:249–273.
of bacterial flagella. Crit. Rev. Microbiol. 22:67– 47. Belas, R. 1997. Proteus mirabilis and other
100. swarming bacteria, pp. 183–219. In: Bacteria as
40. Charon, N. W., and S. F. Goldstein. 2002. Multicellular Organisms. J. A. Shapiro and M.
Genetics of motility and chemotaxis of a fascinat- Dworkin (Eds.). Oxford University Press, New
ing group of bacteria: the spirochaetes. Annu. Rev. York.
Genet. 36:47–73. 48. Matsuyama, T., K. Kaneda, Y. Nakagawa, K.
41. Several spirochaetes cause disease. Diseases Isa, H. Hara-Hotta, and I. Yano. 1992. A novel extra-
caused by spirochaetes include syphilis, caused cellular cyclic lipopeptide which promotes flagellum-
by Treponema pallidum; Lyme disease, caused by dependent and -independent spreading growth of
Borrelia burgdorferi; relapsing fever, caused by Serratia marcescens. J. Bacteriol. 174:1769–1776.
Borrelia spp.; leptospirosis, caused by Leptospira 49. Kearns, D. B., and R. Losick. 2003. Swarming
spp., and periodontal disease, caused by Treponema in undomesticated Bacillus subtilis. Mol. Microbiol.
denticola. In addition to living in animals, spiro- 49:581–590.
chaetes live in soil, fresh water, and salt water, and
attached to protozoa in the termite gut. The spiro- 50. McCarter, L. M. 2005. Multiple modes of motil-
chaetes attached to the protozoa in the termite gut ity: a second flagellar system in Escherichia coli. J.
enable the protozoa to swim. Bacteriol. 187:1207–1209.
42. Goldstein, S. F., N. W. Charon, and J. A. Kreiling. 51. Jarrell, K. F., D. P. Bayley, and A. S. Kostyukova.
1994. Borrelia burgdorferi swims with a planar 1996. The archaeal flagellum: a unique structure. J.
waveform similar to that of eukaryotic flagella. Proc. Bacteriol. 178:5057–5064.
Natl. Acad. Sci. USA 91:3433–3437.
52. Faguy, D. M., K. F. Jarrell, J. Kuzio, and M. L.
43. One model for spirochaete motility is based upon Kalmokoff. 1994. Molecular analysis of archaeal
the finding that the periplasmic flagella influence the flagellins: similarity to the type IV n-transport super-
shape of the cell. In other words, the periplasmic fla- family widespread in bacteria. Can. J. Microbiol.
gella act as a periplasmic skeleton. As an example, 40:67–71.
www.ebook777.com
53. Faguy, D. M., S. F. Koval, and K. F. Jarrell. 1994. recipient cells (transmissible plasmids). Other genes
Physical characterization of the flagella and flagel- that may be carried by plasmids include genes for
lins from Methanospirillum hungatei. J. Bacteriol. antibiotic resistance (carried by resistance transfer
176:7491–7498. factors or R-factors) and genes for the catabolism
of certain nutrients. For example, Pseudomonas
54. Pearce, W. A., and T. M. Buchanan. 1980.
carries plasmids for the degradation of camphor,
Structure and cell membrane-binding properties
octane, salicylate, and naphthalene. Some plasmids
of bacterial fimbriae, pp. 289–344. In: Bacterial
confer virulence because the plasmids carry genes for
Adherence. E. H. Beachey (Ed.). Chapman & Hall,
toxins or other virulence factors, such as pili. The F
London.
plasmid in E. coli carries genes for self-transmission,
55. Ottow, J. C. G. 1975. Ecology, physiology, and including genes for the F pilus, and is also capable of
genetics of fimbriae and pili. Annu. Rev. Microbiol. integrating into the chromosome and promoting the
29:79–108. transfer of chromosomal DNA.
56. Jones, G. W., and R. E. Isaacson. 1983. 66. Dunny, G. M. 1991. Mating interactions in
Proteinaceous bacterial adhesins and their receptors. gram-positive bacteria. pp. 9–33. In: Microbial Cell–
Crit. Rev. Microbiol. 10:229–260. Cell Interactions. M. Dworkin (Ed.). ASM Press,
Washington, DC.
57. Hultgren, S. J., S. Abraham, M. Caparon, P.
Falk, J. W. St. Geme III, and S. Normark. 1993. Pilus 67. An excellent reference for microbial cell–cell
and nonpilus bacterial adhesins: assembly and func- interactions is the book edited by Dworkin: Microbial
tion in cell recognition. Cell 73:887–901. Cell–Cell Interactions, 1991. M. Dworkin (Ed.).
ASM Press, Washington, DC.
58. Yanagawa, R., and K. Otsuki. 1970. Some
properties of Corynebacterium renale. J. Bacteriol. 68. Costerton, J. W., T. J. Marrie, and K.-J. Cheng.
101:1063–1069. 1985. Phenomena of bacterial adhesion, pp. 3–43. In:
Bacterial Adhesion: Mechanisms and Physiological
59. Cisar, J. O., and A. E. Vatter. 1979. Surface
Significance. D. C. Savage and M. Fletcher (Eds.).
fibrils (fimbriae) of Actinomyces viscosus T14V.
Plenum Press, New York.
Infect. Immun. 24:523–531.
69. Sleytr, U. B., and P. Messner. 1983. Crystalline
60. Eisenstein, B. I. 1987. Fimbriae, pp. 84–90.
surface layers on bacteria. Annu. Rev. Microbiol.
In: Escherichia coli and Salmonella typhimurium:
37:311–339.
Cellular and Molecular Biology. Vol. 1. F. C.
Neidhardt et al. (Eds.). ASM Press, Washington, 70. Costerton, J. W., R. T. Irvin, and K.-J. Cheng.
DC. 1981. The bacterial glycocalyx in nature and disease.
61. Hultgren, S. J., S. Normark, and S. N. Abraham.
1991. Chaperone-assisted assembly and molecu- 71. Roberts, I. S. 1995. Bacterial polysaccharides
lar architecture of adhesive. Annu. Rev. Microbiol. in sickness and in health. Microbiology 141:2023–
45:383–415. 2031.
62. Hultgren, S. J., S. Abraham, M. Caparon, P. 72. Bacterial capsules. 1990. In: Current Topics in
Falk, J. W. St. Geme III, and S. Normark. 1993. Pilus Microbiology and Immunology. K. Jann and B. Jann
and nonpilus bacterial adhesins: assembly and func- (Eds.). Springer-Verlag, Berlin.
tion in cell recognition. Cell 73:887–901.
73. Streptococcus pyogenes (also called group A
63. Type IV pili are also used by certain bacteria, streptococcus) is a human pathogen that causes a
such as Pseudomonas, to form biofilms, and by wide range of diseases such as pharyngitis, impe-
myxobacteria for gliding motility. This is discussed tigo, erysipelas (acute infection of the skin), cellulitis
in Sections 23.1.1 and 23.1.2. (infection of subcutaneous tissues), and necrotizing
fasciitis (the strain that causes necrotizing fasciitis,
64. Pathogenic strains of Escherichia coli cause gas-
called the “flesh-eating bacteria,” also causes subcu-
trointestinal and urinary tract infections, Neisseria
taneous infection resulting in muscle destruction and
gonorrhoeae causes gonorrhea, Bacteroides nodo-
organ failure) and toxic shock syndrome (a drastic
sus causes bovine foot rot, Moraxella bovis causes
decrease in blood pressure, i.e., shock, followed by
bovine keratoconjunctivitis, Vibrio cholerae causes
organ failure). S. pyogenes produces several viru-
cholera, and Pseudomonas aeruginosa causes infec-
lence factors, among which is M protein. This cell
tions of the urinary tract and wounds. The adherence
surface protein is released from the bacteria via a
of these bacteria to their host tissue is the first stage
proteolytic enzyme secreted by S. pyogenes. M pro-
in pathogenesis.
tein binds a plasma protein called fibrinogen, and
65. In addition to the bacterial chromosome, most the complex activates neutrophils (polymorphonu-
bacteria carry extrachromosomal DNA molecules clear leukocytes, or PMNs). The activated PMNs
called plasmids. There are many types of plasmid. release heparin-binding protein, which causes vas-
They all carry genes for self-replication. Conjugative cular leakage resulting in the release of plasma and
plasmids also carry genes for DNA transfer into blood cells from the blood vessels, resulting in shock
that contributes to subsequent organ failure. For adhesin that is thought to bind to the LTA, which in
more information, read: Herwald, H., H. Cramer, turn binds to host cell surfaces.
M. Mörgelin, W. Russel, U. Sollenberg, A. Norrby-
Teglund, H. Flodgaard, L. Lindbom, and L. Björck. 86. Sutcliffe, I. C., and N. Shaw. 1991. Atypical
2004. M protein, a classical bacterial virulence deter- lipoteichoic acids of gram-positive bacteria. J.
minant, forms complexes with fibrinogen that induce Bacteriol. 173:7065–7069.
vascular leakage. Cell 116:367–379. 87. Karakousis, P. C., W. R. Bishai, and S. E.
74. Roberts, I. S. 1996. The biochemistry and genet- Dorman. 2004. Mycobacterium tuberculosis cell
ics of capsular polysaccharide production in bacte- envelope lipids and the host immune response. Cell.
ria. Annu. Rev. Microbiol. 50:285–315. Micobiol. 6:105–116.
75. Kolenbrander, P. E., N. Ganeshkumar, F. J. 88. Hamasaki, N., K. Isowa, K. Kamada, Y. Terano,
Cassels, and C. V. Hughes. 1993. Coaggregation: T. Matsumoto, T. Arakawa, K. Kobayashi, and I.
specific adherence among human oral plaque bacte- Yano. 2000. In vivo administration of mycobacterial
ria. FASEB J. 7:406–409. cord factor (trehalose 6,6′-dimycolate) can induce
lung and liver granulomas and thymic atrophy in
76. Nikaido, H., and M. Vaara. 1987. Outer mem-
rabbits. Infect. Immun. 68:3704–3709.
brane, pp. 7–22. In: Escherichia coli and Salmonella
typhimurium: Cellular and Molecular Biology, 89. Virulence is defined as a quantitative measure
Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press, of a pathogen’s ability to cause disease. Bacterial
Washington, DC. pathogens produce virulence factors of several types.
77. Weidel, W., and H. Pelzer. 1964. Bagshaped These include antiphagocytic capsules, enzymes that
macromolecules—a new outlook on bacterial cell degrade host tissues, and toxins. The lipopolysaccha-
walls. Adv. Enzymol. 26:193–232. ride (LPS) is a toxin that can also contribute toward
virulence. LPS is referred to as an endotoxin because
78. Dmitriev, B. A., F. V. Toukach, K.-J. Schaper, O. it is not secreted by the bacteria into the extracellular
Holst, E. T. Rietschel, and S. Ehlers. 2003. Tertiary medium. It is released from dying bacteria and can
structure of bacterial murein: the scaffold model. J. be shed as blebs from living bacteria. The endotoxic
Bacteriol. 185:3458–3468. activity of the LPS resides in the lipid A portion.
79. Vollmer, W., and J.-V. Höltje. 2004. The archi- Endotoxin causes macrophages to produce tumor
tecture of the murein (peptidoglycan) in gram-nega- necrosis factor (TNF-α), which accumulates in the
tive bacteria: vertical scaffold or horizontal layer(s)? bloodstream. The increased TNF-α in the blood-
J. Bacteriol. 186:5978–5987. stream causes septic shock, which is a drastic drop in
blood pressure due to leakage of blood vessels. This is
80. The amino acid at position three in the peptido- a very dangerous condition because the drop in blood
glycan peptide subunit is referred to as L-R3. It var- pressure results in organ failure.
ies with the species. Gram-negative bacteria usually
have meso-diaminopimelic acid (meso-DAP), or in 90. Although the outer membrane is a permeabil-
the case of some spirochaetes, L-ornithine. There is ity barrier to antibiotics and other toxic substances,
much more variability in gram-positive bacteria, and gram-negative bacteria resist these substances for
L-R3 can be L-alanine, L-homoserine, L-diaminobu- other reasons, as well. A second mechanism is the
tyric acid, L-glutamic acid, L-ornithine, L-lysine, or presence of periplasmic β-lactamases that hydro-
DAP. lyze the earlier β-lactam antibiotics that were used
to treat infections. A third mechanism, and one that
81. Koch, A. L. 1983. The surface stress theory of makes very important contributions to the resistance
microbial morphogenesis. Adv. Microb. Physiol. of pathogenic bacteria to antimicrobial agents, is the
24:301–366. presence of drug efflux pumps. Gram-positive bac-
82. Koch, A. L., and S. Woeste. 1992. Elasticity teria also have drug efflux pumps. A wide range of
of the sacculus of Escherichia coli. J. Bacteriol. toxic substances can be removed from the bacterial
174:4811–4819. cytoplasm of both gram-positive and gram-negative
bacteria by using these pumps. Some pumps transport
83. Weidenmaier, C., and A. Peschels. 2008. a single drug, whereas others transport several unre-
Teichoic acids and related cell-wall glycopolymers in lated drugs. Antimicrobial agents that are pumped
gram-positive physiology and host interactions. Nat. out of cells include basic dyes, quaternary ammo-
Rev. Microbiol. 6:276–287 nium compounds, and a large number of different
84. Neuhaus, F. C., and J. Baddiley. 2003. A con- antibiotics including puromycin, chloramphenicol,
tinuum of anionic charge: structures and functions tetracycline, and erythromycin. For a review of this
of D-alanyl-teichoic acids in gram-positive bacteria. subject, read Section 17.5.
Microbiol. Mol. Biol. Rev. 67:686–723.
91. Leduc, M., K. Ishidate, N. Shakibai, and L.
85. S. pyogenes is an important pathogen that causes Rothfield. 1992. Interactions of Escherichia coli
most streptococcal infections in humans, includ- membrane lipoproteins with the murein sacculus. J.
ing “strep throat.” The M protein is a cell surface Bacteriol. 174:7982–7988.
www.ebook777.com
92. Cross-linking reagents are useful tools for deter- 101. Postle, K., and R. J. Kadner. 2003. Touch
mining the physical association of other molecules. and go: tying TonB to transport. Mol. Microbiol.
These are bifunctional reagents (i.e., they have a 49:869–882.
chemical group at both ends of the molecule that can
102. Merchante, R., H. M. Pooley, and D. Karamata.
form a covalent bond to functional groups on proteins
1995. A periplasm in Bacillus subtilis. J. Bacteriol.
or other molecules such as peptidoglycan); thus they
177:6176–6183.
are able to cross-link two molecules that are within
the length of the cross-linking reagent. For example, 103. Matias, V. R. F., and T. J. Beveridge. 2005.
the reagent dithio-bis-succinimidylpropionate (DSP) Cryo–electron microscopy reveals native polymeric
forms covalent cross-links between amino groups cell wall structure in Bacillus subtilis 168 and the
that are less than 12 angstrom units (1.2 nm) apart. existence of a periplasmic space. Mol. Microbiol.
Therefore, DSP can cross-link proteins that are 56:240–251.
physically associated with the peptidoglycan. After
104. Protoplasts are cells from which the cell wall
treatment with DSP, the peptidoglycan and its cross-
has been removed. They can be stabilized by suspen-
linked proteins can be isolated. DSP has two arms
sion in isotonic buffer (i.e., buffer that is iso-osmolar
held together by a disulfide bond that can be cleaved
with the cell contents, hence preventing water from
with a cleavable cross-linking reagent called mercap-
rushing in and lysing the protoplasts).
toethanol. The cross-linked proteins that were held to
the peptidoglycan by the DSP are released after mer- 105. Booth, I. R., and P. Louis. 1999. Managing
captoethanol treatment and can be analyzed by gel hypoosmotic stress: aquaporins and mechanosen-
electrophoresis. Thus, the proteins cross-linked to the sitive channels in Escherichia coli. Curr. Opin.
peptidoglycan can be identified. In addition to vari- Microbiol. 2:166–169.
ous lipoproteins, including the murein lipoprotein, an
106. Calamita, G., B. Kempf, M. Bonhivers, W. R.
outer membrane protein, OmpA, can be cross-linked
Bishai, E. Bremer, and P. Agre. 1998. Regulation of
to the peptidoglycan, indicating close association. The
the Escherichia coli water channel gene aqpZ. Proc.
cross-linking of OmpA to the peptidoglycan reflects
its extension through the outer envelope.
107. Edwards, M. D., I. R. Booth, and S. Miller.
93. Ferguson, S. J. 1992. The periplasm, pp. 311–
2004. Gating the bacterial mechanosensitive chan-
339. In: Procaryotic Structure and Function: A New
nels: MscS a new paradigm? Curr. Opin. Microbiol.
Perspective. S. Mohan, C. Dow, and J. A. Cole (Eds.).
7:163–167.
Society for General Microbiology Symposium 47,
Cambridge University Press, Cambridge. 108. Kung, C., B. Martinac, and S. Sukharev. 2010.
Mechanosensitive channels in microbes. Annu. Rev.
94. Hobot, J. A., E. Carleman, W. Villiger, and E.
Microbiol. 64:313–329.
Kellenberger. 1984. Periplasmic gel: new concept
resulting from the reinvestigation of bacterial cell 109. Booth, I. R., M. D. Edwards, S. Black, U.
envelope ultrastructure by new methods. J. Bacteriol. Schumann, and S. Miller. 2007. Mechanosensitive
160:143–152. channels in bacteria: signs of closure? Nat. Rev.
95. Bayer, M. E. 1991. Zones of membrane adhe-
sion in the cryofixed envelope of Escherichia coli. J. 110. When a cell is subjected to sudden hypo-os-
Struct. Biol. 107:268–280. motic stress, the rapid flow of incoming water exerts
additional outward pressure against the cell mem-
96. Noina, N., M. Guillier, J. B.Travis, and S. K.
brane, causing distortion of the membrane, resulting
Buchanan. 2010. TonB-dependent transporters:
in the immediate opening of the mechanosensitive
regulation, structure, and function. Annu. Rev.
channels. (Mechanosensitive channels are sometimes
referred to as “gated” channels.) This results in out-
97. Bacteria secrete specific iron chelators called ward efflux of internal solutes along their diffusion
siderophores for scavenging iron from the environ- gradients, thus lowering the internal osmotic pres-
ment. Special transport systems in the outer mem- sure and the influx of water.
brane bind the iron siderophores and bring the
111. Romantsov, T., S. Helbig, D. E. Culham, C.
ligand–iron complex into the periplasm. These are
Gill, L. Stalker, and J. M. Wood. 2007. Cardiolipin
activated by TonB.
promotes polar localization of osmosensory trans-
98. Postle, K. 1990. TonB and the gram-negative porter ProP in Escherichia coli. Mol. Microbiol. 64:
dilemma. Mol. Microbiol. 4:2019–2025. 1455–1465.
99. Braun, V. 1995. Energy-coupled transport and 112. De Rosa, M., A. Gambacorta, and A. Gliozzi.
signal transduction through the gram-negative outer 1986. Structure, biosynthesis, and physiochemical
membrane via TonB-ExbB-ExbD-dependent recep- properties of archaebacterial lipids. Microbiol. Rev.
tor proteins. FEMS Microbiol. Rev. 16:295–307. 50:70–80.
100. Bradbeer, C. 1993. The proton motive force 113. Sternberg, B., C. L’Hostis, C. A. Whiteway,
drives the outer membrane transport of cobalamin in and A. Watts. 1992. The essential role of specific
Escherichia coli. J. Bacteriol. 175:3146–3150. Halobacterium halobium polar lipids in 2D-array
formation of bacteriorhodopsin. Biochim. Biophys. DNA-binding protein from Escherichia coli. Proc.
Acta 1108:21–30. Natl. Acad. Sci. USA 72:3428–3432.
114. Gambacorta, A., A. Gliozzi, and M. De Rosa. 127. Rimsky, S. 2004. Structure of the histone-like
1995. Archaeal lipids and their biotechnological protein H-NS and its role in regulation and genome
applications. World J. Microbiol. 11:115–131. superstructure. Curr. Opin. Microbiol. 7:109–114.
115. Westerhoff, H. V., and G. R. Welch. 1992. 128. Pettijohn, D. E. 1996. The nucleoid, pp. 158–
Enzyme organization and the direction of metabolic 166. In: Escherichia coli and Salmonella: Cellular
flow: physiochemical considerations, pp. 361–390. and Molecular Biology. F. C. Neidhardt et al. (Eds.).
In: Current Topics in Cellular Regulation, Vol. 33. E. ASM Press, Washington, DC.
R. Stadtman and P. B. Chock (Eds.). Academic Press,
New York. 129. Molecular interactions among enzymes is
reviewed by Srivastava, D. K., and S. A. Bernhard.
116. Reviewed by Drews, G. 1991. Intracytoplasmic 1986. Enzyme–enzyme interactions and the regula-
membranes in bacterial cells, pp. 249–274. In: tion of metabolic reaction pathways, pp. 1–68. In:
Prokaryotic Structure and Function, A New Current Topics in Cellular Regulation, Vol. 28. B. L.
Perspective. S. Mohan, D. Dow, and J. A. Coles Horecker and E. R. Stadtman (Eds.). Academic Press,
(Eds.). Cambridge University Press, Cambridge. New York.
117. Overmann, J., S. Lehmann, and N. Pfennig. 130. In bacteria, fatty acid synthesis is catalyzed by
1991. Gas vesicle formation and buoyancy regula- enzymatic reactions similar to those found in yeast
tion in Pelodictyon phaeoclathratiforme (green sul- and mammals, but the enzymes cannot be isolated
fur bacteria). Arch. Microbiol. 157:29–37. as a complex.
118. Codd, G. A. 1988. Carboxysomes and ribulose 131. Motor proteins hydrolyze ATP and use the
bisphosphate carboxylase/oxygenase, pp. 115–164. released energy for internal conformational changes
In: Advances in Microbial Physiology, Vol. 29. A. that allow the proteins to move (“walk”) in one
H. Rose and D. W. Tempest (Eds.). Academic Press, direction along linear macromolecules such as pro-
New York. tein filaments. Motor proteins in eukaryotic cells
include myosin, dynein, and kinesin. Myosin moves
119. Yeates, T. O., C. A. Kerfeld, S. Heinhorst, G.
along protein microfilaments (actin filaments) and
C. Cannon, and J. M. Shively. 2008. Protein-based
plays several roles in doing so. These roles include
organelles in bacteria: carboxysomes and related
(a) the contraction of skeletal muscle cells, (b) the
microcompartments. Nat. Rev. Microbiol. 6:681–
contraction of one cell into two during cytokinesis,
691.
(c) ameboid movement, and (d) the movement of
120. Sawaya, M. R., G. C. Cannon, S. Heinhorst, organelles and vesicles along actin microfilaments
S. Tanaka, E. B. Williams, T. O. Yeates, and C. A. in the cytoplasm during cytoplasmic streaming.
Kerfeld. 2006. The structure of β-carbonic anhy- Dynein is a motor protein that moves along micro-
drase from the carboxysomal shell reveals a distinct tubules in cilia and flagella, and as a consequence,
subclass with one active site for the price of two. J. the cilia and flagella bend. Kinesin, like cytoplasmic
Biol. Chem. 281:7546–7555. dynein, is a motor protein that binds to organelles
and vesicles and moves along cytoplasmic microtu-
121. Jogler, C., and D. Schuler. 2009. Genomics, bules, pulling organelles and vesicles through cyto-
genetics, and cell biology of magnetosome forma- plasm.
tion. Annu. Rev. Microbiol. 63:501–521.
132. Graumann, P. L. 2007. Cytoskeletal elements
122. Komeili, A., Z. Li, D. K. Newman, and G. J.
in bacteria. Annu. Rev. Microbiol. 61:589–618.
Jensen. 2006. Magnetosomes are cell membrane
invaginations organized by the actin-like protein 133. Pichoff, S., and J. Lutkenhaus. 2007. Overview
MamK. Science. 311:242–245. of cell shape: cytoskeletons shape bacterial cells.
Curr. Opin. Microbiol. 10:601–605.
123. Komeili, A. 2007. Molecular mechanisms
of magnetosome formation. Annu. Rev. Biochem. 134. Norris, V., G. Turnock, and D. Sigee. 1996.
76:351–366. The Escherichia coli endoskeleton. Mol. Microbiol.
19:197–204.
124. Simmons, S. L., D. A. Bazylinski, and K. J.
Edwards. 2006. South-seeking magnetotactic bac- 135. Lewis, P. J. 2004. Bacterial subcellular archi-
teria in the Northern Hemisphere. Science 311:371– tecture: recent advances and future prospects. Mol.
374. Microbiol. 54:1135–1150.
125. Schüler, D. 2003. Molecular analysis of a 136. Lutkenhaus, J. 2003. Another cytoskeleton in
subcellular compartment: the magnetosome mem- the closet. Cell 115:648–650.
brane in Magnetospirillum gryphiswaldense. Arch.
Microbiol. 181:1–7. 137. PBPs refer to penicillin-binding proteins
that form and cross-link the peptidoglycan glycan
126. Rouviere-Yaniv, J., and F. Gros. 1975. strands that comprise the murein sacculus. The PBP
Characterization of a novel, low-molecular-weight enzymes are transglycosylases and transpeptidases.
www.ebook777.com
In addition to the transglycosylases and transpep- 141. Mollicutes are part of the normal flora in
tidases, murein hydrolases are required for the healthy people. However, they can cause disease.
growth of the murein sacculus. The murein hydro- Diseases in humans caused by Mollicutes include
lases create gaps in the sacculus into which new primary atypical pneumonia (Mycoplasma pneumo-
peptidoglycan strands are inserted. For a review of niae), pelvic inflammatory disease (M. hominis), and
peptidoglycan synthesis, see Section 12.1. RodA nongonococcal urethritis (Ureaplasma urealyticum).
and FtsW might be involved in transporting pep- Spiroplasma causes diseases in insects and plants.
tidoglycan precursors to the sites of peptidoglycan 142. Gilad, R., Porat, A., and S. Trachtenberg. 2003.
synthesis. Modes of Spiroplasma melliferum BC3: a helical,
138. van den Ent, F., L. A. Amos, and J. Löwe. 2001. wall-less bacterium driven by a linear motor. Mol.
Prokaryotic origin of the actin cytoskeleton. Nature Microbiol. 47:657–669.
413:39–44. 143. Trachtenberg, S., R. Gilad, and N. Geffen.
139. Ausmees, N., J. R. Kuhn, and C. Jacobs- 2003. The bacterial linear motor of Spiroplasma
Wagner. 2003. The bacterial cytoskeleton: an inter- melliferum BC3: From single molecules to swimming
mediate filament-like function in cell shape. Cell cells. Mol. Microbiol. 47:671–697.
115:705–713. 144. Graumann, P. L. 2007. Cytoskeletal elements in
140. Figge, R. M., A. V. Divakaruni, and J. W. bacteria. Annu. Rev. Microbiol. 61:589–618.
Gobel. 2004. MreB, the cell shape-determining bacte- 145. Goldstein, S. F., N. W. Charon, and J. A.
rial actin homologue, co-ordinates cell wall morpho- Kreiling. 1994. Borrelia burgdorferi swims with a
genesis in Caulobacter crescentus. Mol. Microbiol. planar waveform similar to that of eukaryotic fla-
51:1321–1332. gella. Proc. Natl. Acad. Sci. USA 91:3433–3437.
2
Growth and Cell Division
The study of the growth of microbial popula- 2.1.1 Turbidity

tions is at the heart of microbial cell physiol- Turbidity measurements are routinely used
ogy because population growth characteristics to ascertain bacterial growth because of the
reflect underlying physiological events in the simplicity of the procedure. Within limits, the
individual cells. In fact, these cellular events are amount of light scattered by a bacterial cell is
frequently manifested to the investigator only proportional to its mass. Therefore, light scat-
by changes in the growth of populations. It is tering is proportional to the total mass. (The
therefore important to understand the changes relationship between total mass and light scat-
in growth that microbial populations undergo tering deviates from linearity at very high cell
and to be able to measure population growth densities.) If the average mass per cell remains
accurately. Additionally, to investigate cer- a constant, one can also use light scattering to
tain aspects of cell physiology (e.g., the inter- measure changes in cell number. What is actu-
dependencies between the rates of synthesis of ally measured is the fraction of incident light
the individual classes of macromolecules such transmitted through the culture (i.e., not the
as DNA, RNA, and protein), the investigator scattered light). The more dense the culture,
must be able to manipulate population growth the less light is transmitted. This relationship is
(e.g., by placing the culture in balanced growth given by the Beer–Lambert law (eq. 2.1). Thus
or continuous growth). This chapter describes if I0 is the incident light and I is the transmitted
methods to measure growth, presents an analy- light, then the Beer–Lambert law states that in a
sis of exponential growth kinetics, discusses population whose cell density is x, the fraction
growth in batch culture and in continuous cul- of light that is transmitted (I/I0) will decrease as
ture, and introduces some important aspects of the logarithm of x to the base 10.
growth physiology under certain nutritional
conditions. I/I0 = 10−xL (2.1)
That is, if one takes the log to the base 10 of

2.1 Measurement of Growth both sides of eq. 2.1, then log (I/I0) = –xL, where
Growth is defined as an increase in mass, and L is the light path in centimeters.
one can measure any growth parameter pro- The situation can be drawn schematically as
vided it increases proportionally to the mass of shown in Fig. 2.1. Note that the fraction of light
the culture. The most commonly used measure- that is transmitted decreases as a logarithmic
ments of growth are turbidity, total and viable function (base 10); the fraction decreases expo-
cell counts, dry weight, and protein. nentially, that is, with the density of the culture.
55
www.ebook777.com
Fig. 2.1 Illustration of light scattering. The incident Fig. 2.2 Standard curve of optical density versus cells
light is I0 and the transmitted light is I. The spectropho- per milliliter. The line deviates from linearity at very
tometer usually provides the logarithm of the recipro- high cell densities.
cal of the fraction of transmitted light, that is, log(I0/I),
and this is called the optical density or turbidity.
Bacteriologists, however, prefer to measure counting chamber. This is simply a glass slide
something that increases with cell density. divided into tiny square wells of known area
They therefore use the reciprocal of log(I/I0), and depth. A drop of culture is placed on a slide,
which is log(I0/I). The log(I0/I) is called turbid- and a coverslip is applied. Each square well in
ity or absorbancy or optical density and has the the grid then holds a known volume of liquid.
symbol OD (for optical density) or A (for absor- The number of cells is counted microscopically,
bancy). Thus turbidity or OD is directly propor- and the number of cells per milliliter is obtained
tional to cell mass in the culture (or cell density, by using a conversion factor based on the total
if the ratio of mass to cell remains constant) and number of square wells counted and their vol-
is written as follows: ume. Although this is a simple procedure, and
one used widely, it has two limitations:
OD = A = xl (2.2)
1. It does not distinguish between live and dead
Turbidity is measured with a colorimeter or a
cells.
spectrophotometer. In practice, one constructs
2. It cannot be performed on populations
a standard curve by measuring the turbidity of
whose cell density is too low to count micro-
several different cell suspensions, counting the
scopically (<106 cells/mL).
cell number independently and using it as a mea-
sure of cell mass. A straight line is generated, Electronic cell counting is a more sophisticated
as shown in Fig. 2.2. However, the line devi- method. The bacteria, suspended in a saline solu-
ates from linearity at high cell densities. This is tion, are placed in a chamber with an electrode,
because when the cell density is too high, some separated by a second chamber filled with the
of the light is rescattered and directed toward the same saline solution and also provided with an
phototube, thus lowering the turbidity reading. electrode. A microscopic pore separates the two
Whenever turbidity measurements are made chambers. The bacterial suspension is pumped
on an unknown sample, a standard curve must through the pore into the second chamber.
be constructed to determine the cell density. Whenever a bacterium passes through the pore,
Turbidity measurements are the simplest way to the electrical conductivity of the circuit decreases
measure growth. Methods to determine the cell because the electrical conductivity of a bacterial
density by direct cell counts are described next. cell is less than that of the saline solution. This
change in conductivity results in a voltage pulse
2.1.2 Total cell counts that is counted electronically. The electronic
If the mass per cell is constant, one can measure counter has the advantage of allowing one also
growth by sampling the culture over a period to measure the size of the bacterium. This is
of time and counting the cells. There are two because the size of the pulse is proportional to
ways to do this: total cell counts and viable the size of the bacterium. Thus, one can obtain
cell counts. For total cell counts one can use a both a total cell count and a size distribution.
growth and cell division 57
2.1.3 Viable cell counts molecules, the regulation of the timing of DNA
A viable cell count is one in which the cells are synthesis and cell division, and such adaptive
serially diluted and then deposited on a solid physiological responses to nutrient availabil-
growth medium. Each viable cell grows into a ity as changes in gene expression, homeostasis
colony and the colonies are counted. Viable cell adaptations to the external environment, and
counts are routinely performed in microbiology the coupling of the rates of biosynthetic path-
laboratories. However, it is important to recog- ways to the rates of utilization of products for
nize that they may underestimate the number of growth (metabolic regulation). From this point
viable cells for the following reasons: of view, most of this text is concerned with the
physiology of growth, and we therefore return to
1. A viable cell count will underestimate the the topic in subsequent chapters. What follows
number of live cells if the cells are clumped, here is an introduction to growth physiology as
since each clump of cells will give rise to a it pertains to the synthesis of macromolecules
single colony. during steady state growth, and very gen-
2. Some bacteria plate with poor efficiency; eral adaptations of cells to nutrient depletion.
that is, single cells give rise to colonies with We begin with a discussion of the sequence of
low frequency. growth phases through which a population of
bacteria progresses after being inoculated into a
2.1.4 Dry weight and protein flask of fresh media.
The most direct way to measure growth is to
quantify the dry weight of cells in a culture. The
2.2.1 Phases of population growth
cells are harvested by either centrifugation or
When one measures the growth of populations
filtration, dried to a constant weight, and care-
of bacteria grown in batch culture, a progres-
fully weighed. Since protein increases parallel
sion through a series of phases can be observed
with growth, one can also measure growth by
(Fig. 2.3). The first phase is frequently a lag
doing protein measurements of the cells. The
phase, in which no net growth occurs (i.e., no
cells are harvested by either centrifugation or
increase in cell mass). This is followed by a
filtration or (more usually) precipitated first
phase of exponential growth, in which cell mass
with acid or alcohol, then recovered by centrif-
increases exponentially with time. Following
ugation or filtration. A simple colorimetric test
exponential growth, the culture enters the sta-
for protein is then performed.
tionary phase, a phase of no net growth. After
a stationary period, cell death occurs in a final
2.2 Growth Physiology stage called the death phase. Notice in Fig. 2.3
The vast topic of growth physiology includes that, prior to the exponential phase of growth,
the regulation of rates of synthesis of macro- when cell division occurs, the cells increase in
Fig. 2.3 Growth kinetics in batch culture: solid line, mass; dashed line, viable cells. Note that if we define
growth as an increase in mass, then only the solid line accurately reflects the growth of the culture. The dashed
line reflects growth only when it is parallel to the solid line.
www.ebook777.com
mass (solid line); that is, they grow larger. At and accumulation of toxic products (e.g., alco-
the end of exponential growth, the cells con- hols, solvents, bases, acids). The accumulation
tinue to divide after growth has ceased (dashed of toxic products is frequently a problem for
line), but they become smaller. It would seem fermenting cells because instead of being con-
to be an advantage to a population of bacteria verted to cell material, most of the nutrient is
to continue to divide after growth has ceased in excreted as waste products. The excretion
the population, since in this way more cells can of fermentation end products is discussed in
be produced for distribution to new sites, where Chapter 15. For a review of the physiology of
conditions for continued growth may be better. stationary phase cells, read ref. 1.
One consequence of the uncoupling of growth
from cell division during the lag and late log Death
phases is that the size of the cell varies during Cell death can result from several factors.
growth in batch culture. For the investigator, Common causes of death include the depletion
the noncoincidence of growth and cell division of cellular energy and the activity of autolytic
during stages of batch growth has a practical (self-destructive) enzymes. Some bacteria begin
consequence: namely, cell counts are not always to die within hours of entering the stationary
a valid measurement of growth. phase. However, many bacteria remain viable
for longer periods. For example, some bacte-
Lag phase ria sporulate or form cysts when exponential
When cells in the stationary phase of growth growth ceases. The spores and cysts are resting
are transferred to fresh media, a lag phase cells that remain viable and will germinate in
often occurs. In this situation, the lag phase is fresh media. As discussed next, even nonsporu-
due basically to the time required for the physi- lating bacteria can adapt to nutrient depletion
ological adaptation of stationary phase cells in and remain viable for long periods in stationary
preparation for growth. Usually the longer the phase.
cells are kept in the stationary phase, the lon-
ger is the lag phase when they are transferred to
fresh media. Some of the physiological changes 2.2.2 Adaptive responses to
that occur in the stationary phase are described nutrient limitation
in Section 2.2.2. In the natural environment bacteria are fre-
There are several possible reasons for the lag quently faced with starvation conditions and
phase. In some cases the lag phase accommo- enter intermittent periods of no growth or very
dates the requirement for time for the cells to slow growth. In some environments the gen-
recover from toxic products of metabolism, such eration time may be many days or even months
as acids, bases, alcohols, or solvents, that may because the nutrient levels are so dilute. When a
accumulate in the external medium. Sometimes, bacterial culture faces the depletion of an essen-
new enzymes or coenzymes must be synthesized tial nutrient, the culture may keep growing by
before growth can resume. Such synthesis will be inducing specific uptake systems to scavenge
necessary if, for example, the fresh medium is dif- the environment for the nutrient or a source
ferent from the inoculum medium and requires thereof, and/or it may induce the synthesis of
a change in the enzyme composition of the cells. enzymes that can use an alternative source of
If significant cell death occurs in the stationary the nutrient. For example, bacteria starved for
phase, an apparent lag phase will be measured inorganic phosphate may induce the synthesis
because the inoculum includes dead cells that of a high-affinity inorganic phosphate uptake
contribute to the turbidity. The lag phase can be system as well as the synthesis of enzymes capa-
avoided if the inoculum is taken from the expo- ble of degrading organic phosphates to release
nential phase of growth and transferred to fresh inorganic phosphate (phosphatases). (See the
medium of the same composition. discussion of the Ntr and PHO regulons in
Sections 19.4 and 19.5, respectively.) If the
Stationary phase bacteria cannot bring into the cells the essential
Cells stop growing and enter the stationary nutrient, then the population will stop growing
phase for various reasons. Among these are and dividing. Under these circumstances, the
exhaustion of nutrients, limitation of oxygen, population enters stationary phase or, in the
case of certain bacteria, the cells may sporulate from self-digestion of cell material, including
or encyst. (See Section 23.3 for a description of the cell envelope. Some gram-negative bacteria
sporulation in B. subtilis.) (e.g., Pseudomonas) bleb off outer membrane
Lately, increased attention is being given to vesicles as they become smaller. Frequently, the
physiological changes that occur in bacteria decrease in size can be accompanied by a change
that enter the stationary phase when they are from rod-shaped cells to coccoid-shaped cells,
experimentally subjected to starvation.2–5 These as discussed next.
bacteria undergo physiological changes that
result in metabolically less active cells that are Morphological changes
more resistant to environmental hazards. This Along with a reduction in size, some bacteria
property has been associated for a long time (e.g., Arthrobacter) change from rod-shaped
with bacteria that sporulate or form desicca- cells to coccoid cells in stationary phase. Similar
tion-resistant cysts upon nutrient deprivation. morphological changes from rods to small
The spores and cysts represent metabolically coccoid shapes can occur upon starvation in
inactive or less active stages of the life cycle of several other bacteria, including Klebsiella,
the organism. When nutrients become avail- Escherichia, Vibrio, and Pseudomonas.
able once more, the spores and cysts germi-
nate into vegetative cells that grow and divide. Changes in surface properties
More recently, it has been discovered that when When certain marine bacteria are starved, the
faced with nutrient deprivation, even some cell surface becomes hydrophobic and the cells
bacteria that do not form spores (e.g., E. coli, are more adhesive.8 Vibrio synthesizes surface
Salmonella, Vibrio, Pseudomonas) undergo fibrils and forms cell aggregates when starved
adaptive changes and thereupon enter station- for a long time. These changes in the surfaces
ary phase. Some of these changes also result of starved cells make them more adhesive.
in resistant, metabolically less active cells, and Presumably, it is advantageous for nutrient-
such responses may be common in most if not limited cells to adhere to particles that have
all bacteria. However, not all the effects can be adsorbed nutrient on the surfaces.
rationalized in terms of survivability, and their
physiological role is not yet known. Some of
Changes in membrane phospholipids
the changes that occur in cells that are starved
In E. coli all the unsaturated fatty acids in the
are described next. For a review of starvation in
membrane phospholipids become converted
bacteria, see ref. 6.
to the cyclopropyl derivatives by means of the
methylation of the double bonds. The advan-
Changes in cell size tage to the cyclopropane fatty acids is not
As discussed earlier, cells that enter stationary known, since it does not appear that mutants
phase upon carbon exhaustion generally become unable to synthesize cyclopropane fatty acids
smaller because they undergo reductive divi- are at a survival disadvantage when faced with
sion; that is, they keep dividing for 1 to 2 h after environmental stresses such as starvation and
growth has ceased. Reductive division results high or low oxygen tension.
in the production of more cells, which may be
advantageous for dispersing the population. In Changes in metabolic activity
some cases there may be several cell divisions in When bacteria enter stationary phase, their
the absence of growth so that the cell size differs overall metabolic rate slows. In addition, during
radically from what is found in the growing cell. starvation many bacteria experience a signifi-
Some bacteria decrease in length from several cant increase in the turnover (metabolic break-
micrometers (e.g. 5–10 μm) to approximately down and resynthesis) of protein and RNA.
1 to 2 μm or even less. (See note 7.) As discussed Presumably, in starved cells the protein and
in ref. 1, in addition to undergoing size reduc- ribosomal RNA can serve as an energy source
tion due to continued division in the absence to maintain viability and crucial cell functions.
of growth, bacteria can become smaller during The latter would include solute transport sys-
starvation after reductive division has been com- tems, an energized membrane, and ATP pool
pleted. The additional decrease in size results levels.
www.ebook777.com
Changes in protein composition starvation sigma factor. It is also known as σ38

Bacteria may synthesize 50 to 70 or more new because it has a molecular weight of 38 kDa.
proteins under conditions of carbon, nitrogen, Sometimes called the “master regulator of the
or phosphate starvation. Many of the proteins stationary phase response,” RpoS has been
serve specific functions related to the nutrient found in other gram-negative bacteria in addi-
that is in low concentration. For example, phos- tion to E. coli, but not in gram-positive bacte-
phate starvation induces the synthesis of PhoE ria.12 Gram-positive bacteria use a different
porin, which is an outer membrane channel for sigma factor (σB) to regulate the transcription of
anions, including phosphate. This helps the genes homologous to some of the σs -dependent
bacterium bring in more phosphate. Another genes in E. coli.13
example is the nitrogen fixation genes that are Proteins whose synthesis depends upon σs
induced when the cells are starved for nitrogen. include a catalase made during stationary
In addition to the proteins of known function, phase (HP II), which presumably accounts for
there are many proteins made under all condi- resistance to H2O2, an exonuclease III that can
tions of starvation for which there is presently repair DNA damage due to H2O2 and near-UV
no known function, although many of the pro- radiation, and an acid phosphatase. Because
teins are presumed to be involved in resistance wild-type rpoS mutants do not have station-
to stress. ary phase resistance properties (measured as
percentage survival) to heat, high salt (osmotic
Changes in resistance to shock), near-UV, H2O2, or prolonged starva-
environmental stress tion (e.g., several days of starvation for carbon
Cells entering stationary phase also become or nitrogen), it has been concluded that the
more resistant to environmental stresses such products of rpoS-dependent genes are necessary
as high temperatures, osmotic stress, and cer- for survival under stressful conditions during
tain chemicals. For example, E. coli reportedly this phase.8 The transcription of the rpoS gene
becomes more resistant in stationary phase to (measured by using a rpoS–lacZ fusion plas-
high temperature, hydrogen peroxide, high salt, mid) increases dramatically in a nutrient-rich
ethanol, solvents such as acetone and toluene, medium as cells approach and enter stationary
and acidic or basic pH. These resistant prop- phase and during starvation for specific nutri-
erties are due to the synthesis of a starvation ents such as phosphate.14,15 (See note 16 for a
sigma factor that has four names: RpoS, σs, σ38, description of lacZ gene fusions.) Homologues
and KatF. to rpoS exist in other gram-negative bacteria,
including pathogens. Mutations in rpoS in the
1. The rpoS gene in E. coli encodes a intestinal pathogen Salmonella result in the
sigma factor required for transcription of attenuation of virulence in mice.17,18 One might
genes expressed during stationary phase suppose that RpoS aids pathogens to survive
and starvation stress such as oxidative stress, which they might
The reasons for concluding that rpoS encodes encounter inside host cells, or pH stress (e.g., in
a sigma factor are as follows: (1) the predicted the stomach).
amino acid sequence of the rpoS gene product,
based upon nucleotide sequence data, suggests 2. RpoS is a global regulator that is important
that it is a sigma factor, and (2) studies with for stress responses in exponentially growing
purified protein have confirmed that the protein cells as well as in stationary phase
binds to RNA polymerase and directs transcrip- The RpoS subunit of RNA polymerase should be
tion of rpoS-dependent genes. RpoS is required viewed as a global regulator that also functions
for the transcription of a regulon that includes during responses to stress during exponential
at least 50 genes that encode proteins induced growth.19 Increased RpoS levels during expo-
by carbon starvation when cells enter station- nential growth can be caused by slow growth,
ary phase; it is also necessary for the transcrip- temperature upshift from 30 °C to 42 °C, and
tion of several genes whose activities increase high osmolarity. For example, during osmotic
during phosphate and nitrogen starvation.9–12 upshifts in exponentially growing cultures
Accordingly, the protein is also called σs, for RpoS levels increase. Such cultures become not
only osmotolerant but also thermotolerant and limitations of carbon, phosphate, or nitrogen
resistant to H2O2. sources. How bacteria survive such nutritional
stress is fundamental to an understanding of
3. Regulation of RpoS levels can be at the their physiology in natural ecosystems.
transcriptional, post-transcriptional,
and post-translational levels
The control of ribosomal RNA synthesis
Interestingly, the reason for the increased lev-
There is much controversy surrounding the
els of RpoS during stress responses is not sim-
question of what regulates the rate of synthesis
ply increased transcription of the rpoS gene.
of ribosomes, although it is clear that ribosome
The levels of RpoS are regulated not only at the
synthesis is coupled to growth rates (Section
transcriptional level but also at the post-tran-
2.2.3).20 As shown in Fig. 2.4, faster growing
scriptional and post-translational levels. Thus,
cells have more ribosomes (reflected in cellular
the regulation of RpoS levels is very complex.
RNA) per unit cell mass. For example, the num-
For example, the rate of degradation of RpoS
ber of ribosomes in an E. coli cell can vary from
decreases when cells enter stationary phase.
fewer than 20,000 to about 70,000 depending
The half-life of RpoS in exponential cultures
upon the growth rate. In addition to growth
of E. coli growing at 37 °C is less than 2 min.
rate control, amino acid starvation results in the
However, the half-life is greater than 30 min
inhibition of ribosome synthesis. Probably sev-
when cells enter stationary phase. See Box 2.1
eral factors are involved in regulating rates of
and the references therein for discussion of how
ribosome synthesis. For example, Fis, a DNA-
the levels of RpoS are adjusted, including tran-
binding protein whose synthesis is increased in
scriptional regulation of the rpoS gene, regula-
rich media, positively regulates transcription of
tion of translation of the RpoS messenger RNA,
rRNA genes. Another DNA-binding protein,
and the role of proteases in the degradation of
H-NS, antagonizes Fis stimulation (see later).
RpoS.
A very important regulator is guanosine tetra-
phosphate (ppGpp), which increases when cells
Other regulators for stationary phase are subjected to amino acid starvation or carbon
gene expression and energy limitation, and slows the synthesis
Although σs is certainly a major transcription of rRNA (and tRNA). The global regulator
factor for gene expression during station- ppGpp is best known as the effector that inhib-
ary phase, additional regulatory factors exist its rRNA and tRNA synthesis during the strin-
(reviewed in ref. 5). This conclusion is sug- gent response, an effect originally discovered as
gested partly because the variation in the tim- the inhibition of rRNA and tRNA synthesis due
ing of expression during stationary phase of to amino acid starvation.21 See note 22 for other
σs-dependent genes, depending upon the gene physiological effects of ppGpp and ref. 23 for a
in question, clearly points to regulatory fac- recent review.
tors in addition to σs. In fact, there are several
σs-dependent genes that are also controlled 1. The stringent response
by other global regulators, both positive and The stringent response in E. coli is a temporary
negative. One of these, cyclic AMP (cAMP) inhibition in the synthesis of ribosomal RNA
receptor protein (CRP) complex, stimulates and transfer RNA when the cells are shifted to a
the expression of approximately two-thirds medium in which they are starved for an amino
of the genes expressed during carbon starva- acid. It works in the following way. When
tion. Furthermore, not all genes whose expres- cells are shifted from a medium rich in amino
sion increases during stationary phase require acids to a minimal medium, “uncharged”
σs. It is clear from the available data that gene tRNA accumulates, causing an insufficiency of
expression during stationary phase is complex aminoacylated tRNA, which in turn results in
and poorly understood at the present time. “idling” or “stalled” ribsomes. An enzyme on
Nevertheless, expression during stationary the “stalled” ribosomes called RelA [(p)ppGpp
phase is extremely important because bacteria synthetase I, or PSI], which is the product of
in their natural habitat are frequently in this the relA gene, becomes activated by uncharged
phase or growing extremely slowly owing to tRNA that binds to the A site. Activated RelA
www.ebook777.com
BOX 2.1 TRANSCRIPTIONAL, TRANSLATIONAL, AND

POST-TRANSLATIONAL REGULATION OF LEVELS OF rpoS
The regulation of transcription of rpoS ClpXP. This, then, is an example of trans-
(katF) appears to be complicated and is lational and post-translational control over
not well understood. There are probably the levels of a protein.
several signaling pathways that influence The stability of the RpoS protein increases
the transcription of the rpoS gene, allow- during stationary phase. Apparently RpoS is
ing the cells to respond to a variety of envi- sensitive to proteolytic digestion by ClpXP
ronmental challenges. The transcription during exponential growth but less so dur-
of rpoS does not always increase marking stationary phase. The evidence for this
edly when cells enter stationary phase is that RpoS is stabilized during exponen-
because expression during growth may be tial growth in clpP and clpX X mutants. For
high. What happens during transcription a discussion of the role of RpoS as well as
depends upon how the cells are grown. how the levels of RpoS can be regulated, see
Anaerobiosis has been reported to stimu- refs. 1 through 5.
late rpoS expression in exponential cells
grown in a rich medium. Several other fac-
REFERENCES
tors appear to influence the levels of rpoS
expression. For example, depending upon 1. Loewen, P. C., and R. Wenge-Aronis. 1994.
the strain of E. coli, rpoS expression may The role of the sigma factor σs (KatF) in bacte-
be high during exponential growth in a rial global regulation. Annu. Rev. Microbiol.
48:53–83.
minimal medium and increase only slightly
during stationary phase. Additionally, 2. Wilmes-Riesenberg, M. R., J. W. Foster, and
starvation induced by transferring expo- R. Curtiss III. 1997. An altered rpoS allele con-
nential phase cells to a medium lacking tributes to the avirulence of Salmonella typh-
imurium LT2. Infect. Immun. 65:203–210.
a specific nutrient might stimulate rpoS
expression significantly or only margin- 3. Schweder, T., K.-H. Lee, O. Lomovskaya, and
ally, depending on which component is A. Matin. 1996. Regulation of Escherichia coli
starvation sigma factor (σs) by ClpXP protease.
missing from the medium. J. Bacteriol. 178:470–476.
The amounts of RpoS also increase
4. Zhou, Y., and S. Gottesman. 1998. Regulation
in response to an upshift in osmolarity, of proteolysis of the stationary-phase sigma fac-
even in exponentially growing cells. The tor RpoS. J. Bacteriol. 180:1154–1158.
increase is due to an increase in translation 5. Venturi, V. 2003. Control of rpoS transcrip-
of the RpoS mRNA as well as to a decrease tion in Escherichia coli and Pseudomonas: why
in the turnover of RpoS due to the protease so different? Mol. Microbiol. 49:1–9.
synthesizes the signaling molecule ppGpp and, 2. Carbon and energy limitation cause
to a lesser extent, the pentaphosphate (i.e., an increase in the levels of ppGpp in a
pppGpp). Together they are referred to as (p) pathway independent of RelA
ppGpp. The ppGpp somehow slows the tran- The regulator (p)ppGpp also accumulates dur-
scription of ribosomal RNA (and transfer ing carbon and energy limitation on a pathway
RNA). This leads to an inhibition of ribosomal that occurs in relA mutants, hence is indepen-
protein synthesis by an interesting mechanism dent of RelA, and requires the product of the
called translational coupling, as explained in spoT gene [(p)ppGpp synthetase II, or PSII]. The
note 24. SpoT protein is most probably a bifunctional
Fig. 2.4 Macromolecular composition of E. coli grown at different growth rates. The values are amounts
per cell and have been normalized to values at a doubling time of 0.6 doubling/h. Source: Neidhardt,
F. C., J. L. Ingraham, and M. Schaechter. 1990. Physiology of the Bacterial Cell. Sinauer Associates,
Sunderland, MA.
enzyme that both synthesizes and degrades (p) medium and when stationary phase cells, which
ppGpp. The ratio of biosynthetic to degradative have very low Fis levels, are subcultured into a
activity is somehow regulated by carbon and nutrient-rich medium.28 One would expect an
energy limitation. This may account at least in increase in the rate of rRNA synthesis under
part for the lower number of ribosomes in cells these conditions, and Fis accounts in part for
growing slowly in poor medium. It is therefore the increased synthesis.
clear that the inhibition of rRNA and tRNA syn- Fis-dependent activation of ribosomal RNA
thesis coincides with essentially any nutritional genes is antagonized by another protein called
condition (not simply amino acid starvation) H-NS (also called H1), which can therefore act
that slows the growth rate. See note 25 for a as a repressor of rRNA transcription.29 (H-NS,
further discussion of the synthesis of (p)ppGpp discussed in Section 1.2.6 as part of the nucleoid,
and the contributions of RelA and SpoT. does not bind to specific nucleotide sequences
but does bind to curved or bent DNA.) The cel-
3. Other factors that control rRNA lular levels of H-NS are highest under conditions
synthesis: Fis, H-NS of slow rRNA synthesis (e.g., stationary phase).
It must be emphasized that the control of ribo- H-NS proteins belong to a remarkable group of
somal RNA synthesis is complex and that DNA-binding proteins that have other cellular
ppGpp is only one factor in the regulation. functions besides the regulation of rRNA gene
Other proteins involved in transcriptional transcription. (See Section 1.2.6 and note 30.)
regulation of the ribosomal RNA genes include So, although ppGpp somehow slows ribosomal
the Fis and H-NS proteins.26 (See note 27 for RNA synthesis, it appears to be only one com-
a further discussion of Fis.) Fis binds to DNA ponent of one or more complex regulatory
sequences upstream of the rRNA promoter and systems that must control ribosomal RNA syn-
activates the transcription of the operon. (Fis thesis under a variety of growth conditions.
regulates the expression, both positively and
negatively, of many genes besides rRNA genes.) 2.2.3 Macromolecular composition as a
Importantly, Fis synthesis is under growth rate function of growth rate
control. Synthesis increases when exponentially Much has been learned about growth physiol-
growing cells are shifted to a nutritionally richer ogy by observing changes in macromolecular
www.ebook777.com
composition of cells whose growth rates have Why faster growing cells have more mass
been altered by changing the nutrient composi- As seen in Fig. 2.4, cells that are grown at a
tion of the medium, or by manipulation of the faster growth rate have a greater mass (i.e., they
dilution rates of cells growing in continuous are larger). For example, E. coli when growing
culture (Section 2.4). at a doubling rate of 2.5 doublings per hour is
Figure 2.4 illustrates the changes that are due approximately six times larger in mass than
to differences in the nutrient composition of the when it is growing at 0.6 doubling per hour.
growth media as seen in the ratios of cell mass, Why is that? It is because, first, the initiation of
RNA, protein, and DNA of E. coli grown at DNA replication requires a minimum cell mass
increasingly faster growth rates. The ordinate per replication origin, called the critical cell
gives relative amounts of macromolecules per mass (Mi); in addition, faster growing cells have
cell, normalized to cells undergoing 0.6 doubling several replication origins, and for the shorter
per hour. The abscissa gives the growth rate μ, in doubling times (<60 min), replication must
units of doublings per hour. Notice that the mass begin in an earlier cell cycle; see note 33 for a
of the cell as well as the relative concentrations more complete explanation. (The significance
of the macromolecules increase as exponential of Mi was reported by Donachie in 1968.34 See
functions of the growth rate. As shown in Fig. note 35 for how this was done. Whether Mi is
2.4, faster growing cells have a higher ratio of an absolute constant or whether it decreases
RNA to protein, a larger mass, and more DNA. slightly with increasing growth rate is not
known with certainty.36,37 For this discussion,
Why faster growing cells have a larger
we will assume that it is constant.)
RNA-to-protein ratio
Faster growing cells are enriched for RNA 2.2.4 Diauxic growth
with respect to the other cell components.
Many bacteria, including E. coli, will grow
This reflects a higher proportion of ribosomes
preferentially on glucose when presented with
in faster growing cells, as well as the composi-
mixtures of glucose and other carbon sources.
tion of ribosomes: approximately 65% RNA
The result is a biphasic growth curve as shown
and 35% protein by weight. The reason for the
in Fig. 2.5 for glucose and lactose. Preferential
increase in ribosomes is that, over a wide range
growth on one carbon source before growth
of rapid growth rates, a ribosome polymerizes
on a second carbon source is called diauxic
amino acids at an approximately constant rate.
growth. The bacteria first grow exponentially
When a cell increases or decreases its growth
on glucose, then enter a lag period, and finally
rate, and therefore the number of proteins it
grow exponentially on the second carbon
must make per unit time, it adjusts the number
source.
of ribosomes rather than making each ribosome
work substantially slower or faster.
2.2.5 Catabolite repression by glucose
Why faster growing cells have more DNA It is now known that in many bacteria, glucose
Figure 2.4 also shows that faster growing cells represses the synthesis of enzymes required to
have more DNA per cell. The increased DNA grow on certain alternative carbon sources such
per cell is rationalized in the following way. In as lactose, and this can account for the diauxic
E. coli, the minimum time required between the growth discussed.
initiation of DNA replication and completion of Glucose also inhibits the uptake of certain
cell division (cell separation) for rapidly grow- other sugars into the cell. The result is that
ing cells (i.e., those with doubling times between growth on the second carbon source does not
20 and 60 min) is about 60 min.31 (See note 32 proceed until the glucose has been exhausted
for an explanation of the C and D periods.) If from the medium. During the lag period follow-
the generation time is shorter than 60 min, then ing the exponential growth on glucose, the bac-
DNA replication must begin in an earlier cell teria synthesize the enzymes required to grow
cycle and cells are born with partially replicated on the second carbon source. The repression of
DNA, which increases the average amount of genes by glucose is called catabolite repression,
DNA per cell. This is illustrated in Fig. 2.4, or glucose repression, and is indeed widespread
which shows that cells with a generation time of among bacteria. The inhibition of uptake of
less than 60 min have more DNA. alternative sugars is called inducer exclusion.
Fig. 2.5 Diauxic growth on glucose and lactose: x, bacterial mass, y, glucose or lactose concentration in the
medium. There are two phases of growth. Initially the cells grow on the glucose. When the glucose is suffi-
ciently depleted, a lag phase occurs followed by growth on lactose. During the lag phase the cells induce the
synthesis of enzymes necessary for growth on lactose. Glucose prevents the induction of the genes necessary
for growth on lactose. The inhibition by glucose is called catabolite repression and is discussed in Sections
19.10.1 and 19.10.2.
A rationale for glucose repression points out of succinate and glucose, Rhizobium shows
that glucose is one of the most common car- diauxic growth by growing on the succinate
bon sources in the environment and that bac- first and the glucose second. This is because
teria frequently grow more rapidly on glucose in these bacteria, succinate represses key
than on other carbon sources, especially other glucose-degrading enzymes of the Entner–
sugars. Therefore, bacteria in certain ecologi- Doudoroff and Embden–Meyerhof–Parnas
cal habitats that preferentially utilize glucose pathways.38 Pseudomonas aeruginosa offers
may be at a competitive advantage with respect another example of a species in which diauxic
to the other cells that depend upon glucose for growth occurs first on organic acids and then
carbon and energy. Additionally, the enzymes on carbohydrates (e.g., glucose). If presented
required to metabolize glucose are generally with glucose or other carbohydrates and any
constitutive (i.e., present under all growth con- one of several organic acids such as acetate,
dition), and the cell is prepared to grow on glu- or one of the intermediates of the citric acid
cose at any time. cycle, P. aeruginosa will grow on the organic
acid first and will repress the enzymes required
2.2.6 Catabolite repression by molecules to degrade the carbohydrate.39
other than glucose
However, not all bacteria preferentially grow
on glucose, nor are the glucose catabolic 2.3 Growth Yields
enzymes necessarily constitutive. Several When a single nutrient (e.g., glucose) is the
obligately aerobic bacteria, when given a sole source of carbon and energy, and when
mixture of glucose and an organic acid, will its quantity limits the production of bacteria,
grow first on the organic acid. This is also it is possible to define a growth yield constant,
called catabolite repression except that it is Y. The growth yield constant is the amount
the organic acid that is the repressor of glu- of dry weight of cells produced per weight of
cose utilization. One example is presented by nutrient used (i.e., Y is the weight of cells made
the nitrogen-fixing bacteria belonging to the divided by the weight of nutrient used); Y has no
genus Rhizobium. These bacteria, which live units, since the weights cancel out. [The molar
symbiotically in root nodules of leguminous growth yield constant is Ym, which is the dry
plants, grow most rapidly in laboratory cul- weight of cells produced (in grams) per mole
tures on C4 carboxylic acids (e.g., succinate, of substrate used.] For example, the Yglucose for
malate, and fumarate) that are part of the cit- aerobically growing cells is about 0.5, which
ric acid cycle. When presented with a mixture means that about 50% of the sugar is converted
www.ebook777.com
to cell material, and 50% is oxidized to CO2. X = x02Y (2.3)

For certain sugars and bacteria, the efficiency of
where x is anything that doubles each genera-
conversion to cell material can be much lower
tion, x0 is the starting value, and Y is the num-
(e.g., 20%). These differences are thought to be
ber of generations; x can be cell number, mass,
related to the amount of ATP generated from
or some cell component (e.g., protein, DNA).
unit weight of the carbon source.
Taking log10 of both sides,40 we write
The more ATP made, the greater the growth
yields. This makes sense if you consider that log x = log x0 + 0.301Y (2.4)
growth is, after all, an increase in dry weight, and
Let us define g as the generation time (i.e., the
a certain number of ATP molecules is required
time per generation). Therefore, g = t/Y, and
to synthesize each cell component. A value of
thus Y = t/g, where t is time elapsed.
10.5 g of cells per mole of ATP (called the YATP
Therefore, eq. 2.4 becomes
or molar growth yield) has been determined
for fermenting bacterial cultures growing on log x = log x0 + (0.301/g)(t) (2.5)
glucose as the energy source. The experiments
You will recognize this as an equation for a
are performed in media in which all the precur-
straight line. When x is plotted against t on
sors to the macromolecules (e.g., amino acids,
semilog paper (which is to base 10), the slope is
purines, pyrimidines) are supplied, so that the
0.301/g and the intercept is x0 (Fig. 2.6).
glucose serves only as the energy source, and
essentially all the glucose carbon is accounted The generation time, g
for as fermentation end products. The generation time, the time needed for the
Knowing the amount of ATP produced in population of cells to double, is an important
the fermentation pathways, it is possible to cal- parameter of growth and is probably the one
culate the YATP from the Yglucose. For example, most widely used by microbiologists. For this
22 g of Streptococcus faecalis cells is produced reason, the student should learn how to find the
per mole of glucose. This organism ferments a generation time. The simplest way is graphi-
mole of glucose to fermentation end products cally. In practice, one determines the generation
using the Embden–Meyerhof–Parnas pathway, time from inspection of a plot of x versus t on
which produces two moles of ATP (Section 9.1). semilog paper (Fig. 2.3 or 2.5) and simply reads
Thus, the YATP is 22/2 or 11. On the other hand, off the time it takes for x to double. Or, if x, x0,
Zymomonas mobilis produces only 8.6 g of and t are known accurately for the exponential
cells per mole of glucose fermented. These phase of growth, one can use eq. 2.3.
organisms use a different pathway for glucose
The growth rate constant, k
fermentation, the Entner–Doudoroff pathway,
The growth rate constant, k, is a measure of the
which yields only 1 ATP per mole of glucose
instantaneous growth rate and has the units of
fermented (Section 9.6). Thus, the YATP is 8.6.
reciprocal time. Another widely used parameter
As mentioned, a comparison of several differ-
ent fermentations by bacteria and yeast has
produced an average YATP of 10.5. Although
this value can vary, knowledge of the growth
yields and the YATP has allowed some deduc-
tions regarding which fermentation pathway
might be operating. Also, an unexpectedly high
growth yield in fermenting bacteria can point
to unrecognized sources of metabolic energy
(Section 4.8.3).
2.4 Growth Kinetics

2.4.1 The equation for exponential growth
During exponential growth, the mass in the cul-
ture doubles in each generation. The equation Fig. 2.6 Exponential growth plotted on a semilog
for exponential growth is scale: g, generation time (time per generation).
of growth, k is not to be confused with the gen- find 108 = x028 or x0 = 108/28 = 3.9 × 105 cells/
eration time, g, which is the average time needed mL. This means that the initial cell density in
for the population of cells to double. To clarify the growth flask should be 3.9 × 105 cells/mL.
the distinction, recall that since each bacterial Since, however, the cell density of the inoculum
cell gives rise to two cells, the rate at which the is 108/mL, the inoculum must be diluted 108/3.9
population is growing at any instant, that is, × 105 or 2.6 × 102 times. To grow 1 liter of cells,
the instantaneous rate of growth (dx/dt), must the inoculum size would have to be 3.8 mL (see
be equal to the number of cells at that time (x) note 42). Most routine growth problems can be
times a growth rate constant (k). Thus: solved with the following equations:
dx/dt = kx or dx/x = k dt (2.6) x = x02Y
Upon integration, we have Y = t/g
ln x = kt + ln x0 g(k) = 0.693
or
x = x0ekt (2.7) 2.4.2 The relationship between the
growth rate (k) and the nutrient
Taking log10 of both sides of eq. 2.7 converts concentration (S)
the expression from the natural log (loge) to log In the natural environment the concentration
base 10, so that x can be plotted against t on of nutrients is so low that the growth rates are
semilog paper: limited by the rates of nutrient uptake or the
log x = 0.4342(kt) + log x0 rate at which a stored nutrient is used. At very
low nutrient concentrations, the growth rate
or can be shown to be a function of the nutrient
log x = kt/2.303 + log x0 (2.8) (substrate) concentration (Fig. 2.7). The curve
approximates saturation kinetics and is proba-
Note that this gives the same line as in Fig. 2.6. bly due to the saturation, in the membrane, of a
However, the slope is equal to k/2.303. Since the transporter that brings the nutrient into the cell.
slope in Fig. 2.6 is equal to 0.301/g, it follows The curve can be compared with the kinetics of
that k/2.303 must be equal to 0.301/g. Thus solute uptake on a transporter, as described in
k = 0.693/g (or kg = 0.693). Therefore, k and Section 17.2.
g vary as the reciprocal of each other with the One can rationalize the kinetics by assuming
proportionality constant of 0.693. The slope of that at very low concentrations of nutrient, the
the curve in Fig. 2.6 can give you either k (the
instantaneous growth rate) or g (the doubling
time for the population).41
Using the growth equations
The growth equations can be used for finding g
and k. In another, more routine, circumstance,
one is growing a culture to be used at a later
time and must choose the proper size inocu-
lum. Convenient approaches to remember for
this calculation have been given (eq. 2.3 or 2.5).
Suppose you would like to subculture an expo-
nentially growing culture at a density of 108
cells/mL so that 16 h later the density of the new
Fig. 2.7 Variation of growth rate as a function of
culture will also be 108 cells/mL. If g = 2 h, what
substrate concentration: kmax, maximal growth rate
should x0 be? constant; k, specific growth rate constant; ks, nutri-
One way to do this problem is to estimate the ent concentration that gives ½kmax; S, nutrient con-
number of generations (Y) and then use eq. 2.3 centration. The concentrations at which growth rate
or 2.5. For example, since Y = t/g, the number is proportional to nutrient concentration are very
of generations is 16/2, or 8. By using eq. 2.3, we low, in the micromolar range.
www.ebook777.com
growth rate is proportional to the percentage growth chamber. For example, if the flow rate
of transporter that has bound the nutrient. At (F) is 10 mL/h and the volume (V) in the growth
high nutrient concentrations, the rate of nutri- chamber is 1 liter, then the dilution rate (D) is
ent entry approaches its maximal value because 10 mL/h per thousand milliliters, or 0.01 h–1.
all of the transporter has bound the nutrient. Notice that the units of D are reciprocal time. If
Therefore the growth rate becomes independent one multiplies the dilution rate D by x, the num-
of the nutrient concentration and is now limited ber of cells in the growth chamber, the product
by some other factor. The curve in Fig. 2.7 can Dx represents the rate of loss of cells from the
be described by the following equation, where outflow and has the units of cells/time.
k is the specific growth rate constant, kmax is the
Relationship of dilution rate (D) to
maximal growth rate constant, S is the nutrient
growth rate constant (k)
concentration, and Ks is the nutrient concentra-
In the steady state, k = D. To show this, consider
tion that gives 0.5kmax:
that the rate of growth is dx/dt = kx. When a
k = kmaxS/(Ks+ S) (2.9) steady state is reached in the growth chamber,
the rate of formation of new cells (kx) equals
These are the same kinetics used to describe the the rate of loss of cells from the outflow (Dx).
saturation of an enzyme by its substrate, that That is, kx = Dx, and therefore, k = D. Thus in
is Michaelis–Menten kinetics (Section 7.2.1). a chemostat, the growth rate of a culture can
Equation 2.9 can be used to calculate growth be changed merely by changing the flow rate, F,
yields during continuous growth (Section 2.3). which determines the dilution rate, D.
Dependence of cell yield, x, on concentration
2.5 Steady State Growth and of limiting nutrient in reservoir, Sr
Continuous Growth The actual concentration of cells in the growth
A culture is said to be in steady state growth (bal- chamber is manipulated by changing the con-
anced growth) when all its components double centration of limiting nutrient in the reservoir.
at each division and maintain a constant ratio
with respect to one another. Steady state growth
is usually achieved when a culture is maintained
in exponential growth by subculturing (i.e.,
when it is not allowed to enter stationary phase) medium reservoir
or during continuous growth.
2.5.1 The chemostat flow rate regulator

Continuous growth takes place in a device called
a chemostat (Fig. 2.8). A reservoir continuously air
feeds fresh medium into a growth chamber at a
flow rate (F) set by the operator. The chemostat
volume is kept constant. The concentration of
the limiting nutrient in the reservoir is kept very siphon overflow
low so that growth is limited by the availability growth chamber
of the nutrient. When a drop of fresh medium
enters the growth chamber, the growth-limiting
nutrient is immediately used up and the cells Fig. 2.8 A chemostat. The device works because the
cannot continue to grow until the next drop of growth of the cells in the growth chamber is limited
medium enters. This set of conditions allows by the rate at which a particular limiting nutrient
one to manipulate the growth rate by adjust- (e.g., glucose) is supplied. That is, the concentration
ing the rate at which fresh medium enters the of the limiting nutrient in the reservoir is low enough
growth chamber. to ensure that there is never an excess in the growth
chamber. At each moment when fresh medium
The dilution rate, D comes into the growth chamber, the incoming lim-
The dilution rate is equal to F/V, where F is the iting nutrient is rapidly utilized and growth cannot
flow rate and V is the volume of medium in the continue until fresh nutrient enters.
Let Sr be the concentration of limiting nutrient a very complex process, which must be care-
in the reservoir and S be the concentration of the fully coordinated. Most research attention has
nutrient in the growth chamber. The difference, been focused on the gram-negative bacteria E.
Sr – S, must be equal to the amount of nutrient coli (and the closely related Salmonella typh-
used up by the growing bacteria. If we define imurium), and Caulobacter crescentus, and the
the growth yield constant Y as being equal to gram-positive bacterium Bacillus subtilis.
the mass of cells (x) produced per amount of
nutrient used up (Sr – S), then Y = x/(Sr – S) or, 2.6.1 The period before septum
rearranging, formation
x = Y(Sr − S) (2.10) When E. coli reaches a critical cell mass during
growth, DNA synthesis is initiated. During rep-
where S can be calculated from Eq. 2.9 by lication, the sister chromosomes move toward
substituting D for k. Since S is much smaller opposite halves of the cell, so that when cell divi-
than Sr, it can be conveniently ignored in most sion takes place, each daughter cell receives a
calculations. chromosome. Rapidly growing E. coli requires
Varying the cell density (x) and growth rate approximately 40 min (the C period) to repli-
constant (k) independently cate its chromosome, followed by a period (the
Equation 2.10 predicts that if the concentra- D period) of about 20 min before cell division
tion of rate-limiting nutrient in the reservoir (Sr) is complete.
increases, the cell density in the growth cham-
ber will increase. When Sr increases, more nutri- Septum formation
ent enters the growth chamber per unit time. The first morphological sign of cell division is
Initially, S must increase. Since the growth rate the centripetal synthesis of a septum at mid-
is limited by the concentration of S (Fig. 2.7), cell, which forms by inward growth of the cell
the cells will respond initially by growing faster membrane and peptidoglycan layers. In E. coli,
and the cell density will increase. A new steady this begins soon after chromosome replication
state will be reached in which the increased is complete (i.e., at the end of the C period).
number of cells can and does use up all the avail- In many gram-negative bacteria, including E.
able nutrient as it enters. Thus, even though Sr coli, the outer envelope also invaginates at this
is increased, the dilution rate still controls the time (i.e., cell separation and septation occur
growth rate. What has been accomplished, at approximately the same time) so that septa
therefore, when Sr is increased, is a higher steady usually are not seen in thin sections of cells pre-
state value of x, according to eq. 2.10, but no pared for electron microscopy. Invagination is
change in the growth rate constant (k). There manifested as a constriction, which becomes
are two conclusions to be drawn: narrower as septation proceeds (Fig. 2.9A).
Initially there is a double layer of peptidoglycan
1. The only way to change the steady state
in the septum, but this is enzymatically split in
growth rate constant is to change the dilu-
half as constriction proceeds and the cells sepa-
tion rate (D), because this changes the rate
rate. Septation and peptidoglycan synthesis are
of supply of S.
coupled; that is, inhibitors of septal peptidogly-
2. The only way to change the steady state
can synthesis prevent septation.
growth yield is to change the concentration
In E. coli, septation occurs during approxi-
of limiting nutrient in the reservoir, Sr.
mately the first 10 to 13 min of the D period,
depending upon the growth rate (slower grow-
2.6 Cell Division ing cells have a longer period of septation).
Cell division is the splitting of a mother cell into Daughter cells separate within 7 min of comple-
two daughter cells separated by a septum. For tion of the septum.
bacteria that grow as single cells, cell division Cell division in B. subtilis differs from that in
is accompanied by, or followed by, cell separa- E. coli in that septation precedes cell separation,
tion. There are many proteins involved in the although cell separation may begin before sep-
division of a mother cell into two daughter cells tation is complete (Fig. 2.9B). The result of all
and their subsequent separation. This is clearly of this is that under electron microscopy, thin
www.ebook777.com
A E. coli outer membrane

cell membrane
B B. subtilis cell wall

cell membrane
Fig. 2.9 Cell division in gram-negative and gram-positive bacteria. (A) Many gram-negative bacteria divide
by constriction as the outer membrane invaginates with the cell membrane. Thus, septation and cell separa-
tion occur together. (B) In gram-positive bacteria, a septum consisting of cell membrane and peptidoglycan
forms, but there is generally little constriction in the early stages. Thus, septation precedes cell separation.
sections of dividing B. subtilis cells are seen to also important for septum formation. The FtsZ
have partially completed septa, and there is no ring with the other proteins is called a septal
visible constriction at the beginning of septation. ring. Figure 2.10 is a model of how the cell divi-
As in gram-negative bacteria, cell separation in sion proteins might be arranged at the site of
B. subtilis (and other gram-positive bacteria) is septation.
effected by the enzymatic splitting of the sep-
tum. Under certain conditions of rapid growth, 2.6.3 A model for the sequence of events in the
septum formation in B. subitilis can actually be assembly of the proteins at the division site
completed before cell separation begins, and as A model for the sequence of events in the
a consequence, chains of cells grow attached to assembly of the cell division proteins at the
each other via their septa. division site for E. coli postulates an ordered
sequence of events during which the prior
2.6.2 Proteins required for septum
formation and cell division
Most research in the area has focused on E. coli. LP
C
Several of the cell division genes are called fts OM
C
PBP3
genes, for filamentation temperature-sensitive N PG

N
phenotype. At the restrictive temperature, the PT
BT N
PP
cells do not divide but grow into long filaments N N FtsA
IM
that eventually lyse. There are several proteins CP
ZipA
ZipA
in E. coli known to be associated with the septal

ZipA
FtsZ ring
ring and essential for cell division. They include C
C
C
FtsZ, FtsA, ZipA, FtsK, FtsQ, FtsL, FtsB (also
called YgbQ), FtsW, FtsI (also called PBP3),
and FtsN. Fig. 2.10 Septal ring model in which a ring of FtsZ
FtsZ and FtsA are cytoplasmic, and the oth- proteins is connected to the cell membrane by ZipA.
Other abbreviations as follows: PT, FtsK and FtsW
ers are inner membrane proteins associated at
division proteins; BT, FtsL, FtsN, and FtsQ division
the constriction site or the septum. Importantly, proteins; PP, periplasm; CP, cytoplasm; LP, lipopro-
FtsZ forms a ring in the cell center. The FtsZ tein; OM, outer membrane; PG, peptidoglycan; IM,
ring is made before invagination and is believed inner (cell) membrane. Source: Hale, C. A., and P. A.
to contract, leading to the constriction of the J. de Boer. 1997. Direct binding of FtsZ to ZipA, an
cell in the center. As we shall see shortly, other essential component of the septal ring structure that
cell division proteins join the FtsZ ring and are mediates cell division in E. coli. Cell 88:175–185.
localization of the earlier added proteins is A variety of physiological and morphologi-

necessary for the recruitment of the later added cal changes take place in bacteria when they
proteins. First FtsZ forms the ring, with the are subjected to nutrient limitation and enter
help of ZipA.43 Then FtsA is recruited by FtsZ stationary phase. These may include the induc-
and joins the ring, followed by FtsK. Next to tion of specific uptake systems to scavenge the
join is FtsQ, followed by FtsL and FtsB; then environment for limiting nutrients or ions,
FtsW is added, followed by FtsI, and last FtsN. sporulation, or encystment. Even bacteria that
The assembly of FtsL and FtsB is codependent. do not sporulate or encyst undergo significant
FtsQ is found only in bacteria with a cell wall, physiological changes when they are starved.
and it may be involved with cell wall synthesis. Cells may become metabolically less active and
Shortly after or around the time of the addi- more resistant to environmental stresses such
tion of FtsN, constriction begins at the septal as heat, osmotic stress, and certain chemicals
site. In B. subtilis the situation is different, (e.g., hydrogen peroxide). Bacteria starved for a
inasmuch as the proteins are recruited to the required amino acid, or for a carbon and energy
septal ring in a concerted manner, rather than source (stringent response), may also undergo
sequentially. inhibition of ribosomal RNA synthesis, an out-
come that is correlated with increased synthe-
sis of guanosine tetra- and pentaphosphates.
2.7 Summary The response to starvation is mediated in part
Growth is defined as an increase in mass and by an increase in the transcription of certain
can be conveniently measured turbidometri- genes. The transcription of many of these genes
cally. Other methods can be used, provided they requires σs, also called σ38. Cells become smaller
measure something that parallels mass increase, and may change their morphology when they
such as the rate of increase in cell number (via- are starved. There may also be changes in the
ble or total) or specific macromolecules (e.g., chemistry of the cell surface.
protein, DNA, RNA). Steady state (continuous) When cells are grown at different growth rates
growth is defined as the growth of a population because of nutritional alterations or chemo-
of cells during which all the components of the stat growth, the macromolecular composition
cell double at each division. changes. Faster growing cells are larger, have
Growth in batch culture can progress through proportionally more ribosomes, and have more
a lag and a log to stationary phase, where net DNA per cell. This observation can be rational-
growth of the population (measured as mass or ized in terms of an approximate constancy of
its equivalent) has ceased. Eventually, the viable ribosome efficiency in protein synthesis, and an
cells may decrease in number, and this is referred almost constant period between the initiation
to as the death phase. The availability of nutri- of DNA replication and cell division.
ents, the need to synthesize specific enzymes to Diauxic growth is characterized by two
metabolize newly encountered nutrients, and phases of population growth separated by a
the accumulation of toxic end products in the stationary phase during which the cells are
medium can explain the different stages of the incubated with certain pairs of carbon and
growth curve. In addition, when cells enter the energy sources. In diauxic growth with glu-
stationary phase, nutrient depletion causes spe- cose, for example, the cells grow on the glucose
cific adaptive physiological changes to occur. first because it represses the expression of the
Growth during the log phase can be described genes necessary to grow on the second carbon
by a simple exponential equation depicting a source and because it prevents the uptake of
first-order autocatalytic process. That is, the other sugars. Repression under these condi-
mass doubles at each generation. From this tions is called catabolite repression or glucose
equation one can derive a generation time and repression. Prevention of sugar uptake is called
an instantaneous growth rate constant. These inducer exclusion. There are at least two pos-
constants are used to characterize growth under sible rationales for this. Glucose is the most
different physiological situations and to predict widely used carbon and energy source, and cells
growth yields at specific times for experimental are better able to outgrow their neighbors if
purposes. they use the glucose first. In doing so, moreover,
www.ebook777.com
they lower the supply of glucose to other cells. density will be 2 × 108/mL. Assume a genera-
Furthermore, many bacteria always express the tion time of 3.5 h. What should be the dilu-
genes to metabolize glucose (i.e., they are con- tion? What size inoculum should be used?
stitutive); the genes to metabolize other carbon
ans. 1/500 or 2 × 10–3; 2 mL
sources, however, are often not expressed unless
the carbon source is present in the medium. This 5. Assume the yield coefficient for glucose
lowers the energy burden of expressing genetic (Yglucose) is 0.5 g of cells per gram of glucose
information. However, glucose is certainly not consumed. In a glucose-limited chemostat,
a universal catabolite repressor among the bac- what should be the concentration of glu-
teria. For example, several obligately aerobic cose in the reservoir to produce a mass of
bacteria (e.g., Rhizobium) are known to grow cells in the growth chamber (x) of 0.1 mg/
preferentially on organic acids such as succinate, mL? For this problem, ignore S in eq. 2.9
malate, and fumarate, when given a mixture because it is much smaller than Sr.
of glucose and one of these acids. This makes
ans. 0.2 mg/mL
physiological sense, since they grow faster on
the C4 carboxylic acids than on glucose. 6. In a 500 mL chemostat, what should be the
Cells can be grown in continuous culture by flow rate (F) in milliliters per minute for a
using a chemostat. Two advantages to growing generation time (g) of 6 h?
cells this way are as follows: (1) the cells can
ans. 0.95 mL/min
be maintained in balanced growth and (2) the
growth rate constant can be easily changed sim- 7. Suppose you were operating a chemostat
ply by altering the flow rate. The growth yields with S as the limiting nutrient. Assume that
can be separately manipulated by changing the D is 0.2 h–1, Ks is 1 × 10–6 M, and kmax is
concentration of limiting nutrient in the reser- 0.4 h–1.
voir. Growth of continuous cultures is possible
a. What is the concentration of S in the
when the growth rate of the culture is limited by
growth chamber?
the supply of a nutrient that is continuously fed
into the growth chamber. b. If the cell density were 0.25 mg/mL,
what would be the concentration of S
in the reservoir for YS = 0.5?
Study Questions
c. What is the concentration of S in the
1. What is the generation time (g) of a culture growth chamber when the cells are grow-
with a growth rate constant (k) of 0.01 ing only half as fast (i.e., D = 0.1 h–1)?
min–1? ans. a. 1 × 10–6 M; b. 0.5 mg/mL; c. 3.3
ans. 69.3 min × 10–7 M
2. Assume you want to grow a culture to 108 8. During which phases of population growth
cells/mL in 3 h. The generation time is 30 would you not use cell number as an indica-
min. What should be the starting cell den- tor of growth?
sity (x0)? 9. What is the physiological role of RpoS?
ans. 1.6 × 106 When is it made?
3. Assume an inoculum whose cell density is
108/mL. The generation time is 30 min. If REFERENCES AND NOTES
you started with a 10–2 dilution, how many
hours would you have to grow the culture 1. Nyström, T. 2004. Stationary-phase physiology.
to reach 108/mL? Annu. Rev. Microbiol. 58:161–181.
ans. 3.3 h 2. Siegele, D. A., and R. Kolter. 1992. Life after log.
J. Bacteriol. 174:345–348.
4. Assume you have a stock culture at 5 × 109 3. Matin, A. 1991. The molecular basis of carbon-
cells per milliliter and you wish to inoculate starvation-induced general resistance in Escherichia
1 liter of fresh medium so that in 15 h the cell coli. Mol. Microbiol. 5:3–10.
4. Kolter, R., D. A. Siegele, and A. Tormo. 1993. the carboxy-terminal end from β-galactosidase.
The stationary phase of the bacterial life cycle. Annu. The hybrid protein has β-galactosidase activity.
Rev. Microbiol. 47:855–874. Therefore, an assay for β-galactosidase is a measure
of the expression of the target gene. Thus, one can
5. Hengge-Aronis, R. 1996. Regulation of gene
measure the expression of virtually any gene simply
expression during entry into stationary phase, pp.
by constructing the proper gene fusion and perform-
1497–1512. In: Escherichia coli and Salmonella:
ing an assay for β-galactosidase. One can construct
Cellular and Molecular Biology. F. C. Neidhardt et
gene fusions in vitro or in vivo. In vitro construction
al. (Eds.). ASM Press, Washington, DC.
involves using restriction endonucleases to cut from
6. Kjelleberg, S. 1993. Starvation in Bacteria. a plasmid containing the cloned DNA a portion of
Plenum Press, New York and London. the gene with its promoter region. The excised por-
tion is ligated to a lacZ gene, without its promoter,
7. This is especially obvious among the cytophages
or ribosome-binding site, in a second plasmid. The
and flexibacteria, the latter sometimes decreasing in
plasmid containing the fused gene is then intro-
length from over 100 μm to approximately 10 to
duced into the bacterium, and transformants are
30 μm.
selected on the basis of resistance to an antibiotic-
8. Kjelleberg, S., M. Hermansson, and P. Marden. resistant marker on the plasmid and the production
1987. The transient phase between growth and non- of β-galactosidase.
growth of heterotrophic bacteria, with emphasis
17. Fang, F. C., S. J. Libby, N. A. Buchmeier, P. G.
on the marine environment. Annu. Rev. Microbiol.
Loewen, J. Switala, J. Harwood, and D. G. Guiney.
41:25–49.
1992. The alternative sigma factor KatF (RpoS) reg-
9. McCann, M. P., J. P. Kidwell, and A. Matin. 1991. ulates Salmonella virulence. Proc. Natl. Acad. Sci.
The putative σ factor KatF has a central role in devel- USA 89:11978–11982.
opment of starvation-mediated general resistance in
18. Wilmes-Riesenberg, M. R., J. W. Foster, and R.
Escherichia coli. J. Bacteriol. 173:4188–4194.
Curtiss III. 1997. An altered rpoS allele contributes
10. Loewen, P. C., and R. Hengge-Aronis. 1994. The to the avirulence of Salmonella typhimurium LT2.
role of the sigma factor σs (KatF) in bacterial global Infect. Immun. 65:203–210.
regulation. Annu. Rev. Microbiol. 48:53–80.
19. Hengge-Aronis, R. 1996. Back to log phase: σs
11. O’Neal, C. R., W. M. Gabriel, A. K. Turk, S. J. as a global regulator in the osmotic control of gene
Libby, F. C. Fang, and M. P. Spector. 1994. RpoS is expression in Escherichia coli. Mol. Microbiol.
necessary for both the positive and negative regula- 21:887–893.
tion of starvation survival genes during phosphate,
20. Gourse, R. L., T. Gaal, M. S. Bartlett, J. A.
carbon, and nitrogen starvation in Salmonella typh-
Appleman, and W. Ross. 1996. rRNA transcription
imurium. J. Bacteriol. 176:4610–4616.
and growth rate-dependent regulation of ribosome
12. Hengge-Aronis, R. 2000. The general stress synthesis in Escherichia coli. Annu. Rev. Microbiol.
response in Escherichia coli, pp. 161–178. In: 50:645–677.
Bacterial Stress Responses. G. Storz and R. Hengge-
21. Cashel, M., D. R. Gentry, V. J. Hernandez, and
Aronis (Eds.). ASM Press, Washington, DC.
D. Vinella. 1996. The stringent response, pp. 1458–
13. Price, C. W. 2000. Protective function and regu- 1496. In: Escherichia coli and Salmonella: Cellular
lation of the general stress response in Bacillus subtilis and Molecular Biology, Vol. 1. F. C. Neidhardt et al.
and related gram-positive bacteria, pp. 179–197. In: (Eds.). ASM Press, Washington, DC.
Bacterial Stress Responses. G. Storz and R. Hengge-
22. In addition to an inhibition of ribosomal RNA
synthesis, ppGpp seems to be responsible for the fol-
14. Mulvey, M. R., J. Switala, A. Borys, and P. lowing:
C. Loewen. 1990. Regulation of transcription of
a. A large decrease in the rate of synthesis of pro-
katE and katF in Escherichia coli. J. Bacteriol. 172:
tein.
6713–6720.
b. A temporary cessation in the initiation of new
15. Gentry, D. R., V. J. Hernandez, L. H. Nguyen,
rounds of DNA replication.
D. B. Jensen, and M. Cashel. 1993. Synthesis of the
stationary-phase sigma factor σs is positively regu- c. An increase in the biosynthesis of amino acids.
lated by ppGpp. J. Bacteriol. 175:7982–7989.
d. A decrease in the rates of synthesis of phospho-
16. Gene fusions are valuable probes to monitor lipids, nucleotides, peptidoglycan, and carbo-
the expression of genes of interest. Consider a lacZ hydrates.
fusion. The fused gene has the promoter region of
e. A stimulation of development in the myxobac-
the target gene but not the promoter for the lacZ
terium Myxococcus xanthus.
gene. Expression of the fused gene is therefore under
control of the promoter region of the target gene. f. A stimulation of expression of virulence genes
The fusions produce a hybrid protein; its amino- in Salmonella. (See Pizzaro-Cerdá, J., and K.
terminal end is derived from the target gene and Tedin. 2004. The bacterial signal molecule,
www.ebook777.com
ppGpp, regulates Salmonella virulence gene is believed to be able to prevent translation of the
expression. Mol. Microbiol. 52:1827–1844.) entire operon when the protein is present in excess. It
is therefore a translational repressor.
g. Stimulation of the synthesis of sigma factor
σs, which is induced during starvation. (See The repressor ribosomal protein can bind not
Mulvey, M. R., J. Switala, A. Borys, and P. C. only to rRNA but also to the initiating region of the
Loewen. 1990. Regulation of transcription of mRNA of one of the genes (usually the first one) in
katE and katF in Escherichia coli. J. Bacteriol. the operon. It appears that in at least some instances
172:6713–6720.) the sequence in the mRNA to which the repressor
protein binds is similar to the sequence in the rRNA,
One must therefore conclude that the increase in lev-
thus accounting for the ability of the repressor ribo-
els of ppGpp and the responses due to those increases
somal protein to bind to both the ribosomal RNA
can be viewed as general responses to conditions
and the mRNA. As an example of such translational
that limit growth, not simply as results of amino acid
autoregulation, consider translation of the rplK–
starvation. This is because ppGpp levels rise as a con-
rplA operon, which encodes the 50S ribosomal sub-
sequence of any nutrient or energy limitation (e.g.,
unit proteins L11 (rplK) and L1 (rplA). The rplA gene
nitrogen limitation) or of a shift to a poorer carbon
is downstream of the rplK gene, and its translation
or energy source. Despite the obvious importance of
requires prior translation of the rplK gene. The L1
ppGpp or one or more metabolically related com-
protein regulates the translation of the entire operon.
pounds in mediating the stringent response, little is
It can bind to either rRNA or the mRNA in the trans-
known concerning how ppGpp is responsible for
lational initiation region of the rplK gene, and it can
the myriad effects that appear to be correlated with
block translation not only of the rplK gene but also of
increase of this molecule. One might think of ppGpp
the rplA gene that follows it. All the ribosomal pro-
as a second messenger that receives a signal from the
tein operons are regulated in a similar manner by one
environment (i.e., nutrient or energy depletion) and
of the protein products.
transfers the signal to the genome, either activating or
inhibiting transcription of relevant genes, or affect- 25. The ppGpp can be synthesized on ribosomes that
ing some other cellular process such as the activity of are “stalled” because of a restriction in the supply of
an enzyme. Of course, ppGpp need not act directly aminoacylated tRNA. This occurs during amino acid
on the target, and perhaps it is one component in a starvation (e.g., by restricting amino acids to aux-
longer signal transduction sequence. otrophs) and requires the product of the relA gene.
The relA gene was discovered in a search for mutants
23. Potrykus, K., and M. Cashel. 2008. (p)ppGpp:
that failed to respond to amino acid starvation with
Still magical? Annu. Rev. Microbiol. 62:35–51.
the stringent response. These mutants were termed
24. Ribosomal RNA and ribosomal proteins are “relaxed,” hence the name of the gene. The RelA
synthesized independently of one another and then protein [also called (p)ppGpp synthetase I or PSI] is
assembled into ribosomes. However, the synthesis of a ribosome-associated protein that synthesizes either
ribosomal proteins depends on continued synthesis ppGpp or pppGpp by displacing AMP and transfer-
of rRNA. Hence, a decrease in the synthesis of rRNA ring a pyrophosphoryl group from ATP to the 3′-OH
leads to an inhibition of ribosomal protein synthe- of either GDP or GTP:
sis. This is because certain ribosomal proteins, in the
absence of rRNA to which they normally bind, will GDP + ATP → ppGpp + AMP
bind to ribosomal mRNA and inhibit translation.
Because the pool size for GDP in E. coli is small
To understand this, it is important to recognize that
relative to GTP, it is believed that the product is
ribosomal proteins are encoded in operons in which
pppGpp, which is then dephosphorylated by a
the translation of the genes is coupled. E. coli has
5′-phosphohydrolyase (gpp) to ppGpp. This sugges-
at least 20 such operons that encode the ribosomal
tion is in agreement with the kinetics of appearance
proteins, and in certain instances also DNA primase,
of pppGpp and ppGpp. The synthetase is activated
RNA polymerase, and elongation factors.
in starved (stalled) ribosomes when aminoacylated
In translational coupling, the operon produces a tRNA becomes limiting during amino acid starvation.
polycistronic mRNA, and the translation of the gene The levels of (p)ppGpp may also rise upon carbon
immediately upstream is required for translation of and energy starvation (e.g., glucose starvation) even
the downstream gene. A model of translational cou- in null relA mutants, indicating alternative ways to
pling for the downstream gene postulates that the synthesize (p)ppGpp. Thus, even in a relaxed strain,
translational start region, including the start codon, RNA accumulation stops when the cells are shifted
is inside a hairpin loop in the mRNA and thus can- from a good carbon and energy source to a poorer
not bind to a ribosome. However, when the ribo- medium (but not when starved for an amino acid).
some translating the upstream gene comes to the stop This constitutes the evidence for a relA-independent
codon of the upstream gene, the ribosome disrupts (p)ppGpp synthetase, called PS II. PS II is thought to
the secondary structure of the mRNA and “opens be encoded by the spoT gene and is believed to be
it up,” making the translational start region for the a 3′-pyrophosphotransferase that uses GTP as the
downstream gene accessible to a second ribosome. acceptor and synthesizes pppGpp. SpoT is also a
One of the ribosomal proteins encoded by the operon pyrophosphohydrolase that degrades (p)ppGpp to
GDP and GTP. In other words, it has been proposed decreases (but less so) as the growth rate increases:
that SpoT is a bifunctional enzyme that takes part from 30 min at 0.6 doubling per hour to 23 min at 2.5
in either the synthesis or the degradation of ppGpp, doublings per hour. See: Bipatnath, M., P. P. Dennis,
and the ratio of biosynthetic to degradative activity is and H. Bremer. 1998. Initiation and velocity of
somehow controlled by energy starvation. The result chromosome replication in Escherichia coli B/r and
is an increase during energy starvation of ppGpp. For K–12. J. Bacteriol. 180:265–273. Bremer, H., and P.
a discussion of these points, see ref. 21. P. Dennis. 1996. Modulation of chemical composi-
tion and other parameters of the cell by growth rate.
26. Xu, J., and R. C. Johnson. 1995. Identification
In: Escherichia coli and Salmonella typhimurium:
of genes negatively regulated by Fis: Fis and RpoS
Cellular and Molecular Biology, 2nd ed., Vol. 2, pp.
comodulate growth-dependent gene expression in
1553–1569. F. C. Neidhardt et al. (Eds.). ASM Press,
Escherichia coli. J. Bacteriol. 177:938–947.
Washington, DC.
27. Fis negatively regulates the expression of the
fis gene and positively or negatively regulates the 33. For illustrative purposes, let us assume that the C
expression of several other genes. Two-dimensional and D periods are constants at 40 and 20 min, respec-
gel electrophoresis of fis mutants has revealed over tively, so that a cell divides 60 min after it has initi-
20 proteins whose synthesis is decreased (indicating ated DNA replication. We will also assume that each
Fis-dependent transcription) and a similar number of cell grows exponentially and that x = x02Y, where x0
proteins whose synthesis has increased (indicating Fis is the mass at birth and x is the mass at a point in the
repression). The evidence suggests that Fis binds to cell cycle where Y is the fraction of cell cycle time. For
both the rRNA DNA and the RNA polymerase and example, if the cell is halfway through the cell cycle,
stimulates transcription by interacting with the poly- x = x020.5. Let us call a cell dividing every 60 min a 60
merase rather than by causing an alteration in DNA min cell, a cell dividing every 40 min a 40 min cell,
structure such as bringing upstream DNA closer to and a cell that divides every 30 min a 30 min cell.
the polymerase. See refs. 20 and 26. The 60 min cell must be born with one chromosome
(and one replication origin) and will begin replicat-
28. Ball, C. A., R. Osuna, K. C. Ferguson, and R. C. ing that chromosome at the beginning of the cell cycle
Johnson. 1992. Dramatic changes in Fis levels upon so that 60 min later it divides. Let us say that the mass
nutrient upshift in Escherichia coli. J. Bacteriol. of the 60 min cell at the beginning of the cell cycle
174:8043–8056. when it begins replicating the chromosome is x0. This
29. Reviewed in: Wagner, R. 1994. The regula- means that the mass (x) 20 min later (when, as we
tion of ribosomal RNA synthesis and bacterial cell shall see, the 40 min cell begins DNA replication) is
growth. Arch. Microbiol. 161:100–109. about 1.15x0 (x = x020.2). The 40 min cell also needs
60 min between the initiation of a round of DNA
30. In addition to being a transcriptional activator replication and cell division. This means that the 40
for rRNA genes, Fis stimulates Hin-, Gin-, and Cin- min mother cell must initiate DNA synthesis at two
mediated inversion and site-specific recombination replication origins (one for each of the daughter cell
of phage λ (both excision and integration) with the chromosomes) 20 min into the cell cycle. Now if you
bacterial chromosome. Fis also binds to the E. coli compare the 40 min cell with the 60 min cell, you
origin of replication (oriC) and possibly plays a role will realize that the 40 min cell must on average be
in the initiation of DNA replication. H-NS is involved larger. This is because by 20 min, the 40 min cell must
in transcriptional regulation of several genes whose grow to a mass that will initiate replication at two
activity is modulated by cellular stress (e.g., gene reg- origins of replication; that is, it must be 2x0, where x0
ulation in stationary phase, osmotic shock). See the is the size of the 60 min cell at birth. Since that size is
discussion of the nucleoid in Section 1.2.6. reached by the 40 min cell halfway into the cell cycle,
31. Cooper, S., and C. R. Helmstetter. 1968. the 40 min cell must be 1.41 times larger than its mass
Chromosome replication and the division cycle of at birth (x = x020.5 = x01.41). That is, 2x0 is 1.41 times
Escherichia coli B/r. J. Mol. Biol. 31:519–539. larger than the mass at birth of the 40 min cell. The
mass at birth must therefore be 2x0/1.41, or about
32. The length of time between the onset of DNA 1.41x0. This can be contrasted to the 60 min cell that
replication and the completion of cell division is the was born with a mass of x0. Now let’s look at the 30
sum of two time periods [i.e., the time it takes for the min cell. To ensure that the daughter cells can divide
chromosome to replicate (the C period) and the inter- 60 min later, the 30 min cell must begin replicating
val between the end of a round of DNA replication the daughter chromosomes at the beginning of the
and the completion of cell division (the D period)]. cell cycle. Thus, replication must begin at two origins
The shortest C and D periods are approximately 40 (one for each of the daughter chromosomes), and the
and 20 min, respectively, in rapidly growing E. coli. mass at the beginning of the cell cycle must be 2x0 as
Hence 60 min is the minimum time required between opposed to 1.41x0 for the 40 min cell and x0 for the
the onset of a round of DNA replication and the com- 60 min cell.
pletion of cell division. These times do change with
growth rate. The C period decreases from approxi- 34. Donachie, W. D. 1968. Relationship between
mately 67 min at 0.6 doubling per hour to about 42 cell size and time of initiation of DNA replication.
min at 2.5 doublings per hour. The D period also Nature 219:1077–1079.
www.ebook777.com
35. Donachie assumed that each cell grew expo- 39. Collier, D. N., P. W. Hager, and P. V. Phibbs
nentially and doubled in mass during the cell cycle. Jr. 1996. Catabolite repression control in the
Knowing the size of the cell at the time of cell division, pseudomonads. Res. Microbiol. 147:551–561.
he was able to calculate the size at any time during the
cell cycle. With this information and the knowledge 40. Suppose you want to convert log2 to log10. First
that DNA initiation occurred 60 min before cell divi- you would write the exponential equation (e.g., x =
sion, he was able to compute the mass of the cell at the x02Y). Now take log10 of both sides of the equation:
time of initiation of DNA initiation. This mass, divided log10x = log10x0 + log102(Y). Since log102 = 0.301, the
by the number of replication origins, was a constant. equation becomes log x = log x0 + 0.301Y. Note that
log10 is usually written simply as log.
36. Wold, S., K. Skarstad, H. B. Steen, T. Stokke,
and E. Boye. 1994. The initiation mass for DNA 41. Some investigators use the symbol μ for the
replication in Escherichia coli K-12 is dependent on instantaneous growth rate constant and k for the
growth rate. EMBO J. 13:2097–2102. reciprocal of the generation time: k = 1/g.
37. Cooper, S. 1997. Does the initiation mass for 42. The dilution is 2.6 × 102. Assume the volume of
DNA replication in Escherichia coli vary with the inoculum is x. Therefore, 2.6 × 102(x) must equal the
growth rate? Mol. Microbiol. 26:1138–1143. final volume, which is 1000 mL + x. Solving for x
gives 3.8 mL.
38. Chandra, N. M., and P. K. Chakrabartty. 1993.
Succinate-mediated catabolite repression of enzymes 43. Davjkovic, A. Pirchoff, S. J. Lutkenhaus, and
of glucose metabolism in root-nodule bacteria. Curr. D. Wirtz. 2010. Cross-linking FtsZ polymers into
Microbiol. 26:247–251. coherent Z rings. Mol. Microbiol. 78: 651–668.
3
Chromosome Replication and
Partitioning of Chromosomes
From a biochemical point of view, nucleic acids The DNA strands are held together by
are polymerized by donating the subunit from base pairing between A–T and G–C residues
a donor with a high group transfer potential to (Fig. 3.2). DNA is replicated via semiconserva-
the growing chain via a nucleophilic displace- tive replication. This means that each strand
ment reaction. This condensation reaction is acts as a template for the synthesis of a daugh-
generally referred to as chain elongation. For ter strand. Semiconservative replication was
nucleic acids, the donors of the nucleotide suggested by Watson and Crick in their paper
monophosphates, (d)NMP, are the respective on the structure of DNA published in Nature
nucleotide triphosphates, (d)NTP. During the in 1953, and was later experimentally demon-
condensation reaction in nucleic acid biosyn- strated by Meselson and Stahl, as described in
thesis, the α phosphate in (d)NTP is attacked by Fig. 3.3.
the 3′-hydroxyl group on ribose or deoxyribose
(at the 3′ end of the growing polynucleotide)
and the released pyrophosphate (PPi) is subse-
quently hydrolyzed by a pyrophosphatase. This
chapter describes the steps in chromosome rep-
lication, the separation of chromosomes, and
the deliverance of chromosomes to daughter
cells.
3.1 DNA Replication, Chromosome

Separation, and Chromosome
Partitioning
3.1.1 Semiconservative replication
DNA consists of two strands wound around
each other in a right-handed double helix
(Fig. 3.1). For a review of the history of DNA
research that led to the realization that it was
the genetic material, see Box 3.1; for a summary Fig. 3.1 The direction of helical turns. To determine
of the contributions of Francis Crick, Rosalind whether the molecule is in a right- or left-handed
Franklin, James Watson, and Maurice Wilkins helix, sight down the molecule from one end. The
toward the elucidation of the three-dimensional right-handed helix turns clockwise, and the left-
structure of DNA, see Box 3.2. handed helix turns counterclockwise.
77
www.ebook777.com
BOX 3.1 HISTORICAL PERSPECTIVE:

THE DISCOVERY OF DNA AND ITS ROLE
The discovery of DNA (The S stands for smooth, which is how the
colonies look when grown on an agar plate,
The realization that DNA is the genetic and the R stands for rough, which describes
material was preceded by decades of patient the colonies formed by unencapsulated
investigation. After its identification in strains.)
1869 in the nuclei of human white blood Griffith found that if he heat-killed the
cells, DNA was later found in cells of many S bacteria, they did not cause pneumonia
types. Its chemical composition, as well unless they were injected with live R bac-
as that of RNA, was learned in 1910. By teria. Furthermore, he saw that all the bac-
the 1920s, microscopic studies of stained teria recovered from the dead mice had
preparations demonstrated that DNA, capsules. Thus, the R bacteria had been
along with protein, is contained in chro- transformed into S bacteria by dead S bac-
mosomes. However, since DNA is made teria. This is because genes to make capsule
of only four different nucleotides, and its were transferred from the dead S bacteria
enormous diversity was not suspected, this to the live R bacteria, a phenomenon now
nucleic acid was not then considered to be called transformation (since the R bacteria
the genetic material. In fact, it was protein are transformed into S bacteria).
that was thought to be the genetic material Avery, MacLeod, and McCarty were
because of the great diversity in proteins. It able to transform R bacteria into S bacteria
was thought that the DNA played a struc- in a test tube by adding DNA purified from
tural role in the chromosomes. S bacteria. They thus provided the first evi-
dence that genes were made of DNA. The
DNA preparations had small amounts of
DNA is the genetic material protein, and to prove the genetic material
was the DNA and not the protein, Avery
It was not until 1944 that experiments were and his coworkers showed that the trans-
published indicating that the genetic mate- forming activity was not destroyed by pro-
rial was indeed DNA. These experiments, teases, which are enzymes that degrade
aimed at investigating a report published protein, nor was the activity destroyed by
in 1928 by Fred Griffith, a British scien- an enzyme that degrades RNA. However,
tist, were done at the Rockefeller Institute the transforming activity was destroyed
in New York by Oswald Avery, Colin by deoxyribonuclease, an enzyme that
MacLeod, and Maclyn McCarty. Griffith, destroys DNA.
who had studied pneumonia caused by A second experiment, which also showed
the bacterium Streptococcus pneumoniae that the genetic material was made of
(at that time called Pneumococcus), had DNA, was published in 1952 by Alfred
described two strains of S. pneumoniae, Hershey and Martha Chase. Hershey and
called R and S. One difference between the Chase studied the infection of the bacte-
two strains was that when the S strain was rium Escherichia coli by the bacterial virus
injected into mice, the mice died of pneu- T2. When the virus infects the bacterium,
monia. However, when the R strain was the viral genes are injected into the bacte-
injected, the mice lived. Another difference rial cell, and those genes direct the synthesis
between the R and S strains was that the S of new viruses. Hershey and Chase demon-
strain had a capsule and the R strain did not. strated, via radioisotopes that labeled the
chromosome replication and partitioning of chromosomes 79
phosphorus in DNA and the sulfur in pro- Together, the experiments of Avery,
teins, that viral DNA and not viral protein MacLeod, and McCarty, and of Hershey
was injected into the cell, thus providing and Chase, convinced everyone that genes
evidence that its genes were made of DNA are made of DNA. (We now know that
and not protein. genes in some viruses are made of RNA.)
BOX 3.2 HISTORICAL PERSPECTIVE: THE STRUCTURE OF DNA

In 1953 James Watson and Francis Crick called the “A” form, which exists in solu-
at Cambridge University in England pub- tions that have relatively little water to
lished their model of DNA as a double hydrate the DNA. It was immediately clear
helix held together by hydrogen bonding to Watson from Franklin’s photograph that
between the bases. Based upon their model the B form was due to a helical structure.
of the structure of DNA, Watson and In 2004, in an article on the history of
Crick suggested that one of the two strands the discovery of the DNA double helix,2
serves as a template for a new complemen- Aaron Klug revealed that one of Rosalind
tary strand during DNA replication, and Franklin’s laboratory notebooks, written in
that each daughter cell receives a duplex 1952, showed that she also thought that the
consisting of a parental strand and a new B form was helical. Franklin’s photograph
strand. This process is known as semicon- helped Watson and Crick to construct the
servative replication, and its occurrence double-helix model of DNA that was pub-
was later demonstrated experimentally. lished in Nature in 1953. In 1962 Watson,
The Watson–Crick model of the structure Crick, and Wilkins received the Nobel Prize
of DNA was based in part upon the exami- for their contributions in deciphering the
nation of X-ray diffraction photographs of structure of DNA. Rosalind Franklin had
DNA taken by Rosalind Franklin, a research died of cancer in 1958, at the age of 37. For
fellow in the laboratory of Maurice Wilkins more information about the fascinating
at King’s College in London, as well as her history of the elucidation of the structure of
presentation at a colloquium in 1952 that DNA, read refs. 1 through 3.
showed that the phosphate groups were
on the outside of the DNA molecule and
the bases faced inward toward each other. REFERENCES
According to the account by Watson, in his 1. Watson, J. D. 1968. The Double Helix.
book The Double Helix,1 Wilkins showed Penguin Books Ltd., London.
Watson one of Franklin’s recent X-ray pho- 2. Klug, A. 2004. The discovery of the DNA
tographs of DNA molecules surrounded by double helix. J. Mol. Biol. 335:3–26.
a large amount of water. The DNA was in a 3. Kass-Simon, G., and P. Farnes (Eds.). 1990.
new form called the “B” form, which is far Women of Science. Indiana University Press,
simpler than the pattern obtained earlier, Bloomington.
www.ebook777.com
Fig. 3.2 Base pairing in DNA. (A) Two complementary strands of the double helix are in opposite polarity
and are held together by hydrogen bonds between A–T and G–C pairs: A, adenine; T, thymine; G, guanine;
C, cytosine; P, phosphate. The deoxyribose moieties are attached via phosphodiester linkages between the C3
hydroxyl of one sugar and the C5 hydroxyl of another. (B) A more detailed examination showing the structure
of the deoxyribose connected by phosphodiester linkages. The phosphate–oxygen double bond is drawn as a
semipolar bond because of the high electronegativity of oxygen and the low propensity of phosphorus to form
double bonds. (C) The structures of the bases, showing the hydrogen bonds. Note that two hydrogen bonds
hold the A–T base pairs together, whereas three hydrogen bonds hold the G–C base pairs together.
3.1.2 The topological problem in and more tightly coiled into a positive super-
DNA replication coil (Fig. 3.5). Unless something was done to
DNA replication is a complex topological prob- prevent overwinding, the unreplicated portion
lem because the DNA in a typical bacterium of the DNA helix would become bunched up
exists as a covalently closed circle of a right- in positive supercoils and further unwinding
handed double helix that may be 500 to 600 would stop. As described shortly, an important
times longer than the cell and is tightly folded enzyme, DNA gyrase, solves the problem.
into supercoiled loops. (However, not all bacte-
ria have circular chromosomes. See note 1.) Positive versus negative supercoiling
Supercoiling refers to the twisting of the DNA What does this mean? Positively supercoiled
double helix around its central axis (Fig. 3.4). DNA has more than 10.5 base pairs per heli-
To visualize supercoiling, think of a telephone cal turn (i.e., it is overwound); negatively super-
cord wound into secondary coils. (See note 2 for coiled DNA has fewer than 10.5 base pairs per
a further explanation of supercoiling.) helical turn (i.e., it is underwound). The DNA
Pulling apart of the duplex strands at the duplex itself is a right-handed helix, and there-
replication fork in the closed circle makes the fore positive supercoils are twisted in the same
unreplicated portion ahead of the replication direction as the helix (overwound). If the twist
fork twist so that the helix becomes overwound of the coil is counterclockwise (left-handed),
15 15
N N
A
15
N-15N
grow in 14N for one

generation
15 14 15 14
N N N N
15
B N/14N
continued
growth in 14N
15 14 14 14 15 14 14 14
N N N N N N N N
14
N/14N
15
C N/14N
Fig. 3.3. The Meselson–Stahl experiment (Meselson, M. and F. W. Stahl, The replication of DNA in
Escherichia coli, Proc. Natl. Acad. Sci. USA 44:671–82, 1958) that demonstrated that DNA is replicated
semiconservatively. (A) Cells are grown in 15N for many generations so that all the nitrogen in the DNA is
heavy (15N). The DNA is isolated and centrifuged to equilibrium in a cesium chloride (CsCl) density gra-
dient that separates the molecules according to their density. If both strands have 15N, the duplex DNA
sediments to a position near the bottom of the tube. (B) The 15N cells are then grown in 14N (light) media,
and after one generation the DNA is isolated and centrifuged in the CsCl gradient. Semiconservative rep-
lication predicts that “hybrid” DNA would be formed, one strand being labeled with 15N and the other
strand with 14N. The 15N/14N DNA occupies a position in the gradient higher than the 15N/15N DNA. (C)
Further growth in 14N results in the formation of “light” DNA (i.e., 14N/14N), which occupies the highest
position in the gradient.
the coil is negatively supercoiled (opposite to DNA ahead of the replication fork and converts
the helix), hence is underwound. them into negative supercoils (underwound
DNA). This activity is advantageous, since the
DNA gyrase solves the problem of duplex must unwind if DNA replication and
overwinding RNA transcription are to take place. As shown in
The problem of overwinding the double helix Fig. 3.6, DNA gyrase works by making a double-
during DNA replication is solved by using DNA stranded break in the DNA, passing an unbro-
gyrase, also called topoisomerase II. This enzyme ken portion through the gap, and resealing the
continuously removes the positive supercoils break. Note 3 provides more information about
(overwound DNA) that form in the unreplicated topoisomerases and how they work.
www.ebook777.com
Fig. 3.4 Supercoiled DNA. (A) When there are 10.5 base pairs per helical turn, the DNA is relaxed. Native
DNA is generally underwound (i.e., >10.5 base pairs per helical turn). Underwinding introduces a strain in the
molecule; and to reduce the strain, the molecule twists upon itself to form supercoils (B). Supercoiling result-
ing from underwound DNA is referred to as negative supercoiling. Supercoils will also form if DNA is over-
wound (i.e., >10.5 base pairs per turn). Supercoiling due to overwound DNA is called positive supercoiling. In
negative supercoiling the DNA is twisted in a direction opposite to that of the right-handed double helix, and
in positive supercoiling the DNA is twisted in the same direction as the right-handed double helix.
More problems to be solved so that each strand can act as a template.

Not only must the DNA be unwound and cop- However, the unwinding does not begin at just
ied rapidly (to replicate a chromosome in E. coli, any place in the DNA, but rather at a particu-
about 1,000 nucleotides per second must be lar site in the DNA duplex termed the origin,
polymerized, in about 40 min), but the strands which should be remembered as the oriC locus.
must be separated from each other without get- When the duplex is unwound, a Y is created.
ting entangled. Finally, the strands must be par- The arms of the Y are single stranded because
titioned into the daughter cells. We will begin the duplex has become unwound there; but the
with the “replication fork,” which serves in the region downstream of the juncture, where the
initiation of DNA replication. The first ques- two arms come together, is still double stranded
tion is, how is it created? This turns out to be (Fig. 3.7). The juncture should be remembered
a very complicated process indeed, involving as the replication fork. DNA is usually repli-
many different proteins. cated bidirectionally; that is, there are actu-
ally two replication forks (Fig. 3.8). (Note 12
3.1.3 Creating the replication fork explains how directionality can be determined.)
At least 30 different proteins are required to ini- Bidirectional replication, which halves the time
tiate replication and to replicate the DNA in E. needed to replicate the DNA molecule, gener-
coli. For more information, the student should ally takes place with phages, plasmids, bacteria,
consult refs. 4 through 11. and eukaryotic cells. The replication forks are
DNA replication takes place in a complex thought to be stationary, and the unreplicated
DNA-synthesizing “factory” called a repli- DNA appears to thread through them. (This is
some, which consists of many enzymes and discussed in Section 3.1.4.) The detailed steps in
proteins that will soon be described. Within the creation of a replication fork are described
the replisome there are replication forks cre- next. Before DNA synthesis can begin, a pre-
ated on the DNA, and this is where replication priming complex must form. The first step is the
takes place. (Clearly, DNA synthesis must be formation of an “open complex,” which devel-
related to the growth rate of the cell. Review ops into the “prepriming complex.”
Section 2.2.3 for a discussion of the relation-
ship between the timing of initiation of DNA Unwinding the duplex: Creation of the
replication and the growth rate.) So how is the prepriming complex
replication fork made? To make the replica- The prepriming complex is formed first and
tion fork, the DNA strands must first unwind it is created in two stages.9 In stage 1 the open
Fig. 3.5 Supercoiling ahead of the replication fork. As the template strands in the closed circle are pulled
apart, the duplex ahead of the replication fork overwinds as positive supercoils are formed. The twists in the
overwound regions are removed by DNA gyrase, which produces negative supercoils and underwinds the
duplex.
complex is formed, and in stage 2 the open influence nucleoid structure and gene expres-
complex develops into the prepriming complex. sion in Section 1.2.6. Within the origin (oriC)
DNA synthesis, which will be described later, there are multiple sites to which DnaA binds.
actually begins with the prepriming complex. Approximately 30 DnaA molecules bind to
the sites as the DNA wraps around a core of
1. Formation of the open complex DnaA molecules. Then, in an ATP-dependent
Creation of the open complex is initiated at reaction that is aided by HU, the adjacent A–T-
the origin of replication (oriC) with ATP and rich region at the 5′ end of the origin sequence
two DNA-binding proteins: DnaA and a his- unwinds to form the 45 bp open complex.
tonelike protein called HU (Fig. 3.9). See the However, something must be done to prevent
subsection entitled DNA-binding proteins the single strands from coming together again
www.ebook777.com
Fig. 3.6 The action of DNA gyrase: converting positive supercoils into negative supercoils. (A) A positively
supercoiled node; that is, the duplex is twisted around its central axis. For example, this happens during DNA
replication as the duplex is being unwound by helicase at the replication fork and the duplex ahead of the rep-
lication fork spontaneously becomes overwound by being twisted in a clockwise direction. (B) Both strands
are cut by DNA gyrase, and then the gyrase passes the uncut portion through the gap and the gap is sealed. (C)
The duplex is now twisted in the opposite direction; that is, it is negative supercoiled and underwound. DNA
gyrase, an ATP-dependent enzyme, is sometimes referred to as providing a “swivel” that allows the replica-
tion fork to continue.
Fig. 3.7. A replication fork. The two template strands on the left have become unwound and are being copied.
The unwinding is caused by DnaB protein, which is the replication fork helicase (not shown) and requires
ATP. Note that there are single-stranded regions near the fork. A DNA-binding protein, called SSB (not
shown), binds to the single-stranded regions, preventing them from coming together. Now (right) the duplex
has both unwound and double-stranded components.
to re-form the duplex. This task is accomplished bind to the DNA. However, DnaB does not bind
by single-stranded binding (SSB) proteins that to the DNA on its own but must be transferred
coat the strands. to the open complex from a DnaB:DnaC:ATP
complex. After binding to the DNA, DnaB
2. Formation of the prepriming complex further unwinds the strands bidirectionally to
The open complex unwinds into the preprim- form the prepriming complex, with two rep-
ing complex. The unwinding is performed by a lication forks ready for the initiation of DNA
protein called helicase (DnaB), which must first replication.
(Pol II), encoded by polB; DNA polymerase III

(Pol III), encoded by polC (dnaE); DNA poly-
merase IV (Pol IV), encoded by dinB; and DNA
polymerase V (Pol V), encoded by umuC. Here
is what they do.
1. DNA polymerase I has a role in DNA rep-
lication, which will be discussed later. (See
subsection entitled Attaching the Okazaki
fragments to each other.) DNA polymerase I
is also important for DNA repair.
2. DNA polymerase II functions in the repair
Fig. 3.8. Bidirectional replication of DNA: solid of damaged DNA. After UV damage, DNA
lines, template strands; broken lines, daughter polymerase II catalyzes a fast reinitiation of
strands. In most organisms and viruses, the DNA is synthesis called “replication restart,” and
replicated bidirectionally, as indicated by the arrows. then DNA polymerase III takes over.
Bidirectionality reduces the time required to replicate 3. DNA polymerase III replicates DNA at the
the DNA. Available data indicate that the replication fork. See note 14 for a description of the sub-
forks are stationary in a replication factory called units in DNA polymerase III.
a replisome, and the unreplicated DNA threads 4. DNA polymerase IV functions in repairing
through them.
damaged DNA.
5. DNA polymerase V functions during DNA
repair. DNA polymerase V inserts nucle-
Timing of initiation otides nonspecifically opposite lesions, such
Precisely what determines the timing of ini- as abasic sites and UV-induced thymine
tiation of bacterial DNA replication is still a dimers, in the template strand. The process
matter of speculation. As discussed in Section is called translesion synthesis, and it is asso-
2.2.3, the cell mass per chromosomal replica- ciated with increased mutations because
tion origin (as opposed to plasmid replication the nucleotides are inserted nonspecifically.
origins) at the time of initiation is constant, Because of the increase in mutation rate,
and all oriC origins (even plasmid origins), translesion synthesis is referred to as SOS
begin replication at the same time. This mass mutagenesis or error-prone repair.
is called the initiation mass or the initiation
volume, and some have suggested that it cor- The problem in synthesizing
responds to the activity of DnaA. Another strands of opposite polarity at the
question is, what prevents the newly repli- replication fork
cated origins from reinitiating in the same At the replication fork the DNA templates are
cell cycle? For a discussion of what regulates antiparallel; that is, one strand is 5′ to 3′ and
the initiation of DNA replication at oriC, see the other strand is 3′ to 5′. In other words,
note 13. the 5′-phosphate end of one strand is paired
with the 3′-hydroxyl end of the other strand
3.1.4 Replicating the DNA (Fig. 3.7). Since the DNA copy strand is antipar-
DNA polymerases allel to the template strand, this means that the
The enzymes that synthesize DNA by using a new strands of DNA must be of opposite polar-
strand of DNA as a template are called DNA ity too; that is, one is 3′–5′ whereas the other is
polymerases. Two important characteristics 5′–3′. That is, at the replication fork one of the
of DNA polymerases are (1) they can only copy strands has its 3′ end pointed toward the
extend DNA chains (i.e., they cannot initiate fork, whereas the other copy strand has its 3′
new ones), and (2) they add mononucleotides end pointed away from the fork. This presents a
to the 3′ hydroxyl of deoxyribose and therefore problem because the DNA polymerase remains
elongate nucleic acid only at the 3′ end. E. coli at the fork, whereas the growing end (the 3′ end)
has five DNA polymerases: DNA polymerase of one of the strands keeps moving further and
I (Pol I) encoded by polA; DNA polymerase II further away from the fork.
www.ebook777.com
oriC
binding site
A+T for DnaA
rich
HU DnaA
ATP
open complex
DnaB: DnaC: ATP
DnaC, ADP, Pi
SSB
prepriming complex
DnaB (helicase) SSB
Fig. 3.9. Creation of the prepriming complex. A multimer of about 20 DnaA proteins recognizes specific
sequences and binds to the duplex DNA at the origin adjacent to an A–T-rich region. The A–T region adja-
cent to the DnaA proteins then unwinds to form the open complex. This step is aided by the HU protein and
requires ATP. DnaB protein (helicase) then binds in a DnaC-dependent reaction and further unwinds the
duplex. Single-stranded DNA-binding proteins (SSBs) bind to the single strands behind the helicase, prevent-
ing the single strands from coming together again.
Okazaki fragments 3′ end then returns to the fork to initiate rep-

We have seen that the DNA polymerase man- lication of another short fragment. (See note
ages to remain at the replication fork and yet 15 for a description of the Okazaki experi-
synthesize a strand of DNA whose 3′ end keeps ments.) The strand copied in short fragments
moving further and further away from the rep- is called the lagging-strand template. The other
lication fork. How is this accomplished? As will strand, which is copied in one piece, is called the
be explained next, the answer lies in synthesiz- leading-strand template. As discussed later, in
ing the strand whose 3′ end faces away from the the subsection entitled Model explaining how
fork. The synthesis proceeds in short fragments the DNA polymerase III synthesizing the lag-
(about 1,000 nucleotides long), called Okazaki ging strand can stay with the replication fork,
fragments after two of the investigators who the DNA polymerases that synthesize the lead-
discovered them. As implied in Fig. 3.10, the ing and lagging strands are actually associated
Fig. 3.10 Synthesis of leading and lagging strands. Both template strands must be copied in the 3′-to-5′
direction so that the copy strands, which grow at the 3′ ends, are of opposite polarity to their respective
templates. Since the template strands are of opposite polarity, the copy strands must be synthesized in
opposite directions: one strand in the direction of replication fork movement and the other strand in the
opposite direction. However, it is known that the DNA polymerase, which is at the 3′ end of the growing
strand, remains at the replication fork. The strand whose 3′ end faces the replication fork is synthesized
by a polymerase that stays attached to the template and moves with the replication fork. This template
is called the leading-strand template, and the strand being synthesized is called the leading strand. The
strand whose 3′ end faces away from the replication fork is synthesized in short fragments called Okazaki
fragments, which are about 1,000 nucleotides long. Upon completion of an Okazaki fragment, the DNA
polymerase leaves the template and returns to the replication fork to synthesize another fragment. The
template that is copied in short fragments is called the lagging-strand template, and the strand being
synthesized is called the lagging strand. Each Okazaki fragment begins with the synthesis of an RNA
primer that is subsequently elongated by the DNA polymerase. The RNA primers are drawn as wavy
lines. Eventually, the RNA primers are removed and the fragments are elongated by DNA polymerase and
sealed. See text for details.
with each other as a dimer and do not leave the tracks. However, as we shall describe shortly,
replication fork. in Section 3.1.5, it now appears that the rep-
lication fork and the replicative DNA poly-
Synthesis of leading strand merase reside in a replication “factory” that
Recall that DNA polymerase cannot initi- does not move along the DNA. The replication
ate new polynucleotide strands but can only factory is called a replisome, and the unrepli-
elongate the 3′ end of an existing strand. This cated DNA threads through it; the terminus is
means that the initiation of replication of the the last part to enter. In connection with topics
leading strand at oriC (as well as the lagging covered later, we describe the experiments that
strand) must begin with the synthesis of a rela- support this model (see note 34, which relates
tively short RNA primer (5–60 nucleotides) primarily to Section 3.1.6).
by an RNA polymerase. An RNA polymerase
called primase (or DnaG) synthesizes the Synthesis of the lagging strand
RNA primer for the Okazaki fragments, and The opposite template strand is called the lag-
perhaps also the RNA primer for the leading ging template strand. It cannot be copied in the
strand. DNA polymerase III then extends the same way as the leading template strand is cop-
RNA primer at its 3′ end and continues the ied because it is of opposite polarity, and the 3′
elongation of the DNA (Fig. 3.10). The DNA end of the RNA primer (i.e., the growing end) is
polymerase that synthesizes the leading strand facing away from the replication fork. Instead,
stays with the replication fork as it adds nucle- the polymerase periodically disengages from the
otides; it dissociates rarely, if at all. At one template strand to return to a newly synthesized
time it was thought that the DNA polymerase primer at the replication fork, thus synthesizing
moved along the DNA with the replication fork short polynucleotide fragments, the Okazaki
to the terminus, much as a train moves along fragments (Fig. 3.10).
www.ebook777.com
The DnaB and DnaG complex unwinds the of DNA in the lagging strand, it stops and leaves
DNA at the replication fork and synthesizes the 3′ end of the newly synthesized DNA. The
the RNA primer for the lagging strand result is a nick in the DNA strand between the
As replication proceeds, RNA primers for the DNA and RNA. The very short RNA primer
Okazaki fragments are synthesized by the spe- fragments (10–12 nucleotides long) in the
cial RNA polymerase called the primase or Okazaki fragments are removed by the 5′- to-3′-
DnaG (Fig. 3.11). The primase is associated exonuclease activity of DNA polymerase I, and
with the helicase (DnaB) to form a complex each ribonucleotide is simultaneously replaced
that stays with the replication fork as the DNA with a deoxyribonucleotide polymerized by
is replicated. The primase must synthesize the DNA polymerase I. When this happens, the nick
Okazaki primers in the direction away from the moves in the direction of DNA synthesis and the
replication fork. The DnaB helicase unwinds process is called nick translation. The forma-
the duplex. Then the DNA polymerase III elon- tion of the phosphodiester bond between the 3′
gates the RNA primer at its 3′ end to synthesize hydroxyl at the end of one strand of DNA and
the Okazaki fragment. the 5′ phosphate of the other strand has been is
catalyzed by DNA ligase. The overall scheme is
Model explaining how the DNA polymerase shown in Fig. 3.12. The ligase makes the cova-
III synthesizing the lagging strand lent bond between the 3′ phosphate of the newly
can stay with the replication fork synthesized DNA and the 3′ hydroxyl of the pre-
The two DNA polymerases III (Pol III) that viously synthesized DNA by transferring AMP
synthesize the leading and lagging strands are from either ATP or NAD+ (depending upon the
physically associated with each other as a dimer
that does not leave the fork. This ensures that
both strands are replicated simultaneously and
DNA polymerase III
at the same rate. However, there is a problem.
SSB DNA gyrase
The template strands are of opposite polarity,
and this implies that the two DNA polymerases 5'
III must move in opposite directions. Yet, they (1) 3'

stay together as a dimer at the replication fork.
primosome (helicase, primase)
It may be that the dimer stays with the replica-
tion fork and still synthesizes the lagging strand, RNA primer
whose growing 3′ end extends away from the

fork because the lagging-strand template loops Fig. 3.11 Replication fork. The model explains
around one of the polymerases at the replica- how DNA polymerase might synthesize the lagging
strand and stay with the replication fork. It is sug-
tion fork so that its 5′ end faces the replication
gested that the lagging-strand template loops around
fork rather than away from it. This suggested a dimeric polymerase (two DNA polymerases physi-
mechanism is illustrated in Fig. 3.11. If the pos- cally associated with each other) so that the poly-
tulated scheme is accurate, then the Okazaki merase bound to the lagging strand moves along it
fragments will be elongated at their 3′ ends at in the direction of the replication fork. Thus, one of
the replication fork and the dimeric DNA poly- the polymerases of the dimer extends an RNA primer
merase III can elongate both the leading and (wavy line) on the lagging strand, and the other poly-
lagging strands at the same time. According to merase elongates the leading strand. Another RNA
the model, after polymerizing approximately primer, labeled (1), is made by primase ahead of the
1,000 nucleotides, or one Okazaki fragment, polymerase at the replication fork. The polymerase
the polymerase synthesizing the lagging strand keeps moving on the template until it encounters the
5′ end of a previously made Okazaki fragment (wavy
would disengage from the template and a new
line). Upon encountering the Okazaki fragment, the
loop would form. polymerase that is synthesizing the lagging strand
disengages from the template and reinitiates at the
Attaching the Okazaki fragments to new RNA primer with a new loop. The arrows point
each other in the direction of strand elongation at a stationary
When DNA polymerase III reaches the RNA at replication fork. See Section 3.1.3 for a discussion of
the 5′ end of the previously synthesized segment the stationary replication fork.
Fig. 3.12 Attaching DNA fragments to each other in the lagging strand. When the growing Okazaki fragment
reaches the 5′ end of the previously synthesized Okazaki fragment, DNA polymerase III that is synthesizing
the lagging strand dissociates from the template DNA. DNA polymerase I then removes the RNA via 5′- to
3′-exonuclease activity from the 5′ end of the previously synthesized Okazaki fragment and replaces the RNA
(wavy line) with DNA (straight line). DNA ligase then seals the break in the DNA.
Fig. 3.13 DNA ligase reaction. The enzyme catalyzes the adenylylation of the 5′ phosphate. Depending upon
the ligase, the AMP can be derived either from ATP, in which case pyrophosphate is displaced, or NAD+, in
which case nicotinamide monophosphate (NMN) is displaced. E. coli DNA ligase uses NAD+ as the AMP
donor. Then the 3′ hydroxyl of the ribose attacks the AMP derivative and displaces the AMP. The result is a
phosphodiester bond.
organism) to the phosphate (Fig. 3.13). This and a clamp loader. The sliding β clamp has a
produces an AMP derivative with a high group large central pore through which the DNA tem-
transfer potential. The ligase then catalyzes the plate is threaded. Very importantly, the sliding
displacement of the AMP by the 3′ hydroxyl, clamp complex binds to the DNA polymerase
and the phosphodiester bond is formed. III core, which consists of the α, ε, and θ sub-
units, and keeps it tethered to the DNA, allow-
Sliding clamps ing the processing capacity of the polymerase to
DNA polymerase III, which is the replicative increase from approximately 12 nucleotides to
polymerase in E. coli, consists of three subassem- thousands of nucleotides before the DNA poly-
blies. The total complex is referred to as DNA merase dissociates from the template. Thus, the
polymerase III holoenzyme. The subassemblies sliding clamp is very important for the efficiency
are the polymerase itself, a sliding β clamp, of DNA replication. Since DNA is synthesized
www.ebook777.com
continuously on the leading strand, the clamp inhibiting the replicative helicase (DnaB).17–19
is loaded only once. However, on the lagging Both Ter sites and the Ter-binding protein
strand, where DNA is synthesized in Okazaki have been well studied in B. subtilis. See note
fragments of approximately 1,000 nucleotides, 20 for similarities and differences with respect
the clamp is unloaded and loaded frequently. to E. coli.
For a discussion of the clamps and clamp load-
ers, see note 16.
3.1.5 Termination
Ter sites and Tus protein
In E. coli the two replication forks that begin
at oriC and polymerize bidirectionally stop
in a region of the chromosome opposite oriC,
called Ter, the termination region (Fig. 3.14).
After termination, the sister chromosomes
separate and partition (segregate) to opposite
halves of the cell. The termination region con-
tains specific sequences of nucleotides called
Ter sequences.
The Ter sites are quite unusual because they
allow replication to proceed in only one direc-
tion! (Note that in Fig. 3.14 the arrows refer to
the direction of replication, not the movement Fig. 3.14 Termination of DNA replication in E. coli.
The arrows refer to the direction of replication,
of the replication forks, which are station-
rather than to movement of the replication forks. A
ary within the replisome.) This property can
terminator region exists 180° opposite the origin. In
account for termination at the Ter sites. For the terminator region there are nucleotide sequences
example, assume that there are two Ter sites, that allow replication in only one direction. Thus, the
which are located next to each other, with per- terminator sequences are said to be polar. E. coli has
haps a short segment of DNA between them. six terminator (Ter) sites. These are, in the sequence
Suppose Ter site 1 allows replication to proceed in which they exist in the DNA, TerE, TerD, TerA,
clockwise but not counterclockwise, and Ter TerC, TerB, and TerF. On one side of the terminator
site 2 allows counterclockwise replication but region are TerE, TerD, and TerA, and TerC, TerB,
not clockwise replication. Thus, if clockwise and TerF occupy the opposite side. The figure shows
replication arrives at site 2, it will stall; and if two replication forks. Data discussed earlier in this
section support a model in which the replication
counterclockwise replication arrives at site 1, it
forks are stationary within a DNA synthesizing “fac-
will stall. Consequently, bidirectional replica-
tory” called the replisome. Unreplicated DNA enters
tion will meet at site 1 or site 2 or somewhere the replisome and passes through the stationary rep-
between them. lication forks. TerC, TerB, and TerF inhibit clock-
As shown in Fig. 3.14, the situation in wise replication, and TerA, TerD, and TerE inhibit
E. coli is a little bit more complicated because counterclockwise replication. Therefore, there are
it has six Ter sites, sites are divided into two three chances to stop replication in the termination
groups of three: one group (TerA, TerD, and region. For example, if counterclockwise replica-
TerE) prevents counterclockwise replication, tion moves past TerA, it will stop at TerD or TerE.
and the second group (TerC, TerB, and TerF), Consequently, replication from both directions will
prevents clockwise replication. As explained stop at one of the termination sites or a site between
them. A protein, called the Tus protein in E. coli,
in the legend to Fig. 3.14, the presence of
binds to Ter and prevents the helicase from proceed-
three Ter sites per replication fork direction
ing in one direction, thus accounting for the polar-
provides a backup in case the replication fork ity of inhibition of movement of the replication fork.
gets by one of these sites. A protein, which in Notice the site marked dif. It is here that recombina-
E. coli is called the Tus (terminator utilization tion takes place to separate chromosomes that have
substance) protein binds to the Ter site and undergone an unequal number of recombinations
imposes the one-way travel. Tus does this by during replication.
3.1.6 Chromosome separation and reflecting the fact that because the sister chro-
partitioning; some general principles mosomes do not partition to opposite cell poles,
First, some definitions: chromosome separation they can both be in the same daughter cell upon
refers to the detachment of the newly replicated cell division. Under these conditions, the nucle-
chromosomes from one another, and chromo- oids are single nucleoids and are not larger than
some partitioning refers to the segregation of normal.
sister chromosomes to opposite halves of the
cell prior to cell division. Chromosome separation
Chromosome separation can take place in Chromosome separation requires the separa-
one of two ways. If the linkage between the tion of the two sister chromosome dimers, said
daughter chromosomes is noncovalent, then to be catenated, by topoisomerase IV. The pro-
topoisomerases such as topoisomerase IV sepa- cess is called decatenation. If there has been an
rate the sister chromosomes, as we shall discuss unequal number of crossover events during
shortly. If an unequal number of recombina- DNA replication, then the sister chromosomes
tions have occurred between the sister chro- are covalently linked and must be separated by
mosomes, then the sister chromosomes are recombination by XerC and XerD.
covalently linked and a site-specific recombina- Recombination occurs at dif sites. A key pro-
tion must occur. (See later subsection entitled 2. tein here is the cell division protein FtsK, which
Site-specific recombination at dif.) is localized to the septum and recruits other
Chromosome partitioning is not a well-un- cell division proteins before cell division. FtsK
derstood process in prokaryotes because a spin- ensures that XerCD and topoisomerase IV act
dle apparatus similar to that which separates on the DNA at midcell during the final stages
sister chromosomes in eukaryotes is not pres- of cell division. A discussion of how this might
ent. This section discusses current ideas about occur can be found later (see subsection 3, which
how separation and partitioning of daughter considers the catalytic activity on this enzyme).
chromosomes occur in prokaryotes. Section
1. Topoisomerase IV
3.1.7 describes chromosome partitioning in B.
When replication is complete (at about 180°
subtilis, and Section 3.1.8 describes chromo-
from the start site), the two completed DNA
some partitioning in C. crescentus.
molecules are linked as two monomeric circles
in a chain; that is, they are catenated and must
Phenotypes of partition mutants be separated from each other (Fig. 3.15). The
The genes involved in chromosome partition- process is called decatenation. In E. coli and
ing have been discovered by analyzing mutants. Salmonella typhimurium, the decatenation
Mutants defective in chromosome partitioning of two monomeric interlinked chromosomes
can be defective in any one of several genes: some requires the action of a DNA gyrase–like enzyme
are involved in detaching the chromosomes called topoisomerase IV.21 Topoisomerase IV
from each other, some are involved in the par- mutants are temperature sensitive for growth;
tition of the detached chromosomes, and some that is, they grow at 30 °C but die at 42 °C.
are even involved in the replication of DNA, At the restrictive temperature, cell division
because only fully replicated DNA molecules is inhibited and the mutants form elongated
can detach from one another. Depending upon cells, often with unusually large nucleoids
whether the defect is in detachment or partition (revealed by DNA staining) in midcell or as
of chromosomes, the phenotype differs. If the several nucleoid masses unequally distributed
defect is in detaching the sister chromosomes in the elongated cells.22 The nucleoids are large
from each other, then the phenotype includes because the two daughter nucleoids are not able
elongated cells with an unusually large nucleoid to separate. The filamentous cells may divide
mass positioned near the cell center. Anucleate in regions where there is no DNA to produce
cells may also form, as discussed next. anucleate cells. Inhibition of cell division result-
On the other hand, if the sister chromosomes ing in elongated or filamentous cells is typical
can detach but cannot partition to opposite of mutants that have a block in DNA replica-
poles, the phenotype is an increased production tion or separation of daughter chromosomes.
of anucleate cells of the size of a newborn cell, This reflects the coupling between cell division
www.ebook777.com
they chromosomes are covalently linked as a cir-

cular dimer to each other and decatenation can-
not separate them. Under these circumstances,
the DNA molecules must be separated by site-
specific recombination by using enzymes called
recombinases. E. coli has a locus called dif (dele-
tion-induced filamentation), which lies in the
replication terminus (Ter) region.25 The recom-
binational event that separates the two sister
chromosomes that have undergone an unequal
number of recombinations during replication
takes place in this Ter region (Fig. 3.16).
The site-specific recombination is catalyzed
by two recombinase proteins, XerC and XerD,
both of which are required.26 The recombinases
are activated by an E. coli cell division protein
called FtsK, which is localized at midcell with
other cell division proteins. (See Section 2.6.2
for a discussion of FtsK and the other cell divi-
sion proteins in E. coli, and ref. 27 for a discus-
sion of the multiple roles of FtsK in cell division
and chromosome segregation.) Homologues
of XerC and XerD have been found in a wide
variety of bacteria. Mutations in dif are viable,
and most of the cells are normal. This is to be
Fig. 3.15 Separating catenated daughter circles after expected, since dif is required only when there
DNA replication by using type IV topoisomerase. has been an unequal number of recombinations.
The reaction is catalyzed by type IV topoisomerase However, approximately 10% of the cells are
(similar to DNA gyrase). The enzyme catalyzes a filamentous with unusually large nucleoids,
double-stranded cut in one of the duplexes. Then the indicative of a failure of the newly replicated
unbroken duplex passes through the gap. The gap is chromosomes to separate. As expected, xerC
sealed, and the DNA molecules have become sepa- and xerD mutants show the same phenotype as
rated. Mutants defective in topoisomerase IV are dif– mutants.
temperature sensitive for growth and show abnor-
mal nucleoid segration. The phenotype is called Par–.
See ref. 21. 3. Topoisomerase IV catalyzes decatenation
primarily at dif in the presence of XerC,
XerD, and FtsK
The activity of topoisomerase IV can take place
and DNA metabolism as described in note 23.
at several chromosomal sites; but in the presence
(Topoisomerase IV mutants have been isolated
of FtsK, XerC, and XerD, decatenation takes
in C. crescentus, but the phenotype differs from
place preferentially at midcell at the dif site.28
that of E. coli and S. typhimurium mutants
However, unlike site-specific recombination,
because checkpoints exist in the Caulobacter
which requires XerC and XerD, decatenation at
cell cycle.24) Topoisomerase IV makes a double-
dif does not require the cell division protein FtsK,
stranded break in one of the DNA molecules,
but is simply favored at dif because of FtsK. Why
and the unbroken molecule is passed through
should topoisomerase IV preferentially cleave
the break (Fig. 3.15). This is followed by seal-
at dif? There have been two suggestions.28 One
ing the break (see section entitled Attaching the
suggestion is that the dif/XerC/XerD complex
Okazaki fragments to each other).
might increase the affinity of topoisomerase IV
2. Site-specific recombination at dif for the dif site. A second suggestion is that topoi-
If an unequal number of recombinations have somerase IV is part of a decatenation/resolution
occurred between sister circular chromosomes, complex that forms at dif.
Chromosome partitioning in eukaryotes

It is worthwhile to summarize how chromosome
partitioning takes place in eukaryotes before
discussing how the process occurs in prokary-
otes. In eukaryotes, chromosome replication is
completed to form two sister chromatids bound
along their length in the S phase of the cell cycle.
The sister chromatids remain together dur-
ing the G2 phase, which follows the S phase.
Following the G2 phase is the M phase (mito-
sis), and this is when the sister chromatids parti-
tion. The M phase consists of nuclear division
and cytokinesis. Partitioning occurs because
the sister chromatids are attached to microtu-
bule spindle fibers that pull them apart. This
occurs during the anaphase portion of mitosis.
Specifically, the sister chromatids are separated
(and are now called chromosomes) and pulled
to opposite cell poles by the shortening of the
microtubular spindle fibers in the mitotic spin-
dle. Each chromatid is attached to the microtu-
bule spindle fibers via a protein complex called
a kinetochore, which binds to a subset of micro-
tubules in the mitotic spindle. The kinetochore
is located at a highly condensed, constricted
region called a centromere.
Chromosome partitioning in prokaryotes

In prokaryotes, “chromosome partitioning”
refers to the segregation of sister chromosomes
toward opposite cell poles in preparation for
cell division. (Bacterial chromosomes are often
referred to as nucleoids.) There is no evidence
in bacteria for a device similar to the eukaryotic
mitotic spindle to partition sister chromosomes.
So how do bacteria partition sister chromo-
Fig. 3.16 Site-specific recombination at termina- somes faithfully? This is a very active area of
tion to separate chromosomes that have undergone research, and the literature provides detailed
unequal numbers of recombinations. Simplified information about the similarities and differ-
drawing of recombination at dif to resolve a DNA ences with respect to what has been learned
dimer into monomers. As discussed in Section 3.1.5, about B. subtilis, E. coli, and C. crescentus.29–33
the replication forks are stationary and remain close Certain generalizations can be made. First, it is
to each other in the replisome, a DNA-synthesizing
“factory.” The unreplicated DNA is fed through the
forks. The arrows depict the direction of polymer- recombination event at dif can proceed in both direc-
ization of DNA. (A) Unequal numbers of recom- tions. However, monomer formation is favored.
binations have taken place covalently, linking the Perhaps the segregation of the monomers ensures
two circles to form a dimer. (B) Recombination is that the newly replicated dif sites cannot recombine.
catalyzed by recombinases acting at the dif sites in For recombination at dif to resolve the dimers, dif
the termination (Ter) regions. The chromosomes are must be at its original site in the terminus region. If a
then segregated to opposite cell poles. A dif− mutant copy of dif is reinserted in a different chromosomal
produces filaments and anucleate cells as well as nor- site in a dif– strain, the cells still show the dif– phe-
mal cells. As the figure indicates, it is believed that the notype.
www.ebook777.com
noted that partitioning (segregation) takes place replication in the previous cell cycle, the student
concurrent with replication of the DNA. is referred to ref. 36.
The DNA is replicated in a stationary “fac- There are important differences between
tory,” called the replisome, which in E. coli or the positioning of the replisome in E. coli and
B. subtilis is assembled at or near midcell when B. subtilis on the one hand, and C. crescentus
DNA replication is initiated and is thought to on the other hand. These differences, which
be anchored there. For a review of replisomes, are related to differences in the cell cycle, are
see ref. 34. When DNA synthesis is finished, the described later. (See Sections 3.1.7 and 3.1.8.)
replisome is disassembled. The unreplicated
DNA threads through the replisome, and the What moves the sister chromosomes apart,
newly duplicated portions move away from the and what directs their movement?
anchored replication forks within the replisome Clearly, prokaryotic cells do not have a spindle
and partition toward opposite poles of the cell. apparatus. What then is responsible for the
Since the origins (ori sites) are replicated first, movement of sister chromosomes to opposite
they leave the replisome first and very quickly poles of the cell prior to cell division? Recently,
are moved to opposite halves of the cell near the there has been a great deal of research and
cell poles. much speculation. Several proteins have been
As will be discussed later, the “extrusion– implicated. Some of the proteins are thought
capture” model proposes a mechanism(s) that to comprise the “motor” that moves the chro-
provides the force that “extrudes” the newly mosomes, some of the proteins may attach the
replicated origins from the replisome and “cap- oriC regions to the opposite cell pole regions,
tures” the origins at or near opposite cell poles. and some of the proteins may compact the
As will also be discussed, for this to occur suc- chromosomes and “guide” the chromosomes
cessfully, certain proteins must compact the toward the opposite poles. Which proteins are
chromosomes and guide the origin regions these? We present an overview of current ideas,
toward the cell poles. Eventually, the terminus is followed by a more detailed examination of the
drawn into the replisome, where it becomes rep- proteins thought to be involved in chromosome
licated; then the sister chromosomes separate, partitioning.
and segregation is completed as the duplicated
termini remain near the cell center. As a result, 1. Sister chromosomes might be pushed
newborn cells in slowly growing cultures of E. out by the replisome to opposite halves
coli, B. subtilis, and C. crescentus have a termi- of the cell
nus near the new cell pole and an origin near Because of the absence of a spindle apparatus
the old cell pole (Fig. 3.17). Finally, the termi- similar to that found in eukaryotes, the mecha-
nus in newborn cells is drawn toward the center nism for the partitioning of sister chromosomes
of the cell by DNA replication in the centrally toward opposite poles of the cell in prokaryotes
located replisome factory. There can be small is not well understood. Various models have
differences among bacteria in the cellular loca- been proposed. One model postulates the repli-
tion of chromosome regions. For example, in some as a DNA-synthesizing “factory” that has
E. coli the origin regions localize at quarter-cell motor proteins generating force that pushes
positions instead of at the cell poles, whereas in the newly replicated sister chromosomes apart
B. subtilis growing vegetatively or sporulating, toward opposite cell poles. According to this
they localize at the cell poles. model, referred to as the “extrusion–capture”
For a description of how the replisome as well model (Fig. 3.17), in E. coli and B. subtilis the
as specific regions of the chromosome such as replisome is anchored to the membrane at the
origins, termini, and regions between them, can cell center; that unreplicated DNA enters the
be microscopically visualized, see note 35, as replisome and threads through it, whereupon
well as the references cited earlier concerning the newly replicated sister chromosomes exit the
chromosome segregation in B. subtilis, E. coli, replisome and partition toward opposite poles
and C. crescentus. For more information about of the cell. In the slightly different situation of
the location of the origins and termini, specifi- C. crescentus, (discussed later), the replisome
cally for fast-growing cells that initiate DNA is initially at the stalked pole of the cell, where
A newborn cell
origin terminus
origin, terminus move to

B
midcell position
DNA replication initiated,

C duplicated origins move to opposite
sides of the cell
D terminus duplicated, sister

chromosomes separate
and segregate to
opposite sides of cell, cell
begins to divide
midcell
origin origin
....
:::: replisome
replication fork
E
terminus
Fig. 3.17 General model for duplication, position, and partitioning of bacterial chromosomes in slowly
growing cells. (A) Newborn cell showing chromosome with single origin (circle) and single terminus (square)
at opposite cell poles. (B) Origin and terminus move and locate to the midcell position, where the replisome
is assembled and DNA replication is initiated. (C) DNA replication is initiated, and copies of origin segregate
to opposite halves of cell. In E. coli and C. crescentus, the origins are near opposite cell poles for most of the
cell cycle and move to the midcell position in the newborn cell. In B. subtilis, origin copies segregate to cell
quarter-positions that will become midcell following cell division. (D) Cell division begins. (E) Stationary
replisome at midcell within which are two replication forks. DNA is pulled into the replisome, origin first and
terminus last. The sister chromosomes are pushed toward opposite cell halves. (F) Cell division completed.
DNA synthesis is initiated, and it moves toward mechanism has been proposed whereby energy
midcell during DNA replication. As mentioned released by the combined action of the DNA
earlier, the replisome is assembled when DNA polymerase and the DNA helicase, both of
replication is initiated and disassembled when which can be considered to be force-generating
DNA replication is finished. motors, “pulls” the unreplicated DNA into
the stationary replisome and “pushes” the
2. Candidate proteins that might be involved replicated sister chromosomes, origin regions
in providing the force to move the DNA first, out of the replisome into opposite halves
What are the postulated motor proteins that seg- of the cell. It has been suggested that because
regate the sister chromosomes? A cooperative the inhibition of transcription also inhibits
www.ebook777.com
sister chromosome partitioning, the force-gen- A more detailed description of proteins that
erating motor may also include stationary RNA play a role in chromosome partitioning
polymerases. As mentioned, several different proteins appear
to have a role in chromosome partitioning.
3. What confers directionality to
This is clearly not a well-understood process.
chromosome movements during partitioning?
However, the following discussion and the ref-
In addition to the “motor” that moves sister
erences cited should spark the student’s curi-
chromosomes apart, one may consider what
osity to explore this very important area of
confers directionality to chromosome move-
research.
ment. One may ask, in other words, why sis-
ter chromosomes move to opposite poles. 1. RacA
One region of the chromosome that is directed RacA (remodeling and anchoring of the
toward opposite cell poles is the oriC region, Chromosome A) is a DNA-binding protein in
located at the origin of replication. In addi- B. subtilis that functions in chromosome par-
tion to oriC, the chromosomes of B. subtilis titioning only during sporulation.39 It binds
have another region near the origin, the polar preferentially at the PLR region, near oriC, and
localization region (PLR), which directs the less specifically throughout the DNA. RacA is
sister chromosomes to cell poles. It may work required for two reasons: (1) for the extension
independently of oriC, or it may be responsible of the nucleoid into an axial filament during spo-
for the migration of the nearby oriC to the cell rulation and (2) for the anchoring of the chro-
poles when chromosomes are partitioned dur- mosome near its origin region to the cell pole
ing sporulation.37 during sporulation (Fig. 3.18A). RacA works in
The question now is: What might “drag” the conjunction with Soj in moving the Spo0J–oriC
localization regions to the cell poles? One pos- complex to the cell pole via interaction with
sibility is that proteins bind to the DNA at the DivIVA at the pole. (See Section 3.1.7.) During
localization regions and that these proteins vegetative growth of B. subtilis, the two chro-
tether the chromosome to receptor proteins mosomal origins become anchored at cell quar-
in the membrane near or at the cell poles. Two ter-positions rather than at the extreme poles,
proteins mentioned in this regard are the RacA and RacA is not involved.
protein in B. subtilis, which binds to the PLR as
well as to regions near oriC, and the Par proteins 2. The Par proteins
(partitioning proteins) in many bacteria, includ- In most bacteria, and in low copy number plas-
ing B. subtilis and C. crescentus (but not E. coli or mids, an early stage of the partitioning process
Haemophilus influenzae), which bind to oriC. involves the Par proteins, which bind to the
In addition to proteins that have been sug- chromosome near the origin of replication and
gested to tether the localization regions to the may contribute toward the directionality of
pole, other proteins function in chromosome movement of the chromosome origins to oppo-
partitioning. These include SMC (structural site cell poles. (For a review, see ref. 40, and for
maintenance of chromosomes) proteins in B. a discussion of the role of Par proteins in plas-
subtilis and C. crescentus, and MukB proteins mid partitioning, see note 41.)
in E. coli. The SMC and Muk proteins condense The Par proteins discussed here that are
the chromosome. It appears that condensation encoded by the bacterial chromosome are called
of the chromosome might be important in mov- ParA and ParB. ParB is a DNA-binding protein
ing it toward the cell pole. In addition, there are that binds to specific sequences, called parS
the SetB and MreB proteins. SetB is an integral sequences, near the origin of replication. The
membrane protein that physically interacts binding is cooperative with ParA, an ATPase that
with MreB, a cytoplasmic actinlike filament. It interacts with ParB and results in a large nucleo-
has been suggested that MreB forms dynamic protein complex that is important for the posi-
filament cables that might be a mitotic-like tioning of sister chromosome and plasmid copies
DNA segregation apparatus.38 Exactly how this in opposite halves of the cell. As reviewed in ref.
works is not known. However, mutations in 21, ParA is membrane associated and may help
mreB in E. coli result in severe effects on nucle- tether the ParB–parS complex to the membrane.
oid segregation. Proteins homologous to the chromosomally
i ii
iv iii
B ii
i iii
iv
Fig. 3.18 Partitioning of chromosomes in Bacillus subtilis. (A) Sporulating cell. (i) Stage 0. The two chromo-
somes are a single axial filament attached by their origins to opposite poles of the cell. (ii) A septum forms,
trapping the region proximal to the origin in the forespore. (iii) DNA translocation moves the rest of the
chromosome into the forespore. (iv) After engulfment, the origins detach from the poles. (B) Vegetative cell.
(i) Prior to DNA replication there is one origin. (ii) The DNA replicates, and both origins move toward the
opposite cell halves. (iii) The chromosomes condense, moving away from each other toward the poles. (iv) A
septum forms in the middle, dividing the cell into two daughter cells. Source: Adapted from Webb, C. D., A.
Teleman, S. Gordon, A. Straight, A. Belmont, D. C.-H. Lin, A. D. Grossman, A. Wright, and R. Losick. 1997.
Bipolar localization of the replication origin regions of chromosomes in vegetative and sporulating cells of B.
subtilis. Cell 88:667–674.
encoded ParA and ParB proteins have been 3. Bacterial SMC and Muk proteins
detected in many bacterial species, including Regardless of the mechanism of partitioning,
B. subtilis, where they are called Soj and Spo0J, the chromosome partitioning cannot take place
respectively, as well as in Caulobacter crescentus unless the chromosomes are compacted and
and Pseudomonas putida. However, E. coli and maintained in a higher order structure. One
H. influenzae do not have chromosomal parABS speculation is that condensation of the DNA
genes. Other proteins are presumably present in contributes toward a “pulling force” that com-
these bacteria that serve a function similar to plements the “pushing force” generated in the
that of the Par proteins. replisome.27 Proteins important for compaction
www.ebook777.com
and higher order structure are called structural MukB forms a complex with two other Muk
maintenance of chromosome (SMC) proteins, proteins, MukE and MukF, and the three act
which are analogues of similar proteins in together. For more information about the phe-
eukaryotes. Not every bacterium has an SMC notype of mutations in the muk and smc genes,
protein. Bacteria such as E. coli, which lack see note 45.
an SMC protein, usually have a protein called
MukB, which is similar in structure and function 4. SetB and MreB
to SMC and sometimes is referred to as a “dis- Another protein that appears to be involved in
tant relative.” MukB (see Fig. 3.19) has thus far chromosome partitioning is SetB, an integral
been detected in E. coli and H. influenzae.42–44 membrane protein in E. coli.46 The deletion of
Fig. 3.19 Model of MukB. Based upon amino acid sequence studies and electron microscopic data, the model
postulates that MukB is a homodimer with a rod-and-hinge structure. At the C-terminal end lies a pair of large
globular domains. The N-terminal end has a pair of smaller globular domains. Electron micrographs suggest
that MukB can bend at a hinge site in the middle of the rod section. MukB binds to DNA, ATP, and GTP. The
amino acid sequence of MukB indicates homology with dynamin, a microtubule-associated protein from rats.
It has been speculated that MukB attaches to the DNA and to some cellular structure and moves the DNA to
opposite poles. Source: Niki, H., R. Imamura, M. Kitaoka, K. Yamanaka, T. Ogura, and S. Hiraga. 1992.
E. coli MukB protein involved in chromosome partition forms a homodimer with a rod-and-hinge structure
having DNA binding and ATP/GTP binding activities. EMBO J. 11:5101–5109.
setB results in a delay in sister chromosome sep- fork has been visualized by means of replicative
aration from the cell center, and overproduc- DNA polymerase fused in frame to the green
tion causes stretching and breaking up of the fluorescent protein (GFP).
nucleoid. Interestingly, SetB physically interacts
with MreB, a member of the family of actinlike Some of the factors important for
proteins that form cytoplasmic helical filaments chromosome structure and partitioning
along the inner cytoplasmic membrane of cylin- in B. subtilis
drical cells and are believed to be important Some of the proteins previously discussed are
for maintaining cell shape.47 (One reason for important for chromosome orientation and
believing that MreB is important for cell shape partitioning during sporulation and vegetative
is that deletion of the gene mreB causes the rod- cell division.
shaped E. coli to become spherical.) The MreB
protein, which is discussed in Section 1.2.7, has 1. ParA (Soj) and ParB (Spo0J)
been found in several different bacteria, includ- The ParA homologue in B. subtilis is Soj, and
ing E. coli, B. subtilis, Thermotoga maritima, the ParB homologue is Spo0J. Spo0J binds to
and Caulobacter crescentus.48 at least eight parS sites located on either side of
oriC, and in the presence of Soj the Spo0J–parS
3.1.7 Chromosome partitioning in complexes are organized into a condensed,
Bacillus subtilis higher order oriC supercomplex, which persists
throughout the cell cycle (growth, sporulation,
Sister chromosome partitioning in
germination) and is one of the factors that is
Bacillus subtilis during sporulation
required for chromosome partitioning dur-
(Fig. 3.18A)
ing vegetative growth as well as sporulation.
It is proposed that at stage 0 (Spo0), prior to
Mutations in soj–spo0J result in cells that are
the initiation of sporulation, the two chromo-
deficient in prespore chromosome partitioning
somes form a single axial filament with their
(segregation), as well as a small fraction of anu-
origin regions attached to opposite poles of the
cleate cells during vegetative growth. The Soj–
cell (Fig. 3.18Ai). Septation occurs close to one
Spo0J system and RacA (described next) appear
pole of the predivisional cell, trapping approxi-
to have partially redundant roles in partition-
mately a third of the forespore chromosome in
ing the prespore chromosome in that they both
the forespore (Fig. 3.18Aii). The origin of the
function to move oriC to the prespore cell pole
chromosome must be very close to the pole of
via an interaction with DivIVA at the pole.
the cell to ensure that it is captured in the pre-
spore when the septum forms. DNA transloca- 2. RacA
tion across the forespore septum takes place, RacA, known to be important in B. subtilis,
resulting in the movement of the forespore chro- is synthesized for a period during the begin-
mosome being entirely into the forespore (Fig. ning of sporulation but not during vegetative
3.18Aiii). Translocation of the DNA requires growth.39,50 This protein binds nonspecifically
a transporter called SpoIIIE DNA translocase, throughout the chromosome and is required
which forms a pore in the septum. Engulfment for formation of the axial filament. Mutants of
occurs, and the origins are detached from the racA fail to form an extended axial filament and
poles (Fig. 3.18Aiv). instead form a “stubby” nucleoid. RacA also
binds preferentially near oriC and is important
Sister chromosome partitioning during for efficient trapping of the oriC region at the
vegetative growth of Bacillus subtilis cell pole in the prespore. Mutations in racA
Prior to replication, a single chromosome with result in approximately 50% of prespores with-
one origin exists (Fig. 3.18Bi). After replica- out DNA. RacA also appears to be important
tion, the origins (called oriC, as in E. coli) are for the formation of the septum at the cell pole
moved to opposite halves of the cell, where during sporulation. Formation of the polar sep-
the origin regions become attached at the cell tum is significantly delayed in racA mutants.
quarter-positions (Fig. 3.18Bii).49 The chromo- It has been suggested that the binding of RacA
somes condense (Fig. 3.18Biii), and a septum to DivIVA at the cell pole displaces the divi-
forms (Fig. 3.18Biv). The stationary replicative sion inhibitor, MinCD, thus allowing septum
www.ebook777.com
formation at the pole.51 The suggestion is based (called Cori or ori) and can be seen to localize
upon an earlier publication showing that to both cell poles when chromosomal replica-
DivIVA sequesters MinCD at the poles so that tion is complete in the stalked cell.53 This can
cell division occurs at midcell during vegetative be demonstrated by using immunofluorescence
growth.52 (See Chapter 23, Box 23.1.) microscopy with anti-ParA or anti-ParB anti-
body. (See note 54 for an explanation of how
3. DivIVA
these experiments were done.)
DivIVA is an anchor protein at the cell poles.
It probably binds directly or indirectly to RacA
and to Spo0J, which themselves are bound to A moving replisome
the chromosome in the area of the oriC region. Rather than being initiated in a centrally located,
Thus, DivIVA, RacA, and Spo0J recruit the stationary replisome as in B. subtilis and E. coli,
oriC region to the cell poles during sporula- DNA synthesis in C. crescentus is initiated in
tion. When RacA is absent, anucleate prespore a replisome assembled at the stalked-cell pole.
compartments form, resulting in a 50% drop in Soon after replication has been initiated, one of
the frequency of sporulation. In racA mutant the origin copies rapidly moves to the opposite
cells that do sporulate, the DivIVA–Spo0J sys- cell pole. During DNA replication, the repli-
tem recruits the chromosome origins to the some slowly moves toward the midcell posi-
cell poles. Thus, the functions of RacA and the tion, where DNA replication, including the
Spo0J system are partially redundant. replication of the terminus, continues.55 It has
been suggested that perhaps the replisome is
3.1.8 Chromosome partitioning in moved passively from the pole to midcell by the
Caulobacter crescentus newly replicated DNA accumulating at the cell
Chromosome partitioning in C. crescentus is pole.55 When DNA replication is complete, the
similar in some ways to the same procedure in B. replisome disassembles. The result is that prior
subtilis and E. coli, and different in other ways. to cell division there is a copy of each origin near
In all three species, when DNA replication is the opposite cell poles.
complete, the origins have moved toward oppo-
site cell poles and the termini copies are segre- 3.1.9 Inhibitors of DNA replication
gated from each other closer to the cell center. A variety of antibiotics made by microorgan-
In C. crescentus, however, replication begins at isms, as well as chemically synthesized anti-
a cell pole (the stalked cell pole) rather than at microbial agents, inhibit DNA replication
midcell, and the replisome is not stationary but itself or inhibit the synthesis of substrates for
moves from the pole to the cell center during DNA replication. Two antibiotics produced by
DNA replication. Before this is aspect of parti- Streptomyces include novobiocin, which inhib-
tioning described further, a brief account of the its DNA gyrase, and mitomycin C, which cross-
cell cycle in C. crescentus will be presented. links the guanine bases in DNA, sometimes in the
same strand and sometimes in opposite strands.
The Caulobacter cell cycle Cross-linking guanine in opposite strands pre-
Normal cell division in C. crescentus produces vents strand separation. Another antibacterial
two cell types, a motile swarmer cell with pili agent is nalidixic acid, a synthetic quinolone
and a polar flagellum, and a stalked cell. (See compound that inhibits DNA gyrase. Its more
Section 23.2 for a detailed description.) The potent fluoroquinolone derivatives (norfloxa-
stalked cell synthesizes DNA and undergoes cin, ciprofloxacin) are used to treat urinary tract
repeated divisions to give rise to swarmer cells, infections—for example, those caused by uro-
which do not synthesize DNA and do not divide. pathogenic E. coli.
After a while the swarmer cell sheds the flagel- There are several other chemically produced
lum, grows a stalk, begins synthesizing DNA, drugs that are widely used to inhibit DNA
and becomes a dividing cell. synthesis. These include acridine dyes (e.g.,
ethidium, proflavin, chloroquine), which insert
ParA and ParB between the bases in the same strand of DNA
As with B. subtilis, ParA and ParB form a com- and distort the double helix. At low concentra-
plex attached to the Caulobacter origin region tions the dyes inhibit plasmid DNA replication
but not chromosomal replication, although a methyl (–CH3) and thus convert dUMP to
they can cause frameshift mutations. (See note dTMP. As a consequence of losing the hydride
56 for an explanation.) If the concentration is ion, the tetrahydrofolate is oxidized to dihydro-
sufficiently high, chromosomal replication is folate. The THF is regenerated by reducing the
inhibited. Chemicals that inhibit the synthesis dihydrofolate with NADPH in a reaction cata-
of precursors to DNA include the monophos- lyzed by dihydrofolate reductase. The dihydro-
phate of 5-fluorodeoxyuridine, aminopterin, folate analogues aminopterin and methotrexate
and methotrexate. inhibit dihydrofolate reductase.
5-Fluorodeoxyuridine-5′-phosphate [FdUMP,
made by cells from fluorouracil (FU)] inhibits 3.1.10 Repairing errors during
thymidylate synthase, which is the enzyme that replication
converts dUMP to dTMP. (See Fig. 3.20.) In this All cells have repair mechanisms that fix DNA
reaction a methylene group (–CH2) and hydride that has been damaged or has suffered errors
(H:) are transferred from methylenetetra- during replication. Repair systems in bacteria
hydrofolate (methylene–THF) to dUMP to form that operate during DNA replication and cor-
rect erroneous insertions of nucleotides are
described next. Such errors cause relatively
minor topological changes in the DNA.
Editing repair
Occasionally the wrong base is inserted in the
copy strand (e.g., a T rather than a C opposite
a G). This can occur if a base tautomerizes to a
form that allows hydrogen bonding to the wrong
partner. See Fig. 3.21 for an explanation. If the
wrong nucleotide is added to the growing chain
(e.g., a T opposite a G), the T must be removed;
otherwise, when the strand containing the G–T
pair is replicated, one of its progeny duplexes
will have an A–T pair in place of the G–C pair:
that is, a mutation will result. When the wrong
base is inserted in the growing chain so that a
mismatched pair results (e.g., G–T), DNA repli-
cation stops: that is, the DNA polymerase does
not continue to the next position. Presumably
this occurs because of the resultant topological
change in the DNA (distortion in the double-
stranded helix). The incorrect nucleotide is
Fig. 3.20 Inhibition of dTMP synthesis by anti- then removed via a 3′-exonuclease. DNA poly-
cancer drugs. 5-Fluorouracil is converted to merases have 3′-exonuclease activity that func-
5-fluorodeoxyuridine-5′-phosphate (fluorodeoxyu- tions in proofreading and removes mismatched
ridylate, or FdUMP), which inhibits thymidylate nucleotides when they are added. Mutations
synthase (1), the enzyme that converts dUMP to in the dnaQ gene (mutD), which codes for a
dTMP. A methylene group (–CH2) and a hydride 3′-exonuclease subunit in the DNA polymerase
(H:) are transferred from methylenetetrahydrofolate III holoenzyme, result in greatly increased rates
(methylene-THF) to dUMP to form a methyl (–CH3), of spontaneous mutations.
which converts dUMP to dTMP. As a consequence
of losing the hydride, the THF is oxidized to dihy-
drofolate. The dihydrofolate is reduced back to THF
Methyl-directed mismatch repair
by NADPH in a reaction catalyzed by dihydrofolate The proofreading system is not perfect, and a
reductase (2). The dihydrofolate reductase is inhib- certain number of wrong bases do get inserted
ited by dihydrofolate analogues such as methotrex- into the newly replicated DNA. These can be
ate and aminopterin. removed, and the fidelity of DNA replication
www.ebook777.com
region of DNA containing the GATC sequence.

Thus, for a few seconds or minutes after replica-
tion, the template strand is fully methylated but
the copy strand is undermethylated. During this
brief period, the copy strand can be repaired.
What happens is that the copy strand with the
incorrect base is cut at the 5′ side of the G in the
unmethylated GATC and the newly synthesized
DNA is removed by an exonuclease to a point
just beyond the mismatch (Fig. 3.22). The gap is
then filled with DNA polymerase III and sealed
with DNA ligase.
Several proteins are involved in mismatch
repair. In E. coli, these include the products
of mutH, mutL, and mutS. MutH, MutL, and
MutS are thought to form a complex with the
Fig. 3.21 Tautomerization can lead to incorrect base DNA. In this complex, MutH is the endonu-
pairing. Two isomers in equilibrium that differ in the clease. According to the model, MutS binds
arrangements of their atoms are called tautomers. to the single base pair mismatch. Then MutL
Commonly, tautomers differ in the placement of binds to MutS. MutH binds to a nearby GATC
hydrogen. For example, one isomer might exist in sequence, and its endonuclease activity is stimu-
the enol form (–OH attached to a carbon–carbon
lated by the MutS/MutL complex. (Another mut
double bond), whereas the other isomer might exist
in the keto form (contains a C=O group). Keto–enol
gene, mutD, encodes the DNA polymerase III
tautomerizations greatly favor the keto form. When 3′ exonuclease, which is important for editing,
thymine tautomerizes from the keto to the enol form, as described earlier in the subsection entitled
a proton dissociates from the nitrogen in the ring and Editing repair.) DNA helicase II (the product of
moves to the oxygen in the keto group to form the the mutU gene) is also required, and it is thought
hydroxyl. When this happens, two electrons shift in that its role is to unwind the cut strand so that it
from the nitrogen to form the C=N. (A) Adenine (A) can be degraded by exonuclease.
correctly base pairs with the keto form of thymine More than one exonuclease can be used. If
(T). (B) Thymine has tautomerized to the enol form the endonucleolytic cut by MutH is made 3′
and base pairs with guanine (G) to form a mismatch. to the mismatch, then exonuclease I (ExoI) or
ExoX, which is a 3′- to 5′-exonuclease is used.
If the cut is made 5′ to the mismatch, then either
improved 102- to 103-fold (Fig. 3.22), by what exonuclease VII (ExoVII) or RecJ, which are
is called the methyl-directed mismatch repair 5′- to 3′-exonucleases, are used. DNA poly-
(MMR) system (reviewed in refs. 57 and 58). merase III fills in the gap, and DNA ligase seals
The nucleotide that is removed is the incorrect the gap.
nucleotide in the copy strand rather than the Mismatch repair can also be used to repair
template strand; thus the template strand is not DNA damaged by the incorporation of base
changed, and a mutation does not occur. analogues or by certain types of alkylating agent
The repair system can distinguish the copy as long as the distortion of the double helix
strand from the template strand because the is not severe. Otherwise, repair mechanisms
template strand is marked by methylation. described in Section 19.2.1 are used. Null muta-
E. coli has an enzyme called deoxyadenosine tions in any of the mut genes involved in mis-
methylase (Dam methylase) that methylates all match repair result in a 102- to 103-fold increase
adenines at the N6 position within 5′-GATC-3′ in mutation frequency in E. coli. Homologues
sequences. (The sequence is a palindrome and of the mut genes can be found in eukaryotes
therefore is present in both strands but in oppo- such as yeast, mice, and humans, where loss of
site orientation, i.e., 3′-CTAG-5′.) However, function also results in increased mutation rates
the enzyme does not begin to methylate the and, in the case of humans, is correlated with
DNA until a short period after replication of the cancer in tumor cell lines from certain tissues
Fig. 3.22 Mismatch repair can occur when the wrong nucleotide has been inserted. For example, a T instead
of a C might be inserted opposite a G. The template strand is methylated at an adenine in a CTAGsequence.
The newly synthesized strand is not yet methylated, and this aids the repair enzymes in distin-guishing between
the newly synthesized strand and the template strand. (There is a slight delay in the methy-lation of newly syn-
thesized DNA. However, the A in GATC eventually becomes methylated.) (A) Anendonucleolytic cut is made
either on the 5′side or the 3′side of the mismatch in the GATC sequence in thenonmethylated strand. (B) An
exonuclease then removes the newly synthesized DNA past the point of themismatch. This requires helicase
II (MutU). (Helicase II is not identical to DnaB, which unwinds the strandsduring DNA replication.) If the
mismatch is on the 5′side of the cut, then exonuclease I degrades the DNA 3′to 5′through the mismatch. If the
mismatch is on the 3′side of the cut, then exonuclease VII or RecJ proteindegrades the DNA 5′to 3′through
the mismatch. (C) DNA polymerase III then fills in the missing DNA, andthe gap is sealed with DNA ligase.
Single-stranded DNA-binding protein is also required. The proteinsinvolved in recognizing the mismatch and
making the cuts are MutH, MutS, and MutL. The MutS proteinrecognizes the mismatch. The MutH protein
is the endonuclease that makes the cut. It recognizes the GATCsequence and cleaves the unmethylated DNA
on the 5′side of the G in the GATC. MutS and MutH form acomplex in which they are linked by MutL.
(e.g., colorectal carcinomas). (For reviews, see in a positive supercoil in the unreplicated por-
refs. 59–61.) tion, creating a primer that can be extended
by DNA polymerase III, and keeping the poly-
merase that synthesizes the lagging strand at
3.2 Summary the replication fork. These and other factors
To replicate DNA, several problems must necessitate using about 30 different proteins to
be attended to. These include unwinding the replicate the DNA. The replication fork is cre-
double helix without causing it to overwind ated at a specific sequence in the DNA (called
www.ebook777.com
the origin) when DNaA, HU, and ATP begin Muk, Par, SetB, and MreB proteins. The MreB
to unwind the helix. Further unwinding of the proteins are related to actin.
DNA during replication requires replication Errors made during replication are repaired
fork helicase (DnaB). Single-stranded binding in two ways. Editing repair involves the imme-
protein binds to the single strands to prevent diate removal of the incorrect nucleotide by the
them from coming together again, thus ensur- 3′-exonuclease activity of DNA polymerase III.
ing that they can serve as templates for new The second method of repair, called mismatch
DNA. Meanwhile, DNA gyrase binds to the repair, can occur if the polymerase fails to
DNA downstream of the replication fork and remove the incorrect nucleotide and proceeds
continuously converts the positive supercoils beyond the error site. During mismatch repair
into negative supercoils so that the DNA does an exonuclease removes the segment of copy
not become overwound. DNA that includes the mismatched nucleotide
Replication of the DNA is initiated when the pair, and the gap is filled with DNA polymerase
enzyme primase synthesizes a short RNA oligo- III and sealed with DNA ligase. The copy strand
nucleotide at the replication fork on both the is recognized because it is undermethylated for
single strands. One copy strand, called the lead- a short period after synthesis.
ing strand, is synthesized continuously by elon-
gating the 3′ end of the RNA primer, the end
Study Questions
that faces the replication fork. The other copy
strand, called the lagging strand, is synthesized
1. Describe the roles of the following proteins
in short fragments of about 1,000 nucleotides
in DNA replication: helicase, DNA gyrase,
called Okazaki fragments, growing in the direc-
DNA ligase, DNA polymerase III, DNA
tion opposite to the movement of the replica-
polymerase I, primase, DnaA, SSB.
tion fork. Each fragment begins with a short
oligonucleotide RNA primer. The 3′ end of the 2. Compare the biochemical reactions cata-
lagging strand faces away from the replication lyzed by DNA polymerase and DNA ligase.
fork. The same DNA polymerase would be How do they differ?
able to synthesize both copy strands if the poly- 3. Describe the Meselson–Stahl experiment,
merase were dimeric and the template for the and explain how it proves that DNA is rep-
lagging strand looped around the polymerase licated semiconservatively.
so that its 5′ end faced the replication fork. The
RNA primers at the 5′ end of the copy strands 4. DNA exists in a negative supercoil. How
are removed by DNA polymerase I, which also might this be advantageous to DNA repli-
fills the gap. The break is sealed by DNA ligase. cation and RNA synthesis? During DNA
Replication is usually bidirectional. replication the unreplicated portion of the
Termination occurs at ter sites and requires DNA winds tighter. Why is this the case?
Tus protein. The enzyme topoisomerase IV 5. What are Okazaki fragments, and why are
separates the newly synthesized duplexes. In they necessary?
the event that an unequal number of recombi-
nations have taken place between sister chro- 6. Outline an experimental approach to
mosomes during replication, the daughter determine whether DNA synthesis takes
chromosomes are covalently linked and a site- place in a centrally located “factory,” and
specific recombination must take place at the dif how the origins and termini move during
site, which is in the Ter region. partitioning.
The process of partitioning nucleoids to 7. Why is the 3′-exonuclease activity of DNA
opposite poles of the cells is not understood. polymerase III important?
One model posits a stationary, centrally located
replicon that provides the force to partition sister
chromosomes to opposite poles. Additionally, REFERENCES AND NOTES
several different proteins outside the replisome
1. Most bacteria have circular chromosomes.
appear to be involved in chromosome position- Exceptions include Borrelia burgdorferi, which
ing and partitioning. These include the SMC, causes Lyme disease, Rhodococcus fasciens, and
Streptomyces lividans. See: Hinnebursch, J., and the number of times two strands are crossed per cata-
K. Tilly. 1993. Linear plasmids and chromosomes in lytic event, they alter the linking number in multiples
bacteria. Mol. Microbiol. 10:917–922. of 2. An example of a type II topoisomerase is DNA
gyrase from E. coli. It produces negative supercoils,
2. Supercoiling of DNA is an important aspect of unwinding the double helix in the following way: (1)
its structure in part because it makes the DNA more It binds to the DNA, wrapping approximately 120
compact. Consider DNA as consisting of two strands base pairs of duplex around itself. (2) It cleaves both
coiled around a central axis. Supercoiling refers to strands with a four-base-pair stagger. When it does
the twisting or coiling of the central axis. This is anal- this, the enzyme covalently bonds to the 5′ ends of
ogous to a coiled telephone cord that twists on itself each strand. (3) It moves the uncut complementary
to form additional coils. Suppose the DNA exists as strands through the break. The region that is moved
a closed double-helical circle. If one strand is cut and is either within the region that is wrapped around the
unwound while the other strand is prevented from enzyme or close by. (4) It reseals the breaks. Passing
turning, and then the cut is sealed, the circular DNA the unbroken portion through the broken portion to
will have fewer helical turns. Moreover, the DNA produce negative supercoils unwinds the DNA. ATP
will then be under structural strain and, as a conse- is required for DNA gyrase activity.
quence, will supercoil to relieve the strain. If the DNA
is underwound (i.e., fewer helical turns), it will twist To see how negative supercoiling can unwind a
on itself into negative (left-handed) supercoils to helix, take two pieces of rubber tubing and wrap them
relieve the stress. If the DNA is overwound (i.e., more around each other in a right-hand helix. This would
helical turns are introduced before the cut is sealed), be clockwise, sighting down the helix, away from you.
it will twist upon itself to produce positive (right- Clamp both ends. Now twist the linear helix so that the
handed) supercoils to relieve the strain. Most cellular central axis twists counterclockwise. This produces
DNA is underwound and negatively supercoiled. It negative supercoils. The helix will unwind. Twisting
is believed that the strands of the underwound DNA the helix so that the central axis turns clockwise (posi-
are more readily separated for replication and tran- tive supercoils) makes the helix wind tighter.
scription. The production of negative supercoils is Topoisomerase I and DNA gyrase adjust the super-
accomplished by a topoisomerase II enzyme called coiling of DNA. Gyrase adds negative supercoils and
DNA gyrase and does not occur by unwinding the topoisomerase I removes negative supercoils. For a
cut strands before rejoining them. (See text for how review of topoisomerases, see: Luttinger, A. 1995.
DNA gyrase introduces negative supercoils.) The twisted “life” of DNA in the cell: bacterial topoi-
3. Topoisomerases change the linking number somerases. Mol. Microbiol. 15:601–608.
of a covalently closed duplex of DNA. The link- 4. Marians, K. J. 1992. Prokaryotic DNA replica-
ing number is the number of times that the chains tion. Annu. Rev. Biochem. 61:673–719.
in the duplex of a covalently closed circle cross one
another; that is, it is the number of helical turns. As 5. Baker, T. A., and S. H. Wickner. 1992. Genetics
noted in the text, DNA is said to be overwound when and enzymology of DNA replication in Escherichia
it has more than 10.5 base pairs per helical turn and coli. Annu. Rev. Genet. 26:447–477.
underwound when it has fewer than 10.5 base pairs 6. Messer, W., and C. Weigel. 1996. Initiation
per helical turn. The original model proposed that of chromosome replication, pp. 1579–1601. In:
the topoisomerase cut one strand of the duplex and, Escherichia coli and Salmonella: Cellular and
while holding both ends of the nicked strands, rotated Molecular Biology, Vol. 1. F. C. Neidhardt et al.
a free end around the intact strand, thus changing the (Eds.). ASM Press, Washington, DC.
number of helical turns. This was followed by sealing
the break. A more recent model suggests that rota- 7. Marians, K. J. 1996. Replication fork propa-
tion is not necessary and that the number of helical gation, pp. 749–763. In: Escherichia coli and
turns can be changed simply by passing the unbro- Salmonella: Cellular and Molecular Biology, Vol. 1.
ken strand through the break in the complementary F. C. Neidhardt et al. (Eds.). ASM Press, Washington,
strand. An example is type I topoisomerase. The DC.
type I topoisomerases catalyze a break in one of the 8. Ryan, V. T., J. E. Grimwade, J. E. Camara, E.
strands, which allows the passage through the break Crooke, and A. C. Leonard. 2004. Escherichia coli
of the unbroken partner strand. The break is then pre-replication complex assembly is regulated by
sealed. This results in changing the linking number dynamic interplay among Fis, IHF, and DnaA. Mol.
(number of helical turns) by multiples of one. Type Microbiol. 51:1347–1359.
II topoisomerases do it somewhat differently and
9. Leonard, A. C., and J. E. Grimwade. 2005.
change the linking number by multiples of 2. Type
Building a bacterial orisome: emergence of new regu-
II topoisomerases catalyze breakage of both strands
lator features for replication origin unwinding. Mol.
of the duplex and allow the passage of two unbro-
Microbiol. 55:1365–2958.
ken complementary strands from a nearby region of
the molecule through the breaks before sealing. (The 10. Lopez de Saro, F. J. 2009. Regulation of inter-
breaks are staggered breaks, i.e., not directly opposite actions with sliding clamps during DNA replication
each other.) Because type II topoisomerases change and repair. Curr. Genom. 10:206–215.
www.ebook777.com
11. Johnson, A., and M. O’Donnell. 2005. Cellular boxes are nucleotide repeats containing 9 base pairs;
DNA replicases: components and dynamics at accordingly, they are called 9-mers. The three non-R
the replication fork. Annu. Rev. Biochem. 2005. boxes are called I boxes (because, as we shall see, the
74:283–315. binding of DnaA to these sites is stimulated by integra-
tion host factors), and these are 11-mers. These sites
12. Bidirectional replication can be demonstrated in
differ in their affinity for DnaA. In vitro studies indi-
two ways. One way is to insert into the chromosome
cate that binding initiates at the tighter binding sites
prophage I at the att site and prophage Mu within vari-
and progresses to the weaker binding sites in the fol-
ous mapped genes around the chromosome. The Mu
lowing order: R4,R1 > R2 > R5(M) > I3 > I2 > I1 > R3.
can be located because when it inserts into a gene, it
Studies indicate that DnaA is bound to R4, R1, and
causes a mutation in that gene, and the location of the
R2 throughout most of the cell cycle, and binds to the
gene is known. If such mutants are also auxotrophic
weaker sites just before DNA replication is initiated.
for certain amino acids, the experimenter can stop the
The binding of DnaA at the weaker sites (R5, R3, and
initiation of new rounds of replication by removing the
the integration sites) results in the unwinding of DNA
required amino acid. By adding the amino acid back
to form an open complex within three contiguous
again, the experimenter can initiate new rounds of rep-
13-mers at the left boundary of oriC; all these nucle-
lication at the replication origin. If new rounds of repli-
otide repeats are rich in A–T. Is there is a mechanism
cation are begun with bromouracil in the medium, then
that ensures that DnaA–ATP binds to the weaker sites
bromouracil is incorporated into the DNA in place
at the same time in all the oriC copies? The answer is
of thymine. The presence of the bromouracil makes
yes. Two histonelike DNA-bending proteins, Fis (fac-
the newly synthesized DNA denser than the parental
tor for inversion stimulation) and IHF (integration
strand. The newly synthesized strands can then be sep-
host factor) are involved. These have binding sites in
arated at various times after the initiation of replica-
the right and left half of oriC, respectively.
tion by density gradient centrifugation and hybridized
with both prophages, I and Mu, whereupon the ratio Fis and IHF bind at oriC and bend the DNA.
of Mu DNA to I DNA can be measured. Knowing the Presumably this is important for the mechanism of
map position of the various Mu inserts, it is possible to their action at oriC. Mutants that do not make either
demonstrate that replication proceeds bidirectionally of these proteins are viable, but they cannot initiate
from the origin of replication. Bidirectional replication replication synchronously at multiple origins in fast-
forks have also been observed by using radioautogra- growing E. coli. In vitro studies have shed light on the
phy and electron microscopy after a pulse of tritiated mechanism of action of IHF and Fis. It was reported
thymidine. Both replication forks can be seen to be that Fis binds to oriC in vitro and forms a complex
labeled when replication is examined in several bacte- that blocks the binding of DnaA to low-affinity sites
ria. If DNA replication were unidirectional, only one in oriC, as well as the binding of IHF to oriC, possi-
replication fork would be labeled. bly inhibiting oriC unwinding. It has been suggested
13. This note summarizes what is known about the that increased levels of active DnaA displace Fis and
regulation of the initiation of DNA replication in allow IHF to bind to DNA, resulting in the redistribu-
E. coli. tion of DnaA from high-affinity sites to low-affinity
sites. Whereas Fis binds to oriC during most of the
Several DNA-binding proteins are involved. One cell cycle, IHF binds to oriC only at the beginning of
critical protein is DnaA, which binds to the origin DNA replication. The regulation of the binding of
first, begins the process of strand separation, and IHF to DNA is not understood. As mentioned, the
then recruits helicase. The active form of DnaA that binding of DnaA to the low-affinity sites is necessary
initiates unwinding is DnaA–ATP. The binding of for unwinding and, as a consequence, IHF promotes
ATP to DnaA promotes an allosteric modification unwinding at oriC. Thus, Fis and IHF have opposite
in DnaA, allowing it to function. Nonhydrolyzable effects and help to regulate the initiation and syn-
analogues of ATP also work. After the initiation of chrony of DNA replication.
DNA replication, DnaA hydrolyzes the bound ATP
to bound ADP. DnaA–ADP is inactive in initiating It has been suggested that because Fis is growth
DNA replication. When the activity of DnaA in the rate regulated (lower levels at slow growth rates), the
cell is sufficiently high, DNA replication is initiated. protein functions only in rapidly growing E. coli with
One line of evidence for the role of DnaA in initia- more than one copy of oriC. The levels of IHF are not
tion is derived from studying a temperature-sensitive growth rate related, and it has been suggested that
dnaA mutant harboring a plasmid containing wild- IHF functions under all growth conditions to pro-
type dnaA under the control of the inducible lac mote DnaA binding to the weaker sites in oriC. For
promoter. When the cells are grown at 42 °C, replica- more discussion, see ref. 8.
tion is dependent upon the expression of the plasmid Interestingly, the levels of DnaA do not change very
dnaA gene. The timing of initiation varies with the much during the cell cycle, and part of the regulation
levels of expression of the gene. When DnaA levels are of the initiation of DNA replication is due to the reg-
increased, initiation occurs earlier in the cell cycle. ulation of the activity of existing DnaA. One model
What does DnaA do? DnaA binds to eight sites proposes that the phospholipids in the membrane
in oriC, five of which are called R boxes. The five R keep sequestered DnaA in an inactive form until it is
bound to oriC. In support of this model is the obser- 14. DNA polymerase III has 10 different protein
vation that a fraction of the cellular DnaA can be subunits coded for by separate genes. One of these,
isolated with the membrane fractions. Furthermore, the α subunit, is encoded by the polC (dnaE) gene
DnaA forms complexes with phospholipids and the and is responsible for polymerization. The subunit
phospholipid can inhibit DnaA activity. responsible for the 3′-exonuclease activity is the ε
subunit encoded by the dnaQ (mutD) gene. These
Another factor that blocks the reinitiation of new
two subunits, along with the ϕ subunit, comprise
rounds of DNA replication is sequestration of oriC,
the core of the polymerase. The τ subunit dimerizes
which refers to any mechanism that prevents reinitia-
the core. The other subunits form a clamp that keeps
tion at oriC. To discuss sequestration, it must be pointed
the polymerase on the template. The subunits that
out that the origin region is rich in GATC/CTAG
are not part of the core are sometimes called acces-
sequences and that these sequences are methylated at
sory proteins. The core plus the accessory proteins is
the adenine residues by deoxyadenosine (Dam) methy-
referred to as the holoenzyme.
lase shortly after a strand is copied. For reinitiation to
occur, both strands at oriC must be methylated. 15. The realization that DNA copied from one
template is made in short fragments came from
However, for a period of time after initiation (about
experiments in which bacteriophage T4 that was
one-third of the cell cycle), the origin remains hemi-
infecting E. coli was labeled with tritiated thymidine
methylated; that is, the template strand is methylated
for various lengths of time (2–60 s). The DNA was
but the copy strand is not. This is because newly rep-
denatured with base, and the intermediates were sep-
licated origins are sequestered from Dam methylase.
arated from the template DNA by centrifugation in
The sequestration is due to a protein called SeqA,
a sucrose gradient. It was found that when the DNA
which binds to oriC and prevents methylation of the
was labeled for a short period of time, substantial
new strand. One of the lines of evidence for this is
radioactivity was recovered in small pieces of DNA
that purified SeqA binds to oriC and prevents DNA
(about 1,500 nucleotides long). As the labeling time
replication in vitro. There is also some evidence that
was increased, the radioactivity in the smaller pieces
in E. coli, sequestration of oriC may also involve
remained constant but accumulated in high molecu-
binding of the SeqA/DNA complex to outer mem-
lar weight DNA, as would be expected if the smaller
brane components. Cell fractionation studies have
pieces were precursors to the high molecular weight
revealed that certain subcellular fractions containing
DNA. Okazaki fragments are made during the syn-
outer membrane components are enriched for hemi-
thesis of all DNA molecules in viruses, prokaryotes,
methylated oriC sequences and that hemimethylated
and eukaryotes.
oriC regions bind more readily to membrane frac-
tions than do fully methylated oriC regions. 16. The E. coli sliding β clamp, which is part of the
DNA polymerase III holoenzyme, is actually a ring of
Eventually however, newly replicated DNA
protein with a hollow core through which the DNA
becomes fully methylated by Dam methylase. This
template is threaded. The β clamp tethers the DNA
takes place well before the next round of initiation,
polymerase to the template, thus increasing the pro-
and therefore other controls besides sequestration
cessivity of replication. It does this by binding to the
by SeqA prevent reinitiation during the cell cycle.
catalytic subunit (α subunit) of the core polymerase,
The nature of these other controls in E. coli is a mat-
tethering it to the template DNA. The clamp itself
ter of speculation. However, there is evidence for a
is assembled by a clamp loader, also a part of the
repressor (CtrA) of the initiation of replication in
holoenzyme, which consists of seven different pro-
Caulobacter crescentus. See Section 23.2.1.
tein subunits: three copies of the dnaX gene product,
For reviews, see: Bravo, A., G. Serrano-Heras, and and a single copy each of four other. The dnaX gene
M. Salas. 2005. Compartmentalization of prokary- encodes two proteins, a full-length protein and a
otic DNA replication. FEMS Microbiol. Rev. truncated version formed by a translational frame-
29:25–47, and Funnel, B. E. 1996. The Role of the shift. A subunit of the clamp loader binds to the DNA
Bacterial Membrane in Chromosome Replication and polymerase, tethering it to the clamp loader, and also
Partition. Chapman & Hall, London. Sequestration interacts with DnaB helicase to coordinate helicase
of the origin also plays a role in regulating transcrip- activity with fork progression. In addition, the sub-
tion in the oriC region, including the transcription of unit in the clamp loader stimulates the release of the
dnaA. For a discussion of this point, see: Bogan, J. A., polymerase from the Okazaki fragment, allowing the
and C. E. Helmstetter. 1997. DNA sequestration and lagging-strand polymerase to be used for the exten-
transcription in the oriC region of Escherichia coli. sion of the next primer.
Mol. Microbiol. 26:889–896. For information on
17. Kuempel, P. L., A. J. Pelletier, and T. M. Hill.
the role of SeqA in sequestering oriC, read: Lu, M., J.
1989. Tus and the terminators: the arrest of replica-
L. Campbell, E. Boye, and N. Kleckner. 1994. SeqA:
tion in prokaryotes. Cell 59:581–583.
a negative modulator of replication initiation in E.
coli. Cell 77:413–426, and Bach, T., M. A. Krekling, 18. Khatri, G. S., T. MacAllister, P. R. Sista, and D.
and K. Skarstad. 2003. Excess SeqA prolongs seques- Bastia. 1989. The replication terminator protein of
tration of oriC and delays nucleoid segregation and E. coli is a DNA sequence-specific contrahelicase.
cell division. EMBO J. 22:315–323. Cell 59:667–674.
www.ebook777.com
19. Manna, A. C., K. S. Pai, D. E. Bussiere, C. 29. Errington, J. 1998. Dramatic new view of bac-
Davies, S. W. White, and D. Bastia. 1996. Helicase– terial chromosome segregation. ASM News 64:
contrahelicase interaction and the mechanism of ter- 210–217.
mination of DNA replication. Cell 87:881–891.
30. Webb, C. D., A. Teleman, S. Gordon, A. Straight,
20. B. subtilis also has 6 Ter sites (TerI–TerVI), and and A. Belmont. 1997. Bipolar localization of the
they are polar as in E. coli. The protein in B. sub- replication origins of chromosomes in vegetative and
tilis that binds to the Ter sequences is called RTP sporulating cells of B. subtilis. Cell 88:667–674.
(replication terminator protein). The B. subtilis Ter
31. Yongfang, L., B. Youngren, K. Serqueev, and S.
sites and RTP do not seem to be related in sequences
Austin. 2003. Segregation of the Escherichia coli chro-
to the E. coli terminators or Tus.
mosome terminus. Mol. Microbiol. 50:825–834.
21. Kato, J.-I., Y. Nishimura, R. Imamura, H. Niki,
32. Mohl, D. A., and J. W. Gober. 1997. Cell cycle–
S. Hiraga, and H. Suzuki. 1990. New topoisomerase
dependent polar localization of chromosome par-
essential for chromosome segregation in E. coli. Cell
titioning proteins in Caulobacter crescentus. Cell
63:393–404.
88:675–684.
22. Kato, J.-I., Y. Nishimura, M. Yamada, H.
33. Jensen, R. B., S. C. Wang, and L. Shapiro. 2001.
Suzuki, and Y. Hirota. 1988. Gene organization in
A moving DNA replication factory in Caulobacter
the region containing a new gene involved in chro-
crescentus. EMBO J. 20:4952–4963.
mosome partition in Escherichia coli. J. Bacteriol.
170:3967–3977. 34. Reyes-Lamonthe, R., D. J. Sherratt, and M. C.
23. Cell division and DNA metabolism are coupled. Leake. 2010. Stoichiometry and architecture of
That is, cell division is inhibited whenever DNA repli- active DNA replication machinery in Escherichia
cation or partitioning is blocked. This is advantageous coli. Science 328:498–501.
in that it lowers the frequency of formation of anucle- 35. The replisome can be visualized via fluores-
ate cells. When cell division is inhibited, long cells or cence microscopy by fusing the DNA polymerase
filaments form. If chromosome separation or segrega- to the green fluorescent protein (GFP) from the jel-
tion is impaired, then the nucleoids are abnormally lyfish Aequorea victoria. These experiments support
placed in the elongated cells or filaments. There are the idea of a stationary replisome in B. subtilis and
two mechanisms responsible for inhibiting cell divi- E. coli. For more information on how the fusions are
sion. One of these requires SulA. Blocking DNA repli- made and introduced into the cell, review the discus-
cation results in the synthesis of SulA as part of the SOS sion of ZipA in Section 2.6.3.
response. (See Section 16.5.2 for a discussion of the
SOS response and SulA.) SulA inhibits the formation DNA can be visualized by using the fluorescent
of the FtsZ septal ring; hence cell division is inhibited. protein GFP. The DNA can be labeled in specific
There is also a SulA-independent block in cell division regions by inserting into the DNA a cassette of many
whenever DNA replication or segregation is inhibited. copies of lac operators (lacO) and labeling the lacO
copies with a fusion of the Lac repressor (Lacl) to
24. Ward, D., and A. Newton. 1997. Requirement GFP (Lacl–GFP). This has been done to label and fol-
of topoisomerase IV parC and parE genes for cell low unreplicated DNA into the replisome and repli-
cycle progression and developmental regulation in cated DNA out of the replisome. This technique has
Caulobacter crescentus. Mol. Microbiol. 26:897– allowed investigators to demonstrate that the origin
910. is replicated first and the copies move away from the
25. Kuempel, P. L., J. M. Henson, L. Dirks, M. replisome toward opposite cell poles, that chromo-
Tecklenburg, and D. F. Lim. 1991. dif, a recA- somal regions located midway between oriC and
independent recombination site in the terminus terC move toward the centrally located replisome,
region of the chromosome of Escherichia coli. New and that the termini follow last into the replisome
Biol. 3:799–811. and, after replication, become positioned in opposite
halves of the cell, closer to the cell center than the ori-
26. Blakely, G., G. May, R. McCulloch, L. K. gins. The terminus has also been visualized by insert-
Arciszewska, M. Burke, S. T. Lovett, and D. J. ing a single parS site near it and labeling the parS site
Sherratt. 1993. Two related recombinases are with GFP–ParB. ParA and ParB have also been visu-
required for site-specific recombination at dif and cer alized via immunofluorescence microscopy bound as
in E. coli K12. Cell 75:351–361. a complex to the parS site at the origin in C. cres-
27. Pogliano, K., J. Pogliano, and E. Becker. 2003. centus. The bacteria were fixed with glutaraldehyde
Chromosome segregation in eubacteria. Curr. Opin. and formaldehyde and then washed and treated with
Microbiol. 6:586–593. lysozyme. After further treatment with methanol
and acetone, they were incubated with antibody. The
28. Hojgaard, A., H. Szerlong, C. Tabor, and P. antibody was visualized with secondary antibody,
Kuempel. 1999. Norfloxacin-induced cleavage which was goat anti-rabbit IgG conjugated to either
occurs at the dif resolvase locus in Escherichia coli fluorescein or biotin. The biotin-conjugated anti-
and is the result of interaction with topoisomerase body was visualized with streptavidin-conjugated
IV. Mol. Microbiol. 33:1027–1036. Texas red dye, which binds to avidin.
36. Draper, G. C., and J. W. Gober. 2002. Bacterial The par system in the E. coli multidrug-resistant
chromosome segregation. 2002. Annu. Rev. Microbiol. plasmid TP228 consists of (1) an ATPase called ParF,
56:567–597. which is part of a subgroup of the ParA superfamily
of proteins (not related to ParM), (2) a DNA-binding
37. Wu, J. L., and J. Errington. 2002. A large dis- protein called ParG, which is not related to ParB, and
persed chromosomal region required for chromo- (3) a region upstream of the parFG genes to which
some segregation in sporulating cells of Bacillus ParG attaches. References that can be consulted
subtilis. EMBO J. 21:4001–4011. include the following: Golovanov, A. P., D. Barillà,
38. Gerdes, K., J. Møller-Jensen, G. Ebersbach, T. M. Golovanova, F. Hayes, and L.-Y. Lian. 2003.
Kruse, and K. Nordström. 2004. Bacterial mitotic ParG, a protein required for active partition of bacte-
machineries. Cell 116:359–366. ria plasmids, has a dimeric ribbon–helix–helix struc-
ture. Mol. Microbiol. 50:1141, and Barillà, D., and F.
39. Ben-Yehuda, S., D. Z. Rudner, and R. Losick. Hayes. 2003. Architecture of the ParF/ParG protein
2003. RacA, a bacterial protein that anchors chro- complex involved in prokaryotic DNA segregation.
mosomes to the cell poles. Science 299:532–536. Mol. Microbiol. 49:487–499.
40. Wheeler, R. T., and L. Shapiro. 1997. Bacterial 42. Niki, H., A. Jaffé, R. Imamura, T. Ogura, and
chromosome segregation: is there a mitotic appara- S. Hiraga. 1991. The new gene mukB codes for a
tus? Cell 88:577–579. 177 kD protein with coiled-coil domains involved
41. Research with low-copy plasmids has revealed in chromosome partitioning of E. coli. EMBO J.
plasmid segregation systems based upon par systems 10:183–193.
encoded by the plasmid. Generally, the par system 43. Niki, H., R. Imamura, M. Kitaoka, K. Yamanaka,
consists of three components: (1) a Par protein that T. Ogura, and S. Hiraga. 1992. E. coli MukB protein
has ATPase activity, which will be referred to here involved in chromosome partition forms a homodi-
as ParA, (2) a Par DNA-binding protein, referred to mer with a rod-and-hinge structure having DNA
here as ParB, and (3) a centromere-like region in the binding and ATP/GTP binding activities. EMBO J.
DNA to which ParB attaches. As will be explained 11:5101–5109.
later, plasmids can differ in their Par proteins, and
the reference to ParA and ParB here is for the sake of 44. Yamanaka, K., T. Ogura, H. Niki, and S.
convenience. Plasmid pairs move to the midregion of Hiraga. 1996. Identification of two new genes, mukE
the cell prior to septation and then are moved bidirec- and mukF, involved in chromosome partitioning in
tionally (segregated) to opposite cell poles. The Par Escherichia coli. Mol. Gen. Genet. 250:241–251.
proteins are required for the segregation of the plas- 45. Muk is derived from the Japanese word mukaku,
mids. ParA and ParB assemble at the centromere-like which means anucleate. When grown at 22 °C,
region, and the resulting complex of DNA and Par mukB mutants form normal nucleated rods with a
proteins interacts with a bacterial scaffoldlike struc- small percentage (5%) of anucleate cells. At higher
ture that is involved in moving the plasmids bidirec- temperatures (e.g., 42 °C) the cells die, and anucleate
tionally toward the cell poles. Some specific examples cells and multinucleate filaments with abnormal pat-
of plasmid segregation will now be described. terns of nucleoids accumulate (clumps and isolated
The system for the E. coli R1 plasmid consists of the nucleoids). Examination of two paired daughter cells
ParM ATPase, the ParR DNA-binding protein, and revealed that some of the anucleate cells were paired
parC, the putative centromere-like region. The sepa- with cells that appeared to have more than one nucle-
rated paired plasmids at midcell move to opposite cell oid, reflecting the fact that both daughter chromo-
poles. The ParM protein is not related to ParA. Rather, somes remained with one of the daughter cells upon
it is a member of the MreB family and forms actinlike division.
helical filaments at the nucleation point of ParR/parC. Mutations in two other genes also lead to the pro-
(In addition to the ParR/parC complex, polymeriza- duction of anucleate cells. These are the mukF and
tion requires ATP and Mg2+, and depolymerization mukE genes. There probably encode alternatives to
requires ATP hydrolysis.) It has been suggested that the Muk proteins. This conclusion is based upon the
the polymerization of ParM provides the driving force finding that null mutants in mukB, mukF, or mukE
that moves the plasmid bound to the ParM protein show a decreased plating efficiency and a very large
toward a cell pole so that each daughter cell receives at number of anucleate cells only at temperatures
least one copy of the plasmid. This could occur if the higher than 22 °C, indicating that the proteins are
plasmids were located at the tips of the filaments (which not absolutely essential at the lower temperatures.
they are), and if polymerization occurred between Similarly, strains of B. subtilis and C. crescentus
plasmid copies. Another possibility is that the ParM bearing null mutations in smc have a severe plat-
filaments are a scaffolding along which the plasmids ing deficiency in rich media at higher temperatures.
move toward opposite cell halves. For a discussion of The B. subtilis mutants form a significant number
how ParM might be involved in plasmid segregation, of anucleate cells, and the majority of C. crescentus
see: Møller-Jensen, J., R. B. Jensen, J. Löwe, and K. mutants are “stalled” at a predivisional stage, do
Gerdes. 2002. Prokaryotic DNA segregation by an not elongate, and are defective in nucleoid distribu-
actin-like filament. EMBO J. 21:3119–3127. tion.
www.ebook777.com
46. Espeli, O., P. Nurse, C. Levine, C. Lee, and K. J. 54. Immunofluorescence microscopy to visual-
Marians. 2003. SetB: an integral membrane protein ize ParA and ParB was performed with synchro-
that affects chromosome ion in Escherichia coli. Mol. nized cultures and the appropriate antibodies. The
Microbiol. 50:495–509. procedure used is detailed in the last paragraph of
note 34.
47. Van den Ent, F., L. A. Amos, and J. Löwe. 2001.
Prokaryotic origin of the actin cytoskeleton. Nature 55. Jensen, R. B., S. C. Wang, and L. Shapiro. 2001.
413:39–44. A moving DNA replication factory in Caulobacter
crescentus. EMBO J. 20:4952–4963.
48. Figge, R. M., and J. W. Gober. 2003. Cell shape,
division and development: the 2002 American Society 56. Frameshift mutations are mutations due to the
for Microbiology (ASM) Conference on Prokaryotic insertion or removal of one or more (but not three)
Development. Mol. Microbiol. 47:1475–1483. bases from DNA. Acridine dyes such as ethidium
bromide and proflavine cause frameshift mutations.
49. Real, G., S. Autret, J. E. Harry, J. Errington, and They insert between the bases, thus increasing the
A. O. Henriques. 2005. Cell division protein DivlB distance between bases and preventing them from
influences the SpoOJ/Soj system of chromosome aligning correctly. As a consequence, the two strands
segregation in Bacillus subtilis. Mol. Microbiol. of DNA slip with respect to each other. This leads to
55:349–367. the insertion or deletion of a base.
50. Wu, L. J., and J. Errington. 2003. RacA and the 57. Modrich, P. 1991. Mechanisms and biologi-
Soj–SpoOJ systems combine to effect polar chromo- cal effects of mismatch repair. Annu. Rev. Genet.
some segregation in sporulating Bacillus subtilis. 25:229–253.
Mol. Microbiol. 49:1463.
58. Schofield, M. J., and P. Hsieh. 2003. DNA mis-
51. Ben-Yehuda, S., D. Z. Rudner, and R. Losick. match repair: molecular mechanisms and biological
2003. RacA, a bacterial protein that anchors chro- function. Annu. Rev. Microbiol. 57:579–608.
mosomes to the cell poles. Science 299:532–536.
59. Kolodner, R. 1996. Biochemistry and genet-
52. Marston, A. L., H. B. Thomaides, D. H. Edwards, ics of eukaryotic mismatch repair. Genes Dev. 10:
M. E. Sharpe, and J. Errington. 1998. Polar localiza- 1433–1442.
tion of the MinD protein of Bacillus subtilis and its
role in selection of the mid-cell division site. Genes 60. Modrich, P. 1995. Mismatch repair, genetic sta-
Dev. 12:3419–3430. bility and tumour avoidance. Philos. Trans. R. Soc.
Lond. B 347:89–95.
53. Mohl, D. A., and J. W. Gober. 1997. Cell cycle-
dependent polar localization of chromosome par- 61. Modrich, P. 1994. Mismatch repair, genetic sta-
titioning proteins in Caulobacter crescentus. Cell bility and cancer. Science 266:1959–1960.
88:675–684.
4
Membrane Bioenergetics:
The Proton Potential
A major revolution in our conception of mem- the principles of the chemiosmotic theory will
brane bioenergetics has taken place in the last give the reader a deeper understanding of how
50 years as a result of the theoretical ideas of the bacterial cell uses ion gradients to couple
Peter Mitchell referred to as the chemiosmotic energy-yielding (exergonic) reactions to energy-
theory.1–6 Briefly, the chemiosmotic theory requiring (endergonic) reactions. This chapter
states that energy-transducing membranes explains the principles of the theory. For a brief
(i.e., bacterial cell membranes; mitochondrial historical account, see Box 4.1.
and chloroplast membranes) pump protons
across the membrane, thereby generating an
electrochemical gradient of protons across the 4.1 The Chemiosmotic Theory
membrane (the proton potential) that can be According to the chemiosmotic theory, protons
used to do useful work when the protons return are translocated out of the cell by exergonic
across the membrane to the lower potential. In (energy-producing) driving reactions, which
other words, bacterial, chloroplast, and mito- are usually biochemical reactions (e.g., respira-
chondrial membranes are energized by proton tion, photosynthesis, ATP hydrolysis). Some of
currents. the translocated protons leave behind negative
Of course, the return of the protons across counterions (e.g., hydroxyl ions), thus estab-
the membrane must be through special proton lishing a membrane potential, outside positive.
conductors that couple the translocation of pro- Protons may also accumulate electroneutrally
tons to do useful cellular work. These proton in the extracellular bulk phase, establishing a
conductors are transmembrane proteins. Some proton concentration gradient that is high on
membrane proton conductors are solute trans- the outside (outside acid). When the protons
porters, others synthesize ATP, and others are return to the inside, moving down the concen-
motors that drive flagellar rotation. The proton tration gradient and toward the negative pole of
potential provides the energy for other mem- the membrane potential, work can be done.
brane activities besides ATP synthesis, solute Figure 4.1 illustrates the proton circuit in a
transport, and flagellar motility (e.g., reversed bacterial cell membrane. The cell membrane
electron transport and gliding motility). is similar to a battery in that it maintains a
Because the chemiosmotic theory is central potential difference between the inside and out-
to energy metabolism, it lies at the foundation side, except that the current that flows is one
of all bacterial physiology. This chapter shows of protons rather than electrons. In Fig. 4.1,
how the chemiosmotic theory brings together the potential difference is maintained by reac-
principles of physics and thermodynamics in tions 1, which translocate protons to the out-
explaining membrane bioenergetics. A study of side. Reactions 1 include redox reactions that
111
www.ebook777.com
BOX 4.1 HISTORICAL PERSPECTIVE:

OXIDATIVE PHOSPHORYLATION
The first mechanism suggested to explain electron carrier protein, and this energy is
how electron transport was coupled to then used to drive the synthesis of ATP from
ATP synthesis was published in 1953. It ADP and inorganic phosphate.
was postulated that mitochondria make a Even as late as 1977 there were some
high-energy phosphorylated derivative that reservations. Mitchell’s hypothesis was
donates a phosphoryl group to ADP. The termed “an attractive possibility” that had
model for this mechanism was based upon not been proven.1
the mechanism of ATP synthesis in the cyto- It was, however, accepted as being cor-
sol catalyzed by triosephosphate dehydro- rect and of far-reaching importance, and
genase and phosphoglycerate kinase (see in 1978 Mitchell won the Nobel Prize in
later: Fig. 9.2, reactions 6 and 7). However, Chemistry. For more information about the
no investigators were able to find the postu- prevalent ideas in 1977 regarding electron
lated phosphorylated intermediate despite transport and oxidative phosphorylation,
many attempts. the student is referred to a series of review
A second postulated mechanism was pub- articles.2
lished in 1961 by the English scientist Peter
Mitchell (1920–1992). Mitchell hypoth- REFERENCES
esized that electron transport is coupled to
1. Boyer, P. D. 1977. Coupling mechanisms in
the generation of an electrochemical proton capture, transmission, and use of energy. Annu.
gradient, which in turn drives ATP synthesis Rev. Biochem. 46:957–966.
(the chemiosmotic theory). In 1973 a third
2. Boyer, P. D., B. Chance, L. Ernster, P. Mitchell,
mechanism was suggested, namely, that the E. Racker, and E. C. Slater. 1977. Oxidative
energy released during electron transport is phosphorylation and photophosphorylation.
trapped in a conformational change in an Annu. Rev. Biochem. 46:955–1026.
occur during electron transport (Fig. 4.2, 1) (Fig. 4.2, 2). (See Sections 4.7.1 and 4.8.1.) The
and an ATP-driven proton pump (the ATP syn- reentry of sodium ions can also be coupled to
thase) (Fig. 4.2, 7). These will be discussed later the performance of work (e.g., solute uptake
(Sections 4.7.1 and 4.7.2). The cell membrane Fig. 4.2, 5). Once established, the membrane
does work via reactions 2. The work that is potential can energize the secondary flow of
done by the protons that enter the cell includes other ions. For example, the influx of potassium
the extrusion of sodium ions (Fig. 4.2, 3), sol- ions can be in response to a membrane potential,
ute transport (Fig. 4.2, 4), flagellar rotation inside negative, created by proton extrusion.
(Fig. 4.2, 6), and the synthesis of ATP via the Mitochondrial and chloroplast membranes are
ATP synthase (Fig. 4.2, 7, and Section 4.6.2). also energized by proton gradients. Therefore,
As Mitchell emphasized (see Box 4.1), the mem- this is a widespread phenomenon, not restricted
brane’s low permeability to protons is impor- to prokaryotes.
tant because the major route of proton reentry is
via the energy-transducing proton transporters
rather than by general leakage. This, of course, 4.2 Electrochemical Energy
would be expected of a lipid bilayer that is rela- When bacteria translocate protons across the
tively nonpermeable to protons. Some bacteria membrane to the outside surface, energy is
couple respiration or the decarboxylation of conserved in the proton gradient that has been
carboxylic acids to the extrusion of sodium ions established. The energy in the proton gradient
membrane bioenergetics: the proton potential 113
H+ The same description applies to chemical

energy. Energy is required to move the proton
1
against its concentration gradient. This energy
is stored in the concentration gradient. The
energy energy that is stored in a concentration gradi-
ent is called chemical energy. When the proton
2
returns to the lower concentration, the energy
in the concentration gradient is dissipated and
work can be done.
work
The sum of the changes in electrical and
chemical energies is called electrochemical
energy. The symbol for electrochemical energy
is ∆μ, which is equal to μin − μout. For the proton,
Fig. 4.1 The proton current. There is a proton circuit it would be ∆μH+ . Electrochemical energy, dis-
traversing the bacterial cell membrane. Protons are cussed in more detail in the sections that follow,
translocated to the cell surface, driven there by either is now expressed in joules per mole.
chemical or light energy through a proton pump
(1) and returned through special proton transport- 4.2.1 The electrochemical energy
ers (2) that do work. The accumulation of protons of protons
on the outside surface of the membrane establishes
a membrane potential, outside positive. A pH gradi-
The proton motive force
ent can also be established, outside acid. In several The electrochemical work that is performed
gram-negative bacteria oxidizing certain inorganic when an ion crosses a membrane is a function of
compounds (lithotrophs), or single-carbon com- both the membrane potential, ∆Ψ, and the dif-
pounds such as methanol, protons that are released ference in concentration between the solutions
into the periplasm via periplasmic oxidations con- separated by the membrane. For example, for
tribute to the proton current (Chapter 13). In some one mole of protons:
cases, periplasmic oxidations are the sole provider of
protons for inward flux. ∆μH+ = F∆Ψ + RT ln([H+]in/[H+]out) joules
(4.1)
is both electrical and chemical. The electrical
energy exists because a positive charge (i.e., In eq. 4.1, F∆Ψ represents the electrical energy
the proton) has been moved to one side of the when one mole of protons moves across a poten-
membrane, creating a charge separation, and tial difference of ∆Ψ volts, and RT ln([H+]in/
therefore a membrane potential. When the [H+]out) represents the chemical energy when
proton moves back into the cell toward the one mole of protons moves across a concentra-
negatively charged surface of the membrane, tion gradient of [H+]in/[H+]out.
the membrane potential is dissipated (i.e., To express eq. 4.1 in millivolts (mV), we sim-
energy has been given up and work can be ply divide by the Faraday constant F (≈96,500
done). The energy dissipated when the proton coulombs/mole; see later). Since RT/F ln([H+]in/
moves to the inside of the cell is equal to the [H+]out) = –60 ∆pH at 30 °C, where ∆pH =
energy required to translocate the proton to pHin – pHout, eq. 4.1 is expressed in millivolts as
the outside. follows:
Stated more precisely, energy is required to
move a charge against the electric field (i.e., ∆p = ∆μH+ /F
to the side of the same charge). This energy is = ∆Ψ − 60 ∆pH mV (at 30 oC) (4.2)
stored in the electric field. The energy that is
stored in the electric field is called electrical Usually, ∆μH+/F is called the proton motive force
energy. Conversely, the electric field gives up and is denoted as ∆p. The ∆p is the potential
energy when a charge moves with the electric energy in the electrochemical proton gradient.
field (i.e., to the opposite pole) and work can When protons move toward the lower electro-
be done. The amount of energy is the same, but chemical potential, the ∆p gives up energy (is
opposite in sign. dissipated) and work can be done (e.g., flagellar
www.ebook777.com
Fig. 4.2 An overview of the proton and sodium ion currents in a generalized bacterial cell Driving reactions
(metabolic reactions) deliver energy to create proton (1) and sodium ion (2) electrochemical gradients, high on
the outside. The major driving reactions encompassed by reaction 1 are the redox reactions that occur during
electron transport. The establishment of sodium ion potentials coupled to metabolic reactions such as respira-
tion is not widespread. When the ions return to the lower electrochemical potentials on the inside, work can
be done. Built into the membrane are various transporters (porters) that translocate protons and sodium ions
back into the cell, completing the circuit, and in the process doing work. There are three classes of porters:
(a) antiporters, which carry two solutes in opposite directions, (b) symporters, which carry two solutes (S) in
the same direction, and (c) uniporters, which carry only a single solute. The Na+/H+ antiporter (3) is the major
mechanism for extruding Na+ in bacteria and also functions to bring protons into the cell for pH homeostasis
in alkaliphilic bacteria. In most bacteria the Na+/H+ antiporter creates the sodium potential necessary for the
Na+/solute symporter because a primary Na+ pump is not present. The antiporter uses the proton electro-
chemical potential as a source of energy. The H+/solute symporter (4) uses the proton potential to accumulate
solutes, and a Na+/solute symporter (5) uses the sodium electrochemical potential to accumulate solutes. Also
shown are a flagellar motor that turns at the expense of the proton electrochemical potential (6) and an ATP
synthase that synthesizes ATP at the expense of the proton electrochemical potential (7). The ATP synthase is
reversible and can create a proton electrochemical potential during ATP hydrolysis.
rotation, ATP synthesis, solute transport). several exergonic reactions (reactions that give
Cells must continuously replenish the ∆p as it up energy), the most widely used being oxida-
is used for doing work. One should view the ∆p tion–reduction reactions in the membrane that
as a force pulling protons across the membrane occur during respiration, and ATP hydrolysis.
into the cell toward their lower electrochemi- Bacteria maintain an average ∆p of approxi-
cal potential. To replenish the ∆p, an equal but mately –140 to –200 mV. The values for respir-
opposite force must be exerted to push protons ing bacteria tend to be a little more positive or
out of the cell toward the higher electrochemical less positive than those for fermenting bacteria.7
potential (against the ∆p). As will be discussed Equation 4.2 is a fundamental equation in cell
later, the force that generates the ∆p (translo- biology. It is derived in more detail later. (See
cates protons out of the cell) can result from eqs. 4.3–4.10.)
Units: what is meant by volts, as the potential energy in the concentration gra-
electron volts, and joules? dient, 0.180 V (17,400/F). We can also think
It is important to distinguish between volts (V), about volts or millivolts as a force that pushes
electron volts (eV), and joules (J). Potential a molecule down its electrical, electrochemical,
energy differences (e.g., ∆E, ∆Ψ, ∆p) are or chemical gradient. The ∆p, in millivolts, is
expressed as volts or millivolts. As long as the force that pushes protons; thus it is called
charges are not moving, these remain as poten- the proton motive force.
tial energy and work is not done. When charges
are moving, work is being done, either on the A more detailed explanation of and
charges or by the charges, depending upon derivation of eq. 4.2
whether the charges are moving toward a 1. The electrical component of the Δ μH+
higher potential or a lower potential. The quan- A membrane potential, ∆Ψ, exists across the cell
tity of work that is done is proportional to the membrane where ∆Ψ = Ψin – Ψout. By conven-
product of the amount of charge that moves and tion, the ∆Ψ is negative when the inner mem-
the potential difference over which the charge brane surface is negative. The volt is the unit of
moves. ∆Ψ. The work done on or by the electric field
The units of work are either joules or elec- when charges traverse the membrane potential
tron volts. One joule (J) is defined as the energy is equal to the total charges carried by the ions
required to raise a charge of one coulomb (C) or electrons multiplied by the ∆Ψ. If a single
through a potential difference of one volt (i.e., electron or monovalent ion such as the proton
J = C × V). (The older literature used calorie units moves across the membrane, then the work
instead of joules: 1 calorie = 4.184 J.) To calcu- done is ∆Ψ electron volts. (For a divalent ion,
late the change in joules when a mole of mon- the work done would be 2∆Ψ electron volts.)
ovalent ions or electrons travels over a voltage The amount of work that is done per mole of
gradient, one multiplies volts by F, the Faraday protons that traverses the ∆Ψ is
constant, since this is the number of coulombs
∆G = F∆Ψ joules (4.3)
of charge per mole of electrons or monovalent
ions. (See note 8.) One electron volt is (rather Bearing in mind that coulombs × volts = joules
than a coulomb) the increase in energy of a (i.e., V = J/F), eq. 4.3 is often expressed as electri-
single electron or monovalent ion when raised cal potential energy (or force) in volts by divid-
through a potential difference of one volt. The ing both sides of the equation by the Faraday
charge on the electron or monovalent ion is constant:
approximately 1.6 × 10–19 coulomb. Therefore,
one electron volt is equal to 1.6 × 10–19 joule. If ∆G/F = ∆Ψ volts (4.4)
the electrons are moving toward a lower energy 2. The chemical component of the Δ μH+
level (i.e., toward a higher electrode potential), Of course if a concentration gradient of protons
then work (e.g., the generation of a ∆p) can be exists, we must add the chemical energy to the
obtained from the system. The energy available electrical energy in eq. 4.3 to obtain the expres-
from the electron flow is n∆E eV or nF∆E joules sion for the electrochemical energy, ∆μH+. The
(if n refers to moles of electrons). An equivalent chemical energy of the proton (or any solute) as
amount of work must be done to move the elec- a function of its concentration is
trons to a higher energy level.
Similarly, if y protons move over a potential G = G0 + RT ln[H+] joules (4.5)
of ∆p volts, then y∆p electron volts of work is
done. If y moles of protons move, then yF∆p (See note 9 for a more complete discussion.)
joules of work is done. Often work units are The free energy change accompanying the
converted to volts or millivolts when one wishes transfer of one mole of protons between a solu-
to express the potential energy in a concentration of protons outside the cell [H+]out and inside
the cell [H+]in is the difference between the free
tion gradient. For example, 17,400 J is required
energies of the two solutions,
to move one mole of solute against a concentra-
tion gradient of 1,000. This can be expressed ∆G = RT ln[H+]in − RT ln[H+]out joules (4.6)
www.ebook777.com
or motive force, or ∆p) by dividing by the Faraday

constant (or by adding eqs. 4.4 and 4.8):
∆G = RT ln([H+]in /[H+]out) joules
∆p = ∆Ψ − 60 ∆pH millivolts at 30 oC (4.10)
Equation 4.6 refers to the free energy change
when one mole of protons moves from one con- The same equation is used to express the elec-
centration to another, in which the concentra- trochemical potential for any ion (e.g., Na+),
tion gradient does not change (i.e., as applies
∆μNa+ /F = ∆Ψ − 60 ∆pNa millivolts (4.11)
to a steady state). It does not refer to the total
energy released when the concentration of pro- where pNa is −log(Na+).
tons comes to equilibrium. Equation 4.6 can be By convention, the values of the potentials
used for the movement of any solute over a con- are always negative when the cell membrane is
centration gradient, not simply protons, and energized for that particular ion.
we will see this equation again when we discuss
solute transport. (In describing the movement 4.2.2 Generating a ∆Ψ and a ∆pH
of solutes other than protons, the symbol S may We now consider some biophysical aspects of
be substituted for H+. Otherwise, the equation generating the proton motive force. First we
used is identical to eq. 4.6.) will examine the formation of the ∆Ψ, and then
Usually, eq. 4.6 is expressed in electrical units we will consider the establishment of a ∆pH.
of potential (volts). To do this, one substitutes
8.3144 J K–1 mol–1 for R, 303 K (30 °C) for T, Electrogenic flow creates a ∆Ψ
converts ln to log10 by multiplying by 2.303, The movement of an uncompensated charge
and divides by F to convert joules to volts, thus creates the membrane potential. When this
deriving eq. 4.7: happens, the charge movement is said to be
∆G/F = 0.06 log([H+]in/[H+]out) volts electrogenic. The moving charge can be either a
proton or an electron (or in fact any uncompen-
= 60 log([H+]in/[H+]out) millivolts (4.7) sated charge). For example, a membrane poten-
tial is generated when a proton is translocated
Since log([H+]in/[H+]out) = pHout – pHin, eq. 4.7 through the membrane to the outer surface,
can be written as leaving behind a negative charge on the inner
surface:
∆G/F = 60 (pHout – pHin)
H+in → H+out
= −60 (pHin – pHout)
A membrane potential can also develop if a
= −60 ∆pH millivolts (4.8)
molecule that has been reduced on the inner
membrane surface, picking up cytoplasmic
3. Proton electrochemical energy, ΔμH+
protons in the process, then diffuses across the
The sum of the electrical (1) and chemical
membrane. This molecule is oxidized on the
energies (2) of the proton is the proton electro-
outer surface, releasing the protons to the exte-
chemical energy. We are now ready to derive
rior, while the electrons return electrogenically
an expression for the proton motive force.
across the membrane to the inner surface:
The total energy change accompanying the
movement of one mole of protons through the A + 2e− + 2H+in → AH2
membrane is the sum of the energy due to the
membrane potential (eq. 4.3) and the energy AH2 → A + 2e− + 2H+out
due to the concentration gradient (eq. 4.6). This
sum is called the electrochemical energy, ∆μH+: This is also electrogenic flow, but here the
electron is the moving charge rather than the
∆μH+ = F∆Ψ + RT ln([H+]in/[H+]out) joules proton. The work done is the same because
the electron carries the same charge as the pro-
(4.9)
ton (i.e., 1.6 × 10–19 C). It makes no difference
One can also express the electrochemical energy from the point of view of calculating the ∆p
as a potential in volts or millivolts (proton whether one thinks in terms of protons moving
as positive charges across the membrane from limits electrogenic proton pumping and ulti-
the inside to the outside, or simply as being car- mately the size of the membrane potential itself.
ried as hydrogen atoms in a reduced organic This is because the membrane potential, which
compound (e.g., AH2). There is a net transloca- is negative on the inside, pulls protons back into
tion of protons in either case. In both instances, the cell.
the total energy necessary to move y protons
against the ∆p is y∆p electron volts. We will Generating a ∆pH
see that bacteria use both electrogenic electron
When a proton is translocated across the mem-
flow and electrogenic proton flow to create a
brane, a ∆Ψ and a ∆pH cannot be created
membrane potential.
simultaneously. This statement is summarized
The material between the two faces of the
graphically in Fig. 4.3. Let us first discuss cre-
membrane consists in part of long hydrophobic
ating a ∆pH (i.e., accumulating protons in the
regions of the phospholipids in the lipid bilayer
external bulk phase). Remember that the bulk
that prevent the protons from rapidly leaking
external medium can become acidified during
back into the cell, and so the cell membrane
proton translocation only if electrical neutral-
stores positively charged protons on its outside
ity is conserved: that is, each proton in the bulk
surface. It is only when the protons traverse the
phase must have a negative counterion. This
membrane via the appropriate protein machin-
can happen if the proton is pumped out with
ery that the flow of charges can be coupled to
an anion (i.e., H+/R−) or if a cation enters the
doing work (e.g., the synthesis of ATP via the
cell from the external bulk phase in exchange
ATP synthase).
for the proton (i.e., H+ in exchange for R+)
(Fig. 4.3). Since, however, these would be elec-
The membrane potential that develops troneutral events, a charge separation, hence
when even a small number of protons move a ∆Ψ, would not develop. Therefore, in the
across the membrane can be more than
100 mV
The relationship between the positive charge
due to protons that accumulates on one face of
IN OUT
the membrane and the membrane potential is
H+ H+
ΔΨ (volts) = en/c 1 H+ A ΔΨ develops.
H+
–19
where e is the charge per proton (1.6 × 10 A−
H+A−
coulomb), n is the number of protons, and C is 2 H+ A− A ΔpH develops.
a constant termed the capacitance of the mem- A− H+ H+
brane measured in farads (F). The capacitance is H+ H+ A−
related to the conductivity of material separat- 3 A− R + A
− A ΔpH develops.
ing the two surfaces of the membrane, the area R+ A−
R H+
+
of the membranes, and the distance between

the membranes. Assuming a membrane area e−
4 A ΔΨ develops.
− +
of about 3 × 10–8 cm2 (for a spherical cell the
size of a typical bacterium), then C ≈ 3 × 10–14 F.
Therefore, the translocation of only 40,000 Fig. 4.3 A ∆pH in the bulk phase and a ∆Ψ can-
protons to the cell surface is sufficient to gener- not develop simultaneously by the movement of a
ate a membrane potential of –200 mV.10,11 (By proton. (1) Electrogenic movement of a proton. (2)
Establishment of a ∆pH by the extrusion of both pro-
convention, the ∆Ψ is said to be negative when
tons and counterions. (3) Establishment of a ∆pH by
the inside potential is negative.)
the exchange of protons for cations in the medium.
The membrane potential that actually devel- Theoretically, a ∆pH could develop if the proton on
ops varies in magnitude from approximately the outer surface of the membrane exchanged with a
–60 mV to about –200 mV, depending upon cation from the bulk phase. But even under these cir-
the bacterium and the growth conditions.6 The cumstances, the membrane potential (positive out-
membrane potential that develops when a rela- side) would limit further efflux of protons. (4) A ∆Ψ
tively small number of protons is translocated can also develop via electrogenic flow of an electron.
www.ebook777.com
absence of compensating ion flow, the protons 4.3 The Contributions of the ∆Ψ
that are pumped out of the cell remain on or and the ∆pH to the Overall ∆ p in
very close to the membrane, and a ∆Ψ rather
Neutrophiles, Acidophiles, and
than a ∆pH (measurable with a pH electrode)
is created. (Theoretically, however, a ∆pH and Alkaliphiles
a ∆Ψ could develop if a cation in the exter- Partly for the reasons stated in Section 4.2.2, the
nal medium exchanged for the proton on the contributions of the ∆Ψ and the ∆pH to the ∆p
membrane.) are never equal. Additionally, the relative con-
Of course, some of the protons could be elec- tributions of the ∆pH and the ∆Ψ to the ∆p vary,
troneutrally released into the bulk phase to cre- depending upon the pH of the environment in
ate a ∆pH, and some might remain at the outer which the bacteria naturally grow. Sections
membrane surface to create a ∆Ψ; but even 4.3.1, 4.3.2, and 4.3.3 summarize, respectively,
under these circumstances, a large ∆pH cannot the relative contributions of the ∆Ψ and the
be generated in the face of a large ∆Ψ. This is ∆pH to the ∆p in neutrophiles, acidophiles, and
because the positive charge on the outside sur- alkaliphiles.12 Notice that in acidophiles and
face inhibits further efflux of protons. In fact, alkaliphiles, the ∆Ψ (acidophiles) or the ∆pH
to demonstrate the formation of a ∆pH experi- (alkaliphiles) has the wrong sign and actually
mentally as a result of proton pumping, a large detracts from the ∆p.
∆Ψ must not be allowed to develop. These con-
ditions are achieved experimentally by making 4.3.1 Neutrophilic bacteria
the membrane permeable either to a cation, so For neutrophilic bacteria (i.e., those that grow
that the incoming cations can compensate elec- with a pH optimum near neutrality), the ∆Ψ
trically for the outgoing protons, or to an anion contributes approximately 70 to 80% to the
that moves in the same direction as the proton. ∆p, with the ∆pH contributing only 20 to 30%.
For example, in many experiments the K+ iono- This is reasonable when one considers that the
phore valinomycin is added to make the mem- intracellular pH is near neutrality and therefore
brane permeable to K+ (Section 4.4). When this the ∆pH cannot be very large.
is done, K+ exchanges for H+, and a ∆pH can
develop.
Although bacteria cannot make both a 4.3.2 Acidophilic bacteria
∆Ψ and a ∆pH with the same proton, they For acidophilic bacteria (i.e., those that grow
can create a ∆Ψ during proton translocation between pH 1 and pH 4, and not at neutral pH),
and then convert it to a ∆pH. Suppose a ∆Ψ the ∆Ψ is positive rather than negative (below
is created because a few protons are trans- an external pH of 3, which is where they are usu-
located to the cell membrane outer surface. ally found growing in nature) and thus lowers
Proton translocation cannot proceed for very the ∆p. Under these conditions the force in the
long because a membrane potential devel- ∆p is due entirely to the ∆pH. Let us examine an
ops quickly, which limits further efflux of example of this situation. Thiobacillus ferroox-
protons. However, a cation such as K+ might idans is an aerobic acidophile that can grow at
enter the cell electrogenically. This would pH 2.0 (see Section 13.4.1). Because it has an
result in a lowering of the membrane poten- intracellular pH of 6.5, the ∆pH is 4.5, which
tial because of the positive charge moving in. is very large (remember, ∆pH = pHin – pHout).
Now more protons can be translocated out of When the aerobic acidophiles grow at low pH,
the cell. The protons can leave the outer mem- however, they have a positive ∆Ψ, and for T.
brane surface and accumulate in the external ferrooxidans the ∆Ψ is +10 mV.13 (See note 14.)
phase because they are paired with the anion The contribution of the ∆pH to the ∆p is –60
that was formerly paired with the K+. Thus, a ∆pH or –270 mV. Since ∆p = ∆Ψ – 60 ∆pH, the
membrane potential can form during proton actual ∆p would be –260 mV. In this case, the
translocation and be converted into a ∆pH by ∆Ψ lowered the ∆p by 10 mV. As discussed in
the influx of K+. This is an important way for Section 16.1.3, the inverted membrane poten-
bacteria to maintain a ∆pH, as described later tial is necessary for maintaining the large ∆pH
in Section 16.1.3. in the acidophiles.
4.3.3 Alkaliphilic bacteria hand, will initially dissipate the ∆Ψ because it

An opposite situation holds for the aerobic electrogenically carries K+ into the cell, thus set-
alkaliphilic bacteria (i.e., those that grow above ting up a potential opposite to the membrane
pH 9, often with optima between pH 10 and potential.
pH 12). For these organisms the ∆pH is one It is also possible to create a ∆Ψ by using
to two units negative (because the cytoplasmic valinomycin. The addition of valinomycin
pH is less than 9.6) and consequently lowers to starved cells or vesicles loaded with K+ will
the ∆p, by 60 to 120 mV.15 Therefore, in the induce a temporary ∆Ψ predicted by the Nernst
alkaliphiles, the potential of the ∆p may come equation, ∆Ψ = –60 log[Sin]/[S]out mV. (This is
entirely from the ∆Ψ. In fact, as we shall explain the concentration gradient expressed as mil-
in Section 4.10, because of the large negative livolts at 30 °C.) What happens is that the K+
∆pH, the calculated ∆p in these organisms is so moves from the high internal concentration to
low that it raises conceptual problems regard- the lower external concentration in the pres-
ing whether there is sufficient energy to synthe- ence of valinomycin (Fig. 4.5). Because the K+
size ATP. moves faster than its counterion, a temporary
diffusion potential, outside positive, is created.
The diffusion potential is temporary because
4.4 Ionophores of the movement of the counterions. The use
Before we continue with the discussion of the of ionophores has helped researchers deter-
proton motive force, we must explain iono- mine which ions are carrying the primary cur-
phores and their use. Ionophores are important rent that is doing the work; it has also assisted
research tools for investigating membrane bio- them investigating the relative importance of
energetics, and their use has contributed to an the membrane potential and ion concentration
understanding of the role of electrochemical gradients in providing the energy for specific
ion gradients in membrane energetics. As men- membrane functions.
tioned earlier, membranes are poorly permeable
to ions, and this is why the membrane can main-
4.4.1 The effect of uncouplers on
tain ion gradients. Ionophores perturb these ion
respiration
gradients. Most ionophores are organic com-
pounds that form lipid-soluble complexes with Uncouplers are ionophores that have the fol-
cations (e.g., K+, Na+, H+) and rapidly equilibrate lowing effects:
these across the cell membrane (Fig. 4.4). The 1. They collapse the ∆p and thereby inhibit ATP
incorporation of an ionophore into the mem- synthesis coupled to electron transport.
brane is equivalent to short-circuiting an elec- 2. They stimulate respiration.
trical device with a copper wire. Another way
of saying this is that the ionophore causes the Why should uncouplers stimulate respiration?
electrochemical potential difference of the ion The flow of electrons through electron carriers
to approach zero. Since some ionophores are in the membrane is obligatorily coupled to a
specific for certain ions, it is sometimes possible, flow of protons in a closed circuit (Section 5.5).
by means of the appropriate use of ionophores, This occurs at the coupling sites discussed in
to identify the ion current that is performing Section 4.7.1. Protons are translocated to the
the work. For example, if ATP synthesis is pre- outer surface of the membrane and then reenter
vented by a proton ionophore, this implies that via the ATP synthase. The reentry of the protons
a current of protons carries the energy for ATP through the ATP synthase can be viewed as rate
synthesis. limiting for the circular flow of protons through
One can also preferentially collapse the ∆Ψ the circuit. In the presence of uncouplers such
or ∆pH with judicious use of ionophores, per- as dinitrophenol, protons rapidly enter the cell
haps gaining information about the driving on the uncoupler rather than through the ATP
force for the ion current. For example, nigeri- synthase, thus stimulating electron transport.
cin, which catalyzes an electroneutral exchange Consistent with this model, if proton reentry is
of K+ for H+, will dissipate the pH gradient but blocked by inhibitors of the ATP synthase [e.g.,
not the ∆Ψ.16 Valinomycin plus K+, on the other dicyclohexylcarbodiimide (CDDC)], or slowed
www.ebook777.com
H+
gramicidin
K+
valinomycin
K+
nigericin
Na+
H+
R–
dinitrophenol H+ R–
H+
monensin
Fig. 4.4 Examples of ionophores and the transport processes that they catalyze. All the reactions are revers-
ible. Valinomycin, which transports K+ and Rb+, transports only one cation at a time. Since K+ is positively
charged, valinomycin carries out electrogenic transport (i.e., creates a membrane potential). If valinomycin
is added to cells with high intracellular concentrations of K+, then the K+ will rush out of the cell ahead of
counterions and create a transient membrane potential, outside positive, as predicted by the Nernst equation
(given shortly and further discussed later in connection with eq. 4.13). In the presence of excess extracellular
K+ and valinomycin, the K+ will rapidly diffuse into the cell, collapsing the existing potential. Nigericin car-
ries out an electroneutral exchange of K+ for H+. When nigericin is added to cells, one can expect a collapse
of the pH gradient as the internal K+ exchanges for the external H+, but the membrane potential should not
decrease. The combination of nigericin and valinomycin will collapse both the ∆pH and the membrane poten-
tial. Monensin carries out an electroneutral exchange of Na+ or K+ for H+. There is a slight preference for Na+.
Dinitrophenol is an anion (R−) that carries out electroneutral influx of H+ and R− into the cell and returns to
the outside without H+ (i.e., R−). Therefore this anion will collapse both the ∆pH and the ∆Ψ. This is the classic
uncoupler of oxidative phosphorylation. Gramicidin carries out electrogenic transport of H+ > Rb+, K+, Na+.
Gramicidin differs from the other ionophores in that it forms polypeptide channels in the membrane. That is,
it is not a diffusible carrier. Since the addition of gramicidin results in the equilibration of protons across the
cell membrane, it will collapse the ∆Ψ and the ∆pH (i.e., the ∆p). Carbonyl cyanide-p-trifluoromethylhydra-
zone (FCCP) (not shown) is a lipophilic weak acid that exists as the nonprotonated anion (FCCP−) and as the
protonated form (FCCPH), both of which can travel through the membrane. Protons are carried into the cell
in the form of FCCPH. Inside the cell, the FCCPH ionizes to FCCP−, which exits in response to the membrane
potential, outside positive. The result is a collapse of both the ∆pH and the ∆Ψ.
by depletion of ADP, then respiration is slowed. in the overall physiology of the cell. The two
One possible explanation for the slowing of res- components of the ∆p, the ∆Ψ and the ∆pH, are
piration is that, as the protons are translocated to measured separately.17,18
the outside surface, the ∆p rises and approaches
the ∆G of the oxidation−reduction reactions. 4.5.1 Measurement of ∆Ψ
This might slow respiration, since the oxidation–
The membrane potential is measured indirectly
reduction reactions are reversible. One can view
because bacteria are too small for the inser-
this as the ∆p producing a “back pressure.”
tion of electrodes. Suppose a membrane poten-
tial exists and an ion is allowed to come to its
4.5 Measurement of the ∆p electrochemical equilibrium in response to the
Measurements of the size of the ∆p are a neces- potential difference. Then, at equilibrium, the
sary part of analyzing the role that the ∆p plays electrochemical energy of the ion is zero, and we
Fig. 4.5 Valinomycin-stimulated K+ efflux can impose Fig. 4.6 Measurement of membrane potential by
a temporary membrane potential. When valinomy- the accumulation of a lipophilic cation. The cation
cin is added to cells or membrane vesicles loaded with accumulates in response to the membrane potential
K+, the valinomycin dissolves in the membrane and until the internal concentration reaches equilibrium,
carries the K+ out of the cell. The efflux of K+ creates that is, the point of which efflux equals influx. At this
a diffusion potential, since the negative counterions time the electrochemical energy of the cation is zero
lag behind the K+. The membrane potential is pre- and ∆Ψ = –60 log[(R+)in/(R+)out] mV at 30 °C. Any
dicted by the Nernst equation, ∆Ψ = –60 log(K+)in/ permeant ion can be be used as long as it diffuses pas-
(K+)out. The membrane potential is transient because sively, is used in small amounts, is not metabolized,
it is neutralized by the movement of counterions. and accumulates freely in the bulk phase on either
side of the membrane. Permeant ions that have been
used include lipophilic cations such as tetraphenyl-
can use the equation for electrochemical energy phosphonium (TPP+).
(for a monovalent ion):
∆μS/F = 0 = ∆Ψ + 60 log10([S]in/[S]out) millivolts

the collapse of the membrane potential by the
(4.12) influx of the ion.
Cationic or anionic fluorescent dyes have also
Solving for ΔΨ, we write
been used to measure the ∆Ψ. The dyes partition
∆Ψ = −60 log10([S]in/[S]out) millivolts at 30 °C between the cells and the medium in response to
the membrane potential, and the fluorescence is
(4.13) quenched. The fluorescent dye will monitor rela-
Equation 4.13, a form of the Nernst equa- tive changes in membrane potential (Fig. 4.7).
tion, states that the measurement of the intra- When a fluorescent dye is used to measure the
cellular and extracellular concentrations of absolute membrane potential, it is necessary to
a permeant ion at equilibrium allows one to produce a standard curve. This involves measur-
calculate the membrane potential. The bacte- ing the fluorescence quenching in a sample for
rial cell membrane is relatively nonpermeable which the membrane potential is known—that is,
to ions; therefore, to measure the membrane has been measured by independent means, such
potential, one must use either an ion plus an as the distribution of a permeant ion (Fig. 4.8).
appropriate ionophore (e.g., K+ and valino-
mycin) or a lipophilic ion (i.e., one that can 4.5.2 Measurement of ∆pH
dissolve in the lipid membrane and pass freely A common way to measure ∆pH is to measure
into the cell). When the inside is negative with the distribution of a weak acid or weak base at
respect to the outside, a cation (R+) is cho- equilibrium between the inside and the outside
sen because it accumulates inside the cells or of the cell. The assumption is that the uncharged
vesicles (Fig. 4.6). When the inside is positive, molecule freely diffuses across the membrane but
an anion (R−) is used. It is important to use the ionized molecule cannot. Inside the cell the
a small concentration of the ion to prevent acid becomes deprotonated or the base becomes
www.ebook777.com
Fig. 4.7 The use of fluorescence to measure the mem- Fig. 4.8 Relationship between the membrane poten-
brane potential. If valinomycin were added to bacteria tial and fluorescence change of 1,1′-dihexyl-2,2′-
in buffer containing low concentrations of potassium oxacarbocyanine (CC6). The changes in fluorescence
ion in the presence of a fluorescent cationic lipophilic (ordinate) are plotted as a continuous line corre-
probe, one might expect the potassium inside to rush sponding to different membrane potentials imposed
out, creating a diffusion potential as predicted by by the addition of valinomycin to Streptococcus
the Nernst equation, whereupon the cationic probe lactis cells. The membrane potentials were calcu-
would enter in response to the membrane potential, lated by using the Nernst equation from potassium
and the fluorescence would be quenched. It has been concentration ratios (in/out) in parallel experiments,
suggested that the quenching of fluorescence is due to where the intracellular K+ concentrations were about
the formation of dye aggregates with reduced fluores- 400 mM and the extracellular concentrations were
cence inside the cell. varied. Also shown is the membrane potential caused
by the addition of glucose (rather than valinomycin)
to the cells. Source: Adapted from Maloney, P. C.,
E. R. Kashket, and T. H. Wilson. 1975. Methods for
protonated, the extent of which depends upon studying transport in bacteria, pp. 1–49. In: Methods
the intracellular pH. For example, in Membrane Biology, Vol. 5. E. D. Korn (Ed.).
Plenum Press, New York.
AH → A− + H+ or B + H+ → BH+
On addition of a weak acid (or base) to a cell sus-

pHin − pHout = ∆pH
pension, the charged molecule accumulates in
the cell. One uses a weak acid (e.g., acetic acid) = log10 [A−]in/[A−]ou t (4.14)
when pHin exceeds pHout because the acid will
Thus, the log10 of the ratio of concentrations
ionize more extensively at the higher pH, and a
inside to outside the weak acid is equal to the
weak base (e.g., methylamine) when pHin is less
∆pH. In practice one uses radioactive acids or
that pHout because such a base will become more
bases as probes and measures the amount of
protonated at the lower pH. At equilibrium, for
radioactivity taken up by the cells. A more com-
a weak acid,
plex equation must be used when the concen-
Ka = [H+]in[A−]in /[AH]in tration of the un-ionized acid (AH) cannot be
ignored.19 Cytoplasmic pH is sometimes also
= [H+]out[A−]out/[AH]out measured by 31P nuclear magnetic resonance
(31P NMR) of phosphate whose spectrum is pH
where Ka is the dissociation constant of the
dependent. (See note 20.)
acid, HA.
If pHin and pHout are at least 2 units higher than
the pK, most of the acid is ionized on both sides 4.6 Use of the ∆p to Do Work
of the membrane and there is no need to take The ∆p provides the energy for several mem-
into account the un-ionized acid. Therefore, the brane functions, including solute transport and
∆pH can be calculated by using eq. 4.14. ATP synthesis discussed in this section. (See
Chapter 17 for a more complete discussion of but more details are provided shortly, when the
solute transport.) mechanism of its function is discussed. (See next
subsection, entitled Model of the mechanism of
4.6.1 Use of the ∆p to drive ATP synthase.) Mitochondria, chloroplasts,
solute uptake and bacteria have a similar ATP synthase with
As an example of how the ∆p can be used for homologous proteins. The ATP synthase is a
doing work, consider symport of an uncharged membrane-embedded rotary engine consisting
solute, S, with protons (Fig. 4.2, 4). The total of two regions, the F0 and F1 (Fig. 4.9).23
driving force on S at 30 °C is The F0 region spans the membrane and serves
as a proton channel through which the protons
y∆p + 60 log([S]in/[S]out) millivolts (4.15) are pumped. The F1 region is located on the
where y is the ratio of H+ to S, and 60 log[S]in/ inner surface of the membrane and is the cata-
[S]out, the force involved in moving S from one lytic subunit responsible for the synthesis and
concentration to another, is the same as eq. 4.7 hydrolysis of ATP. The F1 unit is also called the
but written for S, the solute, rather than for the coupling factor. The F1 subunit from E. coli is
proton. made from five different polypeptides with the
At equilibrium, the sum of the forces is zero, following stoichiometry: α3, β3, γ1, δ1, and ε1.
and therefore, Each of the β-polypeptides contains a single
catalytic site for ATP synthesis and hydrolysis,
y∆p = –60 log([S]in/[S]out ) (4.16) although the sites are not equivalent at any one
time. The F0 portion has three different polypep-
For y = 1 and ∆p = –0.180 V, we write tides, which in E. coli are stoichiometrically a1,
3 = log[S]in/[S]out and [S]in/[S]out = 103 b2, and c10. All in all, E. coli uses polypeptides of
8 different types to construct a 22-polypeptide
Therefore, a ∆p of –180 mV could maintain a machine that acts as a reversible pump driven
103 concentration gradient of Sin/Sout if the ratio by the proton potential or by ATP hydrolysis.
of H+/S were 1. If the ratio of H+/S were 2, then As discussed more completely in Section 4.7.2,
a concentration gradient of 106 could be main- the amount of energy to synthesize one mole of
tained. It can be seen that very large concentra-
tion gradients can be maintained by using the ∆p.
4.6.2 The ATP synthase IN OUT

Built into the cell membranes of prokaryotes
and mitochondrial and chloroplast membranes
is a complex protein complex that couples the ADP + Pi
translocation of protons down an electrochemi- 3H+
cal proton gradient (∆p) to the phosphorylation F0
F1
ATP + H2O
of ADP to make ATP. It is called the proton-
translocating ATP synthase, or simply ATP
synthase. As discussed later, in Section 4.7.2,
the ATP synthase is reversible and can pump
protons out of the cell, generating a ∆p. As
discussed in Section 4.8.1, some bacteria such
as Propionigenium modestum have a Na+- Fig. 4.9 ATP synthase. The proton channel F0 spans
the membrane. It consists of polypeptides of types a,
dependent ATP synthase and utilize the flow
b, and c. The catalytic subunit F1, on the inner mem-
of Na+ rather than H+ down an electrochemi- brane surface, catalyzes the reversible hydrolysis of
cal gradient through the ATP synthase to make ATP. It consists of polypeptides of types α, β, γ, δ,
ATP. and ε. Under physiological conditions, the ATP syn-
thase reaction is poised to proceed in either direction.
Description of the ATP synthase When ∆p levels decrease relative to ATP, they can
For an overview of the ATP synthase, see refs. 21 be restored by ATP hydrolysis. When the ATP levels
and 22. The structure is briefly described here, decrease relative to ∆p, more ATP can be made.
www.ebook777.com
ATP is given by ∆Gp. Assume that ∆Gp is equal (T, or tight, conformation), and ATP released
to approximately 50,000 joules, or (dividing by from the enzyme (O, or open, conformation).
the Faraday constant) 518 mV. Thus, y∆p must What drives the conformational changes?
be ≥ –518 mV, where y is the number of enter- A remarkable mechanism has come to light.
ing protons. If y = 3, then to synthesize one mole Evidence suggests that the ATP synthase con-
of ATP, the ∆p must be at least –518/3 or –173 tains a centrally located rotary motor driven by
mV. (By convention the ∆p is negative when proton translocation and that the rotation of
energy is available to do work.) (See note 24.) the motor causes the conformational changes at
the catalytic sites in the β subunits. The motor
Model of the mechanism of ATP synthase consists of the cn subunits comprising a ring in
The mechanism of ATP synthesis/hydrolysis by F0, linked to the γ subunit and the ε subunit in
the ATP synthase is complex. One model, the F1. As mentioned, the γ subunit along with the
binding-change mechanism, is based upon the ε subunit forms a central stalk in F1. The idea is
finding that purified F1 will synthesize tightly that proton flux through F0 causes the cn ring to
bound ATP in the absence of a ∆p.25 (The Kd rotate, and this in turn rotates γε in F1. When
for the high-affinity site is 10–12 M.) The equi- the γε rotor turns as a consequence of the influx
librium constant between enzyme-bound ATP of protons, the γ subunit sequentially makes
and bound hydrolysis products (ADP and Pi) is contact with the three different αβ subunits,
approximately one, with the use of soluble F1.26 which are prevented from rotating with γε,
The tight binding of ATP and the equilibrium and force is exerted sequentially on each of the
constant of approximately one suggest that the three β subunits, causing each one to undergo
energy requirement for net ATP synthesis is for three sequential conformational changes (L, T,
the release of ATP from the enzyme rather than or O). As mentioned, each site binds ADP and
for its synthesis. The model proposes that when Pi while in the L conformation, synthesizes ATP
protons move down the electrochemical gradi- while in the T conformation, and releases ATP
ent through the F0F1 complex, a conformational while in the O conformation. It takes approxi-
change occurs in F1 that results in the release of mately three H+ ions traversing the F0 portion of
newly synthesized ATP from the high-affinity the ATP synthase to make one ATP. The molec-
site. This is illustrated in Fig. 4.10A. ular structure of the ATP synthase, as well as
To understand the proposed mechanism of evidence that it contains a rotating motor, can
how ATP synthesis is energized, it is necessary be found in refs. 27 through 30.
to add some more details about the structure As discussed later in Section 4.7.1, certain
of the ATP synthase. (See Fig. 4.10B.) In the F1 bacteria use a sodium motive force, rather than
region the three α and three β subunits alternate a proton motive force, to provide energy to
to form a hollow cylinder that surrounds the. γ the ATP synthase. The rotating motor model
subunit, which interacts with the three α and applies to the sodium motive force as well as
three β subunits. The. γ subunit forms a central the proton motive force. Recall that bacteria
stalk with the ε subunit and protrudes from the have another membrane-bound rotating motor
bottom of the cylinder, where it attaches to the driven by proton influx. It is the flagellum motor
c subunits, which are in the form of a cylindrical described in Section 1.2.1.
ring in the F0 part embedded in the membrane.
The a and b subunits are connected laterally
(peripherally) to the c ring. The b subunits form
4.7 Exergonic Reactions That
a peripheral stalk connecting the a subunit to a Generate a ∆p
complex of subunits α and β. Section 4.6 described the influx of protons down
As Fig. 4.10A points out, the “binding- a proton potential gradient coupled to the per-
change mechanism” postulates that there are formance of work (e.g., solute transport or ATP
three catalytic sites for synthesizing ATP, one synthesis). These activities dissipate the ∆p. Let
on each of the three β subunits, each of which us now consider driving forces that generate a
is paired with an α subunit. Each catalytic site ∆p, that is, forces that move protons out of the
on a β subunit must cycle through three confor- cell toward a higher potential. The major driv-
mational changes for ADP and Pi to be bound ing reactions in most prokaryotes are the redox
(L, or loose, conformation), ATP synthesized reactions in the cell membranes of respiring
c ring
H+ periplasm
cell
F0 membrane
a
epsilon
H+
gamma cytoplasm
b2
beta ADP + Pi
beta
F1
ATP
alpha alpha
beta
delta
Fig. 4.10 (A) Binding-change mechanism for ATP synthase. Because there are three β subunits encoded by a
single gene, and because there is only one catalytic site per β subunit, the maximum possible number of cata-
lytic sites is three. However, the three catalytic sites are not functionally equivalent and are thought to cycle
through conformational changes driven by the electrochemical proton gradient. As a result of the conforma-
tional changes, newly synthesized ATP is released from the catalytic site. The three conformational states are
O (open, very low affinity for substrates), L (loose binding,), and T (tight binding, the active catalytic site).
ATP is made spontaneously at site T, which converts to site O in the presence of a ∆p and releases the ATP.
The ∆p also drives the conversion of the preexisting site O to site L, which binds ADP and Pi, as well as the
conversion of the preexisting site L to site T. In this model, the only two catalytic sites that bind substrates
are L and T. Source: Adapted from Cross, L. R., D. Cunningham, and J. K. Tamura. 1984. Curr. Top. Cell
Regul. 24:335–344. (B). Rotary model for E. coli ATP synthase. The ten c subunits in the F0 portion form a
cylindrical ring in the lipid bilayer. The ε and γ subunits interact with each other and form a central stalk in
the F1 portion that is connected to the c ring. Surrounding the εγ stalk are alternating α3β3 subunits. Subunits
a and b are at the periphery of the c ring. The a subunit is thought to be confined to the membrane, and the two
b subunits extend as a second stalk peripherally located between F0 and F1. The b stalk connects to a complex
of subunits, δ and α in the F1 portion. The rotor is considered to be the c ring and the εγ stalk, and α3β3δab2
is proposed as the stator against which the rotor moves. It is thought that protons move through the a and c
subunits across the membrane. The flux of H+ (or Na+ in the case of certain bacteria that use a sodium motive
force) through the F0 portion is believed to cause the rotor to move with respect to the stator. This is believed
to induce conformational changes in the catalytic subunits, α3, resulting in ATP synthesis. Source: Adapted
from Nicholls, D. G., and S. J. Ferguson. 2000. Bioenergetics 3. Academic Press, San Diego, CA.
www.ebook777.com
organisms (electron transport), ATP hydroly- or

sis in fermenting organisms, and several other
Eh = E0 + 2.3[RT/nF] log10([ox]/[red])
less frequently used driving reactions, which we
shall consider in Section 4.8. At this time, let us In these equations n is the number of electrons
consider some general thermodyamic features transferred per molecule, R is the gas constant,
common to all the driving reactions that gen- and F is the Faraday constant. Usually standard
erate the proton gradient. The translocation of potentials at pH 7 are quoted and the symbol is
protons out of cells is an energy-requiring pro- E″0
(See note 31 for more discussion of the rela-
cess that can be written as tionship of E0 to pH.)
You can see that it is important to know the
yH+in → yH+out
concentrations of the oxidized and the reduced
∆G = yF∆p joules or y∆p eV forms to be able to find the actual electrode
potential. Molecules with more negative Eh
That is, the amount of energy required to trans-
values are better donors (i.e., electrons sponta-
locate y moles of protons out of the cell is yF∆p
neously flow to molecules with more positive
joules, or for y protons, y∆p electron volts.
Eh values). It is important to understand that
(This is the same energy, but of opposite sign,
when an electron moves over a ∆Eh (which is
that is released when the protons enter the cell.)
Eh,acceptor – Eh,donor) to a higher electrode poten-
Therefore, the ∆G of the driving reaction must
tial, the electron is actually moving toward a
be equal to or greater than yF∆p joules or y∆p
lower potential energy and, as a consequence,
electron volts. If the reaction is near equilibrium,
energy is released. The energy released is pro-
which is the case for the major driving reactions,
portional to the potential difference (∆Eh volts)
the energy available from the driving reaction
over which the electron travels. When n elec-
is approximately equal to the energy available
trons move over a potential difference of ∆Eh
from the proton gradient and ∆Gdriving reaction =
volts, the energy released is equal to n∆Eh elec-
yF∆p. This relationship allows one to calculate
tron volts. (Recall that potential differences
the ∆p if y and ∆Gdriving reaction are known.
are given in volts but work units are in elec-
The most common classes of driving reac-
tron volts.) Since the total charge carried by
tions, which we shall discuss first, are oxidation–
one mole of electrons is F coulombs, the work
reduction reactions that occur during respiration
done per n moles of electrons is ∆G = –nF∆Eh
and photosynthetic electron transport, and ATP
joules, where ∆G is the Gibbs free energy. The
hydrolysis. In all cases we will write the ∆G for
negative sign in the equation shows that energy
the driving reaction. Then we will equate the
is released (work is done) when the ∆Eh is posi-
∆Gdriving reaction with yF∆p. Finally, we will solve
tive. As described next, the movement of elec-
for ∆p.
trons during an oxidation–reduction reaction
toward a lower potential energy level (higher
4.7.1 Oxidation–reduction reactions electrode potential) can be coupled to the
as driving reactions extrusion of protons toward a higher potential
energy level (higher ∆p), that is, to the genera-
Electrode potentials and energy change during tion of a ∆p.
oxidation–reduction reactions
The tendency of a molecule (A) to accept an Oxidation–reduction reactions can
electron from another molecule is given by its
generate a ∆p
electrode potential E, also called the reduction
An important method of generating a ∆p is
potential, the redox potential, or the oxidation–
to couple proton translocation to oxidation–
reduction potential. Electrons spontaneously
reduction reactions that occur during electron
flow toward molecules with a higher electrode
transport. The details of electron transport are
potential. Under standard conditions (1 M for
discussed in Chapter 5; here we will simply com-
all solutes and gases at 1 atm), the electrode
pute the ∆p that can be created. Energy released
potential has the symbol E0 and is related to the
from oxidation–reductions can generate a ∆p
actual electrode potential Eh as follows:
because the oxidation–reduction reactions are
Eh = E0 + [RT/nF] ln([ox]/[red]) coupled to proton translocation. We postpone
a description of the mechanism of coupling until approximately +0.2 V (200 mV). Substituting
Section 5.5, concentrating here on the energetic this value for ∆Eh, and by using n = 2 and y = 4,
relationship between the ∆Eh and the ∆p. we obtain a ∆p of –0.1 V (–100 mV). That is,
Let us do a simple calculation to illustrate how when two electrons travel down a ∆Eh of 200
to estimate the size of the ∆p that can be gener- mV and four protons are translocated, –100
ated by an oxidation–reduction reaction that is mV is stored in the ∆p. For another way of
coupled to proton translocation. Consider what looking at this, assume that when two electrons
happens in mitochondria and in many bacteria. travel down a redox gradient of 0.2 V, there is
A common oxidation–reduction reaction in the 0.4 eV available for doing work (2 × 0.2 V). The
respiratory chain is the oxidation of reduced 0.4 eV is used to raise each of four protons to a
ubiquinone, UQH2, by cytochrome c1 (cyt c1) in ∆p of 0.1 V. Oxidation–reduction reactions in
an enzyme complex called the bc1 complex. The the respiratory chain that are coupled to proton
oxidation–reduction is coupled to the translo- translocation are called coupling sites, of which
cation of protons across the membrane, hence the bc1 complex is an example. Coupling sites
the creation of a ∆p. As stated earlier, the total are discussed in more detail in Section 5.5.
energy change during an oxidation–reduction Coupling of redox reactions during electron
reaction is ∆G = –nF∆Eh joules, where n is the transport to proton translocation is the main
number of moles of electrons. The total energy process by which a ∆p is created in respiring
change during proton translocation is yF∆p bacteria, in phototrophic prokaryotes, in chlo-
joules, where y is the number of moles of trans- roplasts, and in mitochondria.
located protons. [One can convert to electrical
potential (volts) and describe these reactions in Reversed electron transport
terms of forces by dividing by the Faraday con- Some of the redox reactions in the respiratory
stant.] This is summarized as follows: pathway are in equilibrium with the ∆p, and
1. UQH2 + 2 cyt c1(ox) this has important physiological consequences.
→ UQ + 2 cyt c1(red) + 2H+ For example, it supports the expectation that
2. yH+in→ yH+out the ∆p can drive electron transport in reverse.
That is, protons driven into the cell by the ∆p
Reaction 1 gives the Gibbs free energy, in joules, at coupling sites can drive electrons to the more
and in reaction 2, for yF∆p, the units are joules, negative electrode potential (e.g., the reversal of
as well. reactions 2 and 1, shown earlier for the oxida-
These reactions (1 and 2) are coupled; that is, tion of UQH2 by cyt c1 coupled to the extrusion
one cannot proceed without the other. What is of protons). In fact, one test for the functioning
the size of the ∆p that can be generated? of reversed electron transport is its inhibition by
Reactions 1 and 2 are close to equilibrium ionophores that collapse the ∆p (Section 4.4).
and can proceed in either direction. Therefore Reversed electron transport commonly occurs
one can write that the total force available from in bacteria that use inorganic compounds (e.g.,
the redox reaction is equal to the total force of ammonia, nitrite, sulfur) as a source of electrons
the proton potential: to reduce NAD+ for biosynthesis of cell material
(chemolithotrophs), since these electron donors
∆G/F = −n∆Eh = y∆p (4.17)
are at a potential higher than NAD+. (A list of
Equation 4.17 summarizes an important relation- electrode potentials of biological molecules is
ship between the ∆Eh of an oxidation−reduction provided later: see Table 5.1.) The chemolitho-
reaction during respiration and the ∆p that can trophs are discussed in Chapter 13.
be generated at a coupling site. We will return
to this point, and to this equation, in Chapter 5 Respiration coupled to sodium ion efflux
when coupling sites are discussed in more Although respiratory chains coupled to pro-
detail. ton translocation appear to be the rule in most
Now we solve for ∆p. Four protons (y = 4) are bacteria, a respiration-linked Na+ pump (a Na+-
translocated for every two electrons (n = 2) that dependent NADH–quinone reductase) has been
travel from reduced quinone to cytochrome c1. reported in several halophilic marine bacteria
The ∆Eh between quinone and cytochrome c1 is that require high concentrations of Na+ (0.5 M)
www.ebook777.com
for optimal growth.32,33 The situation has been is usually created by Na+/H+ antiport rather
well studied with Vibrio alginolyticus, an alka- than by a primary Na+ pump. For example, the
lotolerant marine bacterium that uses a ∆μNa+ marine sulfate reducer Desulfovibrio salexi-
for solute transport, flagellar rotation, and ATP gens, which uses a ∆μNa+ for sulfate accumula-
synthesis (at alkaline pH). tion, generates the ∆μNa+ by electrogenic Na+/
V. alginolyticus creates the ∆μNa+ in two ways. H+ antiport driven by the ∆μH+, which is cre-
At pH 6.5, a respiration-driven H+ pump generated by a respiration-linked proton pump.37
ates a ∆μH+ , which drives a Na+–H+ antiporter Similarly, nonmarine aerobic alkaliphiles
that creates the ∆μNa+. The antiporter creates the belonging to the genus Bacillus, which rely
sodium ion gradient by coupling the influx of on the ∆μNa+ for most solute transport and
protons (down the proton electrochemical gra- for flagella rotation, create the ∆μNa+ by using
dient) with the efflux of sodium ions. However, a Na+/H+ antiporter driven by a ∆μH+ that is
at pH 8.5, the ∆μNa+
is created directly by a res- created by a respiration-linked proton pump
piration-driven Na+ pump, the Na+-dependent (Section 4.10). Furthermore, in D. salexigens
NADH–quinone reductase. In agreement with as well as the alkaliphilic Bacillus species, the
this conclusion, the generation of the membrane membrane-bound ATP synthase is H+ linked
potential at alkaline pH (pH 8.5) is resistant to rather than Na+ linked.
the proton ionophore m-carbonylcyanide phe-
nylhydrazone (CCCP), which short-circuits the 4.7.2 ATP hydrolysis as a driving reaction
proton current (Section 4.4) but is sensitive to for creating a ∆ p
CCCP at pH 6.5.34 Electron transport reactions are the major
It has been suggested that switching to a energy source for creating a ∆p in respir-
Na+-dependent respiration at alkaline pH ing organisms, but not in fermenting bacte-
may be energetically economical.35 The rea- ria. (However, see note 38.) A major energy
soning is that when the external pH is more source for the creation of the ∆p in fermenting
alkaline than the cytoplasmic pH, the only bacteria is ATP hydrolysis catalyzed by the
part of the ∆p that contributes energy to the membrane-bound, proton-translocating ATP
antiporter is the ∆Ψ, and therefore the anti- synthase, which yields considerable energy
porter must be electrogenic. That is, the H+/ (Fig. 4.2, 7). Consider the following coupled
Na+ must be greater than one. If one assumes reactions:
that the antiporter is electroneutral when the
external pH is acidic, then the continued use 1. ATP + H2O → ADP + Pi ∆Gp
of the antiporter at alkaline pH would neces- 2. yH+in → yH+out yF∆p
sitate increased pumping of protons out of the To refer to the energy of ATP hydrolysis or syn-
cell by the primary proton-linked respiration thesis by means of physiological concentrations
pumps. Rather than do this, the cells simply of ADP, Pi, and ATP, the term ∆Gp (phosphory-
switch to a Na+-dependent respiration pump lation potential) is used instead of ∆G:
to generate the ∆μNa+. This argument assumes
that the ratios Na+/e− and H+/e– are identical so ∆Gp = ∆G0′ + 2.303RT log[ATP]/[ADP][Pi]
that the energy economies of the respiration- (4.18)
linked cation pumps are the same. (See note
36 for additional information regarding pro- The ATP synthase reaction is close to equi-
ton and sodium ion pumping in V. alginolyti- librium and can operate in either direction.
cus.) Several other bacteria have been found to Therefore, the total force available from the
generate sodium ion potentials by a primary proton potential equals the force available from
process. For example, primary Na+ pumping ATP hydrolysis:
is also catalyzed by sodium ion translocating
∆Gp/F = y∆p volts 4.19)
decarboxylases in certain anaerobic bacteria,
as described shortly (Section 4.8.1). It must be The value of ∆Gp is about –50,000 joules (518 mV),
pointed out, however, that although the ∆μNa+ and the consensus value for y is 3. Under these
is relied upon for solute transport and motil- circumstances, the hydrolysis of one ATP would
ity in many other Na+-dependent bacteria, it generate a maximum ∆p of –173 mV.
4.8 Other Mechanisms for Creating a to the electrogenic translocation of two moles
∆Ψ or a ∆ p of sodium ions per mole of methylmalonyl–
CoA decarboxylated46:
Redox reactions and ATP hydrolysis are the
most common driving reactions for creating a Methylmalonyl–CoA + 2Na+in
proton potential. However, other mechanisms → propionyl–CoA + CO2 + 2Na+out
exist for generating proton potentials, and even
P. modestum differs from most known bacte-
sodium ion potentials. These driving reactions
ria in that it has a Na+-dependent ATP synthase
are not as widespread among the prokary-
rather than a H+-dependent ATP synthase, and
otes as the others. Nevertheless, they are very
thus it relies on a Na+ current to make ATP. The
important for certain groups of prokaryotes,
consequence of this is that, when P. modestum
especially anaerobic bacteria and halophilic
grows on succinate, the decarboxylation of
archaea. In some instances, they may be the only
methylmalonyl–CoA is its only source of ATP.
source of ATP. Some of these driving reactions
Another example is V. alcalescens. This is an
are considered next.
anaerobic gram-negative bacterium that grows
in the alimentary canal and mouth of humans
4.8.1 Sodium transport decarboxylases and other animals. Like P. modestum, it is unable
can create a sodium potential to ferment carbohydrates, but it does ferment
Although chemical reactions directly linked to the carboxylic acids lactate, malate, and fumar-
Na+ translocation (primary transport of Na+) ate to propionate, acetate, H2, and CO2. During
are not widespread among the bacteria, since the fermentation, pyruvate and methylmalo-
most primary transport is of the proton, they nyl–CoA are formed as intermediates, as in P.
can be very important for certain bacteria.39–44 modestum.47 Some of the pyruvate is oxidized
An example of primary Na+ transport is the Na+ to acetate, CO2, and H2 with the formation of
pump coupled to respiration in Vibrio alginolyt- one ATP via substrate-level phosphorylation
icus, discussed in Section 4.7.1. There is also (Section 8.3). Pyruvate will then be converted
the decarboxylation of organic acids coupled to methylmalonyl–CoA that is decarboxylated
to sodium ion efflux. This occurs in anaerobic to propionyl–CoA coupled to the extrusion of
bacteria that generate a sodium gradient by cou- sodium ions. The sodium ion gradient that is
pling the decarboxylation of a carboxylic acid created might be used as a source of energy for
to the electrogenic efflux of sodium ions. The solute uptake via a process called Na+-coupled
decarboxylases include methylmalonyl–CoA symport, described in Section 17.3.1.
decarboxylase from Veillonella alcalescens and Acidaminococcus fermentans is an anaer-
Propionigenium modestum, glutaconyl–SCoA obe that ferments the amino acid glutamate to
decarboxylase from Acidaminococcus fermen- acetate and butyrate. During the fermentation,
tans, and oxaloacetate decarboxylase from glutaconyl–CoA is decarboxylated to crotonyl–
Klebsiella pneumoniae and Salmonella typh- CoA by a decarboxylase that is coupled to the
imurium. A description of these bacteria and translocation of sodium ions, as just shown for
their metabolism will explain the importance methylmalonyl–CoA48:
of the sodium-translocating decarboxylases as
Glutaconyl–CoA + yNa+in
a source of energy.
→ crotonyl–CoA + CO2 + yNa+out
Propionigenium modestum is an anaerobe
isolated from marine and freshwater mud, and
from human saliva.45 It grows only on carboxy- The oxaloacetate decarboxylase in
lic acids (i.e., succinate, fumarate, L-aspartate, Klebsiella pneumoniae
L-malate, oxaloacetate, and pyruvate). The car- We will examine the Na+-dependent decar-
boxylic acids are fermented to propionate and boxylation of oxaloacetate by oxaloacetate
acetate, forming methylmalonyl–CoA as an decarboxylase from the facultative anaerobe K.
intermediate. (The description of the propionic pneumoniae because this process has been well
acid fermentation in Section 15.7 explains these studied. A substantial sodium potential develops
reactions.) The methylmalonyl–CoA is decar- because the decarboxylation of oxaloacetate is
boxylated to propionyl–CoA and CO2 coupled coupled to the electrogenic efflux of sodium ion.
www.ebook777.com
(The standard free energy for the decarboxyla- sodium ion was prevented by avidin, an inhibi-
tion reaction is approximately –29,000 J.) The tor of the oxaloacetate decarboxylase.
enzyme from K. pneumoniae translocates two
2. The structure of the oxaloacetate
Na+ ions out of the cell per oxaloacetate decar-
decarboxylase
boxylated according to the following reaction:
Oxaloacetate decarboxylase consists of two
2Na+in + oxaloacetate parts, a peripheral catalytic portion attached
→ 2Na+out + pyruvate + CO2 to the inner surface of the membrane and an
integral membrane portion that serves as a Na+
1 Evidence that oxaloacetate decarboxylase channel (see note 36). The integral membrane
is a sodium pump protein also takes part in the decarboxylation
Inverted membrane (inside-out) vesicles were step, along with the peripheral membrane pro-
prepared from Klebsiella and incubated with tein. How the decarboxylase translocates Na+
22
Na+ and oxaloacetate. (See note 49 for a to the cell surface is not understood.
description of how to prepare the vesicles.)
3. Use of the sodium current
As seen in Fig. 4.11, oxaloacetate-dependent
What does Klebsiella do with the energy con-
sodium ion influx took place. The uptake of
served in the sodium potential? It uses the energy
to actively transport the growth substrate, cit-
rate, into the cell, as well as to drive the reduction
of NAD+ by ubiquinol (i.e., via a Na+-dependent
NADH:ubiquinone oxidoreductase). The latter
is an example of reversed electron transport
driven by the influx of Na+. This all occurs dur-
ing the fermentation of citrate and is depicted
in Fig. 4.12.
4.8.2 Oxalate:formate exchange

can create a ∆ p
Oxalobacter formigenes, an anaerobic bacte-
rium that is part of the normal flora in mamma-
lian intestines, uses dietary oxalic acid as its sole
source of energy for growth. The organism has
evolved a method for creating a proton potential
Fig. 4.11 Sodium ion influx into vesicles driven by at the expense of the free energy released from
oxaloacetate decarboxylation. Inside-out vesicles the decarboxylation of oxalic acid to formic
were prepared from Klebsiella pneumoniae and acid and carbon dioxide.50–52 What is especially
incubated with 22Na+ in the presence (curve A) and interesting is that the enzyme is not in the mem-
absence (curve B) of oxaloacetate. Because of their brane and therefore cannot act as an ion pump,
large diameter, the vesicles could be isolated after one yet a ∆Ψ is created. The reaction catalyzed by
minute of incubation by using size exclusion chroma- oxalate decarboxylase is
tography. The fractions from the Sephadex columns
that contained vesicles were detected by absorbance –OOC–COO− + H+ → CO2 + HC, OO−
at 280 nm (protein). Only when oxaloacetate was
oxalate formate
present did the vesicle fraction contain 22Na+ (curve
A). This demonstrates the dependency of sodium For every mole of oxalate that enters the cell,
uptake upon oxaloacetate. In separate experiments one mole of formate leaves (Fig. 4.13). Since a
it was demonstrated that oxaloacetate decarboxy-
dicarboxylic acid crosses the cell membrane in
lase was present in the vesicles, that the oxaloac-
exchange for a monocarboxylic acid, there is
etate was decarboxylated, and that inhibition of
the oxaloacetate decarboxylase by avidin prevented net movement of negative charge toward the
uptake of 22Na+. Source: Adapted from Dimroth, inside, thus creating a ∆Ψ, inside negative. Also,
P. 1980. A new sodium-transport system energized a proton is consumed in the cytoplasm dur-
by the decarboxylation of oxaloacetate. FEBS Lett. ing the decarboxylation, which can contribute
122:234–236. to a ∆pH, inside alkaline. The antiporter was
Na+
Na+/H+
citrate
B
A UQ
Na+ FDH
Na+
CO2 UQH2
citrate OAA pyruvate formate
Na+
CO2
CL PFL C
NAD+
CoASH +
acetate Na
acetyl~SCoA H+ + NADH
AK Pi
acetate acetyl~ P PTA
ADP
ATP
Fig. 4.12 Citrate fermentation and proposed Na+ ion currents in Klebsiella pneumoniae. (A) Na+–citrate
symporter. Citrate enters the cell via symport with Na+. Uptake of citrate is not dependent upon the ∆Ψ
and is therefore electroneutral and dependent upon the ∆pNa+. (It has been suggested that only two of the
carboxyl groups of citrate are neutralized with sodium ions, and the third is protonated.) The citrate is then
cleaved by citrate lyase (CL) to oxaloacetate and acetate. (B) Na+-pumping oxaloacetate decarboxylase. Na+
is pumped out of the cell during the decarboxylation of oxaloacetate to pyruvate and CO2, creating a sodium
ion electrochemical gradient that is used for citrate uptake (Na+–citrate symporter) and NADH production
(Na+-pumping NADH:ubiquinone oxidoreductase). The pyruvate is used to make ATP via substrate-level
phosphorylation. First the pyruvate is cleaved by pyruvate–formate lyase (PFL) to acetyl–CoA and formate.
Then the acetyl–CoA is converted to acetyl–phosphate by phosphotransacetylase (PTA). Finally, the acetyl–
phosphate donates the phosphoryl group to ADP to form ATP and acetate in a reaction catalyzed by acetate
kinase (AK). NADH is produced as described next. Some of the formate is oxidized to CO2 by a membrane-
bound formate dehydrogenase (FDH). The electron acceptor is ubiquinone. Ubiquinol is reoxidized by NAD+
to form NADH + H+ in a reaction catalyzed by a Na+-pumping NADH:ubiquinone oxidoreductase (C). This
is a case of reversed electron transport driven by the electrochemical Na+ gradient. There is therefore a Na+
circuit. Sodium ion is pumped out of the cell via the decarboxylase and enters via the citrate transporter and
the NADH:ubiquinone oxidoreductase. Source: Adapted from Dimroth, P. 1997. Primary sodium ion trans-
locating enzymes. Biochim. Biophys. Acta 1318:11–51.
purified in 1992 and shown to be a 38 kDa hydro- The decarboxylation of other acids
phobic polypeptide that catalyzes the exchange may also create a ∆ p
of oxalate and formate in reconstituted proteo- In principle, the influx and decarboxylation of
liposomes.53 (For a discussion of proteolipo- any dicarboxylic acid coupled to the efflux of the
somes, see note 54 and Section 17.1.) monocarboxylic acid can create a proton poten-
tial (Fig. 4.14). For example, a decarboxylation
ATP synthesis and electrogenic antiport was reported for the
Assuming that the stoichiometry of the ATP decarboxylation of malate to lactate acid dur-
synthase reaction is 3H+/ATP and for oxalate ing malolactate fermentation in Lactobacillus
decarboxylation it is 1H+/oxalate, then a steady lactis.55
proton current requires the decarboxylation of
three moles of oxalate per mole of ATP synthe- 4.8.3 End-product efflux as the
sized. In other words, one-third mole of ATP driving reaction
can be synthesized per mole of oxalate decar- Theoretically, it is possible to couple the excre-
boxylated. This is apparently the only means of tion of fermentation end products down a
ATP synthesis in Oxalobacter formigenes. concentration gradient to the translocation of
www.ebook777.com
Fig. 4.13 The electrogenic oxalate:formate exchange and the synthesis by means of a proton current of ATP
in Oxalobacter formigenes. Oxalate2– enters via an oxalate2–/formate– antiporter. The oxalate2– is decarboxy-
lated to formate– and CO2, while a cytoplasmic proton is consumed. (The oxalate is first converted to oxalyl–
coenzyme A, which is decarboxylated to formyl–coenzyme A. The formyl–CoA transfers the coenzyme A to
incoming oxalate, thus forming formate and oxalyl–CoA.) An electrogenic exchange of oxalate2– for formate−
creates a membrane potential, outside positive. It is suggested that the stoichiometry of the ATP synthase is
3H+/ATP. Since the decarboxylation of one mole of oxalate results in the consumption of one mole of protons,
the incoming current of protons during the synthesis of one mole of ATP is balanced by the decarboxylation
of three moles of oxalate. In other words, a steady state current of protons requires that of a mole of ATP be
made per mole of oxalate decarboxylated. Source: Anantharam, V., M. J. Allison, and P. C. Maloney. 1989.
Oxalate:formate exchange. J. Biol. Chem. 264:7244–725. Also, see refs. 45 and 46.
protons out of the cell, thereby creating a ∆p. gradient; or the efflux of a solute down a con-
This scheme, the reverse of solute uptake by centration gradient can create a ∆p. The driving
proton-coupled transport systems, is called force for solute transport in symport with pro-
the energy recycling model.56,57 In other words, tons is the sum of the proton potential and the
the direction of solute transport depends upon electrochemical potential of the solute, that is,
which is greater: the energy from the proton
potential, ∆p, which drives solute uptake, cre- Driving force (mV) = y∆p + ∆μs/F (4.22)
ating an electrochemical solute gradient ∆μ/F, where y is the number of protons in symport
or the electrochemical solute gradient, which with the solute, S. The term ∆μs/F represents the
drives solute efflux, creating a ∆p. Solute efflux electrochemical potential of S:
coupled to proton translocation can spare ATP
because in fermenting bacteria, the hydrolysis ∆μs/F = m∆Ψ + 60 log[S]in/[S]out millivolts
of ATP (catalyzed by the ATP synthase) is used (4.23)
to pump protons to the outside to create the ∆p. where m is the charge of the solute. Therefore,
As the ∆p rises, ATP hydrolysis is diminished. the overall driving force (substituting for ∆μs/F
and ∆p) is
Energetics
Consider solute/proton symport to be revers- Driving force (mV)
ible. Under these circumstances, the ∆p can drive = 60 log[S]in/[S]out + (y + m)∆Ψ − y60 ∆pH
the uptake of a solute against a concentration (4.24)
OUT IN
_ _ _
O2C R CO 2 O2C R CO 2
H+ A
_ _ CO2
HR CO 2 HR CO2
_ _
HOOC R CO 2 HOOC R CO2
H+ B
CO2
HOOC RH HOOC RH
ATP + Pi
3H+ C
ATP + H2O
Fig. 4.14 Scheme for the coupling of the decarboxylation of a dicarboxylic acid to the development of a pro-
ton potential: dicarboxylic acid enters as a divalent anion (A) or a monovalent anion (B). In the cytoplasm,
a decarboxylase cleaves off CO2, consuming a proton in the process and producing a monocarboxylic acid
with one negative charge fewer than the original dicarboxylic acid. Exchange of the dicarboxylic acid and the
monocarboxylic acid via the antiporter is electrogenic and produces a membrane potential, inside negative.
The consumption of the proton during the decarboxylation creates a ∆pH. (C) Influx of protons via the ATP
synthase completes the proton circuit and results in ATP synthesis. The energetics can be understood in terms
of the decarboxylase removing the dicarboxylic acid from the inside, thus maintaining a concentration gradi-
ent that stimulates influx.
Upon substituting m = –1 for lactate, eq. 4.24 Symport of protons and sodium ions with
becomes fermentation end products
Driving force Coupled translocation of protons and lactate
= (y − 1)∆Ψ − y60∆pH + 60 log[S]in/[S]out can be demonstrated in membrane vesicles pre-
pared from lactic acid bacteria belonging to the
(4.25)
genus Streptococcus.58–61 Similarly, coupled
During growth, lactate transport is near equil- translocation of sodium ions and succinate is
brium and the net driving force is close to zero. catalyzed by vesicles prepared from a rumen bac-
Thus by setting eq. 4.25 equal to zero, one can terium belonging to the genus Selenomonas.62 In
solve for y: both cases a transporter exists in the membrane
that simultaneously translocates the organic
∆Ψ − 60 log([S]in/[S]out) acid (R−) with either protons or sodium ions out
y= (4.26)
∆p of the cell (symport) (Fig. 4.15). (See note 63.)
Lactate efflux will be described next and is sum-
Note that eq. 4.25 states that when y, which marized in Fig. 4.16. Succinate efflux is similar
is the number of moles of H+ translocated per and is summarized in Fig. 4.17.
mole of lactate, is one, then the translocation is
electroneutral and a ∆Ψ does not develop; only
a small ∆pH develops, owing to the acidifica- Lactate efflux
tion of the external medium by the lactic acid. L-Lactate-loaded membrane vesicles were pre-
We will return to these points later when we pared from Streptococcus cremoris in the fol-
discuss the physiological significance of energy lowing way. A concentrated suspension of cells
conservation via end-product efflux. Let us now was treated with lysozyme in buffer to degrade
consider some data that support the hypothesis the cell walls. The resulting cell suspension was
that lactate/H+ symport and succinate/Na+ sym- gently lysed (broken) by adding potassium sul-
port can create a membrane potential in mem- fate. The cell membranes spontaneously resealed
brane vesicles derived from cells. into empty vesicles. The vesicles were purified
www.ebook777.com
R R
R
R nNa+
R R
R
R R
R
nH+
R
Fig. 4.15 End-product efflux in symport with protons Fig. 4.16 Lactate efflux can produce a membrane
or sodium ions. The high intracellular concentrations potential. Lactate-loaded membrane vesicles from
of R− may drive the efflux of Na+ or H+ via symport- Streptococcus cremoris were incubated with the
ers. If the ratio of protons or sodium ions to carboxyl lipophilic cation Ph4P+ without lactate (curve 1)
(i.e., n/carboxyl) exceeds 1, then the symport is elec- and with 50 mM lactate in the external medium
trogenic and a membrane potential develops. (curve 2). In the absence of external lactate, the lipo-
philic probe accumulated inside the vesicles, suggest-
ing that a membrane potential developed as lactate
and then incubated for 1 h with 50 mM L-lac- left the cells. The addition of lactate to the external
tate, which equilibrated across the membrane, medium prevented the formation of the membrane
potential because the lactate concentration (in/out)
thus loading the vesicles with L-lactate. Then
was lowered. Additional experiments showed that
the membrane vesicles loaded with L-lactate the electrogenic ion was the proton. Source: Adapted
were incubated with the lipophilic cation tetra- from Otto, R., R. G. Lageveen, H. Veldkamp, and
phenylphosphonium (Ph4P+) in solutions con- W. N. Konings. 1982. Lactate efflux-induced electri-
taining buffer (Fig. 4.16, curve 1) and buffer + cal potential in membrane vesicles of Streptococcus
50 mM L-lactate (Fig. 4.16, curve 2). Samples cremoris. J. Bacteriol. 149:733–738.
were filtered to separate the vesicles from the
external medium, and the amount of Ph4P+ that
accumulated was measured. The Ph4P+ accumu- developed during succinate efflux (Fig. 4.17,
lated inside the cells when the ratio of lactatein to curve 1).
lactateout was high. When the external buffer contained a high
The accumulation of Ph4P+ implies that the concentration of succinate, the membrane
efflux of lactate along its concentration gradient potential did not develop, indicating that the
imposed a membrane potential on the vesicles. energy to establish the potential was derived
The membrane potential was calculated by from the efflux of succinate along its concen-
using the Nernst equation and the Ph4P+ accu- tration gradient, [succ]in/[succ]out (Fig. 4.17,
mulation ratio as explained in Section 4.5.1. curve 2). The molar growth yields of S. rumi-
nantium are higher when succinate production
Succinate efflux is at a maximum, implying that more ATP is
An experiment similar to the one just described available for biosynthesis as a result of the suc-
for lactate efflux was done with vesicles from the cinate efflux.
rumen bacterium Selenomonas ruminantium
loaded with succinate (Fig. 4.17). The Ph4P+ Physiological significance of end-product
was taken up by the vesicles when there was a efflux as a source of cellular energy
concentration gradient of succinate, indicat- To demonstrate the generation of a membrane
ing that a membrane potential, inside negative, potential due to end-product efflux, cells or
Brink et al., who used experimentally measured

values of lactate concentrations, membrane
potential, and ∆pH.64 Only when the external
lactate concentrations were very low was lac-
tate/proton efflux electrogenic (i.e., y > 1). For
example, when the cells were grown at pH 6.34
and the lactate accumulated in the medium over
time, the value of y decreased from about 1.44
to 0.9 while the external lactate concentrations
increased from 8 mM to 38 mM. This means
that only under certain growth conditions (low
external lactate and external pH high so that the
∆pH is 0 or inverted) would one expect lactate
efflux effectively to generate a ∆p. This may
occur in the natural habitat, where growth of
the producer may be stimulated by a population
of bacteria that utilize lactate, thus keeping the
Fig. 4.17 Succinate efflux can produce a membrane external concentrations low.
potential. Succinate-loaded membrane vesicles from
Selenomonas ruminantium were incubated with the
lipophilic cation Ph4P+ without succinate (curve 1) or 4.8.4 Light absorbed by
with succinate (curve 2) in the external medium. The bacteriorhodopsin can drive the
uptake of Ph4P+ indicates that a membrane potential creation of a ∆ p
developed when the ratio of succinatein to succina-
Certain archaea, namely, the extremely halo-
teout was high. Source: Adapted from Michel, T. A.,
and J. M. Macy. 1990. Generation of a membrane
philic archaea, have evolved a way to produce
potential by sodium-dependent succinate efflux a ∆p by using light energy directly (i.e., without
in Selenomonas ruminantium. J. Bacteriol. 172: the intervention of oxidation–reduction reac-
1430–1435. tions and without chlorophyll).65–67 See note 68
for a more complete discussion of the extreme
halophiles. (It has been reported that under the
membrane vesicles must be “loaded” with high proper conditions, these microorganisms can be
concentrations of the end product by incubat- grown photoheterotrophically, i.e., on organic
ing de-energized cells or membrane vesicles for carbon with light as a source of energy, but such
several hours before dilution into end-product- conclusions have been questioned.69,70)
free media. These are not physiological condi- Halophilic archaea are heterotrophic
tions, and the physiological relevance of the organisms that carry out an ordinary aerobic
conditions under which the electrogenic extru- respiration, creating a ∆p driven by oxidation–
sion of lactate coupled to protons has been reduction reactions during electron transport.
questioned. (See note 71.) The ∆p is used to drive ATP
Assuming a given value of y, one can use synthesis via a membrane ATP synthase. (See
eq. 4.26 to calculate the necessary lactate Sections 4.7.1 and 4.6.2.) However, conditions
concentration gradient from experimentally for respiration are not always optimal and, in
determined values of ∆Ψ and ∆pH. When one the presence of light and low oxygen levels, the
substitutes y = 2 (for electrogenic lactate extru- halophiles adapt by making photopigments
sion), a major problem appears. To substitute a (rhodopsins), one of which (bacteriorhodopsin)
measured external lactate concentration of 30 functions as a proton pump that is energized
mM, a ∆Ψ of –100 mV, a –60 ∆pH of –30 mV directly by light energy. (See note 72.) Whereas
(at pH 7.0), and y = 2, the calculated internal photosynthetic electron flow is an example of
lactate concentration would have to be 14 M an indirect transformation of light energy into
for lactate secretion to occur.64 This is a non- an electrochemical potential (via redox reac-
physiological concentration. (Internal lactate tions), bacteriorhodopsin illustrates the direct
concentrations reach concentrations of about transformation of light energy into an electro-
0.2 M.) In fact, the values of y were calculated by chemical potential. We will first consider data
www.ebook777.com
that demonstrate the light-dependent pumping

of protons. We will then describe the proton
pump, the photocycle, and a model for the mech-
anism of pumping protons. Bacteriorhodopsin
is examined in detail because it is the best char-
acterized ion pump.
Evidence that halophiles can use light energy

to drive a proton pump
Figure 4.18 models the results of subjecting a
suspension of Halobacterium halobium to light
of different intensities and for different periods
of time, during which the pH of the external
medium was monitored. As illustrated in Fig.
4.18A, the light produced an efflux of protons
from the cell. Higher light intensities produced
a greater rate of proton efflux (compare Ic to
Ia.) In Fig. 4.18B, the rate of proton efflux, in
nanograms per second, as determined from
the slopes in Fig. 4.18A, is plotted as a func-
tion of the light intensity in nanoeinsteins per Fig. 4.18 Proton pumping by bacteriorhodopsin.
second (an einstein is equal to a “mole” of pho- Expected results if Halobacterium cells were illu-
tons). From data such as these, a quantum yield minated by light. To measure proton outflow accu-
(i.e., protons ejected per photon absorbed) rately, proton inflow through the ATP synthase must
be blocked either with uncouplers or nigericin, which
was calculated to be 0.52 proton per photon
collapse the proton electrochemical potential, or
absorbed.73 The reported values for the maxi-
with an ATP synthase inhibitor such as DCCD. The
mum quantum yield for proton efflux is a little extruded protons can be quantitated with a pH meter.
higher (0.6–0.7 proton/photon absorbed). The (A) Proton efflux measured as a function of time at
quantum yield for the photocycle (i.e., the frac- increasing light intensities Ia < Ib < Ic, expressed as
tion of bacteriorhodopsin molecules absorbing nanoeinsteins per second. The slope of each line is the
light that undergoes the photocycle described rate of proton efflux. (B) The rate of proton efflux is
in the next subsection) is 0.64 ± 0.04.74,75 These plotted as a function of the light intensity. From these
values suggest that one proton is pumped per data one can calculate the quantum yield (i.e., the
photocycle. number of protons extruded per photon absorbed).
Source: Adapted from data by Bogomolni, R. A.,
R. A. Baker, R. H. Lozier, and W. Stoeckenius. 1980.
Bacteriorhodopsin is the proton pump Action spectrum and quantum efficiency for proton
Built into the cell membrane of the halophilic pumping in Halobacterium halobium. Biochemistry
archaebacteria is a pigment protein called bac- 19:2152–2159.
teriorhodopsin, which is a pump responsible
for the light-driven electrogenic efflux of pro-
tons. It consists of one large polypeptide (248 The photocycle and a model for
amino acids, 26,486 Da) folded into seven α proton pumping
helices that form a transmembrane channel The photoevents occurring in halophiles can
(Fig. 4.19). (See Ref. 76 for a review.) Located be followed spectroscopically because when
in the middle of the channel, and attached to bacteriorhodopsin (bR) absorbs light, it loses
the bacteriorhodopsin, is a pigment called reti- its absorption peak at 568 nm (bleaches) and is
nal (a C20 carotenoid), which is attached via a converted in the dark to a series of pigments that
Schiff base to a lysine residue on the protein. have absorption peaks at different wavelengths.
(Note 77 tells what a Schiff base is.) When the This sequence of events is called the photocycle.
retinal absorbs light, the bacteriorhodopsin Also, site-specific mutagenesis of bacteriorho-
remarkably translocates protons out of the cell, dopsin is being used to identify the amino acid
and a ∆p is created. side chains that transfer protons across the
N. Aspartate-96 acquires a proton from the

cytoplasm in going from N to O. Thus, a proton
has moved from the cytoplasmic side through
aspartate-96 to the Schiff base to aspartate-85.
From aspartate-85, the proton moves to the
outside membrane surface. (See note 82.)
Precisely how the proton travels through the
bacteriorhodopsin channel from the cytoplas-
mic side to the Schiff base in the center of the
channel, and from there to the external sur-
face of the membrane, is not known. Probably
the proton is passed from one amino acid
side group that can be reversibly protonated
to another. Site-specific mutagenesis experi-
ments have suggested that two of these amino
acids are aspartate-85 and aspartate-96. The
protonatable residues would extend along the
protein from the cytoplasmic surface to the out-
side. In the case of bacteriorhodopsin, where
the three-dimensional structure is known, the
protonatable residues are oriented toward the
center of the channel. Any bound water might
also participate in proton translocation. For
example, there might be a chain of water and
Fig. 4.19 Diagram of bacteriorhodopsin. The seven
hydrogen-bonded protons connecting protona-
helices, shown as rods, form a central channel; the
retinal is attached to a lysine residue on helix G.
table groups. Since the Schiff base gives up its
Source: Henderson, R., J. M. Baldwin, T. A. Ceska, proton to the extracellular side of the channel
F. Zemlin, E. Beckmann, and K. H. Downing. 1990. (transition L to M) and becomes protonated
Model for the structure of bacteriorhodopsin based with a proton from the cytoplasm (transition
on high-resolution electron cryomicroscopy. J. Mol. M to N), it has been assumed that there is a
Biol. 213:899–929. switch that reorients the Schiff base, causing it
to face the cytoplasmic and extracellular sides
of the channel alternately. Logically, the switch
membrane.78–81 The photocycle is shown in Fig. would be at M. The mechanism for the switch
4.20. The retinal is attached via a Schiff base to is unknown, although a conformational change
the ε amino group of lysine-216 (K216). Before in the bacteriorhodopsin has been suggested.80
absorbing light, the retinal is protonated at the It is not obviously correlated with the retinal
Schiff base and exists in the all-trans (13-trans) isomerizations because the retinal is in the cis
configuration (bR568). The subscript refers configuration throughout most of the photocy-
to the absorption maximum, in nanometers. cle, including the protonation and deprotona-
Upon absorbing a photon of light, the retinal tion of the Schiff base. The transfer of protons
isomerizes to the 13-cis form (K625). All subse- along protonatable groups has been suggested
quent steps do not require light and represent for several proton pumps besides bacteriorho-
the de-energization of the bacteriorhodopsin dopsin, including cytochrome oxidase and the
via a series of intermediates that have absor- proton-translocating ATP synthase.83
bance maxima different from those of the origi-
nal unexcited molecule. The transitions are
very fast and occur in the nanosecond and mil- 4.9 Halorhodopsin, a Light-Driven
lisecond ranges. These intermediates are, in the Chloride Pump
order of their appearance, K, L, M, N, and O. The halophiles have a second light-driven
In converting from L to M, the Schiff base electrogenic ion pump, but one that does not
loses its proton to aspartate-85 but regains a energize the membrane. The second pump,
proton from aspartate-96 in going from M to called halorhodopsin, is structurally similar
www.ebook777.com
Fig. 4.20 The photochemical cycle of bacteriorhodopsin. Upon absorption of a photon of light, bR568 under-
goes a trans-to-cis isomerization and is converted to a series of intermediates with different absorbance max-
ima. The Schiff base becomes deprotonated during the L-to-M transition and reprotonated during the M-to-N
transition. A recent photocycle postulates two M states: L to M1 to M2 to N. Although not indicated, some of
the steps are reversible. (See Lanyi, J. K. 1992. Proton transfer and energy coupling in the bacteriorhodpsin
photo-cycle. J. Bioenerg. Biomemb. 24:169–179.) Source: Krebs, M. P., and H. Gobind Khorana. 1993.
Mechanism of light-dependent proton translocation by bacteriorhodopsin. J. Bacteriol. 175:1555–1560.
Reproduced with permission from American Society for Microbiology.
to bacteriorhodopsin and is used to accumu- Bacillus pasteurii, B. firmus, and B. alcalophi-

late Cl– intracellularly, to maintain osmotic lus. The ∆p has been measured in obligate aero-
stability.59,73,84 Recall that the halophiles live bic alkaliphiles that grow optimally at pH 10
in salt water, where the extracellular concen- to 12, and it appears to be too low to drive the
trations of NaCl can be 3 to 5 M. The osmotic synthesis of ATP. The problem is that because
balance is preserved by intracellular concen- the external pH is so basic, the internal pH is
trations of KCl that match the extracellular generally at least 2 pH units more acid. This
Cl− concentrations. (As discussed in Chapter 16, gives the ∆pH a sign opposite to that found in
K+ is important for osmotic homeostasis in the other bacteria, that is, negative. A ∆pH of 2
eubacteria as well.) Since the membrane is nega- is equivalent to 60 × 2 or 120 mV. Thus, the ∆p
tively charged on the inside, energy must be used can be lowered by about 120 mV in these organ-
to bring the Cl− into the cell, and halorhodop- isms. Typical membrane potentials for aerobic
sin accomplishes this purpose. In the dark the alkaliphiles are approximately –70 mV (posi-
halophiles use another energy source, probably tive out). Therefore, the ∆p values can be as low
ATP, to accumulate chloride ions actively.85 as (–170 + 120) or –50 mV.87
The ATPase in the alkaliphiles is a proton-
translocating enzyme, as in most bacteria. Even
4.10 The ∆p and ATP Synthesis if the ATPase translocated as many as four pro-
in Alkaliphiles tons, this would generate only 0.2 eV, far short
Alkaliphilic bacteria grow in habitats having of the approximately 0.4 to 0.5 eV required to
a very basic pH, usually around pH 10. These synthesize an ATP under physiological condi-
habitats include soda lakes, dilute alkaline tions. The energy to synthesize an ATP (i.e.,
springs, and desert soils, where the alkalinity ∆Gp) is 40,000 to 50,000 J. Dividing this num-
is usually due to sodium carbonate. Obligate ber by the Faraday constant gives 0.4 to 0.5 eV.
alkaliphiles are organisms that cannot grow How can this dilemma be resolved? It is pos-
at pH values of 8.5 or less and usually have an sible that the protons in the bulk extracellu-
optimum around 9. (See note 86.) These include lar phase are not as important for the ∆p as
protons on the membrane or a few angstrom purpose in most bacteria. Thus, although the
units away from the membrane. One sug- major ion circuit is a proton circuit, the sodium
gestion is that random collisons within the circuit is also important.
membrane may put proton pumps in frequent A ∆p is created when an exergonic chemical
contact with the ATPases, and as soon as a reaction is coupled to the electrogenic flow of
proton is pumped out of the cell, it may reen- charge across the cell membrane and the libera-
ter via an adjoining ATPase without entering tion of protons on the outer membrane surface.
the pool of bulk protons.88 This suggestion Energy input of at least yF∆p joules is neces-
emphasizes the activity of protons at the face sary to raise the electrochemical potential of y
of the membrane rather than the ∆pH, which is moles of protons to ∆p volts. The three most
due to the concentration of protons in the bulk widespread reactions that provide the energy to
phase, and raises questions about the details of create the ∆p are oxidation–reduction reactions
the proton circuit. during electron transport in membranes (respi-
One important tenet of the chemiosmotic ration), oxidation–reduction reactions during
theory may not apply in this situation, and that electron transport stimulated by light absorp-
is that a delocalized ∆p is used. That is, the che- tion (photosynthesis), and ATP hydrolysis via
miosmotic theory postulates that proton cur- the membrane ATP synthase. Respiration and
rents couple any exergonic reaction with any ATP hydrolysis are reversible, and the ∆p can
endergonic reaction (i.e., proton circuits are drive reversed electron transport as well as the
delocalized over the entire membrane). In that synthesis of ATP. During reversed electron
way, proton extrusion during respiration can transport, protons enter the cells rather than
provide the energy for several different reac- leave the cells. The ATP synthase is an enzyme
tions, not simply the ATP synthase (Fig. 4.2). complex that reversibly hydrolyzes ATP and
However, it may be that in some alkaliphiles pumps protons out of the cell. When protons
there is direct transfer of protons extruded dur- enter via the ATP synthase, ATP is made.
ing respiration to the ATP synthase (i.e., local- Light energy can also be used directly to
ized proton circuits). create a ∆p without the establishment of a
redox potential. This occurs in the halophilic
archaea, which use a light-driven proton
4.11 Summary pump called bacteriorhodopsin to create a ∆p.
The energetics of bacterial cell membranes can Bacteriorhodopsin, which forms a proton chan-
be understood for most bacteria in terms of an nel through the membrane, is being studied as a
electrochemical proton potential established model system to investigate the mechanism of
by exergonic chemical reactions or light. The ion pumping across membranes.
protons are raised from a low electrochemical Fermenting bacteria have evolved additional
potential inside the cell to a high electrochemi- ways to generate a ∆p. Lactic acid bacteria can
cal potential outside the cell. When the protons create a ∆p via coupled efflux of protons and
circulate back into the cell through appropriate lactate (in addition to ATP hydrolysis) under
carriers, work can be done (e.g., the synthesis certain growth conditions. Another anaerobic
of ATP via the membrane ATP synthase, solute bacterium, Oxalobacter, creates a ∆p by the
transport, flagellar rotation). oxidation of oxalic acid to formic acid coupled
The proton potential is due to a combination with the electrogenic exchange of oxalate for
of a membrane potential (∆Ψ), outside positive, formate and the consumption of protons during
and a ∆pH, outside acid. The membrane poten- the decarboxylation. Other examples similar to
tial seems to be the dominant component in the these will no doubt be discovered in the future.
∆p for most bacteria, except for acidophiles, Sodium potentials are also important in the
which can have a reversed membrane potential. prokaryotes, especially for solute transport.
Other cations, especially sodium ions, can use Although most sodium potentials are created
the established membrane potential for doing secondarily from the proton potential via anti-
work, principally solute accumulation. The porters, there are some prokaryotes that couple
sodium ions must be returned to the outside a chemical reaction to the creation of a sodium
of the cell, and Na+/H+ antiporters serve this potential. For example, some marine bacteria,
www.ebook777.com
as exemplified by Vibrio alginolyticus, couple Study Questions

respiration to the electrogenic translocation
of sodium ions out of the cell at alkaline pH. 1. The E′0 for ubiquinone (ox)/ubiquinone
Because these bacteria can couple the sodium (red) is +100 mV, and for NAD+/NADH it
potential to the membrane ATP synthase, they is –320 mV. What is the ∆E0?
rely on a sodium current rather than a proton
ans. 420 mV
current when growing in basic solutions. When
growing in slightly acidic conditions, they use a 2. In the electron transport chain, oxidation–
proton potential. reduction reactions with a ∆Eh of about
Several different fermenting bacteria can cre- 200 mV appear to be coupled to the extru-
ate a sodium potential by coupling the decarbox- sion of protons. Assume that 100% of the
ylation of organic acids with the translocation oxidation–reduction energy is converted
of sodium ions out of the cell, or by coupling the to the ∆p. For a two-electron transfer and
efflux of end products of fermentation with the the extrusion of two protons, what is the
translocation of sodium ions to the outside. In expected ∆p?
some bacteria (e.g., Klebsiella pneumoniae), the
ans. –200 mV
sodium potential is used to drive the influx of
the growth substrate into the cell but not for the 3. What is the maximum ∆p (i.e., 100%
generation of ATP. Rather, ATP is synthesized energy conversion) when the hydrolysis of
by a substrate-level phosphorylation during the one mole of ATP is coupled to the extru-
conversion of pyruvate to formate and acetate. sion of four moles of protons? Three moles
These bacteria may also generate a membrane of protons? Assume that the free energy of
potential by coupled efflux of the end products hydrolysis of ATP is –50,000 J.
of citrate degradation (formate and acetate)
ans. –130 mV, –173 mV
with protons.
It is emphasized that in fermenting bacteria, 4. Design an experimental approach that
ATP is hydrolyzed via the ATP synthase to cre- can show that the efflux of an organic acid
ate the ∆p that is necessary for membrane activ- along its concentration gradient is coupled
ities (e.g., solute transport, flagellar rotation). to proton translocation and can generate a
A decrease in the ∆p should result in even more membrane potential. (You must be able not
ATP hydrolysis. Thus, reactions are expected only to demonstrate the membrane poten-
to conserve ATP if they create a membrane tial but also to show that the proton is the
potential (e.g., electrogenic efflux of sodium conducting charge.)
ions or protons in symport with end products
5. Explain how the decarboxylation of oxalic
of fermentation, or during decarboxylation
acid by Oxalobacter creates a ∆pH and a
reactions).
∆Ψ.
Measurements of the ∆Ψ and the ∆pH are
necessarily indirect because of the small size 6. What is the reason for stating that light cre-
of the bacteria. The ∆Ψ is measured by using ates a ∆p indirectly in photosynthesis but
cationic or anionic fluorescent dyes that equili- directly in the extreme halophiles?
brate across the membrane in response to the
7. Lactate efflux was in symport with pro-
potential. The distribution of the dyes is moni-
tons, whereas succinate efflux was in sym-
tored by fluorescence quenching. A second way
port with sodium ions. Which ionophores
to measure the membrane potential is by the
might you use to distinguish the cations
equilibration of a permeant ion that achieves
involved?
electrochemical equilibrium with the membrane
potential. The membrane potential is computed 8. A reasonable figure for the actual free
by using the Nernst equation and the intracel- energy of hydrolysis of ATP inside cells is
lular and extracellular ion concentrations. The –50,000 J/mol. It is believed that the hydro-
∆pH is measured by using a weak acid or weak lysis of ATP is coupled to the extrusion of
base whose log ratio of concentrations inside the three protons in many systems. If a ∆p of
cell to outside the cell is a function of the ∆pH. –150 mV were generated, what would be
the efficiency of utilization of ATP energy in millivolts, at 30 °C? What is the common
to create the ∆p? expression used for the force due to the con-
centration gradient when S is a proton?
ans. 87%
18. Assume that an ion traverses a charged
9. Assume a reduction potential of +400 mV
membrane with a potential of ∆Ψ volts.
for an oxidant and a potential of –100 mV
Further assume that there is no concentra-
for a reductant. How many joules of energy
tion gradient. What is the expression that
will be released when two moles of electrons
denotes the driving force? What will the
flow from the reductant to the oxidant?
sign be if the ion moves toward the side of
ans. 96,500 J opposite charge?
10. Assume that 45 kJ is required to synthe- 19. What is the expression, in millivolts at 30
size one mole of ATP. What would be the °C, for the ∆p? What is the expression in
required ∆p, assuming that three moles of joules?
H+ entered via the ATPase per mole of ATP
made?
REFERENCES AND NOTES
ans. –155 mV
11. Assume that the ∆p is –225 mV. If the 1. Harold, F. M. 1986. The Vital Force: A Study of
∆pH at 30 °C is 1.0, what is the membrane Bioenergetics. W. H. Freeman, New York.
potential? 2. Nicholls, D. G., and S. J. Ferguson. 2002.
Bioenergetics 3. Academic Press, San Diego, CA.
ans. –165 mV
3. Mitchell, P. 1961. Coupling of phosphorylation
12. How much energy in joules is required to to electron and hydrogen transfer by a chemiosmotic
move a mole of uncharged solute into the cell type of mechanism. Nature 191:144–148.
against a concentration gradient of 1,000 at 4. Mitchell, P. 1966. Chemiosmotic coupling in oxi-
30 °C? If transport were driven by the ∆p, dative and photosynthetic phosphorylation. Biol.
what would be the minimal value of the ∆p Rev. Cambridge Philos. Soc. 41:445–502.
required if one H+ were cotransported? 5. Mitchell, P. 1977. Vectorial chemiosmotic pro-
cesses. Annu. Rev. Biochem. 46:996–1005.
ans. 17,370 J, 180 mV
6. Mitchell, P. 1979. Compartmentation and com-
13. Assume membrane vesicles loaded with K+ munication in living systems. Ligand conduction: a
so that Kin+ / K+out = 1,000. The temperature is general catalytic principle in chemical, osmotic and
chemiosmotic reaction systems. Eur. J. Biochem.
30 °C, and valinomycin is added. What is
95:1–20.
the predicted initial membrane potential in
millivolts? 7. Kashket, E. R. 1985. The proton motive force
in bacteria: a critical assessment of methods. Annu.
ans. –180 mV Rev. Microbiol. 39:219–242.
14. What is the rationale for adding valino- 8. A single proton (or any monovalent ion, or elec-
tron) carries 1.6 × 10–19 C of charge. If we multiply
mycin and K+ to an experimental system in this by Avogadro’s number, we arrive at the charge
which the number of protons being pumped carried by a mole, which is approximately 96,500
out of the cell is measured? C/mol, or the Faraday constant (F).
15. Briefly, what does the chemiosmotic theory 9. The actual equation is G = G0 + RT ln a, where a
is activity. The activity is a product of the molal con-
state?
centration (c) and the activity coefficient (γ) for the
16. What is meant by the ∆Eh? What is the rela- particular compound, a = γc. In practice, concentra-
tionship between the ∆Eh and the ∆p that is tions are usually used instead of activities, and the
concentrations are in molar units instead of molal.
generated at coupling sites? The symbol T is the absolute temperature in degrees
17. Assume that a concentration gradient of an kelvin (273 + °C). The symbol R is the ideal gas con-
stant, and G0 is the standard free energy (i.e., when
uncharged solute, S, exists across the cell mem- the concentration of all reactants is 1 M). When one
brane. What is the formula that expresses the uses 8.3144 J K–1 mol–1 as the units of R, then the free
driving force in the concentration gradient, energy (G) is given in J/mol.
www.ebook777.com
10. Cecchini, G., and A. L. Koch. 1975. Effect of 22. Ballmoos, C. von, G. M. Cook, and P. Dimroth.
uncouplers on “downhill” β-galactoside transport in 2008. Unique rotary ATP synthase and its biological
energy-depleted cells of Escherichia coli. J. Bacteriol. diversity. Annu. Rev. Biophys. 37:43–64.
123:187–195.
23. Cross, R. L. 1992. The reaction mechanism of
11. Gould, J. M., and W. A. Cramer. 1977. F0F1-ATP synthases, pp. 317–330. In: Molecular
Relationship between oxygen-induced proton efflux Mechanisms in Bioenergetics. L. Ernster (Ed.).
and membrane energization in cells of Escherichia Elsevier Science Publishers, Amsterdam.
coli. J. Biol. Chem. 252:5875–5882.
24. Another way of stating this is in terms of the total
12. E. Padan, D. Zilberstein, and S. Schuldiner. force. The total force is equal to y∆p +∆Gp/F, where
1981. pH homeostasis in bacteria. Biochim. Biophys. ∆Gp/F = 518 mV. At equilibrium the total force is 0;
Acta 650:151–166. therefore y∆p = –518 mV, and when y = 3, ∆p = –173
mV.
13. Reviewed in: Cobley, J. G., and J. C. Cox.
1983. Energy conservation in acidophilic bacteria. 25. Kandpal, R. P., K. E. Stempel, and P. B. Boyer.
Microbiol. Rev. 47:579–595. 1987. Characteristics of the formation of enzyme-
bound ATP from medium inorganic phosphate
14. The pumping of protons out of the cell or the
by mitochondrial F1 adenosine triphosphatase in
electrogenic influx of electrons will create a mem-
the presence of dimethyl sulfoxide. Biochemistry
brane potential, positive outside. However, in the
26:1512–1517.
aerobic acidophilic bacteria [i.e., bacteria that live
in environments of extremely low pH (pH 1–4)], 26. Grubmeyer, C., R. L. Cross, and H. S. Penefsky.
other events act to reverse the membrane potential. 1982. Mechanism of ATP hydrolysis by beef heart
These bacteria have positive membrane potentials mitochondrial ATPase: rate constants for elemen-
(i.e., inside positive with respect to outside, at low tary steps in catalysis at a single site. J. Biol. Chem.
pH). It is not clear why the aerobic acidophiles have 257:12092–12100.
a positive ∆Ψ. One possibility is that they have an
27. Vonck, J., T. K. von Nidda, T. Meier, U. Matthey,
energy-dependent K+ pump that brings K+ into the
D. J. Mills, W. Kuhlbrandt, and P. Dimroth. 2002.
cells at a rate sufficient to establish a net influx of
Molecular architecture of the undecameric rotor
positive charge, creating an inside positive mem-
of a bacterial Na+-ATP synthase. J. Mol. Biol. 321:
brane potential. This point is discussed further in
307–316.
Section 17.1.3.
28. Kaim, G., M. Prummer, B. Sick, G. Zumofen, A.
15. Krulwich, T. A., and A. A. Guffanti. 1986.
Renn, U. P. Wild, and P. Dimroth. 2002. Coupled
Regulation of internal pH in acidophilic and alka-
rotation within single F0F1 enzyme complexes dur-
lophilic bacteria, pp. 352–365. In: Methods in
ing ATP synthesis or hydrolysis. FEBS Lett. 525:
Enzymology, Vol. 125. S. Fleischer and B. Fleischer
156–163.
(Eds.). Academic Press, New York.
29. Capaldi, R. A., and R. Aggeler. 2002. Mechanism
16. Actually, what happens in the presence of nigeri-
of the F1F0 ATP synthase, a biological rotary motor.
cin is that an equalization of the K+ and H+ gradients
Trends Biochem. Sci. 27:154–160.
occurs.
30. Fillingame, R. H., and O. Y. Dmitriev. 2002.
17. Padan, E., D. Zilberstein, and S. Schuldiner.
Structural model of the transmembrane F0 rotary
1981. pH homeostasis in bacteria. Biochim. Biophys.
sector of H+-transporting ATP synthase derived by
Acta 650:151–166.
solution NMR and intersubunit cross-linking in situ.
18. Rottenberg, H. 1979. The measurement of Biochim. Biophys. Acta 1565:232–245.
membrane potential and ∆pH in cells, organelles,
31. The ∆E0 can be a function of the pH. This is
and vesicles. Methods Enzymol. 55:547–569.
the case for the following reaction, where m is not
19. Bakker, E. P. 1990. The role of alkali–cation zero: in some reactions n = 1 and m = 0 (cytochrome
transport in energy coupling of neutrophilic and aci- redox reactions), n = 2 and m = 1 (NAD+ or NADP+
dophilic bacteria: an assessment of methods and con- redox reactions), n = 2 and m = 2 (fumarate–succi-
cepts. FEMS Microbiol. Rev. 75:319–334. nate redox reactions).
20. This is because the resonance frequency of inor- In reactions involving protons, if the pH is not
ganic phosphate or of the γ-phosphate of ATP in zero, then the ∆E0 is more negative. When m = n, the
a high magnetic field is a function of the degree to ∆E0 is –60 mV/pH. When m = 1 and n = 2, the ∆E0
which the phosphate is protonated. (See: Ferguson, S. is –30 mV/pH. For more discussion of this subject,
J., and M. C. Sorgato. 1982. Proton electrochemical see ref. 2.
gradients and energy-transduction processes. Annu.
32. Reviewed in Unemoto, T., H. Tokuda, and M.
Rev. Biochem. 51:185–217.)
Hayashi. 1990. Primary sodium pumps and their
21. Nicholls, D. G., and S. J. Ferguson. 2002. signficance in bacterial energetics, pp. 33–54. In: T.
Bioenergetics 3, pp. 195–217. Elsevier Science Ltd., A. Krulwich (Ed.). The Bacteria, Vol. XII. Academic
London. Press, New York.
33. Reviewed in: Skulachev, V. P. 1992. Chemiosmotic 41. Dimroth, P. 1980. A new sodium-transport sys-
systems and the basic principles of cell energetics, pp. tem energized by the decarboxylation of oxaloac-
37–73. In: Molecular Mechanisms in Bioenergetics. etate. FEBS Lett. 122:234–236.
Ernster, L. (Ed.). New Comprehensive Biochemistry:
42. Dimroth, P., and A. Thomer. 1988. Dissociation
Molecular Mechanisms in Bioenergetics, Vol. 23.
of the sodium-ion-translocating oxaloacetate decar-
Elsevier, Amsterdam.
boxylase of Klebsiella pneumoniae and reconstitu-
34. Tokuda, H., and T. Unemoto. 1982. Character- tion of the active complex from the isolated subunits.
ization of the respiration-dependent Na+ pump in Eur. J. Biochem. 175:175–180.
the marine bacterium Vibrio alginolyticus. J. Biol.
Chem. 257:10007–10014. 43. Hilpert, W., and P. Dimroth. 1983. Purification
and characterization of a new sodium transport
35. Maloney, P. C., and F. C. Hansen III. 1982. decarboxylase. Methylmalonyl–CoA decarboxy-
Stoichiometry of proton movements coupled to ATP lase from Veillonella alcalescens. Eur. J. Biochem.
synthesis driven by a pH gradient in Streptococcus 132:579–587.
lactis. J. Membrane Biol. 66:63–75.
44. Buckel, W., and R. Semmler. 1983. Purification,
36. In his review of primary sodium ion translo- characterization and reconstitution of glutaconyl–
cating enzymes, Dimroth points out that V. algi- CoA decarboxylase. Eur. J. Biochem. 136:427–434.
nolyticus has two different NADH:ubiquinone
oxidoreductases: NQR1, which is Na+ dependent 45. Schink, B., and N. Pfennig. 1982. Propionigenium
and functions at pH 8.5 but not at pH 6.5, and modestum gen. nov. sp. nov. A new strictly anaerobic
NQR2, which is Na+ independent and is not a cou- nonsporing bacterium growing on succinate. Arch.
pling site. There is apparently no H+-dependent Microbiol. 133:209–216.
NADH:ubiquinone oxidoreductase. However, these 46. Hilpert, W., B. Schink, and P. Dimroth. 1984.
bacteria do have a cytochrome bo oxidase that oxi- Life by a new decarboxylation-dependent energy
dizes the quinol, is not Na+ dependent, and is believed conservation mechanism with Na+ as coupling ion.
to be a proton pump as in other bacteria. The pres- EMBO J. 3:1665–1670.
ence of both pumps can explain how V. alginolyti-
cus operates a Na+-dependent respiratory pump at 47. De Vries, W., R. Theresia, M. Rietveld-Struijk,
pH 8.5 and a H+-dependent respiratory pump at pH and A. H. Stouthamer. 1977. ATP formation asso-
6.5. The cytochrome bo proton pump must function ciated with fumarate and nitrate reduction in grow-
at both acidic and basic pH values because mutants ing cultures of Veillonella alcalescens. Antonie van
lacking the Na+-dependent NADH:ubiquinone oxi- Leeuwenhoek 43:153–167.
doreductase extrude Na+ at pH 8.5, using a Na+/H+ 48. Buckel, W., and R. Semmler. 1982. A biotin-de-
antiporter in combination with a primary proton pendent sodium pump: glutaconyl–CoA decarboxy-
pump, and the wild type is known to extrude Na+ at lase from Acidaminococcus fermentans. FEBS Lett.
pH 6.5, using the Na+/H+ antiporter in combination 148:35–38.
with the primary proton pump. (Dimroth, P. 1997.
Primary sodium ion translocating enzymes. Biochim. 49. Inside-out vesicles are prepared by sonicating
Biophys. Acta 1318:11–51.) whole cells or shearing them with a French pres-
sure cell. Right-side-out vesicles are prepared by first
37. Kreke, B., and H. Cypionka. 1994. Role of removing the cell wall with lysozyme in a hypertonic
sodium ions for sulfate transport and energy metab- medium, and then osmotically lysing the protoplasts
olism in Desulfovibrio salexigens. Arch. Microbiol. or spheroplasts in hypotonic medium.
161:55–61.
50. Anantharam, V., M. J. Allison, and P. C.
38. Many nonfermenting anaerobic bacteria carry Maloney. 1989. Oxalate:formate exchange. J. Biol.
out electron transport by using as electron accep- Chem. 264:7244–7250.
tors either organic compounds such as fumarate or
inorganic compounds such as nitrate. Thus, electron 51. Baetz, A. L., and M. J. Allison. 1990. Purification
flow in these bacteria can be coupled to proton efflux and characterization of oxalyl–coenzyme A decar-
and the establishment of a ∆p. Furthermore, even boxylase from Oxalobacter formigenes. J. Bacteriol.
fermenting bacteria can carry out some fumarate res- 171:2605–2608.
piration generating a ∆p. However, the major source 52. Baetz, A. L., and M. J. Allison. 1990. Purification
of energy for the ∆p in most fermenting bacteria is and characterization of formyl–coenzyme A trans-
ATP hydrolysis. ferase from Oxalobacter formigenes. J. Bacteriol.
39. Dimroth, P. 1997. Primary sodiumion translocat- 171:3537–3540.
ing enzymes. Biochim. Biophys. Acta 1318:11–51.
53. Ruan, Z., V. Anantharam, I. T. Crawford,
40. Dimroth, P. 1990. Energy transductions by an S. V. Ambudkar, S. Y. Rhee, M. J. Allison, and P.
electrochemical gradient of sodium ions, pp. 114– C. Maloney. 1992. Identification, purification, and
127. In: The Molecular Basis of Bacterial Metabolism. reconstitution of OxIT, the oxalate:formate antiport
G. Hauska and R. Thauer (Eds.). Springer-Verlag, protein of Oxalobacter formigenes. J. Biol. Chem.
Berlin. 267:10537–10543.
www.ebook777.com
54. To prepare proteoliposomes, one disperses phos- growing and nongrowing cells of Streptococcus cre-
pholipids (e.g., those isolated from E. coli) in water, moris. J. Bacteriol. 162:383–390.
where they spontaneously aggregate to form spheri-
65. Oesterhelt, D., and J. Tittor. 1989. Two pumps,
cal vesicles consisting of concentric layers of phos-
one principle: light-driven ion transport in halobac-
pholipid. These vesicles, called liposomes, are then
teria. Trends Biochem. Sci. 14:57–61.
subjected to high-frequency sound waves (sonic
oscillation), which break them into smaller vesicles 66. Bogomolni, R. A., R. A. Baker, R. H. Lozier, and
surrounded by a single phospholipid bilayer resem- W. Stoeckenius. 1980. Action spectrum and quan-
bling the lipid bilayer found in natural membranes. tum efficiency for proton pumping in Halobacterium
Then purified protein (e.g., the OxIT antiporter) is halobium. Biochemistry 19:2152–2159.
mixed with the sonicated phospholipids in the pres-
67. Henderson, R., J. M. Baldwin, and T. A. Ceska.
ence of detergent, and the suspension is diluted into
1990. Model for the structure of bacteriorhodopsin
buffer. The protein becomes incorporated into the
based on high-resolution electron cryomicroscopy.
phospholipid bilayer, and membrane vesicles called
J. Mol. Biol. 213:899–929.
proteoliposomes are formed. When the proteoli-
posomes are incubated with solute, they catalyze 68. The extreme halophiles require unusually high
uptake of the solute into the vesicles, provided the external NaCl concentrations [at least 3–5 M (i.e.,
appropriate carrier protein has been incorporated. In 17–28%)] to grow. They inhabit hypersaline envi-
addition, one can “load” the proteoliposomes with ronments such as the solar salt evaporation ponds
solutes (e.g., oxalate) by including these in the dilu- near San Francisco and salt lakes (e.g., the Great Salt
tion buffer. Lake in Utah and the Dead Sea). There are now six
recognized genera, two of them being the well-known
55. Poolman, B., D. Molenaar, E. J. Smid, T. Ubbink,
Halobacterium and Halococcus. The best studied is
T. Abee, P. P. Renault, and W. N. Konings. 1991.
Hb. salinarium (halobium). The other four genera
Malolactic fermentation: electrogenic malate uptake
are Haloarcula, Haloferax, Natronobacterium, and
and malate/lactate antiport generate metabolic
Natronococcus. The majority of the known halo-
energy. J. Bacteriol. 173:6030–6037.
philic archaea are aerobic chemo-organotrophs
56. Konings, W. N. 1985. Generation of metabolic and can grow on simple carbohydrates as well as
energy by end-product efflux. Trends Biochem. Sci. long-chain saturated hydrocarbons. They generally
10:317–319. grow best at pH values between 8 and 9. However,
Natronobacterium and Natronococcus are also alka-
57. Konings, W. N., J. S. Lolkema, and B. Poolman.
liphilic and grow well at pH values up to 11. When
1995. The generation of metabolic energy by solute
oxygen is not present, the halobacteria will grow
transport. Arch. Microbiol. 164:235–242.
anaerobically by using several electron acceptors in
58. Michels, J. P., J. Michel, J. Boonstra, and W. place of oxygen. These include fumarate, dimethyl
N. Konings. 1979. Generation of an electrochemi- sulfoxide (DMSO), and trimethylamine N-oxide
cal proton gradient in bacteria by the excretion of (TMAO). Members of the genera Haloarcula and
metabolic end-products. FEMS Microbiol. Lett. Haloferax can grow on nitrate as the terminal elec-
5:357–364. tron acceptor. Some of the halobacteria reduce the
nitrate to nitrite and some reduce it completely to
59. Otto, R., et al. 1982. Lactate efflux–induced elec-
nitrogen gas. Some halobacteria can also grow fer-
trical potential in membrane vesicles of Streptococcus
mentatively in the absence of oxygen. These include
cremoris. J. Bacteriol. 149:733–738.
Hb. salinarium (halobium), which can ferment argi-
60. Brink, B. T., and W. N. Konings. 1982. nine to citrulline.
Electrochemical proton gradient and lactate con-
69. Oesterhelt, D., and G. Krippahl. 1983.
centration gradient in Streptococcus cremoris cells
Phototrophic growth of halobacteria and its use for
grown in batch culture. J. Bacteriol. 152:682–686.
isolation of photosynthetically deficient mutants.
61. Driessen, A. J. M., and W. N. Konings. 1990. Ann. Microbiol. (Inst. Pasteur). 134B:137–150.
Energetic problems of bacterial fermentations: extru-
70. Gest, H. 1993. Photosynthetic and quasi-photo-
sion of metabolic end products, pp. 449–478. In:
synthetic bacteria. FEMS Microbiol. Lett. 112:1–6.
Bacterial Energetics. T. A. Krulwich (Ed.). Academic
Press, New York. 71. Some halophilic archaea can use nitrate as an
electron acceptor to carry out anaerobic respiration.
62. Michel, T. A., and J. M. Macy. 1990. Generation
of a membrane potential by sodium-dependent succi- 72. Respiration can be severely limited under cer-
nate efflux in Selenomonas ruminantium. J. Bacteriol. tain growth conditions because the oxygen content
172:1430–1435. of hypersaline waters, the normal habitat of these
organisms, is usually 20% or less than is found in
63. The lactic and succinic acids are presumed to be
normal seawater, and in unstirred ponds oxygen
in the ionized form because the intracellular pH is
becomes even more scarce. The halobacteria can
much larger than the pKa values.
derive energy from the fermentation of amino acids;
64. Brink, B. T., R. Otto, U. Hansen, and W. N. but in the absence of a fermentable carbon source
Konings. 1985. Energy recycling by lactate efflux in and respiration, light is the only source of energy.
73. Bogomolni, R. A., R. A. Baker, R. H. Lozier, and 81. Krebs, M. P., and H. G. Khorana. 1993.
W. Stoeckenius. 1980. Action spectrum and quan- Mechanism of light-dependent proton translocation
tum efficiency for proton pumping in Halobacterium by bacteriorhodopsin. J. Bacteriol. 175:1555–1560.
halobium. Biochemistry 19:2152–2159.
82. Since aspartate-85 remains protonated while a
74. Tittor, J., and D. Oesterhelt. 1990. The quan- proton is released into the aqueous phase, we must
tum yield of bacteriorhodopsin. FEBS Lett. 263: conclude that the immediate source of the released
269–273. proton is a different amino acid residue.
75. Govindjee, R., S. P. Balashov, and T. G. Ebrey. 83. Senior, A. E. 1990. The proton-translocating
1990. Quantum efficiency of the photochemical cycle ATPase of Escherichia coli. Annu. Rev. Biophys.
of bacteriorhodopsin. Biophys. J. 58:597–608. Chem. 19:7–41.
76. Lanyi, J. K. 1997. Mechanism of ion trans- 84. Lanyi, J. K. 1990. Halorhodopsin, a light-driven
port across membranes. J. Biol. Chem. 272: electrogenic chloride-transport system. Physiol. Rev.
31209–31212. 70:319–330.
77. A Schiff base is an imine and has the following 85. Duschl, A., and G. Wagner. 1986. Primary and
structure: R–CH5N–R′. It is formed between a car- secondary chloride transport in Halobacterium halo-
bonyl group and a primary amine, as follows: R–CHO bium. J. Bacteriol. 168:548–552.
+ H2N–R′ → R–CH5N–R′ + H2O. In bacteriorho-
86. There are many nonobligate alkaliphilic bacteria
dopsin, the amino group is donated by lysine in the
whose optimal growth pH is 9 or greater that never-
protein and the carbonyl is donated by the retinal.
theless can grow at pH 7 or below. These microor-
78. Lanyi, J. K. 1992. Proton transfer and energy ganisms, widely distributed among the bacterial
coupling in the bacteriorhodopsin photocycle. J. genera, include both bacteria and archaea. There are
Bioenerg. Biomemb. 24:169–179. also alkaliphilic-tolerant bacteria that can grow at
pH values of 9 or more, but whose optimal growth
79. Oesterhelt, D., J. Tittor, and E. Bamberg. 1992.
pH is around neutrality.
A unifying concept for ion translocation by retinal
proteins. J. Bioenerg. Biomemb. 24:181–191. 87. Reviewed in: Krulwich, T. A., and D. M. Ivey.
1990. Bioenergetics in extreme environments,
80. Fodor, S. P. A., J. B. Ames, R. Gebhard, E. M.
pp. 417–447. In: The Bacteria, Vol. XII. T. A.
M. van den Berg, W. Stoeckenius, J. Lugtenburg,
Krulwich (Ed.). Academic Press, New York.
and R. A. Mathies. 1988. Chromophore structure
in bacteriorhodopsin’s N intermediate: implications 88. Krulwich, T. A., and A. A. Guffanti. 1989.
for the proton-pumping mechanism. Biochemistry Alkalophilic bacteria. Annu. Rev. Microbiol.
27:7097–7101. 43:435–463.
www.ebook777.com
5
Electron Transport
Electron transport refers to the current of elec- membrane as part of reduced quinone (QH2).
trons in the cell membranes of prokaryotes, and Sometimes (e.g., Fig. 5.12), reduced quinone is
in mitochondrial and chloroplast membranes. referred to as reduced ubiquinone (UQH2). The
The electrons flow spontaneously down an quinone is picked up as a proton (H+) from the
energy gradient through a series of electron car- cytosol on one side of the membrane, translo-
riers, and at specific sites in the chain of carri- cated through the membrane as part of reduced
ers, the energy that is released is conserved as quinone, and released as H+ on the outside of the
a proton motive force (∆p) that can be used membrane surface. At other coupling sites, the
to make ATP and supply energy for other cel- proton itself (H+) is actually pumped through
lular tasks. This chapter describes the entire the membrane. The net result, regardless of
process as it occurs in bacteria and in mito- the mechanism, is the translocation of protons
chondria. Electron transport in photosynthetic from one side of the membrane to the other side,
membranes is described in Chapter 6. Archaeal as shown in Fig. 5.1. This is important because
electron transport chains appear to be similar a membrane potential is created during proton
in many ways to bacterial electron transport translocation (due either to the transmembrane
chains but are quite diverse. For reviews, see movement of electrons or of H+, as explained
refs. 1 and 2. later), and also because the reentry of protons
In what direction do electrons spontaneously through the membrane-bound ATP synthase
flow? As discussed in Section 4.7.1, electrons results in the synthesis of ATP. As explained in
flow spontaneously down a potential energy Chapter 4, the energy to synthesize the ATP is
gradient toward acceptors that have a more derived from a combination of the membrane
positive electrode potential. The electrons potential and the proton concentration gradi-
flow from primary electron donors to termi- ent, called the ∆p.
nal electron acceptors through a series of elec- Mitochondria have three coupling sites,
tron carrier proteins and a class of lipids called whereas bacteria can have one to three coupling
quinones. An important point is that proton sites depending upon the bacterium and the
translocation across the membranes takes place growth conditions. In other words, prokaryotic
during electron transport at specific sites in the cell membranes (and mitochondria and chloro-
electron transport chain called coupling sites; plasts) convert an electrode potential difference
an electrochemical proton potential, called a ∆Eh into a proton electrochemical potential dif-
proton motive force, or ∆p, is created at these ference (∆p). (As discussed in Section 4.7.1, Eh is
sites. In some cases (i.e., the Q loop and Q the actual electrode potential of a compound at
cycle, explained in Sections 5.6.1 and 5.6.2), the specified concentrations of its oxidized and
the hydrogen is actually transported across the reduced forms.) The proton electrochemical
146
electron transport 147
OUT IN the final electron acceptors, the mechanism is

referred to as anaerobic respiration. The final
Ared
electron acceptors can be inorganic, such as
Aox
oxygen, nitrate, or sulfate. They can also be
yH+ organic, for example, fumarate. The name of
the respiration is derived from the final electron
Box acceptor. Thus, we have oxygen respiration,
Bred nitrate respiration, sulfate respiration, fumar-
ate respiration, and so on.
Fig. 5.1 Electron transport in membranes. Electrons 5.2 The Electron Carriers
flow from A to B, through a series of electron carri-
ers in the membrane, from a low electrode potential The electrons flow through a series of electron
toward a higher potential. The individual reactions carriers. Some of these carry hydrogen as well
that transfer the electrons are referred to as redox as electrons, and some carry only electrons. The
reactions. The intermediate redox reactions between electron carriers are as follows:
A and B are not shown. Certain of the redox reac-
tions in the series are coupled to the translocation of 1. Flavoproteins (hydrogen and electron carriers)
protons across the cell membrane, involving either 2. Quinones (hydrogen and electron carriers)
reduced quinone (UQH2) or H+ pumps. These occur 3. Iron–sulfur proteins (electron carriers)
at coupling sites. In this way a redox potential (∆Eh), 4. Cytochromes (electron carriers)
which in this case refers to the energy that is released
when the electrons move down an energy gradient
The quinones are lipids, whereas the other
from the donor to the acceptor, is converted into a electron carriers are proteins, which exist in
proton potential (∆p): that is, n∆Eh = y∆p, where n multiprotein enzyme complexes called oxi-
is the number of electrons transferred from donor to doreductases. (See note 3 for examples of oxi-
acceptor, and y is the number of protons extruded. doreductases.) The electrons are not carried in
When the protons return across the membrane, work the protein per se, but in a nonprotein molecule
(e.g., the synthesis of ATP or solute transport) can be bound to the protein. The nonprotein portion
done. The type of work that is done depends upon that carries the electron is called a prosthetic
the protein machinery through which the protons group. (See note 4 for useful definitions.) The
return. prosthetic group in iron–sulfur proteins is a
cluster of iron–sulfide, which is abbreviated as
FeS. The prosthetic group in flavoproteins (Fp)
potential is then used to drive solute transport, is a flavin, which can be either flavin adenine
ATP synthesis, flagellar rotation, and other dinucleotide (FAD) or flavin mononucleotide
membrane activities. In mitochondria, all elec- (FMN). The prosthetic group in cytochromes
tron transport pathways are much the same. is heme. The chemistry of the prosthetic groups
However, prokaryotes are diverse creatures, is described in Section 5.2.1. Some of the pros-
and their electron transport pathways differ thetic groups (flavins) carry hydrogen as well as
depending upon the primary donor and termi- electrons, and they are referred to as hydrogen
nal acceptor. This chapter describes electron carriers. The quinones are also hydrogen car-
transport pathways in mitochondria and bacte- riers. Some of the prosthetic groups (FeS and
ria and how they are coupled to the formation heme) carry only electrons, and they are referred
of a ∆p. to as electron carriers.
Each of the electron carriers, described in
Sections 5.2.1 through 5.2.4, has a different
5.1 Aerobic and Anaerobic electrode potential, and the electrons are trans-
Respiration ferred sequentially to a carrier of a higher poten-
Electron transport in membranes is referred tial to the final acceptor, which has the highest
to as respiration. If oxygen is the final electron potential.
acceptor, then the mechanism is referred to as The standard potentials at pH 7 of the elec-
aerobic respiration. If other compounds are tron carriers and other electron donors and
www.ebook777.com
acceptors in metabolic pathways are shown in two ring nitrogens. There are many different
Table 5.1. flavoproteins, and they catalyze diverse oxida-
tion–reduction reactions in the cytoplasm, not
5.2.1 Flavoproteins merely those of the electron transport chain in
A flavoprotein (Fp) is an electron carrier that the membranes. Although all the flavoproteins
has as its prosthetic group an organic molecule have FMN or FAD as their prosthetic group,
called a flavin. The term is derived from the they catalyze different oxidations and have dif-
Latin word flavius, which means yellow, in ref- ferent redox potentials. These differences are
erence to the color of flavins. The flavins FAD due to differences in the protein component of
and FMN are synthesized by cells from the vita- the enzyme, not in the flavin itself.
min riboflavin (vitamin B2).
What is the difference between FMN and 5.2.2 Quinones
FAD? This is shown in Fig. 5.2. Phosphorylation Quinones are lipid electron carriers. Owing
of riboflavin at the ribityl 5′-OH yields FMN, to their hydrophobic lipid nature, some are
and adenylylation (addition of ADP) yields believed to be highly mobile in the lipid phase
FAD. of the membrane, carrying hydrogen and elec-
As Fig. 5.2 illustrates, when flavins are trons to and from the complexes of protein
reduced they carry 2H• (equivalent to two electron carriers that are not mobile. Quinone
electrons and two protons), one on each of structures and oxidation–reduction reactions
are shown in Fig. 5.3. All quinones have hydro-
phobic isoprenoid side chains that contribute
Table 5.1 Standard electrode potentials at pH 7 to their lipid solubility. The number of isoprene
units varies but is typically 6 to 10. Bacteria
Couple Potential (mV)
make two types of quinone that function dur-
Fdox/Fdred (spinach) –432 ing respiration: ubiquinone (UQ), a quinone
CO2/formate –432 also found in mitochondria, and menaquinone
H+/H2 –410 (MQ, or sometimes MK). Menaquinones (Fig.
Fdox/Fdred (Clostridium) –410
NAD+/NADH –320
5.3C), which are derivatives of vitamin K, differ
FeS (ox/red) in mitochondria –305 from ubiquinones in being naphthoquinones in
Lipoic/dihydrolipophilic –290 which the additional benzene ring replaces the
Sº/H2S –270 two methoxy groups present in ubiquinones
FAD/FADH2 –220 (Fig. 5.3A,B). Menaquinones also have a much
Acetaldehyde/ethanol –197
FMN/FMNH2 –190
lower electrode potential than ubiquinones and
Pyruvate/lactate –185 are used predominantly during anaerobic res-
Oxaloacetate/malate –170 piration, where the electron acceptor has a low
Menaquinone (ox/red) –74 potential (e.g., during fumarate respiration). A
cyt b558 (ox/red) –75 to –43 third type of quinone, plastoquinone (Fig. 5.3D),
Fumarate/succinate + 33
Ubiquinone (ox/red) +100
occurs in chloroplasts and cyanobacteria, and
cyt b556 (ox/red) +46 to +129 functions in photosynthetic electron transport.
cyt b562 (ox/red) +125 to +260 In plastoquinones, the two methoxy groups are
cyt d (ox/red) +260 to +280 replaced by methyl groups.
cyt c (ox/red) +250
FeS (ox/red) in mitochondria +280
cyt a (ox/red) +290 5.2.3 Iron–sulfur proteins
cyt c555 (ox/red) +355 Iron–sulfur proteins contain nonheme iron and
cyt a3 (ox/red) in mitochondria +385 usually acid-labile sulfur (Fig. 5.4). The term
NO−3/NO−2 +421 “acid-labile sulfur” means that when the pH is
Fe3+/Fe2+ +771
O2 (1 atm)/H2O +815 lowered to approximately 1, hydrogen sulfide is
released from the protein. This is because there
Sources: Thauer, R. K., K. Jungermann, and K. Decker. is sulfide attached to iron by bonds that are rup-
1977. Energy conservation in chemotrophic anaerobic
tured in acid. Generally, the proteins contain
bacteria. Bacteriol Rev. 41:100–180. Metzler, D. E. 1977.
Biochemistry: The Chemical Reactions of Living Cells. clusters in which iron and acid-labile sulfur are
Academic Press, New York. present in a ratio of 1:1. However, there may
Fig. 5.2 Structures of riboflavin (X = H), FMN (X = PO3H2), and FAD (X = ADP). For convenience, the reduc-
tion reaction is drawn as proceeding via a hydride ion even though this is not the actual mechanism in all flavin
reductions.
Fig. 5.3 The structure of quinones: (A) oxidized ubiquinone, (B) reduced ubiquinone, (C) oxidized menaqui-
none, and (D) oxidized plastoquinone. The value of n can be 4 to 10 and is 8 for both quinones in E. coli. In
E. coli ubiquinone plays a major role in aerobic and nitrate respiration, whereas menaquinone is dominant
during fumarate respiration. One reason for this is that ubiquinone has a potential of +100 mV, versus +30
mV for fumarate. It is therefore at too high a potential to deliver electrons to fumarate. Menaquinone has a
low potential, –74 mV, and is thus able to deliver electrons to fumarate. Plastoquinone is used in chloroplast
and cyanobacterial photosynthetic electron transport.
be more than one iron–sulfur cluster per pro-

tein. For example, in mitochondria the enzyme
complex that oxidizes NADH has at least four
FeS clusters (see later, Fig. 5.9). The FeS clusters
have different Eh values, and the electron trav-
els from one FeS cluster to the next toward the
higher Eh. It appears that the electron may not
be localized on any particular iron atom, and
the entire FeS cluster should be thought of as
carrying one electron, regardless of the number
of Fe atoms. Fig. 5.4 Scheme for FeS clusters: this is a Fe2S2 clus-
Iron–sulfurs proteins also contain cysteine ter. More than one cluster may be present per pro-
sulfur, which is not acid labile, and bonds the tein. The sulfur atoms held only by the iron are acid
iron to the protein. There are several different labile. The iron is bonded to the protein via sulfur in
types of iron–sulfur protein, and these catalyze cysteine residues.
www.ebook777.com
numerous oxidation–reduction reactions in the oxygen. (In naming cytochromes, sometimes

cytoplasm as well as in the membranes. (See the O2-binding heme is given the subscript 3.)7
note 5 for more information on iron–sulfur As the names imply, each cytochrome contains
proteins.) The iron–sulfur proteins have char- two types of heme, one being heme b and the
acteristic electron spin resonance (ESR) spectra other being heme d or o.7 As mentioned previ-
because of an unpaired electron in either the ously, the hemes can be distinguished according
oxidized or reduced form of the FeS cluster in to the side groups that they possess, as summa-
different FeS proteins. (See note 6 for a descrip- rized in Fig. 5.5. For example, heme o differs
tion of electron spin resonance.) The iron–sulfur from heme b in having an hydroxyethylfarnesyl
proteins cover a very wide range of potentials, group substituted for a vinyl group. However,
from approximately –400 mV to +350 mV. the only difference between heme b and heme
They therefore can carry out oxidation–reduc- c is that the latter is covalently bound to pro-
tion reactions at both the low-potential end and tein by thioether linkages between the two vinyl
the high-potential end of the electron transport groups and cysteine residues in the protein.
chain, and indeed are found in several locations. Hemes can usually be distinguished spectro-
In the FeS cluster shown in Fig. 5.4, note that photometrically. When cytochromes are in the
each Fe is bound to two acid-labile sulfurs and reduced state, absorption by the heme produces
two cysteine sulfurs. This would be called an characteristic light absorption bands in the vis-
Fe2S2 cluster. ible range: the α, β, and γ bands. The α bands
absorb light between 500 and 600 nm, the β
5.2.4 Cytochromes bands absorb at a lower wavelength, and the γ
Cytochromes are electron carriers that have bands (also called Soret bands) are in the blue
heme as the prosthetic group. Heme consists region of the spectrum. The spectrum for a cyto-
of four pyrrole rings attached to each other by chrome c is shown in Fig. 5.6. Cytochromes are
methene bridges (Fig. 5.5). Because hemes have distinguished, in part, by the position of the max-
four pyrroles, they are called tetrapyrroles. imum in the α band. For example, cyt b556 has a
Each of the pyrrole rings is substituted by a peak at 556 nm and cyt b558 a peak at 558 nm.
side chain. Substituted tetrapyrroles are called
porphyrins. Therefore, hemes are also called Reduced minus oxidized spectra
porphyrins. (An unsubstituted tetrapyrrole is Because of light scattering and nonspecific
called a porphin.) Hemes are placed in differ- absorption, it is very difficult to resolve the
ent classes, described shortly, on the basis of the different peaks of individual cytochromes in
side chains attached to the pyrrole rings. In the whole cells unless one employs difference spec-
center of each heme there is an iron atom that troscopy. For difference spectroscopy, the cells
is bound to the nitrogen of the pyrrole rings. are placed into two cuvettes in a split-beam
The iron is the electron carrier and is oxidized spectrophotometer, and monochromatic light
to ferric or reduced to ferrous ion during elec- from a single monochromator scan is split to
tron transport. Cytochromes are therefore one- pass through both cuvettes. In one cuvette the
electron carriers. The Eh values of the different cytochromes are oxidized by adding an oxi-
cytochromes vary depending on the protein and dant, and in the second cuvette they are reduced
the molecular interactions with surrounding by adding a reductant. The spectrophotometer
molecules. subtracts the output of one cuvette from the
other to give a reduced minus oxidized differ-
Classes of cytochromes ence spectrum. In this way nonspecific absorp-
Figure 5.5 shows five classes of heme that dis- tion and light scattering are eliminated from the
tinguish the cytochromes: hemes a, b, c, d, and spectrum, and the cytochromes in the prepara-
o. Hemes d and o have been found only in the tion are identified.
prokaryotic cytochrome oxidases. Bacterial
cytochromes include cytochromes bd (some- Dual-beam spectroscopy
times called cytochrome d) and bo (some- A dual-beam spectrophotometer is used to fol-
times called cytochrome bo3 or cytochrome low the kinetics of oxidation or reduction of
o), which are quinol oxidases that reduce a particular cytochrome. The instrument has
Fig. 5.5 The prosthetic groups of the different classes of cytochromes. The hemes vary according to their
side groups. Heme c is covalently bound to the protein via a sulfur bridge to a cysteine residue on the protein.
Source: Adapted from Gottschalk, G. Bacterial Metabolism, Springer-Verlag, New York. 1986.
two monochromators: light from one is set at Table 5.1 shows standard electrode potentials
a wavelength at which absorbance will change at pH 7 (E′0) of some electron donors, acceptors,
during oxidation or reduction, and the second and electron carriers. Notice that redox couples
beam of light is at a nearby wavelength for are generally written in the form “oxidized/
which absorbance will not change. The light is reduced.” Many of the oxidation–reduction
sent alternatively from both monochromators reactions in the electron transport chain can be
through the sample cuvette, and the difference reversed by the ∆p as discussed in Section 4.7.1.
in absorbance between the two wavelengths is This means that the ox/red ratio for several of the
automatically plotted as a function of time. electron carriers (flavoproteins, cytochromes,
www.ebook777.com
Four complexes can be recognized in

mitochondria. They are complex I (NADH–
ubiquinone oxidoreductase), complex II
(succinate dehydrogenase), complex III
(ubiquinol–cytochrome c oxidoreductase,
also called the bc1 complex), and complex IV
(cytochrome c oxidase, which is cytochrome
aa3). Complexes I, III, and IV are coupling sites
(Section 5.5). Each complex can have several
proteins. The most intricate is complex I, from
mammalian mitochondria, which has about
40 polypeptide subunits, at least four iron–sul-
fur centers, one flavin mononucleotide (FMN),
and one or two bound ubiquinones. Analogous
complexes have been isolated from bacteria,
Fig. 5.6 Absorption spectra of oxidized (dashed but in some cases (e.g., NADH–ubiquinone
curve) and reduced (solid curve) cytochrome c. The oxidoreductase and the bc1 complex), they
α band in the reduced form is used to identify cyto- have fewer protein components.8–11 (See note
chromes. 12 for a description of complex II and how it
varies with different bacteria.) Note the pat-
tern in the arrangement of the electron carri-
quinones, FeS proteins) must be close to 1. ers in mitochondria; a dehydrogenase complex
Thus, for these reactions the Eh (actual potential accepts electrons from a primary donor and
at pH 7) values of the redox couples are close to transfers the electrons to a quinone. The qui-
their midpoint potentials, which is the potential none then transfers the electrons to an oxi-
at pH 7 when the couple is 50% reduced: that dase complex via intervening cytochromes. As
is, [ox] = [red]. described next, the same general pattern exists
in bacteria.
5.3 Organization of the Electron
Carriers in Mitochondria 5.4 Organization of the Electron
For a historical perspective, see Box 5.1. Carriers in Bacteria
The electron carriers are organized as an For a review, see ref. 13. Bacterial electron
electron transport chain that transfers electrons transport chains vary among the different bac-
from electron donors at a low electrode poten- teria, and also according to the growth condi-
tial to electron acceptors at a higher electrode tions. These variations will be discussed later
potential (Fig. 5.7). Electrons can enter at the (Section 5.7). First, we shall describe the com-
level of flavoprotein, quinone, or cytochrome, mon features in bacterial electron transport
depending upon the potential of the donor. schemes and compare them with systems of
The carriers are organized in the membrane as mitochondrial electron transport. As with the
individual complexes. The complexes can be mitochondrial electron transport chain, the bac-
isolated from each other by appropriate separa- terial chains are organized into dehydrogenase
tion techniques after mild detergent extraction, and oxidase complexes connected by quinones
which removes the lipids but does not destroy (Fig. 5.8). The quinones accept electrons from
the protein–protein interactions. The separated dehydrogenases and transfer them to oxidase
complexes can be analyzed for their components complexes that reduce the terminal electron
and also can be incorporated into proteolipo- acceptor. Bacteria are capable of using electron
somes to facilitate study of the oxidation–reduc- acceptors other than oxygen (e.g., nitrate and
tion reactions that each catalyzes, in addition fumarate) during anaerobic respiration. The
to proton translocation. (Proteoliposomes are enzyme complexes that reduce electron accep-
artificial constructs of purified lipids and pro- tors other than oxygen are called reductases,
teins. They are described in Section 17.1.) rather than oxidases.
BOX 5.1 HISTORICAL PERSPECTIVE: CELLULAR RESPIRATION

The elucidation of the pathway by which 1887. MacMunn reported the discovery of
electrons flow from organic compounds a pigment in the course of a spectroscopic
to oxygen was the result of many years of examination of the tissues of various species
research by different investigators. The of vertebrates and invertebrates. He called
realization that iron compounds were the pigment histohaematin, or myohae-
electron carriers came from early work matin, and it had a characteristic absorp-
on the effect of cyanide on respiration. tion spectrum consisting of four bands.
Otto Heinrich Warburg (1883–1970), a MacMunn reported that when he added
German biochemist who studied respi- the oxidizing agent hydrogen peroxide, the
ration, reported the inhibitory effect of bands disappeared, only to reappear upon
cyanide on respiration in sea urchin eggs, reduction. He concluded that since the sub-
yeast, and bacteria before and after World stances were capable of being oxidized and
War I. Because he knew that cyanide inhib- reduced, they were respiratory pigments.
its autoxidation reactions (e.g., the oxida- He also concluded that they were protein.
tion of cysteine to cystine, catalyzed by iron However, MacMunn’s discovery was not
compounds), Warburg concluded that cya- accepted by his contemporaries, especially
nide also inhibits respiration, further rea- the famous biochemist Hoppe-Seyler, who
soning that an iron-containing enzyme that thought MacMunn’s belief that he had dis-
he called Atmungsfermentt (“respiratory covered a cellular respiratory pigment was
ferment”) catalyzes the oxidation of the erroneous.
organic substrates. Warburg also showed, MacMunn was clearly frustrated that
in 1926, that carbon monoxide inhibits the his discovery had not been accepted, and
uptake of oxygen by yeast cells. He knew he wrote in a book that was not published
from earlier studies by others that carbon until after he died, in 1911, that “doubtless
monoxide combines with hemoglobin and in time this pigment will find its way into
that it can be dissociated from the ferro– the textbooks.”1 This indeed did occur, pri-
heme complex with visible light. He was marily through the work of David Keilin in
able to show that visible light also reverses the 1920s.
the inhibition of respiration of yeast by David Keilin, who left Poland to study
carbon monoxide. By measuring the effect first in Belgium, then in France, and emi-
of light of different wavelengths on revers- grated to England in 1915, was unaware
ing the inhibition, Warburg was able to of MacMunn’s work, although later when
determine the absorption spectrum of the he learned of it, he acknowledged it. In
pigment. The absorption spectrum, which 1925 Keilin, using a spectromicroscope,
was that of heme, supported his conclusion rediscovered the pigments in the thoracic
that Atmungsfermentt was a heme pigment. muscles of insects, in Bacillus subtilis, and
(Warburg’s Atmungsferment turned out in yeast, and named them cytochromes
to be cytochrome oxidase, later studied (cellular pigments). Keilin discovered cyto-
by David Keilin.) For his discovery of the chromes a, b, and c, and later cytochrome
respiratory enzyme and its mode of action, oxidase. Keilin published several papers
Warburg was awarded the Nobel Prize in in the late 1920s and early 1930s describ-
Physiology or Medicine in 1931. ing the respiratory chain as a chain of car-
Research on respiratory enzymes riers consisting of dehydrogenases that
had actually begun in the late 1800s. remove hydrogen from organic substrates,
Cytochromes were first described by the as well as oxidized cytochromes that are
Englishman Charles Alexander MacMunn reduced by the dehydrogenases, and cyto-
in three papers published between 1884 and chrome oxidase, an autoxidizable heme
www.ebook777.com
compound that oxidizes the cytochromes For more information about the history
and reduces oxygen. In the early 1930s it of biochemical research, see ref. 2.
was recognized that the respiratory chain
transfers electrons, rather than hydrogen, REFERENCES
from the organic substrate to oxygen, and
1. Quoted in Fruton, J. S., and S. Simmonds.
in 1931 Warburg correctly attributed the 1958. General Biochemistry, 2nd ed. John
oxidations and reductions to a change in Wiley & Sons, New York.
the valency of iron. The realization of the
2. Florkin, M. 1975. A history of biochemistry.
role of pyridine-linked dehydrogenases and In: Comprehensive Biochemistry, Vol. 31. M.
flavin-linked dehydrogenases was due to Florkin and E. H. Stotz (Eds.). Elsevier Scientific
the research of Warburg in the 1930s. Publishing, Amsterdam.
Fig. 5.7 Electron transport scheme in mitochondria. Electrons travel in the electron transport chain from a
low to a high electrode potential. Complexes I to IV are enclosed in dashed lines. Complex I is NADH dehydro-
genase, also called NADH–ubiquinone oxidoreductase. Complex II is succinate dehydrogenase, also called
succinate–ubiquinone oxidoreductase. Complex III is the bc1 complex, also called ubiquinol–cytochrome
c oxidoreductase. Complex IV is the cytochrome aa3 oxidase, also called cytochrome c oxidase. There are
several FeS clusters in complexes I and II, and an FeS protein in complex III. Complex II has both peripheral
and integral membrane protein subunits. The flavin (FAD) and FeS centers are in the peripheral membrane
subunits, and hemes (not shown) are in the integral membrane subunits. Electrons flow from FAD through
the FeS centers to quinone, probably via heme. Abbreviations: fp, flavoprotein; FeS, iron–sulfur protein; UQ,
ubiquinone; b, cytochrome b; c1, cytochrome c1; c, cytochrome c; aa3, cytochrome aa3.
Some of the dehydrogenase complexes are synthesis of some dehydrogenases, and stimu-
NADH and succinate dehydrogenase com- lates the synthesis of others. The electron carrier
plexes, analogous to complexes I and II in mito- complexes in bacteria are sometimes referred to
chondria. In addition to these dehydrogenases, as modules, since they can be synthesized and
several others reflect the diversity of substrates “plugged into” the respiratory chain when
oxidized by the bacteria: H2 dehydrogenases needed.
(called hydrogenases), formate dehydrogenase,
lactate dehydrogenase, methanol dehydroge- 5.4.1 The different terminal oxidases
nase, methylamine dehydrogenase, and so on. A word should be said about the many different
Depending upon the source of electrons and terminal oxidases found in bacteria.14 Whereas
electron acceptors, bacteria can synthesize and mitochondria all have the same cytochrome c
substitute one dehydrogenase complex for oxidase (cytochrome aa3), bacteria have a vari-
another, or reductase complexes for oxidase ety of terminal oxidases, often two or three dif-
complexes. For example, when E. coli is grown ferent ones in the same bacterium (i.e., two or
anaerobically, it makes the reductase complexes three branches to oxygen). Some of the bacte-
instead of the oxidase complexes, represses the rial terminal oxidases oxidize quinols (quinol
Fig. 5.8 Generalized electron transport pathways found in bacteria. The details will vary depending upon the
bacterium and the growth conditions. (A) Aerobic respiration. A dehydrogenase complex removes electrons
from an electron donor and transfers these to a quinone. The electrons are transferred to an oxidase complex
via a branched pathway. Many bacteria have bc1, cytochrome c, and cytochrome aa3 in one of the branches,
and in this way they resemble mitochondria. Other bacteria do not have a bc1 complex and may or may not
have cytochrome aa3. (B) Anaerobic respiration: Y represents either an inorganic electron acceptor other than
oxygen (e.g., nitrate) or an organic electron acceptor (e.g., fumarate). Under anaerobic conditions the elec-
trons are transferred to reductase complexes, which are synthesized anaerobically. Several reductases exist,
each one specific for the electron acceptor. More than one reductase can simultaneously exist in a bacterium.
oxidases) and some oxidize cytochrome c (cyto- bo3 oxidase (cytochrome o oxidase), and cyto-
chrome c oxidases). The terminal oxidases dif- chrome bb3 oxidase. It should be emphasized
fer in their affinities for oxygen and in whether that some of these are quinol oxidases and some
they are proton pumps, as well as in the types are cytochrome c oxidases.
of hemes and metals they contain. However,
despite these differences, most belong to the 5.4.2 Bacterial electron transport
heme–copper oxidase superfamily of oxidases, chains are branched
to which the mitochondrial cytochrome c oxi- Two major difference between mitochondrial
dase also belongs. An exception is the cyto- and bacterial electron transport chains are as
chrome bd oxidase, discussed later, which is follows: (1) the routes to oxygen in the bacte-
not a member of the heme–copper oxidase ria are branched, the branch point being at the
superfamily. quinone or cytochrome, and (2) many bac-
All members of the heme–copper oxidase teria can alter their electron transport chains
superfamily share a protein subunit that is depending upon growth conditions (Fig. 5.8).
homologous to subunit I of the mitochon- As noted in Section 5.4.1, under aerobic con-
drial cytochrome c oxidase. This subunit has ditions there are often two or three branches
a bimetallic (binuclear) center that binds oxy- leading to different terminal oxidases. For
gen and reduces it to water, and pumps pro- example, a two-branched electron transport
tons. (Several of the heme–copper oxidases are chain might contain a branch leading to cyto-
known to be proton pumps, whereas it is not yet chrome o oxidase (quinol oxidase) and a branch
known whether certain others pump protons.) leading to cytochrome aa3 oxidase (cytochrome
The bimetallic center contains a heme iron and c oxidase). Other bacteria (e.g., E. coli) have
copper. There is a second heme in all these oxi- cytochrome bo and bd oxidase branches (both
dases, but it is not part of the bimetallic center are quinol oxidases) but lack the cytochrome
and probably functions in transferring electrons aa3 branch. The ability to synthesize branched
to the bimetallic center. Heme–copper oxidases electron transport pathways to oxygen confers
that are mentioned later in the context of bac- flexibility on the bacteria, since not only may
terial respiratory systems include the cbb3-type the branches differ in the Δp that can be gener-
oxidases, cytochrome aa3 oxidase, cytochrome ated (because they may differ in the number of
www.ebook777.com
coupling sites) but their terminal oxidases also anaerobes will ferment when a terminal electron
may differ with respect to affinities for oxygen. acceptor is unavailable. Fermentation is the sub-
For example, in E. coli cytochrome bo has a low ject of Chapter 15.)
affinity for oxygen, whereas cytochrome bd has
a higher affinity for oxygen. Switching to an
oxidase with a higher affinity for oxygen allows
5.5 Coupling Sites
the cells to continue to respire even when oxy- For a review, see ref. 17. Sites in the electron
gen tensions fall to very low values. This abil- transport pathway at which redox reactions
ity is important to ensure the reoxidation of the are coupled to proton extrusion creating a ∆p
reduced quinones and NADH so that cellular are called coupling sites.18 Each coupling site
oxidations such as the oxidation of glucose to is also a site for ATP synthesis, since the pro-
pyruvate or the oxidation of pyruvate to CO2 tons extruded reenter via ATP synthase to make
may continue. ATP. The three coupling sites in mitochondria,
It has also been suggested that under sites 1, 2, and 3, were shown in Fig. 5.7, where
microaerophilic conditions the use of an oxi- site 1 is the NADH dehydrogenase complex
dase with a high affinity for oxygen would (complex I), site 2 is the bc1 complex (complex
remove traces of oxygen that might damage III), and site 3 is the cytochrome aa3 complex
oxygen-sensitive enzymes that are made under (complex IV). The succinate dehydrogenase
microaerophilic or anaerobic growth condi- complex (complex II) is not a coupling site.
tions.15 As an example of a protective role that The ratio of protons translocated for every
an oxidase might play, consider the situation two electrons varies, depending upon the com-
with the strict aerobe Azotobacter vinelan- plex. A consensus value is 10 protons extruded
dii. This organism has a branched respiratory for every two electrons that travel from NADH
chain leading to cytochromes bo and bd as the to oxygen. The bc1 complex translocates four
terminal oxidases. It also fixes nitrogen aerobi- protons for every two electrons; depending
cally. However, as with other nitrogenases, the upon the reported value, complexes I and IV
Azotobacter nitrogenase is inactivated by oxy- translocate 2 to 4 protons for every two elec-
gen. Azotobacter employs two mechanisms to trons. (The consensus value for mitochondrial
protect its nitrogenase, respiratory and confor- complex I is 4H+/2e–.) During reversed electron
mational. (See Section 13.3.2 for a discussion of flow, protons enter the cell through coupling
this point.) It is thought that the rapid consump- sites 1 and 2, driven by the ∆p, and the electrons
tion of oxygen by a terminal oxidase maintains are driven toward the lower redox potential.
the intracellular oxygen levels low enough to This creates a positive ∆E at the expense of the
prevent the inactivation of the nitrogenase. In ∆p. (See eq. 4.17.) Coupling site 3 is not physi-
agreement with this suggestion is the report that ologically reversible. Thus, water cannot serve
mutants of A. vinelandii that are deficient in the as a source of electrons for NAD+ reduction
cytochrome bd complex failed to fix nitrogen in by means of reversed electron flow. However,
air, although they did fix nitrogen when the oxy- during oxygenic photosynthesis, light energy
gen tension was sufficiently reduced.16 (Mutants can drive electrons from water to NADP+.
in cytochrome bo can fix nitrogen in air.) The mechanism of photoreduction of NADP+,
The adaptability of the bacteria with respect to which is different from reversed electron flow,
their electron transport chains can also be seen in is discussed in Chapter 6.
many bacteria that can respire either aerobically
or anaerobically. Under anaerobic conditions 5.5.1 The identification of coupling sites
they do not make the oxidase complexes but For an understanding of the physiology of
instead synthesize reductases. For example, energy metabolism during electron transport,
during anaerobic growth, E. coli synthesizes it is necessary to study the mechanism of pro-
fumarate reductase, nitrate reductase, and ton translocation, and for this the coupling
trimethylamine-N-oxide (TMAO) reductase. sites must be identified and isolated. The cou-
The different reductases enable the bacte- pling sites can be identified by the use of elec-
ria to utilize alternative electron acceptors tron donors that feed electrons into the chain
under anaerobic conditions. (Some facultative at different places, followed by measuring the
amount of ATP made for every two electrons must occur between ubiquinone and oxygen.
transferred through the respiratory chain. The When electrons enter the respiratory chain
number of ATPs made for every 2e– transfer to after the bc1 complex, the P/O ratio is reduced
oxygen is called the P/O ratio. It is equal to the to 1, indicating that the bc1 complex is the sec-
number of ATP molecules formed per atom of ond coupling site and that site 3 is cytochrome
oxygen taken up. When an electron acceptor aa3 oxidase. Site 3 can be demonstrated by
other than oxygen is used, P/2e– is substituted bypassing the bc1 complex. The bc1 complex
for P/O. The P/O ratio is approximately equal can be bypassed via an artificial electron donor
to the number of coupling sites. (However, see that will reduce cytochrome c [e.g., ascorbate
Section 5.5.2.) and tetramethylphenylenediamine (TMPD)],
In mitochondria, the oxidation of NADH thus channeling electrons from ascorbate to
results in a P/O ratio of about 3, indicating that TMPD to cytochrome c to cytochrome aa3.
three coupling sites exist between NADH and Alternatively, one can simply use reduced cyto-
O2. The use of succinate as an electron donor chrome c as an electron donor to directly reduce
results in a P/O ratio of approximately 2. Since cytochrome aa3.
electrons from succinate enter at the ubiqui- Each coupling site is characterized by a drop
none level, this indicates that coupling site 1 in midpoint potential of about 200 mV, which
occurs between NADH and ubiquinone [i.e., is sufficient for generating the ∆p (Fig. 5.9). The
the NADH–ubiquinone oxidoreductase reac- size of the ∆p that can be generated with respect
tion (Fig. 5.7)]. The other two coupling sites to the ∆E was discussed in Section 4.7.
Fig. 5.9 Average midpoint potentials, E′m, for components of the mitochondrial respiratory chain. The com-
plexes are in dashed boxes. The actual potentials (Eh) for most of the components are not very different from
the midpoint potentials. An exception is cytochrome aa3, whose Eh is much more positive than its midpoint
potential. There are changes in potential of 200 mV or more at three sites, which drive proton translocation.
One site is within complex I, a second within complex III, and the third between complex IV and oxygen.
Source: Adapted from Nicholls, D. G., and S. J. Ferguson. 1992. Bioenergetics 2. Academic Press, London.
www.ebook777.com
5.5.2 The actual number of ATPs that can in mitochondria is 6. Therefore, the maximal
be made for every two electrons traveling P/O ratio for this segment of the electron trans-
through the coupling sites port chain may be 6/4, or 1.5. The P/O ratios
By definition, the P/O ratio is the number of in bacteria can be higher, since a proton is not
moles of ATP made at a coupling site per two required to bring Pi into the cell. One significant
electrons that pass through the coupling site. aspect of branched aerobic respiratory chains
It is called the P/2e− ratio when oxygen is not in bacteria is that the number of coupling sites,
the terminal electron acceptor. The P/2e− ratio and therefore H+/O, can differ in the branched
does not need to be a whole number, since the chains. Thus, the different branches are not
amount of ATP that can be made per coupling equally efficient in generating a ∆p or making
site is equal to the ratio of protons extruded at ATP. We will return to this point later.
the coupling site to protons that reenter via the Of course, an ATP can be made when three
ATP synthase (Fig. 5.10). The ratio of protons protons enter via the ATP synthase only if the
translocated for every two electrons traveling ∆p is sufficiently large. As an exercise, we can
through all the coupling sites to O2 is called ask how large the ∆p must be. Recall that y∆p
the H+/O ratio. It can be measured by admin- is the work that can be done in units of electron
istering a pulse of a known amount of oxygen volts when y protons traverse a proton potential
to an anaerobic suspension of mitochondria or of ∆p volts. If H+/ATP is 3, then the number of
bacteria and measuring the initial efflux of pro- electron volts made available by proton influx
tons with a pH electrode as the small amount of through the ATP synthase is 3∆p. How many
oxygen is used up. The experiment requires that electron volts is needed to synthesize an ATP?
valinomycin plus K+ or a permeant anion such as The free energy of formation of ATP at physi-
thiocyanate, SCN–, be in the medium to prevent ological concentrations of ATP, ADP, and Pi is
a ∆Ψ from developing. (See Section 4.2.2 for a the phosphorylation potential, ∆Gp, which is
discussion of this point.) The reported values approximately 45,000 to 50,000 J. (See note 20
for H+/O for NADH oxidation vary. However, for an explanation of the phosphorylation
there is consensus that the true value is proba- potential.) Dividing by the Faraday constant
bly around 10. The ratio of protons entering via (96,500 C/mol) expresses the energy required
the ATP synthase to ATP made is called the H+/ to synthesize an ATP in electron volts. For
ATP ratio. It can be measured by using inverted 45,000 J this is 0.466 eV. Therefore, 3∆p must
submitochondrial particles prepared by sonic be greater than or equal to 0.466 eV. Thus, the
oscillation. These particles have the ATP syn- minimum ∆p is 0.466/3 or –0.155 V. Values of
thase on the outside and will pump protons ∆p that approximate this are easily generated
into the interior upon addition of ATP. Similar during electron transport. (However, see the
inverted vesicles can be made from bacteria by discussion in Section 4.10 regarding the low ∆p
first enzymatically weakening or removing the in alkaliphiles.)
cell wall and breaking the spheroplasts or pro-
toplasts by passage through a French press at
high pressures. (See note 19 for how to make
right-side-out vesicles.) Values of H+/ATP from
2 to 4 have been reported, and a consensus value
of 3 can be used for calculations. For intact
mitochondria, an additional H+ is required to
bring Pi electroneutrally from the cytosol into
the mitochondrial matrix in symport with H+,
so H+/ATP would be 4. A value of 10 for H+/O
predicts a maximum P/O ratio of 2.5 (10/4) for
mitochondria. This means that the often-stated
value of 3 for a P/O ratio for NADH oxidation
by mitochondria may be too high. Fig. 5.10 The ratio of protons extruded to protons
The number of protons ejected for every translocated through the ATPase determines the
2e– traveling between succinate and oxygen amount of ATP made.
5.6 How a Proton Potential Might Be and quinones) and those that carry only elec-
Created at the Coupling Sites: Q Loops, trons (iron–sulfur proteins and cytochromes).
This is illustrated in Fig. 5.11. The electron
Q Cycles, and Proton Pumps
carriers and their sequence are flavoprotein
The preceding discussion points out that pro- (H carrier), FeS protein (e– carrier), quinone
ton translocation takes place at coupling sites (H carrier), and cytochromes (e– carriers). The
when electrons travel “downhill” over a poten- flavoprotein and FeS protein comprise the
tial gradient of at least 200 mV. (See Fig. 5.9.) NADH dehydrogenase, which can be a cou-
However, the mechanism by which the redox pling site (Section 5.6.3). When electrons are
reaction is actually coupled to proton transloca- transferred from the FeS protein to quinone
tion was not explained. This coupling is thought (Q) on the inner side of the membrane, two
to occur (1) in a Q loop or Q cycle, where H+ is protons are acquired from the cytoplasm.
picked up on the cytosolic side of the membrane According to the model, the reduced quinone
and carried as part of UQH2 and released as (QH2), called quinol, then diffuses to the outer
H+ on the outside surface of the opposite side membrane surface and becomes oxidized,
of the membrane, and (2) by means of a proton releasing the two protons. The electrons then
pump where H+ actually is pumped across the return via cytochromes to the inner membrane
membrane. Even when protons (H+) per se are surface, where they reduce oxygen. This would
not translocated as charged molecules, but are create a ∆Ψ because the protons are left on the
translocated at part of UQH2 during the Q loop outer surface of the membrane as the electrons
or Q cycle, a membrane potential is created, as move electrogenically across the membrane to
will be explained next. the inner surface. Thus, quinol oxidation is a
In the Q loop or Q cycle, quinones are reduced second coupling site.
on the inner membrane surface and carry hydro- As mentioned earlier, the energy to create the
gen across the membrane and becomes oxidized, membrane potential is derived from the ∆Eh
releasing protons on the external face of the between the oxidant and the reductant. One
membrane. Then, the electrons return electro- way to view this is that the energy from the ∆Eh
genically via electron transport carriers to the “pushes” the electron to the negative membrane
inner membrane surface creating a membrane potential on the inside surface. Note that the role
potential. Proton translocation in the Q loop or of the quinone is to ferry the hydrogens across
Q cycle is referred to as scalar translocation. the membrane, by diffusing from a reduction
Proton pumps are electron carrier proteins site on the cytoplasmic side of the membrane to
that couple electron transfer to the electrogenic an oxidation site on the outer side. For the pur-
translocation of protons through the membrane pose of calculating the expected ∆p generated at
creating a membrane potential. Such trans- a coupling site, it makes no difference whether
location of protons through the membrane is one postulates the transmembrane movement
referred to as vectorial. of protons in one direction or electrons in the
Although the two mechanisms, Q loop or opposite direction, since both carry the same
Q cycle and proton pump, are fundamentally charge (eq. 4.17).
different, the result is the same: the net trans-
location of protons across the membrane with 5.6.2 The Q cycle
the establishment of a ∆p. The ∆p is the same
For a review, see ref. 22. Although the linear Q
regardless of whether the moving charges are
loop as just described for the oxidation of qui-
electrons or protons, since both carry the same
nol accurately describes quinol oxidation in E.
charge. The relationship between the ∆p and
coli and some other bacteria, it is inconsistent
the ∆Eh was given by eq. 4.17.
with experimental observations of electron
transport in mitochondria, chloroplasts, and
5.6.1 The Q loop many bacteria. For example, the Q loop pre-
For a review, see ref. 21. The essential fea- dicts that the ratio of H+ released per reduced
ture of the Q loop model is that the electron ubiquinone (UQH2 ) oxidized is 2, whereas
carriers alternate between those that carry the measured ratio in mitochondria and many
both hydrogen and electrons (flavoproteins bacteria is actually 4. To account for the extra
www.ebook777.com
Fig. 5.11 Proton translocation showing a Q loop and a proton pump. It is proposed that the electron carriers
exist in an alternating sequence of the following hydrogen [H] and electron [e–] carriers: flavoprotein (fp),
iron–sulfur protein (FeS), quinone (Q), and cytochromes. Oxidation of the flavoprotein deposits protons
on the outer membrane surface. Bacteria contain two types of NADH dehydrogenase. One of these, called
NDH-1, is a proton pump and, as in mitochondria, deposits four protons on the outer membrane surface. See
Section 5.6.3 for a further discussion. The electrons return to the inner membrane surface, where a quinone
is reduced, taking up two protons from the cytoplasm. The reduced quinone diffuses to the outer surface of
the membrane, where it is oxidized, depositing two more protons on the surface. The electrons return to the
cytoplasmic surface via cytochromes, where they reduce oxygen in a reaction that consumes protons. Some
cytochrome oxidases function as proton pumps. During anaerobic respiration the cytochrome oxidase is
replaced by a reductase, and the electrons reduce some other electron acceptor (e.g., nitrate or fumarate). It
should be noted that electrons can also enter at the level of quinone (e.g., from succinate dehydrogenase).
two protons, Peter Mitchell suggested a new anion UQ– during step 1 (Fig. 5.12 stage I), and
pathway for the oxidation of quinol called the the two protons are released to the membrane
Q cycle.23 The Q cycle operates in an enzyme surface. One electron travels to the FeS protein,
complex called the bc1 complex (complex III). and from there to cytochrome c1 on its way to
The bc1 complex from bacteria contains three the terminal electron acceptor. Thus, the ratio
polypeptides. These are cytochrome b with two of protons released at the membrane surface
b-type hemes, an iron–sulfur protein contain- to electrons traveling to the terminal electron
ing a single 2Fe–2S cluster (the Rieske protein), acceptor is 2:1 (stage 1, step 1, Fig. 5.12). The
and cytochrome c1 with one heme. The complex second electron is removed during step 2 and
spans the membrane and has a site for binding transferred via steps 2 and 3 across the mem-
reduced ubiquinone, UQH2, on the outer sur- brane to the N surface (inside surface) via
face of the membrane called site P (for posi- cytochromes bL and bH. This actually creates
tive), and a second site on the inner surface for a membrane potential, inside surface nega-
binding UQ, site N (for negative) (Fig. 5.12). tive, because it is a moving negative charge.
At site P, UQH2 is oxidized to the semiquinone The result is that UQ− becomes UQ, which
+
A 2H
OUT +
Fe/S c1 terminal
5
electron
1e
acceptor
UQH2 UQ UQ
1 2
P site
1e
bL
UQH2 UQ
pool 3 pool
UQH2
from bH
respiration
UQ UQ
4
N site
IN
+
B 2H
OUT +
Fe/S c1 terminal
5
electron
1e
acceptor
UQH2 UQ UQ
1 2
P site
1e
bL
UQH2
pool 3
UQH2
from bH
respiration
6
UQH2 UQ UQ
4
N site
IN
+
2H
Fig. 5.12 The bc1 complex. The bc1 complex isolated from bacteria contains three polypeptides, which
are a cytochrome b containing two heme b groups, bH and bL, an iron–sulfur protein, Fe/S (the Rieske pro-
tein), and cytochrome c1. It is widely distributed among the bacteria, including the photosynthetic bacteria.
(Mitochondrial bc1 complexes are similar but contain an additional 6–8 polypeptides without prosthetic
groups.) The iron–sulfur protein and cytochrome c1 are thought to be located on the outside (positive, or P)
surface of the membrane, and the cytochromes bH and bL are believed to span the membrane, acting as an elec-
tron conductor. On the outer surface there is a binding site in the bc1 complex for ubiquinol, UQH2 (the P site).
On the inner surface (negative, or N) there is a binding site in the bc1 complex for UQ and UQ− (the N site).
UQ binds to the P site and one electron is removed, forming the semiquinone anion, UQ−, (reaction 1). At this
time two protons are released on the outer membrane surface. The electron that is removed is transferred to
the iron–sulfur protein and from there to cytochrome c1 (step 5). (Cytochrome c1 transfers the electron to the
terminal electron acceptor via a series of electron carriers in other reactions.) In reaction 2 the second electron
is removed from UQ−, producing the fully oxidized quinone, UQ. The second electron is transferred trans-
membrane via cytochrome b to UQ at site N (steps 3 and 4) on the opposite side of the membrane. Because
the electron travels transmembrane, a membrane potential is created, outside positive (P). The result is that
UQH2 is oxidized to UQ, which enters the UQ pool. The second electron that is transferred transmembrane
via cytochrome b generates a membrane potential, and also reduces a UQ from the UQ pool to UQ−. A second
UQH2 is oxidized at the P site, releasing two more protons, and the electron that is transferred transmem-
brane reduces UQ– at the N site to UQH2. The two protons are acquired from the cytoplasmic surface of the
cell membrane. The net result is that for every electron that travels to the terminal electron acceptor at the P
site, two protons are released to the outer membrane surface so the ratio of protons released to electrons that
travel to the terminal electron acceptor is 2:1 as opposed to 1:1 when UQH2 is oxidized via the Q loop shown
in Fig. 5.11 (site 2).
www.ebook777.com
enters the UQ pool (stage I). In step 4 (stage I) 5.6.3 Pumps

UQ molecules in the UQ pool can accept elec- Proton pumps also exist. These catalyze the elec-
trons that travel across the membrane via cyto- trogenic translocation of protons (H+) across the
chromes b (step 4) and become reduced to UQ−. membrane rather than electrons (Fig. 5.11). For
Now focus on stage II. Each UQ− can accept an example, proton extrusion accompanies the
electron (step 4) on the N side of the membrane cytochrome aa3 oxidase reaction when cyto-
and pick up 2H+ from the cytosol (step 6) to chrome c is oxidized by oxygen in mitochon-
become UQH2, which can then be oxidized in dria and some bacteria. This can be observed
stage 1 (step 1). The net result is that for every by feeding electrons into the respiratory chain
2H+ and 2 electrons removed from UQH2, one at the level of cytochrome c, thus bypassing
electron is recycled across the membrane so the quinone. The experimental procedure is to
that the net result is 2H+ released per electron incubate the cells in lightly buffered anaerobic
that travels through the respiratory chain to the media with a reductant for cytochrome c and a
terminal electron acceptor. But, as mentioned, permeant anion (e.g., SCN– or valinomycin plus
there is also another very important result of K+). Changes in pH are measured with a pH elec-
this pathway. Electron flow from the QP site trode. Upon addition of a pulse of oxygen, given
to the QN site is transmembrane and creates a as an air-saturated salt solution, a sharp, tran-
membrane potential, which contributes most sient acidification of the medium occurs, and the
of the energy to the ∆p at that site. The situ- ratio of H+ to O can be calculated. If the electron
ation with respect to released protons can be donor itself releases protons, these scalar pro-
summarized as follows: tons are subtracted from the total protons trans-
located to determine the number of vectorially
P site: 2UQH2 + 2cyt cox translocated protons (due to proton pumping by
→ 2UQ + 2cyt cred + 4H+out + 2e− the cytochrome c oxidase). For example, when
cytochrome c is reduced by means of ascor-
N site: UQ + 2e− + 2H+in → UQH2 bate plus TMPD, the ascorbate is oxidized to
Sum: UQH2 + 2cyt cox + 2H+in dehydroascorbate with the release of one scalar
proton for every two electrons. Any additional
→ UQ + 2cyt cred + 4H+out protons released are due to proton pumping by
the cytochrome c oxidase. When similar experi-
Since the Q cycle translocates four protons for ments were done with Paracoccus denitrificans,
every two electrons that flow to the terminal it was found that the P. denitrificans cytochrome
electron acceptor, it generates a larger pro- aa3 oxidase translocates protons with a stoichi-
ton current than the Q loop that translocates ometry of 1H+/1e–.24,25
only one proton per electron. This can result Cytochrome oxidase pumping activity can
in more ATP synthesis. Consider the situation also be demonstrated in proteoliposomes made
of an ATP synthase that requires the influx of with purified cytochrome aa3. Because the pro-
three protons to make one ATP. If the trans- ton pumps move a positive charge across the
fer of a mole of electrons through the electron membrane, leaving behind a negative charge, a
transport pathway resulted in the transloca- membrane potential, outside positive, develops.
tion of one mole of protons, then a third of The membrane potential should be the same as
a mole of ATP could be made for each mole that recorded when an electron moves inward,
of electrons if the protons reentered via the since the proton and the electron carry the same
ATP synthase. On the other hand, if a mole charge (i.e., 1.6 × 10–19 C). The mechanism of
of electrons transported resulted in the trans- pumping probably requires conformational
location of two moles of protons, then two- changes in cytochrome oxidase resulting from
thirds of a mole of ATP could be made when its redox activity. Conformational changes
the protons reentered via the ATP synthase. probably also occur in bacteriorhodopsin dur-
In other words, the size of the proton cur- ing proton pumping (Section 4.8.4).
rent generated by respiration determines the The mitochondrial NADH dehydrogenase
upper value of the amount of ATP that can complex (NADH:ubiquinol oxidoreductase)
be made. translocates four protons for every NADH
oxidized.26 Two types of NADH:ubiquinol oxi- not complete, but it conveys the diversity of
doreductases exist in bacteria.8–11 One of these, electron transport systems found in bacteria.
called NDH-1, is similar to the mitochondrial The electron transport chains for the respira-
complex I in that it is a multisubunit enzyme tory metabolism of ammonia, nitrite, inorganic
complex (approximately 14 polypeptide sub- sulfur, and iron are described in Chapter 15.
units) consisting of FMN and FeS clusters, and
it translocates protons across the membrane 5.7.1 Escherichia coli
during NADH oxidation (4H+/2e–). The mech- E. coli is a gram-negative heterotrophic faculta-
anism of proton translocation is not known, tive anaerobe. It can be grown aerobically by
but it is referred to as a proton pump. [The using oxygen as an electron acceptor, anaerobi-
marine bacterium Vibrio alginolyticus has a cally (e.g., by using nitrate or fumarate as the
sodium-translocating NDH (Na–NADH). See electron acceptor), or anaerobically via fer-
Section 4.7.1.] mentation (Chapter 15). The bacteria adapt to
A second NADH:ubiquinol oxidoreductase, their surroundings, and the electron transport
called NDH-2, may also be present. NDH-2 system that the cells assemble reflects the elec-
differs from NDH-1 in consisting of a single tron acceptor that is available. (See ref. 29 for
polypeptide and FAD, and in that it is not an a review.)
energy-coupling site. In E. coli, NDH-1 and All the electron transport chains present in
NDH-2 are simultaneously present.27 How E. coli branch at the level of quinone, which
E. coli regulates the partitioning of electrons connects the different dehydrogenases with
between the two NAD dehydrogenases clearly the various terminal reductases and oxidases
has important energetic consequences. that are present under different growth condi-
tions. E. coli makes three quinones, ubiquinone
(UQ), menaquinone (MQ), and demethylme-
5.7 Patterns of Electron Flow in naquinone (DMQ), and their relative amounts
Individual Bacterial Species depend upon the nature of the electron acceptor.
Although the major principles of electron trans- When UQ is growing aerobically, it accounts
port as outlined already apply to bacteria in for 60% of the total quinone, DMQ is 37% of
general, several different patterns of electron the total, and MQ is only about 3%. However,
flow exist in particular bacteria, often within very little UQ is made anaerobically. For anaer-
the same bacterium grown under different con- obic growth on nitrate, the major quinones
ditions.28 The patterns of electron flow reflect are DMQ (70%) and MQ (30%). The major
the different sources of electrons and electron quinone grown anaerobically on fumarate or
acceptors that are used by the bacteria. For DMSO is MQ (74%), with DMQ (16%) and
example, bacteria may synthesize two or three UQ (10%) contributing the rest.28 There is no
different oxidases in the presence of air and bc1 complex and no cytochrome c, which may
several reductases anaerobically. In addition, serve as additional branch points in other bacte-
certain dehydrogenases are made only anaero- rial respiratory chains (e.g., see the discussion of
bically because they are part of an anaerobic Paracoccus denitrificans in Section 5.7.2).
respiratory chain. Furthermore, whereas elec-
tron donors such as NADH and FADH2 gen- Aerobic respiratory chains
erated in the cytoplasm are oxidized inside the When grown aerobically, E. coli makes two dif-
cell, there are many instances of oxidations ferent quinol oxidase complexes, cytochrome
in the periplasm in gram-negative bacteria. bo complex (has heme b and heme o and is also
Substances that are oxidized in the periplasm called bo3 or cytochrome o) and cytochrome
include hydrogen gas, methane, methanol, bd complex (has heme b and heme d and is also
methylamine, formate, perhaps ferrous ion, called cytochrome d), resulting in a branched
reduced inorganic sulfur, and elemental sulfur. respiratory chain to oxygen (Fig. 5.13).30–34 In
In most of these instances periplasmic cyto- these pathways electrons flow from ubiquinol
chromes c accept the electrons from the electron to the terminal oxidase complex, which is why
donor and transfer them to electron carriers in the oxidases are called (ubi)quinol oxidases.
the membrane. The discussion that follows is The cytochrome bo complex from E. coli is a
www.ebook777.com
Fig. 5.13 Aerobic respiratory chain in E. coli. The chain branches at the level of ubiquinone (UQ) to two
alternate quinol oxidases, cytochrome bo and bd. Cytochrome bd complex has a higher affinity for oxygen
and is synthesized under low oxygen tensions, where it becomes the major route to oxygen. Proton transloca-
tion occurs during NADH oxidation, catalyzed by NADH:ubiquinone oxidoreductase (also called NDH-1),
a rotenone-sensitive enzyme complex, and quinol oxidation, catalyzed by cytochrome bo complex or cyto-
chrome bd complex. Additionally, cytochrome bo is a proton pump. E. coli has a second NADH dehydro-
genase, called NADH-2, which does not translocate protons and is not sensitive to rotenone. Therefore, the
number of protons translocated per NADH oxidized can vary from two (NADH-2 and bd complex) to eight
(NADH-1 and bo complex).
proton pump, with a stoichiometry of 1H+/e–. of transcription of the NADH dehydrogenase

Cytochrome bo complex is the predominant genes in E. coli is complex.34,36
oxidase when the oxygen levels are high. When
the oxygen tensions are lowered, E. coli makes Physiological significance of alternate
more cytochrome bd than cytochrome bo. electron routes that differ in the number
The cytochrome bd complex has a higher of coupling sites
affinity for oxygen (lower Km) than does the Since the NDH-1 dehydrogenase may trans-
cytochrome bo complex, which may be why it locate as many as four protons for every two
is the dominant oxidase under low oxygen ten- electrons, whereas the NDH-2 dehydrogenase
sions. Cytochrome bd is not a proton pump. In translocates none, the number of protons trans-
the same sense that the bc1 complex is a cou- located per NADH oxidized can theoretically
pling site because it catalyzes the oxidation of vary from 2 (NDH-2 and bd complex) to 8
quinol resulting in the scalar extrusion of pro- (NDH-1 and bo complex). Assuming a H+/ATP
tons, both the bo and bd complexes are cou- of 3 (i.e., the ATP synthase translocates inwardly
pling sites. The difference between the two, as 3 protons for every ATP made), the ATP yields
mentioned, is that the bo complex is also a proper NADH oxidized can vary fourfold from
ton pump and catalyzes the vectorial extrusion 2/3, or 0.67, to 8/3, or 2.7. It also means that
of protons. The oxidation of quinol by the bo E. coli has great latitude in adjusting the ∆p gen-
complex results in two protons translocated for erated during respiration. Since a large ∆p can
every electron (one scalar and one vectorial), drive reversed electron transport and thus slow
whereas the oxidation of quinol by the bd com- down oxidation of NADH and quinol, it may
plex results in only one proton translocated per be an advantage to be able to direct electrons
electron (scalar). along alternate routes that bypass coupling
E. coli has two NADH dehydrogenases, sites and translocate fewer protons. This could
NDH-1 and NDH-2, encoded by the nuo ensure adequate rates of reoxidation of NADH
operon and ndh, respectively. Only NDH-I is and quinol.
a coupling site (proton pump). It is a complex
enzyme consisting of 14 subunits and translo- Anaerobic respiratory chains
cates two protons per electron. As discussed In the absence of oxygen, E. coli can use either
in Section 5.6.3, NDH-1 is similar to com- nitrate or fumarate as an electron acceptor.
plex I of mitochondria. It has been shown that The nitrate is reduced to nitrite, which is fur-
NDH-1 is used during fumarate respiration, ther reduced to ammonia, and the fumarate is
whereas NDH-2 is used primarily during aer- reduced to succinate, all of which are excreted
obic and nitrate respiration.35 The regulation into the medium.37 (See note 38 for a further
Fig. 5.14 Anaerobic respiratory chains in E. coli. When oxygen is absent, E. coli synthesizes any one of sev-
eral membrane-bound reductase complexes depending upon the presence of the electron acceptors. Nitrate
induces the synthesis of nitrate reductase and represses the synthesis of the other reductases. Menaquinone
(E0′ = –74 mV) (or demethylmenaquinone, E0′ = –40 mV ) must be used to reduce some of the reductases
(e.g., fumarate reductase, because it has a sufficiently low midpoint potential). Ubiquinol (E0′ = +100 mV )
or menaquinone can reduce nitrate reductase because the E0′ of nitrate is 421 mV. Each reductase may be a
complex of several proteins and prosthetic groups through which the electrons travel to the terminal electron
acceptor. The transfer of electrons from the dehydrogenases to the reductases results in the establishment
of a proton potential. If the dehydrogenase has site 1 activity, there can theoretically be two coupling sites,
one at the dehydrogenase step and one linked to quinol oxidation at the reductase step. Abbreviations: cyt b,
cytochrome b; Fe/S, nonheme iron–sulfur protein; FAD, flavoprotein with flavin adenine dinucleotide as the
prosthetic group: Mo, molybdenum; TMANO, trimethylamine N-oxide; DMSO, dimethyl sulfoxide; MQ,
menaquinone; UQ, ubiquinone.
description of fumarate and nitrate reduction.) is zero, indicating that both the release of pro-
The anaerobic respiratory chains consists of a tons when MQH2 is oxidized and the uptake of
dehydrogenase, a reductase, and a diffusible protons when fumarate is reduced take place
quinone to mediate the transfer of electrons on the cytoplasmic side of the membrane. (It
between the dehydrogenase and the reductase does not appear that either fumarate reductase
(Fig. 5.14). or any of the nitrate reductases are proton
The number of coupling sites depends upon pumps.) Thus, for E. coli to make ATP during
whether the electron acceptor is nitrate or the oxidation of NADH by fumarate, it must
fumarate.28 When nitrate is the electron accep- use NDH-1 rather than NDH-2, since only the
tor, there can be two coupling sites: that is, one former is a coupling site. In agreement with this
at the dehydrogenase step if NDH-1 is used conclusion, it has been found that the genes for
(site 1) and one at the quinol oxidation step [i.e., NDH-1 (nuo genes) are essential for anaerobic
the nitrate reductase (scalar)]. During nitrate respiration with fumarate as the electron accep-
however, respiration, NDH-2, which is not a tor. (Reviewed in ref. 39.)
coupling site, is also used.
Quinol oxidation by nitrate results in a ∆p 5.7.2 Paracoccus denitrificans
because the oxidation of one quinol and release P. denitrificans is a nonfermenting gram-
of two protons takes place on the periplasmic negative facultative anaerobe that can obtain
side of the membrane, and the uptake of two energy from either aerobic respiration or
protons during the reduction of one nitrate to nitrate respiration.20,40,41 It is found primar-
nitrite takes place on the cytoplasmic side. Thus, ily in soil and sewage sludge. This bacterium
the oxidation of quinol by nitrate catalyzed by can grow heterotrophically on a wide variety
nitrate reductase yields a proton-to-electron of carbon sources, or autotrophically on H2
ratio of 1.28 See the discussion of the reaction and CO2 under anaerobic conditions using
for nitrate reductase in Section 5.7.2. nitrate as the electron acceptor. It can be iso-
However, the H+/e– ratio for quinol oxidation lated from soil by anaerobic enrichment with
by fumarate in E. coli or Wolinella succinogenes media containing H2 as the source of energy
www.ebook777.com
and electrons, Na2CO3 as the source of carbon, Bradyrhizobium.42 Cytochrome bb3 formerly
and nitrate as the electron acceptor. Electron was called cytochrome o or b0. But no heme o was
transport in P. denitrificans receives a great detected when the hemes were extracted from
deal of research attention because certain fea- membranes and analyzed by reversed-phase high
tures closely resemble electron transport in performance liquid chromatography24,43,44). The
mitochondria. cytochrome aa3 and cytochrome bb3 are proton
pumps. (Whether cytochrome cbb3 pumps pro-
Aerobic pathway tons apparently depends upon the assay con-
P. denitrificans differs from E. coli in that it ditions.39) The aerobic pathways as well as the
has a bc1 complex and a cytochrome aa3 oxi- sites of proton translocation are shown in Fig.
dase (cytochrome c oxidase), and in this way 5.15A. Electrons traveling from NADH to oxy-
resembles mitochondria. In addition to the gen can pass through as many as three coupling
cytochrome aa3, there are two other terminal sites (NDH-1, bc1 complex, cyt aa3 or perhaps
oxidases in the aerobic pathway.24,39 These are cyt cbb3) or as few as two coupling sites (NDH-1
a different cytochrome c oxidase (cytochrome and cyt bb3). Recall that the bc1 complex and
cbb3) and a ubiquinol oxidase (cytochrome cyt bb3 are coupling sites because they oxidize
bb3). (Cytochromes cbb3 are heme–copper quinol, whereas the NDH-1 and cytochrome c
oxidases found in several bacteria including oxidases are proton pumps. As described later,
Thiobacillus, Rhodobacter, Paracoccus, and P. denitrificans also oxidizes methanol, and in
Fig. 5.15 A model for electron transport pathways in Paracoccus denitrificans. (A) Aerobic. The pathway has
two branch points. One branch is at the level of ubiquinone, leading to one of two ubiquinol oxidases (i.e.,
the cyt bc1 complex or cyt bb3). Both these quinol oxidases are coupling sites. The bc1 complex extrudes two
protons per electron via the Q cycle. Whereas cyt bb3 is a proton pump and extrudes one proton per electron
vectorially, the second proton is extruded via a Q loop. A second branch point occurs at the level of the bc1
complex. Electrons can flow either to cyt aa3, which is a proton pump, or to cyt cbb3, which has been reported
to pump protons under certain experimental conditions. NDH-1 is also a coupling site. (B) Anaerobic. When
the bacteria are grown anaerobically, using nitrate as the electron acceptor, the cytochrome aa3 levels are very
low and the electrons travel from ubiquinone to nitrate reductase and also through the bc1 complex to nitrite
reductase, nitric oxide reductase, and nitrous oxide reductase.
this case the electrons enter at a cytochrome c, of the membrane and reduces nitrate on the
thus bypassing the bc1 site. cytoplasmic side. This would create a ∆Ψ as
two electrons from UQH2 flow electrogenically
across the membrane, leaving two protons on
Anaerobic pathway
the outside. In the cytoplasm, protons are taken
P. denitrificans can also grow anaerobically by up during nitrate reduction according to the fol-
using nitrate as an electron acceptor, reducing lowing reaction:
it to nitrogen gas in a process called denitrifica-
tion (Fig. 5.15B).36,45 During anaerobic growth 2H+ + 2e− + NO−3 → NO−2 + H2O
on nitrate, P. denitrificans has a complete citric
acid cycle in which electrons are donated to the The result is the net translocation of protons
electron transport chain; but the electron trans- to the outside, although only electrons moved
port chain is very different from that of aero- across the membrane, not protons. This is the
bic growth. The cells contain cytochrome bb3, same as the Q loop described in Section 5.6.1
nitrate reductase, nitrite reductase, nitric oxide except that the terminal electron acceptor is
reductase, and nitrous oxide reductase. (For a nitrate instead of oxygen. The electrons to the
more complete description of these enzymes, see other reductases flow from UQH2 through the
note 46.) The levels of cytochrome aa3 are very bc1 complex, and a ∆p is generated via the Q
low. The cytochrome c shown in Fig. 5.15 is cycle catalyzed by the bc1 complex which, as
periplasmic, although there is also cytochrome described in Section 5.6.1, is a modification of
c in the membrane associated with some of the the Q loop that results in the translocation of
electron carriers. The nitrate (NO3−) is reduced two protons per electron rather than one.
to nitrite (NO2−) in a two-electron transfer In P. denitrificans, Pseudomonas aeruginosa,
via a membrane-bound nitrate reductase. and many other facultative anaerobes, both the
The NO2− is reduced to nitric oxide (NO) in a synthesis and the activity of the denitrifying
one-electron transfer via a periplasmic nitrite enzymes are prevented by oxygen. However,
reductase. The NO is reduced to a half-mole in certain other facultative anaerobes, includ-
of nitrous acid (1/2N2O) in a one-electron step ing Comamonas spp., certain species of
via a membrane-bound nitric oxide reductase. Pseudomonas, Thiosphaera pantotropha, and
And the 1/2N2O is reduced to a half-mole of Alcaligenes faecalis, denitrifying enzymes are
dinitrogen (1/2N2) in a one-electron step by a made and are active in the presence of oxy-
periplasmic nitrous oxide reductase. Thus to gen.47 In these systems, both oxygen and nitrate
reduce one mole of NO3− to 1/2N2 a total of five are used simultaneously as electron acceptors,
electrons flow in and out of the cell membrane although aeration can significantly decrease the
through membranous electron and periplasmic rate of nitrate reduction. The advantage of co-
electron carriers from ubiquinol to the various respiration using both oxygen and nitrate is not
reductases. obvious.
As shown in Fig. 5.15, the electron trans-
port pathway includes several branches to the Periplasmic oxidation of methanol
individual reductases. The first branch site is at Many gram-negative bacteria oxidize sub-
UQ, where electrons can flow either to nitrate stances in the periplasm and transfer the elec-
reductase or to the bc1 complex. Then there are trons to membrane-bound electron carriers,
three branches to the three other reductases often via periplasmic cytochromes c. (See
after the bc1 complex at the level of cytochrome Section 1.2.4 for a description of the periplasm.)
c. In agreement with the model, electron flow to An example is P. denitrificans, which can grow
nitrate reductase is not sensitive to inhibitors of aerobically on methanol (CH3OH) by oxidiz-
the bc1 complex, whereas electron flow to the ing it to formaldehyde (HCHO) and 2H+ with
other reductases is sensitive to the inhibitors. a periplasmic dehydrogenase, methanol dehy-
The nitrate reductase spans the membrane, and drogenase (Fig. 5.16):
it has been proposed that it creates a ∆p via a Q
CH3OH → HCHO + 2H+
loop, similar to that described in Section 5.6.1.
It is proposed that the nitrate reductase accepts (Growth on methanol is autotrophic, since the
electrons from UQH2 on the periplasmic side formaldehyde is eventually oxidized to CO2,
www.ebook777.com
Periplasm cell membrane cytoplasm 5.7.3 Rhodobacter sphaeroides

Rhodobacter sphaeroides is a purple photo-
CH3OH HCHO + 2H + synthetic bacterium that can be grown photo-
heterotrophically under anaerobic conditions
methanol or aerobically in the dark. When growing aero-
dehydrogenase bically in the dark, the bacteria obtain energy
from aerobic respiration. The respiratory chain
2 x 1e-
resembles the mitochondrial and P. denitri-
cyt c551 1 O + 2H+ ficans respiratory chains in that it consists of
2 x 1e- 2 2 NADH:ubiquinone oxidoreductase (coupling
cyt c
H2O site 1), a bc1 complex (coupling site 2), and a
2H + 2H+ cytochrome aa3 complex (coupling site 3). As in
cyt aa3 E. coli, there are two NADH:ubiquinone oxi-
doreductases: NDH-1 and NDH-2.
Fig. 5.16 Oxidation of methanol by P. denitrificans. 5.7.4 Fumarate respiration in

Methanol is oxidized to formaldehyde by a periplas- Wolinella succinogenes
mic methanol dehydrogenase. The electrons are Fumarate respiration occurs in a wide range of
transferred to periplasmic cytochromes c and to a bacteria growing anaerobically.48 This is prob-
membrane-bound cytochrome aa3, which is also a ably because fumarate itself is formed from
proton pump. A ∆p is created as a result of the elec- carbohydrates and protein during growth.
trogenic influx of electrons and the electrogenic efflux We describe the electron transport pathway in
of protons, accompanied by the release of protons in W. succinogenes, a gram-negative anaerobe iso-
the periplasm (methanol oxidation) and uptake in lated from the rumen, by reference to Fig. 5.17.
the cytoplasm (oxygen reduction).
W. succinogens can grow at the expense of
H2 or formate, both produced in the rumen by
other bacteria. The electron transport pathway
which is assimilated via the Calvin cycle: see is shown in Fig. 5.17A. The active sites for both
Chapter 14.) The electrons are transferred the hydrogenase and the formate dehydroge-
from the dehydrogenase to c-type cytochromes nase are periplasmic, whereas the active site
in the periplasm. The cytochromes c transfer for the fumarate reductase is cytoplasmic. An
the electrons to cytochrome aa3 oxidase in the examination of the topology of the components
membrane, which reduces oxygen to water on of the respiratory chain reveals how a ∆p is
the cytoplasmic surface. A ∆p is established as generated.
a result of the inward flow of electrons and the
outward pumping of protons by the cytochrome Topology of the components of the
aa3 oxidase, as well as the release of protons in electron transport pathway
the periplasm during methanol oxidation and The electron transport chain consists of a
consumption in the cytoplasm during oxygen periplasmic enzyme that oxidizes the elec-
reduction. Since the oxidation of methanol tron donor, a membrane-bound menaquinone
bypasses the bc1 coupling site, the ATP yields (MQ) that serves as an intermediate elec-
are lower. In addition to methanol, the bacteria tron carrier, and membrane-bound fumarate
can grow on methylamine (CH3NH+3), which reductase, which accepts the electrons from
is oxidized by methylamine dehydrogenase to the menaquinone and reduces fumarate on the
formaldehyde, NH+4 , and 2H+: cytoplasmic side of the membrane (Fig. 5.17B).
Both the hydrogenase and the formate dehydro-
CH3NH+3 + H2O → HCHO + NH+4 + 2H+
genase are made of three polypeptide subunits,
Methylamine dehydrogenase is also located in two facing the periplasm and one an integral
the periplasm and donates electrons via cyto- membrane protein (cytochrome b). Note that
chromes c to cytochrome aa3, bypassing the bc1 cytochrome b not only serves as a conduit for
complex. electrons, but also binds the dehydrogenases
Fig. 5.17 A model for the electron transport system of Wolinella succinogenes. (A) Electrons flow from H2
and formate through menaquinone (MQ) to fumarate reductase. (B) Illustration that the catalytic portions
of hydrogenase and formate dehydrogenase are periplasmic, whereas fumarate reductase reduces fumarate
on the cytoplasmic side. Electrons flow electrogenically to fumarate. A ∆p is created because of electrogenic
influx of electrons together with the release of protons in the periplasm and their consumption in the cyto-
plasm. Source: Modified from Kroger, A., V. Geisler, E. Lemma, F. Theis, and R. Lenger. 1992. Bacterial
fumarate respiration. Microbiology 158:311–314.
into the membrane. The two periplasmic sub- Electron flow and the establishment of a Δp
units of the hydrogenase are a Ni-containing Electron flow is from the dehydrogenase or
protein subunit and an iron–sulfur protein. The hydrogenase to cytochrome b to menaquinone
two periplasmic subunits of the formate dehy- to fumarate reductase. Two electrons are elec-
drogenase are a Mo-containing protein subunit trogenically transferred across the membrane
and an iron–sulfur protein. to fumarate for every H2 or formate oxidized,
The fumarate reductase is a complex contain- leaving two protons on the outside, thus estab-
ing three subunits. One subunit of the fumar- lishing a ∆p. In a study of whole cells, a value
ate reductase is a flavoprotein with FAD as the of 1.1 was obtained for the ratio of protons
prosthetic group (subunit A). A second subunit to electrons during fumarate reduction, sug-
has several FeS centers (subunit B). And the gesting perhaps that a mechanism of proton
third subunit has two hemes of the b type (sub- translocation through the membrane may also
unit C), which binds the fumarate reductase to exist.48 If one assumes that 1.1 is a correct num-
the membrane. Fumarate reductase is similar in ber, and also assumes a stoichiometry for the
structure to succinate dehydrogenase isolated ATP synthase of 3H+/ATP, then the theoretical
from several different sources, which catalyzes maximum number of ATPs that can be formed
the oxidation of succinate to fumarate in the cit- from the transfer of two electrons to fumarate is
ric acid cycle. 2.2/3, or 0.73. The actual number measured by
www.ebook777.com
experimentation was 0.56. Note that the qui- used when oxygen levels are high, and to cyto-
none functions as an electron carrier between chrome d, which is used when oxygen becomes
the cytochromes b but does not take part in limiting. Other bacteria may have in addition,
hydrogen translocation across the membrane or instead, an electron transport pathway in
as in a Q loop or Q cycle. which electrons travel from reduced quinone
through a bc1 complex, to cytochrome aa3,
which reduces oxygen. This is the same as the
5.8 Summary electron transport pathway found in mitochon-
All electron transport schemes can be viewed as dria, although the carriers may not be identi-
consisting of membrane-bound dehydrogenase cal. The alternative branches may differ in the
complexes, such as NADH dehydrogenase or number of coupling sites, and this could have
succinate dehydrogenase, that remove electrons regulatory significance regarding the rates of
from their substrates and transfer the electrons oxidation of reduced electron carriers, as well
to quinones, which in turn transfer the electrons as ATP yields.
to oxidase or reductase complexes. The latter Another difference from mitochondria is that
complexes reduce the terminal electron accep- bacteria can have either aerobic or anaerobic
tors. In contrast to mitochondria, which all electron transport chains; or, as is the case with
have the same electron transport scheme, bac- facultative anaerobes such as E. coli, either can
teria differ in the details of their electron trans- be present, depending upon the availability
port pathways, although the broad outlines of of oxgyen or alternative electron acceptors. A
all such schemes are similar. In bacteria, the hierarchy of electron acceptors is used. For E.
dehydrogenase, oxidase, and reductase com- coli, oxygen is the preferred acceptor, followed
plexes are sometimes referred to as modules by nitrate, and finally fumarate.
because specific ones are synthesized under cer- With respect to cytochrome c oxidase, there
tain growth conditions and “plugged into” the are two classes. Cytochrome aa3 is the major class
respiratory pathway. For example, in faculta- and has been reported in many bacteria, includ-
tive anaerobes such as E. coli, the oxidase mod- ing Paracoccus denitrificans, Nitrosomonas
ules are synthesized in an aerobic atmosphere europaea, Pseudomonas AM1, Bacillus subti-
and the reductase modules under anaerobic lis, and Rhodobacter sphaeroides. A different
conditions. cytochrome c oxidase has been reported for
Other dehydrogenases besides NADH dehy- Azotobacter vinelandii, Rhodobacter capsulata,
drogenase and succinate dehydrogenase exist. R. sphaeroides, R. palustris, and Pseudomonas
These oxidize various electron donors (e.g., aeruginosa, as well as P. denitrificans.13,49
methanol, hydrogen, formate, H2, glycerol) Apparently, both classes of cytochrome c oxi-
and are located in the periplasm or the cyto- dase coexist in the same organism and serve as
plasm. The coenzyme or prosthetic groups for alternate routes to oxygen.
these soluble dehydrogenases vary (e.g., they The main energetic purpose of the respira-
may be NAD+ or flavin). The electrons from tory electron transport pathways is to convert
the various dehydrogenases are transferred to a redox potential (∆Eh) into a proton potential
one of the electron carriers (e.g., quinone, cyto- (∆p). This is done at coupling sites. A mem-
chrome) and from there to a terminal reductase brane potential is created by electrogenic influx
or oxidase. of electrons, leaving the positively charged
An important difference between elec- proton on the outside, or during electrogenic
tron transport chains in bacteria and those in efflux of protons during proton pumping, leav-
mitochondria is that the former are branched. ing a negative charge on the inside. Influx of
Branching can occur at the level of quinone or electrons occurs when oxidations take place
cytochrome. The branches lead to different oxi- on the periplasmic membrane surface or in
dases or reductases, depending upon whether the periplasm, and electrons move vectorially
the bacterium is growing aerobically or anaer- across the membrane to the cytoplasmic sur-
obically. Many bacteria, including E. coli, face, where reductions take place. This occurs
transfer electrons from reduced quinone to in two situations: (1) when the substrate (e.g.,
cytochrome o, which is the major cytochrome H2, methanol) is oxidized by dehydrogenases
in the periplasm and the electrons move across 3. What are two features that distinguish
the membrane to reduce the electron acceptor the Q cycle from linear flow of electrons?
and (2) when quinones are reduced on the cyto- How can you verify that the Q cycle is
plasmic side of the membrane, diffuse across operating?
the membrane, and are oxidized on the outside
4. What is the relationship between H+/O, H+/
membrane surface.
ATP, and the number of ATPs that can be
Quinone oxidation can be via a Q loop or a Q
synthesized?
cycle. A Q loop is an electron transport pathway
in which a reduced quinone carries hydrogen to 5. What is the relationship between the ∆p and
the outside surface of the cell membrane and number of ATPs that can be synthesized?
releases two protons as a result of oxidation.
6. Draw a schematic outline of three ways in
The two electrons return to the inner surface,
which a cell might create a membrane poten-
via cytochromes, where they reduce an electron
tial during respiration. You can include both
acceptor. The inward transfer of the electrons is
periplasmic and cytoplasmic oxidations.
electrogenic (i.e., a membrane potential is cre-
ated). In the Q loop, which is a linear pathway 7. Assuming a ∆p of –200 mV and a H+/ATP of
of electron flow, a proton is released for every 3, calculate what the ∆Gp must be in joules
electron. per mole if ∆Gp is in equilibrium with ∆p.
The Q cycle, a more complicated pathway, Assume that the temperature is 30 °C, and
results in the release of two protons for every that ∆G′p is 37 kcal/mol. What will be the
electron. In the Q cycle, QH2 gives up two pro- ratio of ATP to [ADP][Pi]?
tons and two electrons, but one of the electrons
8. What are the similarities and differences
is recycled back to oxidized quinone. Thus, the
between bacterial respiratory pathways and
ratio of H+ to e– is 2, rather than 1. The Q cycle
the mitochondrial respiratory pathway?
is more efficient, since it results in the availabil-
ity of more protons for influx through the ATP 9. Calculate the maximum number of ATPs
synthase per electron, and thus can increase the that can be made per NADH oxidized by E.
yield of ATP. coli by using the following combinations:
In a second method of coupling oxidation– NDH-1 plus cytochrome bo and NDH-2
reduction reactions to the establishment of a plus cytochrome bd.
proton potential, some of the electron carriers
10. What drives reversed electron transport?
act as proton pumps. A membrane potential is
What experiment can be done to support
created when protons are pumped electrogeni-
this conclusion?
cally out of the cell. A well-established example
is cytochrome aa3 oxidase. However, there is 11. What is it about the sequential arrange-
evidence that other oxidases are also proton ment of electron carriers that makes pro-
pumps, including cytochrome bo in E. coli and ton translocation possible in a Q loop or Q
cytochromes bb3 and cbb3 in P. denitrificans. cycle?
Mitochondrial and certain bacterial NADH
12. What is the definition of the P/O ratio?
dehydrogenases are also coupling sites, and
P/2e− ratio?
these may function as proton pumps.
Study Questions REFERENCES AND NOTES
1. What is it about the solubility properties 1. Archaea—Molecular and Cellular Biology. 2007.
and electrode potentials of quinones that R. Cavicchioli (Ed.). ASM Press, Washington, DC.
make them suitable for their role in electron 2. Lubben, M. 1995. Cytochromes of archaeal
transport? electron transfer chains. Biochim. Biophys. Acta
1229:1–22.
2. Design an experiment that can quantify the 3. For example, the enzyme complex that oxidizes
H+/O for the cytochrome aa3 reaction in NADH and reduces quinone is an NADH–quinone
proteoliposomes. oxidoreductase. It consists of a flavoprotein and
www.ebook777.com
iron–sulfur proteins. (Reduced quinone is called qui- 10. Ohnishi, T. 1993. NADH–quinone oxi-
nol.) An enzyme complex that oxidizes quinol and doreductase, the most complex complex. J. Bioenerg.
reduces cytochrome c is called a quinol–cytochrome Biomemb. 25:325–329.
c oxidoreductase. It consists of cytochromes and an
iron–sulfur protein. 11. Yagi, T., T. Yano, and A. Matsuno-Yagi. 1993.
Characteristics of the energy-transducing NADH–
4. Here are some definitions that are useful to quinone oxidoreductase of Paracoccus denitrificans
know. Cofactors are nonprotein molecules, either as revealed by biochemical, biophysical, and molecu-
metals or organic, bound with varying degrees of lar biological approaches. J. Bioenerg. Biomemb.
affinity to enzymes and required for enzyme activ- 25:339–345.
ity. Coenzymes are organic cofactors that shuttle
12. Complex II (succinic dehydrogenase) has four
back and forth between enzymes carrying electrons,
polypeptides, two of which are hydrophilic periph-
hydrogen, or organic moieties (e.g., acyl groups).
eral membrane proteins protruding into the matrix
Prosthetic groups are cofactors (organic or inor-
of mitochondria or the cytoplasm of bacteria; the
ganic) that are tightly bound to the protein and do
other two are hydrophobic integral membrane
not dissociate from the protein. Thus, one difference
proteins. The polypeptide furthest from the mem-
between a coenzyme and a prosthetic group is that
brane, called subunit A, is a flavoprotein with FAD
the former shuttles between enzymes and the latter
as its prosthetic group. Subunit A is attached to the
remains tightly bound to the enzyme. Coenzyme
second peripheral polypeptide, called subunit B,
A (carrier of acyl groups) and NAD+ (electron and
which is closest to the membrane. Subunit B con-
hydrogen carrier) are examples of coenzymes,
tains three FeS centers through which electrons
whereas FAD (electron and hydrogen carrier) and
from the flavoprotein are passed. The two integral
heme (electron carrier) are examples of prosthetic
membrane proteins (subunits C and D) are associ-
groups.
ated with one or two heme groups in many bacteria
5. The first iron–sulfur protein identified was a sol- depending upon the enzyme. The hemes may trans-
uble protein isolated from bacteria and called ferre- fer electrons from the FeS centers to the quinone
doxin. A protein similar to ferredoxin with a low within the membrane. There are variations of this
redox potential can be isolated from chloroplasts and common theme. For example, the succinic dehy-
mediates electron transfer to NADP+ during noncyc- drogenase from certain bacteria such as E. coli and
lic electron flow. Iron–sulfur proteins that have no some others does not contain heme. In addition,
other prosthetic groups are divided into four classes most gram-positive bacteria as well as ε proteobac-
based upon the number of iron atoms per molecule teria have just one very large integral polypeptide
and whether the sulfur is acid labile. Examples from subunit (called C), and this is believed to result
the four different classes are rubredoxin (no acid- from fusion of the genes for the two smaller sub-
labile sulfur), isolated from bacteria; high-potential units (C and D). The student is referred to a special
iron protein (HiPIP), isolated from photosynthetic issue of Biochimica et Biophysica Acta, volume
bacteria; chloroplast ferredoxin; and various bacte- 1553 (2002), which has a collection of articles on
rial ferredoxins. succinic dehydrogenase.
6. Electron spin resonance spectroscopy detects 13. Anraku, Y. 1988. Bacterial electron trans-
unpaired electrons such as the ones that exist in the port chains, pp. 101–132. In: Annual Review of
FeS cluster. Monochromatic microwave radiation is Biochemistry, Vol. 57. C. C. Richardson, P. D. Boyer,
absorbed by the unpaired electron when a magnetic I. B. Dawid, and A. Meister (Eds.). Annual Reviews,
field is applied. The size of the magnetic field required Palo Alto, CA.
for radiation absorption depends upon the molecular 14. García-Horsman, J. A., B. Barquera, J. Rumbley,
environment of the unpaired electron. A spectrum is J. Ma, and R. B. Gennis. 1994. The superfamily of
obtained by varying the magnetic field and keeping heme–copper respiratory oxidases. J. Bacteriol.
the frequency of the microwave radiation constant. 176:5587–5600.
The spectra differ for different FeS clusters. A spectro-
scopic constant, called the g value, is obtained which 15. Hill, S., S. Viollet, A. T. Smith, and C. Anthony.
is characteristic of the FeS cluster in the protein. 1990. Roles for enteric d-type cytochrome oxidase
in N2 fixation and microaerobiosis. J. Bacteriol.
7. Puustinen, A., and M. Wikstrom. 1991. The heme 172:2071–2078.
groups of cytochrome o from Escherichia coli. Proc.
Natl. Acad. Sci. USA 88:6122–6126. 16. Kelly, M. J. S., R. K. Poole, M. G. Yates, and C.
Kennedy. 1990. Cloning and mutagenesis of genes
8. Yagi, T., X. Xu, and A. Matsuno-Yagi. 1992. The encoding the cytochrome bd terminal oxidase com-
energy-transducing NADH–quinone oxidoreductase plex in Azotobacter vinelandii: mutants deficient in
(NDH-1) of Paracoccus denitrificans. Biochim. the cytochrome d complex are unable to fix nitrogen
Biophys. Acta 1101:181–183. in air. J. Bacteriol. 172:6010–6019.
9. Finel, M. 1993. The proton-translocating 17. Hinkle, P. C. 2005. P/O ratios of mitochon-
NADH:ubiquinone oxidoreductase: a discussion of drial oxidative phosphorylation. Biochim. Biophys.
selected topics. J. Bioenerg. Biomemb. 25:357–366. Acta—Bioenergetics 1706:1–11.
18. Reviewed in J. Bioenerg. Biomembr. 25(4), cation of the cytochrome o complex of Escherichia
1993. coli yields a two heme/one copper terminal oxidase
with high specific activity. Biochemistry 31:6917–
19. Right-side-out vesicles are made by osmotic lysis
6924.
of the spheroplasts or protoplasts.
32. Gennis, R. B., and V. Stewart. 1996. Respiration,
20. The phosphorylation potential, ∆Gp, is depen- pp. 217–261. In: Escherichia coli and Salmonella,
dent upon the standard free energy of formation Cellular and Molecular Biology, Vol. 1. F. C.
of ATP and the actual ratios of ATP, ADP, and Pi, Neidhardt et al. (Eds.). ASM Press, Washington,
according to the following equation: ΔGp = ΔG′0 + RT DC.
ln[ATP]/[ADP][Pi] J/mol. The free energy of forma-
tion is proportional to the equilibrium constant (i.e., 33. Anraku, Y., and R. B. Dennis. 1987. The aero-
ΔG′0 = −RT ln K′eq ). bic respiratory chain of Escherichia coli. Trends
Biochem. Sci. 12:262–266.
21. Trumpower, B. L. 2003. Proton motive force
generation by a redox loop mechanism. FEBS Lett. 34. Calhoun, M. W., K. L. Oden, R. B. Gennis, M.
545:25–30. J. Teixeira de Mattos, and O. M. Neijssel. 1993.
Energetic efficiency of Escherichia coli: effects of
22. Hunter, C., H. Palsdottir, and B. L. Trumpower. mutations in components of the aerobic respiratory
2003. Proton motive pathways and mechanisms chain. J. Bacteriol. 175:3020–3025.
in the cytochrome bc1 complex. FEBS Lett. 545:
39–46. 35. Tran, Q. H., J. Bongaerts, D. Vlad, and G. Unden.
1997. Requirement for the proton-pumping NADH
23. Reviewed in: Trumpower, B. L. 1990. dehydrogenase I of Escherichia coli in respiration
Cytochrome bc1 complexes of microorganisms. of NADH to fumarate and its bioenergetic implica-
Microbiol. Rev. 54:101–129. tions. Eur. J. Biochem. 244:155–160.
24. van Verseveld, H. W., K. Krab, and A. H. 36. Green, J., M. F. Anjum, and J. R. Guest. 1997.
Stouthamer. 1981. Proton pump coupled to cyto- Regulation of the ndh gene of Escherichia coli by
chrome c oxidase in Paracoccus denitrificans. integration host factor and a novel regulator, Arr.
Biochim. Biophys. Acta 635:525–534. Microbiology 143:2865–2875.
25. de Gier, J.-W. L., M. Lubben, W. N. M. 37. Reviewed in: Berks, B. C., S. J. Ferguson, J. W.
Reijnders, C. A. Tipker, D-J. Slotboom, R. J. M. van B. Moir, and D. J. Richardson. 1995. Enzymes and
Spanning, A. H. Stouthamer, and J. van der Oost, associated electron transport systems that catalyze
1994. The terminal oxidases of Paracoccus denitri- the respiratory reduction of nitrogen oxides and
ficans. Mol. Microbiol. 13:183–196. oxyanions. Biochim. Biophys. Acta 1232:97–173.
26. Hinkle, P. C., M. A. Kumar, A. Resetar, and D. 38. Fumarate is reduced to succinate by using the
L. Harris, 1991. Mechanistic stoichiometry of mito- membrane-bound enzyme fumarate reductase.
chondrial oxidative phosphorylation. Biochemistry Nitrate is reduced to nitrite by using the membrane-
30:3576–3582. bound nitrate reductase, which resembles the dissim-
27. Matsushita, K., T. Ohnishi, and H. R. Kaback. ilatory nitrate reductase in P. denitrificans. Nitrite is
1987. NADH–ubiquinone oxidoreductases of reduced to ammonia by using an NADH-linked cyto-
the Escherichia coli aerobic respiratory chain. plasmic enzyme (NADH–nitrite oxidoreductase).
Biochemistry 26:7732–7737. Energy is not conserved in the reaction. If NO3− is the
only source of nitrogen, then some of the ammonia is
28. Sled, V. D., T. Freidrich, H. Leif, H. Weiss, S. assimilated and the excess is excreted. Nitrite is toxic
W. Meinhardt, Y. Fukumori, M. W. Calhoun, R. B. to bacteria, and probably the reduction to ammonia
Gennis, and T. Ohnishi. 1993. Bacterial NADH– functions to reduce toxic levels.
quinone oxidoreductases: iron–sulfur clusters and
related problems. J. Bioenerg. Biomemb. 25:347–356. 39. Unden, G., and J. Schirawski. 1997. The oxy-
gen-responsive transcriptional regulator FNR of
29. Unden, G., and J. Bongaerts. 1997. Alternative Escherichia coli: the search for signals and reactions.
respiratory pathways of Escherichia coli: energetics Mol. Microbiol. 25:205–210.
and transcriptional regulation in response to electron
acceptors. Biochim. Biophys. Acta 1320:217–234. 40. van Verseveld, H. W., and A. H. Stouthamer.
1992. The genus Paracoccus, pp. 2321–2334. In:
30. van der Oost, J., A. P. N. de Boer, J.-W. L. de The Prokaryotes, Vol. III, 2nd ed. A. Balows, H. G.
Gier, W. G. Zumft, A. H. Stouthamer, and R. J. M. Truper, M. Dworkin, W. Harder, and K.-H. Schleifer
van Spanning. 1994. The heme–copper oxidase fam- (Eds.). Springer-Verlag, Berlin.
ily consists of three distinct types of terminal oxi-
dases and is related to nitric oxide reductase. FEMS 41. van Spanning, R. J. M., A. P. N. de Boer, W. N.
Microbiol. Lett. 121:1–10. M. Reijnders, J.-W. L. de Gier, C. O. Delorme, A. H.
Stouthamer, H. V. Westerfoff, N. Harms, and J. van der
31. Minghetti, K. C., V. C. Goswitz, N. E. Gabriel, Oost. 1995. Regulation of oxidative phosphorylation:
J. Hill, C. A. Barassi, C. D. Georgiou, S. I. Chan, and the flexible respiratory network of Paracoccus denitri-
R. B. Gennis. 1992. A modified, large-scale purifi- ficans. J. Bioenerg. Biomembr. 27:499–512.
www.ebook777.com
42. Visser, J. M., A. H. de Jong, S. de Vries, L. A. cofactor (Moco), which consists of molybdenum
Robertson, and J. G. Kuenen. 1997. cbb3-Type cyto- bound to a pterin. The Moco prosthetic group is in
chrome oxidase in the obligately chemolithoauto- all molybdenum-containing proteins except nitroge-
trophic Thiobacillus sp. W5. FEMS Microbiol. Lett. nase. [The molybdenum cofactor from nitrogenase
147:127–132. contains nonheme iron and is called FeMo cofactor
or FeMoco. Nitrite reductase is a periplasmic enzyme
43. Cox, J. C., W. J. Ingledew, B. A. Haddock, and that reduces nitrite to nitric oxide. Nitrite reductase
H. G. Lawford. 1978. The variable cytochrome con- from P. denitrificans has two identical subunits.
tent of Paracoccus denitrificans grown aerobically These have hemes c and d1. Nitric oxide reductase is
under different conditions. FEBS Lett. 93:261–265. a membrane-bound enzyme that reduces nitric oxide
44. Puustinen, A., M. Finel, M. Virkki, and M. to nitrous oxide. It has heme b and heme c. Nitrous
Wikstrom. 1989. Cytochrome o (bo) is a proton oxide reductase is a periplasmic enzyme that reduces
pump in Paracoccus denitrificans and Escherichia nitrous oxide to nitrogen gas. It has two identical
coli. FEBS Lett. 249:163–167. subunits containing copper.
45. Ferguson, S. J. 1987. Denitrification: a question 47. Patureau, D., N. Bernet, and R. Moletta.
of the control and organization of electron and ion 1996. Study of the denitrifying enzymatic system of
transport. Trends Biochem. Sci. 12:354–357. Comamonas sp. strain SGLY2 under various aera-
tion conditions with a particular view on nitrate and
46. Nitrate reductase is a membrane-bound molyb- nitrite reductases. Curr. Microbiol. 32:25–32.
denum protein with three subunits, α, β, and γ. The
48. Reviewed in: Kroger, A., V. Geisler, E. Lemma,
γ subunit is a cytochrome b. A proposed route for
F. Theis, and R. Lenger. 1992. Bacterial fumarate
electrons is from ubiquinol to the FeS centers of the
respiration. Microbiology 158:311–314.
β subunit, and from there to the molybdenum cen-
ter at the active site in the α subunit. The α subunit 49. Meyer, D. J., and C. W. Jones. 1973. Oxidative
then reduces the nitrate. The molybdenum is part of phosphorylation in bacteria which contain different
a cofactor (prosthetic group) called the molybdenum cytochrome oxidases. Eu. J. Biochem. 36: 144–151.
6
Photosynthesis
Photosynthesis is the conversion of light that (bacterio)chlorophyll-based photosynthe-

energy into chemical energy used for growth. sis is widely distributed, being found in bacteria,
Organisms that obtain most or all of their green plants, and algae, but is not present in the
energy in this way are called phototrophic or known archaea. (Interestingly, bacteriochloro-
photosynthetic. Depending upon the photosyn- phyll has also been found in nonphototrophic
thetic system, light energy has one or both of the bacteria, i.e., rhizobia.1,2) This chapter begins
following properties. with background information on the different
groups of phototrophic prokaryotes and then
1. In all photosynthetic systems light energy can
describes photosynthetic electron transport in
drive the phosphorylation of ADP to make
all the photosynthetic systems.
ATP (the process of photophosphorylation).
2. In some photosynthetic systems light can
also drive the transfer of electrons from H2O 6.1 The Phototrophic Prokaryotes
(∆Em,7 = +820 mV) to NADP+ (∆Em,7 = –320
The phototrophic prokaryotes are a diverse
mV). (As will be discussed later, green sulfur
assemblage of organisms that share the com-
photosynthetic bacteria use light energy to
mon feature of being able to use light as a source
move electrons from inorganic sulfur com-
of energy for growth. Their classification is
pounds rather than water to NADP+.)
based upon physiological differences, includ-
During photoreduction of NADP+, the water ing whether they produce oxygen, and what
becomes oxidized to oxygen gas and the NADP+ they use as a source of electrons for biosynthe-
becomes reduced to NADPH. The student will sis. For example, there are both oxygenic pho-
recognize that this is the reverse of the direction totrophs (produces of oxygen) and anoxygenic
in which electrons flow spontaneously during phototrophs. Oxygenic phototrophs include
aerobic respiration (Chapter 5). During both the well-known cyanobacteria (formerly
ATP and NADPH synthesis, electromagnetic called blue-green algae) and certain prochlo-
energy is absorbed by photopigments in the rophytes, members of the genera Prochloron,
photosynthetic membranes and converted into Prochlorothrix, and Prochlorococcus.3 The
chemical energy. This chapter explains how latter three genera consist of organisms phy-
that is done. At the heart of the process is the logenetically related to the cyanobacteria but
light-driven oxidation of chlorophyll or bacte- having chlorophyll b as a light-harvesting pig-
riochlorophyll, which initiates electron trans- ment instead of phycobilins.
port that results in the generation of a ∆p and The anoxygenic phototrophs are the purple
subsequent synthesis of ATP, and in the case of photosynthetic bacteria, the green photosyn-
chlorophyll, NADP+ reduction. It is of interest thetic bacteria, and the heliobacteria. (For a
175
www.ebook777.com
more complete discussion of the taxonomy of phycobilins as the light-harvesting pigment.5

the photosynthetic bacteria, see note 4.) These There are three genera among the prochlo-
microorganisms photosynthesize only under rophytes: Prochloron, Prochlorothrix, and
anaerobic conditions. The purple and green Prochlorococcus. Members of the genus
photosynthetic bacteria are further subdivided Prochloron are obligate symbionts of certain
according to whether they use sulfur as a source ascidians (sea squirts). Prochlorothrix is a fila-
of electrons. Thus there are purple sulfur and mentous, free-living microorganism that lives
nonsulfur photosynthetic bacteria and green sul- in freshwater lakes. Prochlorococcus is a free-
fur and nonsulfur photosynthetic bacteria. The living marine microorganism.
various phototrophic prokaryotes and some of
their properties are summarized in Table 6.1.
6.1.2 Anoxygenic phototrophs
The other phototrophic prokaryotes do not pro-
6.1.1 Oxygenic phototrophs
duce oxygen (i.e., H2O is not a source of electrons).
The oxygenic prokaryotic phototrophs use H2O Instead, they use organic compounds, inorganic
as the electron donor for the photosynthetic sulfur compounds, or hydrogen gas as a source
reduction of NADP+. Most of the species belong of electrons. These organisms will grow pho-
to the cyanobacteria, which are widely distrib- totrophically only anaerobically or when oxygen
uted in nature, occurring in fresh and marine tensions are low. There are four major groups of
waters, and in terrestrial habitats. They have anoxygenic phototrophs (see Table 6.1):
only one type of chlorophyll (chlorophyll a), and
light-harvesting pigments called phycobilins. 1. Purple photosynthetic bacteria, which
As stated earlier, another kind of oxygenic includes both the purple sulfur and the pur-
microbial phototroph consists of the prochlo- ple nonsulfur bacteria
rophytes, which are a diverse group of photo- 2. Green sulfur photosynthetic bacteria
synthetic prokaryotes that are evolutionarily 3. Green nonsulfur photosynthetic bacteria,
related to the cyanobacteria but differ from also called green gliding bacteria
the latter in having chlorophyll b rather than 4. Heliobacteria
Table 6.1 Phototrophic prokaryotes
Type Pigmentsa Electron donor Carbon source Aerobic dark

growth
Oxygenic
Cyanobacteria Chl a, phycobilins H2O CO2 No
Prochloron Chl a, Chl b H2O CO2 No
Chlorothrix Chl a, Chl b H2O CO2 No
Prochlorothrix Chl a, Chl b H2O CO2 No
Anoxygenic
Purple sulfur Bchl a or Bchl b H2Sb, S°, S2O32−, CO2, organic Yesc
H2, organic
Purple nonsulfur Bchl a or Bchl b H2, organic, CO2, organic Yes
H2S (some)
Green sulfur Bchl a and c, d, or e H2Sd, S°, S2O32−, H2 CO2, organic No
Green gliding Bchl a and c or d H2S, H2, organic CO2, organic Yes
(nonsulfur)
Heliobacteria Bchl g Organic Organic No
a
Carotenoids are usually present.
b
Members of the genus Chromatium accumulate So intracellularly. Members of the genus Ectothiorhodospira accumulate
extracellular sulfur.
c
In the natural habitat, growth occurs primarily anaerobically in the light. However, several purple sulfur species
(Chromatiaceae) can be grown continuously in the laboratory aerobically in the dark at low oxygen concentrations.
(Overmann, J., and N. Pfennig. 1992. Continuous chemotrophic growth and respiration of Chromatiaceae species at low
oxygen concentrations. Arch. Microbiol. 158:59–67.)
d
Accumulates S° extracellularly.
photosynthesis 177
Purple sulfur phototrophs Green sulfur phototrophs

The purple sulfur phototrophic bacteria, which For a review of photosynthesis in the green
grow photoautotrophically in anaerobic envi- sulfur and green nonsulfur photosynthetic bac-
ronments, use hydrogen sulfide as the electron teria, see ref. 6. The green sulfur phototrophic
donor and CO2 as the carbon source. Their nat- bacteria are strict anaerobic photoautotrophs
ural habitats are freshwater lakes and ponds, or that use H2S, S°, S2O32 − (thiosulfate), or H2 as
marine waters where the presence of sulfate-re- the electron donor and CO2 as the source of
ducing bacteria produces a high sulfide content. carbon. The green sulfur bacteria coexist with
They oxidize sulfide to elemental sulfur, which the purple sulfur phototrophs in sulfide-rich
accumulates as granules intracellularly in all anaerobic aquatic habitats, although they are
known genera except the Ectothiorhodospira, found in a separate layer. Their light-harvesting
which deposits sulfur extracellularly. However, pigments are located in special inclusion bodies
some can grow photoheterotrophically, and called chlorosomes, which will be described
several have been grown chemoautotrophically later, and their reaction centers (the part of
under low partial pressures of oxygen with the photosynthetic membrane where the bac-
reduced inorganic sulfur as an electron donor teriochlorophyll is oxidized) differ from those
and energy source. There are many genera, of the purple photosynthetic bacteria. There
including the well-studied Chromatium. are several genera, including the well-studied
Chlorobium.
Purple nonsulfur phototrophs
The purple nonsulfur phototrophic bacteria Green nonsulfur phototrophs
are extremely versatile with regard to sources (green gliding bacteria)
of energy and carbon. Originally it was thought The best-studied green nonsulfur photosyn-
that these bacteria were not able to utilize sul- thetic bacterium is Chloroflexus, which is a
fide as a source of electrons, hence the name thermophilic, filamentous, gliding green pho-
“nonsulfur.” However, it turns out that the totroph. Chloroflexus can be isolated from
concentrations of sulfide used by the purple alkaline hot springs, whose pH can be as high as
sulfur bacteria are toxic to the nonsulfur pur- 10. Most isolates have a temperature optimum
ples and, provided the concentrations are suf- for growth from 52 to 60 °C. Chloroflexus can
ficiently low, some purple nonsulfur bacteria be grown as a photoheterotroph, photoauto-
(viz., Rhodopseudomonas and Rhodobacter) trophically with either H2 or H2S as the electron
can use sulfide and/or thiosulfate as a source of donor, or chemoheterotrophically in the dark
electrons during photoautotrophic growth. in the presence of air. However, it really grows
The purple nonsulfur phototrophs can grow best as a photoheterotroph.
photoautotrophically or photoheterotrophi- Chloroflexus has chlorosomes, but its reac-
cally in anaerobic environments. When grow- tion center is a quinone type, resembling that
ing photoautotrophically, they use H2 as the of the purple photosynthetic bacteria. It differs
electron donor and CO2 as the source of car- from the reaction center in the green sulfur bac-
bon. Photoheterotrophic growth uses simple teria, which is an iron–sulfur type, resembling
organic acids such as malate or succinate as the reaction center of heliobacteria and reaction
the electron donor and source of carbon, and center I of cyanobacteria and chloroplasts.
light as the source of energy. If these organ-
isms are placed in the dark in the presence of Heliobacteria
oxygen, they will carry out ordinary aerobic The heliobacteria are recently discovered,
respiration and grow chemoheterotrophically strictly anaerobic photoheterotrophs that
(e.g., on succinate or malate). A few can even differ from the other phototrophs in having
grow fermentatively, very slowly, in the dark. bacteriochlorophyll g as their reaction center
Because of their physiological versatility, the chlorophyll, in containing few carotenoids,
purple nonsulfur photosynthetic bacteria have and in having no internal photosynthetic mem-
received much attention in research. They are branes or chlorosomes. The two known genera
found in lakes and ponds with low sufide con- are Heliobacterium and Heliobacillus. The tet-
tent. Representative genera are Rhodobacter rapyrrole portion of bacteriochlorophyll g is
and Rhodospirillum. similar to chlorophyll a and in fact isomerizes
www.ebook777.com
into chlorophyll a in the presence of air. These creates a ΔEm′ of about 1 V between P870 +
/P870,
organisms are placed with the gram-positive which is around +0.45 V, and Bpheo+/
bacteria, both by 16S rRNA analyses and by the Bpheo, which is around –0.6 V.
lack of lipopolysaccharide in the cell wall. It has Step 3. The electron is transferred from the bac-
been suggested that these heliobacteria are phy- teriopheophytin to a quinone called ubiqui-
none A (UQA). The ΔEm′ for UQ A /UQ A is
–
logenetically related to the clostridia. In agree-
ment with this, most species of heliobacteria about –0.2 V.
make endospores that resemble those produced Step 4. UQA transfers the electron to a ubiqui-
by Bacillus or Clostridium. none in site B, called UQB, and UQB becomes
reduced to UQ B– . Steps 1 to 4 are repeated
so that UQB accepts a second electron and
6.2 The Purple Photosynthetic becomes reduced to UQ 2– B .
Bacteria Step 5. UQ 2–B picks up two protons from the
In all photosynthetic systems, light is absorbed cytoplasm and is released from the reaction
by light-harvesting pigments (also called acces- center as UQH2. The UQH2 joins the qui-
sory pigments), and the energy is transferred to none pool in the membrane.
reaction centers where bacteriochlorophyll is Steps 6 and 7. UQH2 transfers the electrons to
oxidized, creating a redox potential, ∆E. The a bc1 complex and the bc1 complex reduces
light-harvesting pigments, the photosynthetic cytochrome c2. The bc1 complex translocates
pigments in the reaction center, and energy protons to the cell surface, thus generating a
transfer to the reaction center are considered in ∆p. (See Section 5.6.2 for a discussion of the
Sections 6.6 and 6.7. The structure of the photo- bc1 complex and the Q cycle.)
synthetic membrane systems in the prokaryotes Step 8. The cytochrome c2 returns the electron to
as revealed by electron microscopy is presented in the oxidized bacteriochlorophyll molecule
Section 6.8. The generation of the ∆E and photo- in the reaction center. Thus light energy has
synthetic electron transport are described next.7,8 caused a cyclic electric current in the membrane.
The reaction center of green nonsulfur filamen-
These eight steps describe cyclic electron flow.
tous photosynthetic bacteria is similar. See note 9
We can follow the path of the electron starting
for a comparison of the two reaction centers.
with cytochrome c2 (Fig. 6.1B). Notice that an
electron moves “uphill” (to a lower electrode
potential) from cytochrome c2 to ubiquinone
6.2.1 Photosynthetic electron transport
when “boosted” or “pushed” in the reaction
This section discusses the oxidation of bacte- center by the energy from a quantum of light.
riochlorophyll, the generation of the ∆E, and As pointed out earlier, for the reaction cen-
electron transport. Photosynthetic electron ter to reduce UQ to UQH2, two electrons and
flow in the purple photosynthetic bacteria is therefore two light reactions are required. This
illustrated in Fig. 6.1A. A detailed discussion of can be written as follows,
the reaction center is given in Section 6.2.2; the
photosynthetic process itself has eight steps, as 2 cyt c 2,red + 2H + + 2 UQ
follows. (6.1)
⎯⎯⎯⎯⎯⎯
2 quanta of light
→ 2 cyt c 2,0x + UQH 2
Step 1. A dimer of two bacteriochlorophyll mol-
ecules (P870) in the reaction center absorbs Because two electrons are required before
light energy, and one of its electrons becomes reduced quinone leaves the reaction center, UQB
excited to a higher energy level ( P870 * ). In is referred to as a two-electron gate.
contrast to P870, P* has a very low reduction As illustrated in Fig. 6.1, the reduced quinone
870
potential ( E′m ). transfers the electrons to cytochrome c2 via a
Step 2. The electron is transferred to an accep- bc1 complex outside the reaction center. The bc1
tor molecule within the reaction center, thus complex translocates protons across the mem-
oxidizing the P870 and reducing the acceptor brane via a Q cycle: UQH2 + 2H+in + 2 cyt c2,ox.
molecule. The acceptor molecule is bacte- UQH 2 + 2H in+ + 2 cyt c 2,0x
riopheophytin (Bpheo), which is bacteri- +
(6.2)
ochlorophyll without magnesium. The light ⎯⎯→
bc1
UQ + 4H out + 2 cyt c 2,red
photosynthesis 179
Fig. 6.1 Photosynthetic electron flow in purple photosynthetic bacteria. (A) Light energizes bacteriochloro-
phyll, here shown as P870 (bacteriochlorophyll a) in the reaction center (RC). (The number in the subscript
gives the major long-wavelength absorption peak in nanometers.) Some purple photosynthetic bacteria have
bacteriochlorophyll b instead, which absorbs at 1020 to 1035 nm. The energized bacteriochlorophyll reduces
bacteriopheophytin (Bpheo), which is bacteriochlorophyll without Mg2+. The electron then travels through
two ubiquinones, UQA and UQB, in the reaction center. (Some species use menaquinone.) After a second light
reaction, UQB becomes reduced to UQH2 and leaves the reaction center. The electron returns to the reaction
center via a proton-translocating bc1 complex and cytochrome c2. ATP is synthesized via a membrane ATP
synthase driven by the ∆p (not shown). Midpoint potentials at pH 7 (approximate): Bchl870 (P870), +450 mV;
Bpheo, –600 mV; UQA, –200 mV; UQB, +80 mV; c cytochromes, +380 mV. (B) A simplified version of (A)
that emphasizes the cyclic flow of electrons. Source: Data from Mathis, P. 1990. Biochim. Biophys. Acta
1018:163–167.
www.ebook777.com
The ∆p that is generated is used to drive the

synthesis of ATP via a membrane-bound ATP
synthase: hυ
+
IN N P OUT
ADP + P1 + 3H out 2 x 1e-
(6.3) UQA c2
⎯⎯⎯⎯⎯
ATP synthase
→ ATP + H 2O + 3H in+ RC
UQB
+
2H
Thus, the net result of photosynthesis by purple
photosynthetic bacteria is the synthesis of ATP. UQH2 UQ 2 x 1e-
This is called cyclic photophosphorylation.
The actual amount of ATP made depends ADP + Pi
upon the number of protons translocated by ATP + H2O
2H+ 4H+
the bc1 complex and the number of protons that
bc1
enter via the ATP synthase. For example, if four
protons for every two electron transfers are
translocated out of the cell by the bc1 complex
and three protons reenter via the ATP synthase,
then the maximum number of ATP molecules
made per two electrons is 4/3 (1.333).
Figure 6.2 illustrates the topographical
relationships between the reaction center, the ATP 3H+
bc1 complex, and the ATP synthase in photo- synthase
synthetic membranes and can be compared to

Fig. 6.2 Relationship between reaction center, bc1
Fig. 6.1, which illustrates the sequential steps complex, and ATP synthase in photosynthetic mem-
in electron transfer. In the reaction center, branes. The reaction center (RC) takes two pro-
electrons travel across the membrane from a tons from the cytoplasm and two electrons from a
periplasmic cytochrome c2 to the quinone at site periplasmic cytochrome c to reduce ubiquinone
A. This creates a membrane potential, which is (UQ) to UQH2. The UQH2 leaves the reaction cen-
negative inside. After a second electron is trans- ter and diffuses to the bc1 complex. The bc1 complex
ferred, the quinone at site B accepts two pro- oxidizes UQH2 and translocates four protons to the
tons from the cytoplasm and leaves the reaction periplasmic surface in a Q cycle, which is described
center as UQH2. The UQH2 diffuses through in Section 5.6.2. The electrons travel back to cyto-
chrome c2 (cyclic flow). The translocated protons
the lipid matrix to the bc1 complex, where it
reenter via the ATP synthase, which makes ATP.
is oxidized. The bc1 complex translocates four
protons to the outside for every UQH2 oxidized
via a Q cycle, creating a ∆p. The electrons are aspects of the structure of the reaction center. A
transferred from the bc1 complex to a periplas- more complete description follows.
mic cytochrome c2, which returns them to the
reaction center. The proton circuit is completed Structure and composition of the
when the protons are returned to the cytoplas- reaction center
mic side via the ATP synthase accompanied by The reaction centers from Rhodopseudomonas
the synthesis of ATP. Thus, light is used to drive viridis and Rhodobacter sphaeroides have been
an electron circuit, which drives a proton cir- crystallized, and the structures determined from
cuit, which in turn drives ATP synthesis. This high-resolution X-ray diffraction studies.10–12
successful method for harnessing light energy The reaction centers are similar. The following
to do chemical work is used in all phototrophic proteins and pigments are found in the reaction
organisms, from bacteria to plants. center from R. sphaeroides:
6.2.2 A more detailed examination of the 1. Reaction center protein. The reaction center
reaction center and what happens there protein has 11 membrane-spanning α heli-
The description just given for electron flow in ces, and a globular portion on the cytoplas-
the reaction center omitted some of the compo- mic side of the membrane. It consists of three
nents and electron carriers, and did not cover all polypeptides: H, L, and M. The reaction
photosynthesis 181
center protein serves as a scaffolding to

which the bacteriochlorophyll and bacterio-
pheophytin are attached, and to which the
quinones and nonheme iron are bound.
2. Four bacteriochlorophyll molecules (Bchl).
There are four bacteriochlorophyll mol-
ecules in the reaction center. Two of the bac-
teriochlorophyll molecules exist as a dimer
(Bchl)2, and two are monomers (BchlA and
BchlB).
3. Two molecules of bacteriopheophytin
(BpheoA and BpheoB), which is Bchl without
Mg2+.
4. Two molecules of ubiquinone (UQA and
UQB).
5. One molecule of nonheme ferrous iron (Fe).
6. One carotenoid molecule. The function of
the carotenoid is to protect the reaction cen-
ter pigments from photodestruction.13 (For a
further explanation, see note 14.)
Electron transfer in the reaction center

The sequence of redox reactions in the reaction
center summarized in Fig. 6.1A (boxed area) is
shown in more detail in Fig. 6.3, along with the
proposed time scale for the electron transfer
events. The terminal electron acceptors in the
reaction center (Fig. 6.1) are quinines, which
are called type II reaction centers and are found
in purple bacteria (Proteobacteria) and green
nonsulfur bacteria (Chloroflexus group). (Type
I reaction centers use iron–sulfide clusters as Fig. 6.3 The reaction center in purple photosynthetic
the terminal electron acceptor, and these are bacteria. The reaction center, which spans the mem-
present in green sulfur bacteria and heliobac- brane, is represented by the boxed areas. The absorp-
tion of light creates a transient membrane potential,
teria, as discussed later.) As shown in Fig. 6.3,
positive (P-phase) on the periplasmic side and negative
the arrangement of the pigment molecules and (N-phase) on the cytoplasmic side. Step 1: absorption
the quinones reveals twofold symmetry, with of a photon of light energizes a bacteriochlorophyll
right and left halves that are very similar. On dimer (Bchl)2. Step 2: the energized (Bchl)2* reduces
the periplasmic side of the membrane there sits BchlA; the question mark indicates that this has not
a pair of bacteriochlorophyll molecules (Bchl)2 been unequivocally demonstrated. Step 3: bacterio-
(Fig. 6.3). These bacteriochlorophyll mole- pheophytin A (BpheoA) is reduced. Step 4: the electron
cules are P870 in Fig. 6.1. When the energy from is transferred to a quinone (UQA). Step 5: the oxidized
a quantum of light is absorbed by (Bchl)2, an (Bchl)2 is reduced via cytochrome c2. Step 6: the elec-
electron becomes excited [i.e., (Bchl)2 becomes tron moves from UQA to UQB. Steps 1 to 6 are repeated,
(Bchl)2*] (Fig. 6.3, step 1). The redox potential of leading to the formation of UQ2– B
. Two protons are
acquired from the cytoplasm to produce UQH2, which
(Bchl)2* is low enough to reduce bacteriochlo-
enters the reduced quinone pool in the membrane and
rophyll A (BchlA), forming (Bchl)2− and (Bchl)A− returns the electrons to oxidized cytochrome c2 via the
(Fig. 6.3, step 2). The electron then moves bc1 complex. It may be that one proton is acquired by
to bacteriopheophytinA (BpheoA), forming UQ–B and the second proton is acquired when the sec-
BpheoA− (Fig. 6.3, step 3). [As reflected by the ond electron arrives. Source: Nicholls, D. G., and S.
question mark in step 2 of Fig. 6.3, the exact J. Ferguson. 1992. Bioenergetics 2. Academic Press,
route of the electron is not known. It has not London.
www.ebook777.com
been unequivocally demonstrated that the elec- The contribution to the ∆p by the
tron moves through the BchlA monomer. Some reaction center
researchers suggest that the electron travels When reaction centers absorb light energy,
from Bchl2* to BchlA and very quickly moves on a ∆Ψ and a proton gradient are created. The
to BpheoA, whereas other investigators pos- reason for the ∆Ψ is that the electrons move
tulate that the electron moves directly from electrogenically from cytochrome c2 to UQB
(Bchl*)2 to BpheoA.] Then the electron moves across the membrane from outside to inside
to ubiquinone bound to site A on the cytoplas- (Fig. 6.2). (This will produce a ∆Ψ, outside
mic side of the reaction center (UQA) forming positive, because the inward movement of a
−
the semiquinone, UQ A (Fig. 6.3, step 4). At negative charge is equivalent to the outward
this time an electron is returned to (Bchl)2+ from movement of a positive charge.) The creation
reduced cytochrome c2, which is periplasmic of the membrane potential is detected by a
(step 5). Then, the electron is transferred from shift in the absorption spectra of membrane
UQ A− to UQB to form UQ B− (Fig. 6.3, step 6). carotenoids. (For additional explanation, see
Electron flow is therefore across the membrane note 15.) The value of this technique is that very
from periplasmic c2 to UQB located on the rapid changes in membrane potential can be
cytoplasmic side. This generates a membrane monitored. The ∆Ψ produced by the reaction
potential, outside positive. A second light reac- center is delocalized, and increases in the ∆Ψ
tion occurs, and steps 1 through 6 in Fig. 6.3 are are made in the bc1 complex. Additionally, two
2−
repeated so that UQB has two electrons (UQ A ) . protons are taken from the cytoplasm to reduce
Two protons are picked up from the cytoplasm UQB to UQH2. Eventually, the two protons will
to form UQH2, which leaves the reaction cen- be released on the outside during the oxidation
ter to join the quinone pool in the membrane. of UQH2 by the bc1 complex. Thus, the reaction
(Although Fig. 6.3 indicates that two protons center contributes to both the ∆Ψ and the ∆pH
enter after both electrons have arrived at UQB, it components of the ∆p, as well as creating a ∆E
has been suggested that one proton enters after between UQ/UQH2 and c2,ox/c2,red.
UQB has received the first electron.) Eventually,
the protons carried by UQH2 are released on Summary of photosynthesis by purple
the periplasmic side during oxidation of UQH2 photosynthetic bacteria
by the bc1 complex.
Thus, the reaction center and the bc1 complex ADP + Pi ⎯⎯⎯⎯⎯
light and Bchl
→ ATP + H 2O
cooperate to translocate protons to the outside.
As explained earlier, UQB is called a two-elec- 6.2.3 Source of electrons for growth
tron gate because it cannot leave the reaction NADH and NADPH are important sources of
center until it has accepted two electrons. The electrons for growth but serve separate func-
length of time that it takes the electron to travel tions. NADH donates electrons to the respi-
from (Bchl)2 to UQA is a little more than 200 ps. ratory chain, resulting in ATP synthesis, and
[One picosecond (ps) is one trillionth of a sec- NADPH donates electrons in biosynthetic reac-
ond, or 10–12 s.] The rates of subsequent steps tions. Although they serve different functions,
are slower (but still very fast), being measured interconversion is possible via phosphorylation
in microseconds. [One microsecond (1 μs) is a of NAD+, dephosphorylation of NADP+, and
millionth of a second, 10–6 s.] transhydrogenation. When the electron donor
Determining the pattern and timing of electron is of a higher potential than the NAD+/NADH
transfer in the reaction center requires the use of couple, energy is required to reduce NAD+. For
picosecond laser pulses and very rapid recording example, this is the case for the purple photo-
of the absorption spectra of the electron carriers. synthetic bacteria that use certain inorganic
Interestingly, only one side of the reaction center sulfur compounds or succinate as a source of
(the A side) appears to be involved in electron electrons. The purple photosynthetic bacte-
transport. For example, there is photoreduction ria use the ∆p created by light energy to drive
of only one of the bacteriopheophytin mole- electron transport in reverse (Fig. 6.4). During
cules. Why only one branch appears to function reversed electron flow, ubiquinol reduces NAD+
in electron transport is not known. via the NADH:ubiquinone oxidoreductase.
photosynthesis 183
Possessing iron–sulfur centers as terminal

electron acceptors
Reactions centers whose terminal electron
acceptors are iron–sulfur centers are called type
I to distinguish them from the reactions centers
in the purple photosynthetic bacteria, which
use quinones as the terminal electron acceptors
and are called type II.
Reducing NAD(P)+ instead of quinone

The reaction centers of green sulfur bacteria are
Fig. 6.4 Relationship between cyclic electron flow similar to the type I reaction centers of cyanobac-
and reversed electron transport in the purple pho- teria and chloroplasts. However, the principles
tosynthetic bacteria. The electron is driven by light
underlying the transformation of electrochemi-
from a cytochrome c2 (c2) to a ubiquinone (UQ) and
cal energy into a ∆E and a membrane potential
then returns via electron carriers to the cytochrome c2
are the same as for the purple photosynthetic
with production of a ∆p. The ∆p is used to drive ATP
synthesis as well as reversed electron transport. bacteria.16,17 Light energizes a bacteriochloro-
phyll a molecule (P840), which reduces a primary
acceptor (A0), establishing a redox potential dif-
(See Section 4.7.1 for a discussion of reversed ference greater than 1 V (Fig. 6.5). The primary
electron transport.) electron acceptor (A0) in the green sulfur bacte-
Figure 6.4 illustrates the situation in which ria has recently been reported to be an isomer of
light energizes the creation of a ∆p (proton motive chlorophyll a called bacteriochlorophyll 663.18
force), which in turn energizes the synthesis of It might be added that the primary acceptor of
ATP as well as reversed electron flow to reduce heliobacteria is also a chlorophyll a derivative,
NAD+. However, another scenario has been hydroxychlorophyll a, reflecting the similari-
suggested, which is useful to consider because ties known to exist between the reaction centers
it illustrates the role that an increase in the ∆p of the green sulfur bacteria, heliobacteria, and
can play in slowing the rate of electron transfer. photosystem I of chloroplasts and cyanobacte-
It should be recalled that ubiquinone can both ria. (See note 19.) The electron flows from A0 to
accept electrons from bacteriopheophytin dur- a quinonelike acceptor called A1. The electron
ing cyclic electron transport and reduce the bc1 is then transferred from A1 through three iron–
complex, which generates a ∆p (Fig. 6.1). It has sulfur centers to ferredoxin. There can be both
been suggested that as the ∆p grows larger, it cyclic and noncyclic electron flow. In cyclic flow
might exert “back-pressure” on the oxidation the electron returns to the reaction center via
of ubiquinol by the bc1 complex, thus slowing menaquinone (MQ) and a bc1 complex, creat-
down the oxidation of ubiquinol via this route ing a ∆p. In noncyclic electron flow inorganic
and making it available for NAD+ reduction via sulfur donates electrons that travel through the
reversed electron transport.7 To the extent that reaction center to NAD+.
this might occur, the electron donor (succinate According to the scheme in Fig. 6.5, electrons
or inorganic sulfur compounds) would replen- from inorganic sulfur enter at the level of cyto-
ish electrons to the bacteriochlorophyll via chrome c and the bc1 complex is bypassed. This
either ubiquinone or cytochrome c. is a widely held view based upon available data
obtained by using isolated oxidoreductases.
6.3 The Green Sulfur Bacteria From an energetic point of view, however, it
(Chlorobiaceae) is wasteful because a coupling site is bypassed,
6.3.1 Photosynthetic electron transport even though the E′m values of some of the sulfur
The green sulfur photosynthetic bacteria and couples are low enough to reduce menaquinone.
the heliobacteria have reaction centers that are For example, the E′m for sulfur/sulfide (n = 2) is
distinguished from the reaction centers of the –0.27; for sulfite/sulfide (n = 6) it is –0.11 V; and
purple photosynthetic bacteria in two respects, for sulfate/sulfite (n = 2) it is –0.54 V. All these
as detailed in the subsections that follow. couples are at a potential low enough to reduce
www.ebook777.com
Fig. 6.5 A model for electron flow in the green sulfur bacteria. Both cyclic and noncyclic flows of electrons
are possible. Reaction center bacteriochlorophyll a (P840) becomes energized and reduces A0, which is bacte-
riochlorophyll 663. The electron then travels to A1, a quinonelike molecule, and then through two or three
iron–sulfur centers (FeS). The first reduced product outside the reaction center is the iron–sulfur protein ferre-
doxin (Fd). The ferredoxin reduces NAD(P)+ in the noncyclic pathway. Cyclic flow occurs when the electron
reduces menaquinone (MQ) instead of NAD(P)+ and returns to the reaction center via a bc1 complex and
cytochrome c555. A ∆p is created in the bc1 complex. In the noncyclic pathway the electron donor is a reduced
inorganic sulfur compound, here shown as hydrogen sulfide. Elemental sulfur or thiosulfate can also be used.
The inorganic sulfur is oxidized by cytochrome c555, which feeds electrons into the reaction center.
menaquinone, which has an E′m of –0.074 V. flow also occurs in the green sulfur bacteria,
However, the point of entry of the electron is as described in Section 6.3.1.)
still an unresolved issue.20
6.4.1 Two light reactions
6.3.2 Summary of photosynthesis by As we shall see, chloroplasts and cyanobacteria
green sulfur bacteria have essentially combined the light reactions of
purple photosynthetic bacteria and green sulfur
H 2 S + NAD+ + ADP + Pi ⎯⎯⎯⎯⎯
light and Bchl
→ photosynthetic bacteria in series, so that two
S° + NADH + H + + ATP light reactions energize a single electron that
energizes ATP synthesis and reduces NADP+.
(These organisms can also use elemental sulfur, The initial evidence for two light reactions came
oxidizing it to sulfate.) from early studies of photosynthesis performed
with algae by Emerson and his colleagues.21,22
6.4 Cyanobacteria and Chloroplasts They observed that the efficiency of photosyn-
thesis, measured as the moles of oxygen evolved
Photosynthesis in cyanobacteria and chloro-
per einstein absorbed (i.e., the quantum yield)
plasts differs from photosynthesis discussed
is high over all the wavelengths absorbed by
thus far in three important respects:
chlorophyll and the light-harvesting pigments,
1. H2O is the electron donor and oxygen is but it drops off sharply at 685 nm even though
evolved. chlorophyll continues to absorb light between
2. There are two light reactions in series, hence 680 and 700 nm (Fig. 6.6). This became known
two different reaction centers. as the “red drop” effect because 700 nm light
3. Electron flow is primarily noncyclic, pro- is red. One can restore the efficiency of pho-
ducing both ATP and NADPH. (Noncyclic tosynthesis of 700 nm light by supplementing
photosynthesis 185
Fig. 6.6 Quantum yield of photosynthesis. When the rate of O2 evolution per quantum absorbed is plotted
against wavelength, the rate drops off sharply above 680 nm. Source: After Stryer, L. 1988. Biochemistry. W.
H. Freeman, New York.
it with light at shorter wavelengths (e.g., 600 Chlorophyll a, with a major absorption peak at
nm). This is explained by pointing out that 680 nm (i.e., P680—probably a dimer), becomes
there are two reaction centers, one energized by energized and reduces an acceptor molecule, which
light at a wavelength of around 700 nm (reac- is pheophytin (pheo). Having lost an electron,
+
tion center I, or RC I) and one energized by the P680 /P680 couple has a redox potential (esti-
lower wavelengths of light (reaction center II, mated at about +1.1 V) high enough to replace the
RC II). These are also called photosystems I and lost electron with one from water, thus oxidizing
2 H 2 O to 4 O2 . The energized electron travels to
1 1
II (PS I and PS II). Photosystems I and II oper-
ate in series, and therefore both must be ener- pheophytin, which has an E m′ of about –0.6 V. The
gized to maintain the electron flow from water pheophytin reduces plastoquinone (PQ), which is
to NADP+. The lower wavelengths of light can structurally similar to ubiquinone (Fig. 5.3D).
energize both reaction centers, but 700 nm light A two-electron gate, similar to the two-elec-
can energize only reaction center I. It is for this tron gate in the reaction center of the purple
reason that 700 nm light is effective only when photosynthetic bacteria, operates during qui-
given in combination with supplemental doses none reduction. Electrons leave the reaction
of shorter wavelengths. center in PQH2 and are transferred to a copper-
containing protein called plastocyanin (Pc)
6.4.2 Photosynthetic electron transport through a b6f complex (structurally and func-
A schematic drawing of the overall pattern of tionally similar to the bc1 complex, except that
electron flow in cyanobacteria and chloroplasts is cytochrome f, which has a c-type heme, replaces
shown in Fig. 6.7. These systems have two reaction cytochrome c1 and there are some differences in
centers, PS I and PS II, connected by a short elec- the cytochrome b). A ∆p is created by the b6f
tron transport chain that includes the analogue to complex via a Q cycle similar to the respiratory
+
the bc1 complex (i.e., the b6f complex) that carries bc1 complex. The plastocyanin reduces P700 in
out a Q cycle similar to that carried out by the bc1 reaction center I, previously oxidized by the
complex discussed in Chapter 5. (For a description light reaction we shall describe next.
of the b6f complex, see ref. 23 and note 24.) Let us Photosystem I is similar to the type I reaction
begin with PS II, in which the pathway of electron center in the green photosynthetic bacteria (see
flow is the same as in the type II reaction center in Fig. 6.5). A photon of light energizes P700, which
the purple photosynthetic bacteria (see Fig. 6.1A). is chlorophyll a with a major absorption peak
www.ebook777.com
Fig. 6.7 Photosynthesis in cyanobacteria and chloroplasts. Light stimulates electron flow from water through
reaction center II (RC II) to reaction center I (RC I) to NADP+. Cyclic flow is possible by using RC I from
ferredoxin through the b6f complex. Abbreviations: P680, chlorophyll a with a major absorption peak at 680
nm; Pheo, chlorophyll pheophytin; PQ, plastoquinone; b6f, cytochrome b6f complex; Pc, plastocyanin; P700,
chlorophyll a with a major absorption peak at 700 nm; A0, chlorophyll a; A1, phylloquinone; FeS, one of sev-
eral iron–sulfur centers; Fd, ferredoxin.
at 700 nm. The P700*

reduces a chlorophyll a mol- which is the pathway for reducing CO2 to car-
ecule, called A0. For this initial redox reaction, bohydrate in oxygenic phototrophs, requires
the energized electron travels from an E m′ of three ATPs for every two NADPHs to reduce
+
about +0.5 V ( P700 /P700 ) to an E m′ of about –1.0 one CO2 to the level of carbohydrate.
–
V( A 0 /A 0 ). Note that this is a far more negative
potential than that generated in reaction center 6.4.3 Summary of photosynthesis by
II. From A0 the electrons travel to A1, which is green plants, algae, and cyanobacteria
a phylloquinone. Phylloquinones have a struc-
H 2O + NADP + + ADP + Pi ⎯⎯⎯⎯⎯
light and Chl
→
ture similar to menaquinone (Fig. 5.3), but with
only one double bond in the isoprenoid chain.
1
2 O2 + NADPH + H + + ATP
The electron is transferred from the quinone
through several iron–sulfur centers (FeS), which 6.5 Efficiency of Photosynthesis
reduce the iron–sulfur protein ferredoxin (Fd) In this section, as an exercise, we calculate the
that is outside the reaction center. Ferredoxin in efficiencies of photosynthesis based upon input
turn reduces NADP+. Thus, the two light reac- light energy and products of photosynthesis.
tions in series energize electron flow from H2O The calculated efficiencies are only approxima-
to NADP+, which is over a net potential differ- tions based upon assumptions regarding ATP
ence of about 1.1 V. This is called noncyclic yields, actual redox potentials, standard free
electron flow because the electron never returns energies, and so on.
to the reaction center. However, cyclic electron
flow is possible. 6.5.1 ATP synthesis
There is a branch point at the ferredoxin step, Basically, photosynthesis is work done by ener-
and it is possible for the electron to cycle back to gized electrons. The work is the phosphoryla-
reaction center I via the b6f complex, augment- tion of ADP and the reduction of NAD(P)+. Each
ing the ∆p, rather than reducing NADP+. This electron is energized by a photon (quantum) of
may be a way to increase the amounts of ATP light, and each mole of electrons by an einstein
made relative to NADPH. The Calvin cycle, (6.023 × 1023 quanta) of light. It is instructive
photosynthesis 187
to ask how much of this energy is conserved in 6CO2 + 12H2O C6H12O6+ 6H2O + 6O2
ATP and NADPH. ∆G′0 = 2,870 kJ
Let us consider the synthesis of ATP. Assume
Therefore, for every mole of O2 produced, the
that each energized electron results in the trans-
standard free energy requirement at pH 7 is
location of two protons (e.g., during the Q cycle
2,870/6, or 478 kJ. The number of einsteins of
in the bc1 complex) but that three protons must
light required to produce one mole of O2 is 8.
reenter through the ATP synthase to make one
An einstein of 680 nm light carries 176 kJ of
ATP. As mentioned before, a value of 3H+/ATP
energy. Therefore, 176 × 8, or 1,408 kJ of light
is reasonable in light of experimental data. Thus,
energy, is used to produce one mole of O2. The
each energized electron (or each photon) results
efficiency is thus (478/1,408)(100), or 34%.
in the synthesis of two-thirds of an ATP. How
much energy is required to synthesize two-thirds
of an ATP? The ∆Gp for the phosphorylation of 6.6 Photosynthetic Pigments
one mole of ADP to make ATP is about 45 kJ. 6.6.1 Light-harvesting pigments
Therefore, two-third of a mole of ATP should The photosynthetic pigments are divided into
require approximately +45( 23 ) kJ = 30 kJ. An two categories: reaction center pigments (pri-
einstein of 870 nm light, which corresponds to marily chlorophylls) and light-harvesting pig-
the absorption maximum of bacteriochloro- ments (carotenoids, phycobilins, chlorophylls).
phyll a (found in purple phototrophs), has 138 The light-harvesting pigments are sometimes
kJ of energy. Therefore, the efficiency is (30/138) called accessory pigments or antennae pigments.
(100), or about 22%. One can also calculate By far, most of the photosynthetic pigments are
the efficiency by using electron volts instead of light-harvesting pigments. The light-harvesting
joules. The energy in a photon of light at 870 nm pigments are critical to photosynthesis because
is 1.43 eV. The synthesis of two-thirds of a mole they absorb light of different wavelengths and
of ATP requires 30,000 J. Dividing this number funnel the energy to the reaction center. Figure 6.8
by the Faraday constant gives the energy in elec- shows whole-cell absorption spectra of a purple
tron volts, (i.e., 0.31 eV). photoysynthetic bacterium and a cyanobacte-
rium, including the absorption wavelengths of
6.5.2 ATP and NADPH synthesis the pigments; the various pigments and their
What about photosystems that reduce NADP+ absorption peaks are listed in Table 6.2. The
as well as make ATP (i.e., photosystems I and nature of the light-harvesting pigments will vary
II)? The E m′ for O2/H2O is +0.82 V, and for with the type of organism, but some important
NADP+/NADPH it is –0.32 V. Therefore, the generalizations about them can be made:
energized electron must have 0.82 – (–0.32), or 1. They absorb light at wavelengths different
1.14 eV, to move from water to NADP+. (The from reaction center chlorophyll and there-
answer would not be very different if Eh values fore extend the range of wavelengths over
were used instead of E m′ .) If two-thirds of an which photosynthesis is possible because
ATP is made, then a total of 0.31 + 1.14 = 1.45 they transfer energy to the reaction center
eV would be required to make two-thirds of an (Table 6.2).
ATP and half an NADPH. A photon of light at 2. In the purple photosynthetic bacteria, they
a wavelength of 680 nm (the major long-wave- are embedded in the membrane as pigment–
length absorption peak of chlorophyll a in reac- protein complexes that are in close physical
tion center II) has 1.82 eV. Two photons, or the association with the reaction center.
equivalent of about 3.6 eV, are used. Therefore, 3. In the green sulfur bacteria and Chloroflexus,
approximately 40% of the light energy is con- they exist in chlorosomes, which are sepa-
served as ATP and NADPH.
rate “organelles,” also called inclusion bod-
ies, attached to the inner surface of the cell
6.5.3 Carbohydrate synthesis and oxygen membrane.
production 4. In the cyanobacteria (and eukaryotic red
One can also estimate the approximate effi- algae), the light-harvesting pigments (phy-
ciency by considering the number of light cobilins) that transfer energy to PS II are
quanta required to produce oxygen: localized in granules called phycobilisomes
www.ebook777.com
Fig. 6.8 Absorption spectra of a purple photosynthetic bacterium, Rhodopseudomonas palustris (A), and a
cyanobacterium (B). In spectrum A, bacteriochlorophyll a peaks at 360, 600, 805, and 870 nm. Carotenoid
peaks at 450–525 nm. In spectrum B, chlorophyll a peaks at 440 and 680 nm. The phycocyanin peak is
around 620 nm. Source: Adapted from Brock, T. D., and M. T. Madigan. 1988. Biology of Microorganisms.
Reprinted by permission of Prentice Hall, Upper Saddle River, NJ.
Table 6.2 In vivo long-wavelength absorption max- that are attached to the photosynthetic
ima of photosynthetic pigments of phototrophic membranes (also called thylakoids). It
prokaryotes should be pointed out that as in other pho-
Pigment Peak wavelength (nm) tosynthetic organisms, cyanobacteria also
use carotenoids as light-harvesting pig-
Chlorophyll
ments. Carotenoids have been found in
A 680 isolated complexes of photosystems I and
B 675 II from cyanobacteria, and fluorescence
Bacteriochlorophyll
data suggest that carotenoids absorb light
A 800–810, 850–910 and transfer energy to reaction center
B 835–850, 1,020–1,035
C 745–760 chlorophyll in PS I.25 (See Section 6.7.2.)
D 725–745 Carotenoids in the reaction centers also
E 715–725 serve to protect the chlorophyll from pho-
G 670, 788 tooxidation. (See note 14.)
Carotenoid
Chlorobactene 458
Isorenieratene 517 Light-harvesting pigments of the purple
Lycopene, rhodopin 463, 490, 524 photosynthetic bacteria
Okenone 521
Rhodopinal 497, 529
The light-harvesting complexes of the purple
Spheroidene 450, 482, 514 photosynthetic bacteria are bacteriochloro-
Spirilloxanthin 486, 515, 552 phyll–protein–carotenoid complexes localized
β-Carotene 433, 483 in the cell membrane in close association with
γ-Carotene 433, 483 the reaction centers. The pigment–protein com-
Phycobilins plexes absorb light in the range of 800 to 1,000
Phycocyanin 620–650 nm (i.e., in the near-infrared range). Upon treat-
Phycoerythrin 560–566 ment of the photosynthetic membranes with a
Source: Stolz, J. F. 1991. Structure of Phototrophic mild detergent [e.g., lauryl dimethylamine-N-
Prokaryotes. CRC Press, Boca Raton, FL. oxide (LDAO)], the light-harvesting complexes
photosynthesis 189
Fig. 6.9 Light-harvesting complexes LHC I and LHC II in the purple nonsulfur photosynthetic bacterium R.
sphaeroides. An arrow shows transfer of energy to reaction center (RC). Each cylinder represents a polypeptide
unit. The small solid circles symbolize bound pigments. Excitation energy begins in the LHC II units and then
travels to the LHC I units. The LHC I units transfer the excitation energy to the bacteriochlorophyll dimer in the
reaction center. Source: Nicholls, D. G., and S. J. Ferguson. 1992. Bioenergetics 2. Academic Press, London.
can be separated from the reaction centers and membrane but rather in chlorosomes.26 These
analyzed. In Rhodobacter sphaeroides there are interesting inclusion bodies are shown later
two light-harvesting complexes (LHC I and LHC (Fig. 6.16), lining the inner surface of the cell
II), each containing two polypeptides (α and β) membranes of Chlorobium, Prosthecochloris,
to which the pigments are attached (Fig. 6.9). and Chloroflexus. Surrounded by a unit mem-
Analysis of the hydrophobic domains of the brane consisting of a lipid called galactolipid
polypeptides suggests that they span the mem- (sugar molecules called galactose are attached
brane. It should be emphasized that pigment to the lipid) and some protein, the chlorosomes
molecules must be positioned correctly with are about 150 × 70 × 30 nm3. They are attached
respect to one another if effective energy trans- to the cell membrane by a baseplate consisting
fer is to take place between them (Section 6.7). of BChla located on the membrane surface of
Organization on protein molecules probably the chlorosome. Baseplates are drawn as the
accomplishes this. When light is absorbed by the shaded portion of the chlorosome in Fig. 6.10.
light-harvesting pigments, the energy is quickly The chlorosomes contain all the bacteriochlo-
transferred to the reaction center (Fig. 6.9). rophylls c, d, or e, much of the carotenoid, and
some of the bacteriochlorophyll a. Light energy
Light-harvesting pigments in the green is captured by the chlorosome, transferred to
photosynthetic bacteria the baseplate, and from there to the reaction
The major light-harvesting pigment of the center (RC) in the cell membrane.
green photosynthetic bacteria is bacteriochlo- Within the chlorosomes are rod-shaped ele-
rophyll c, d, or e, depending upon the species ments that run lengthwise through the chlo-
(see later: Fig. 6.14B). (The green sulfur bacteria rosome; in Chlorobium, they are about 10 nm
may contain Bchl c and/or d or e, but the green in diameter. The rods are composed of a poly-
nonsulfur photosynthetic bacteria have only peptide to which the bacteriochlorophyll c, d,
bacteriochlorophyll c.) In addition, there is bac- or e is thought to be attached. However, some
teriochlorophyll a. As with the purple photo- investigators have suggested that the bacteri-
synthetic bacteria, the bacteriochlorophylls are ochlorophyll exists as self-aggregated clusters
associated in some way with carotenoids and in the chlorosomes rather than attached to pro-
protein. tein. Bacteriochlorophyll a exists as a protein–
In contrast to the purple bacteria, most of pigment complex in the baseplate. In addition,
the light-harvesting pigments in the green there is bacteriochlorophyll a in the membrane,
photosynthetic bacteria are not in the cell which is bound to the reaction centers. Most of
www.ebook777.com
nm owing to open-chain tetrapyrroles (phy-

cobilins) covalently bound to the protein via
thioether linkages to a cysteine residue in the
phycobiliprotein (Fig. 6.12). Examination of
RC RC the absorption spectra of chlorophylls and caro-
tenoids reveals that the absorbance is relatively
Fig. 6.10 Structure of the chlorosome. The chlo- poor, lying in the range (500–600 nm) in which
rosome is surrounded by a galactolipid layer that the phycobiliproteins absorb.
contains some protein and is attached to the reaction There are three classes of phycobiliproteins:
centers in the cell membrane via a baseplate (shaded phycoerythrin, phycocyanin, and allophyco-
area). There is bacteriochlorophyll a in the baseplate cyanin. Allophycocyanin is usually at a much
and in the membrane. One model for the inside of lower concentration than phycoerythrin and
the chlorosomes postulates rod-shaped proteins to
phycocyanin and, as will be described shortly, is
which is attached the major light-harvesting pig-
considered to be an energy funnel to the reaction
ment, bacteriochlorophyll c, d, or e. In an alternative
model, the bacteriochlorophyll exists as an aggregate center. Although most species of cyanobacteria
not attached to protein. It is suggested that light is contain both phycoerythrin (red) and phyco-
absorbed by bacteriochlorophyll in the interior of the cyanin (blue), the phycocyanins generally pre-
chlorosome and the energy is transmitted to bacteri- dominate. The phycobiliproteins are arranged
ochlorophyll a in the baseplate, then to bacteriochlo- in layers around each other with phycoerythrin
rophyll a attached to the reaction centers, and from on the outside, phycocyanin in the middle, and
there to reaction center bacteriochlorophyll a. allophycocyanin in the inside, closest to the
reaction center.
A simplified drawing of a phycobilisome is
the light is absorbed primarily by the bacteri- shown in Fig. 6.11A. Figure 6.11B represents
ochlorophyll c, d, or e in the chlorosome, and a detailed model at the molecular level of the
the energy is transferred to bacteriochlorophyll phycobilisomes from the cyanobacterium
a in the baseplate. The bacteriochlorophyll in Synechococcus, which does not contain phyco-
the baseplate then transfers the energy to mem- erythrin. These are fan-shaped (hemidiscoidal)
brane bacteriochlorophyll a, which transfers the phycobilisomes that consist of a dicylindrical
energy to reaction center bacteriochlorophyll a. core (A and B) containing allophycocyanin and
The green sulfur photosynthetic bacteria have a six cylindrical phycocyanin rods radiating from
reaction center based on an iron–sulfur protein the core. (In other cyanobacteria the core may
and resembling reaction center I of cyanobac- consist of three cylinders.) If phycoerythrin cyl-
teria and chloroplasts; the green nonsulfur inders were present, they would be an extension
(filamentous) photosynthetic bacteria have a of the phycocyanin rods. The shorter wave-
quinone-based reaction center resembling that lengths of light are absorbed by the phycoeryth-
of the purple photosynthetic bacteria. rin, the longer wavelengths by phycocyanin,
and the still longer wavelengths (650 nm) by the
allophycocyanin. These pigments are thus well
Light-harvesting pigments in the
positioned to transfer energy via inductive reso-
cyanobacteria
nance to the reaction center. (See Section 6.7 for
The photosynthetic membranes of cyanobacte-
a discussion of energy transfer.)
ria are actually intracellular membrane-bound
A second kind of light-harvesting complex is
sacs called thylakoids that have both reaction
located in the thylakoids and transmits energy
centers and light-harvesting complexes called
to the closely associated PS I. It consists of chlo-
phycobilisomes.27,28 The phycobilisomes are
rophyll a, carotenoids, and protein.
numerous granules covering the thylakoids
and attached to PS II (Fig. 6.11). They can be
removed from the photosynthetic membranes Light-harvesting pigments in algae
with nonionic detergent (Triton X-100) for All the algae contain chlorophyll a and the caro-
purification and analysis. The phycobilisomes tenoid β-carotene. In addition, there are more
consist of proteins called phycobiliproteins, than 60 other carotenoids found among algae,
which absorb light in the range of 450 to 660 some of which are shown in Fig. 6.13.
photosynthesis 191
A light
thylakoid
PS I PS II PS I
membrane
A B
TM
1 2 3
core rod
Fig. 6.11 Structure of phycobilisomes. (A) Hemispherical phycobilisome common in red algae. These light-
harvesting complexes are approximately 48 nm × 32 nm × 32 nm in size. The major pigment is phycoerythrin.
Light energy absorbed by the pigments in the outer region is transferred to the reaction center through an inner
core: 1, phycoerythrin; 2, phycocyanin; 3, allophycocyanin. (B) Detailed model of phycobilisome from the
cyanobacterium Synechococcus. Phycobilisomes from cyanobacteria are generally fan shaped, about 60 nm
wide, 30 nm high, and 10 nm thick. The major pigment is phycocyanin. The outermost region of the phycobili-
some consists of six cylindrical rods of phycocyanin. Each rod is made from three hexamers of phycocyanin: 1,
2, and 3. (Synechococcus does not have phycoerythrin.) The innermost region of the phycobilisome (the core)
consists of two closely associated cylinders (A and B) of allophycocyanin. Each cylinder of allophycocyanin
consists of two hexamers (not shown). The cross-hatched rectangle within the thylakoid membrane (TM)
represents the reaction center and associated chlorophylls. The solid ovals between the core and the thylakoid
membrane represent high molecular weight polypeptides thought to anchor the phycobilisome to the thyla-
koid membrane. There is a linker polypeptide associated with each phycocyanin hexamer that is drawn as an
open circle between the hexamers. The open triangle at the terminus of each rod represents a polypeptide that
may terminate the rod substructure. Sources: Adapted from Gantt, E. 1986. Phycobilisomes, pp. 260–268.
In Photosynthesis III: Photosynthetic Membranes and Light Harvesting Systems. L. A. Staehelin and C. J.
Arntzen (Eds.). Springer-Verlag, Berlin. Grossman, A. R., M. R. Schaefer, G. G. Chiang, and J. L. Collier.
1993. The phycobilisome, a light-harvesting complex responsive to environmental conditions. Microbiol.
Rev. 57:725–749.
1. The green algae (Chlorophyta) contain chlo- II) associated with PS II. The inner antenna
rophyll b and xanthophylls (modified caro- complex that is part of the reaction centers is
tenes) called lutein. the same in all algae and higher plants, but LHC
2. The brown algae (Chromophyta) have chlo- I and LHC II vary in composition and structure.
rophyll c1 and c2, as well as xanthophylls The main functions of LHC I and II are to har-
called fucoxanthin and peridinin. vest light and to transmit its energy to the pig-
3. The red algae (Rhodophyta) have phyco- ment complex in the reaction center.
bilins (e.g., phycoerythrobilin) (Fig. 6.12),
chlorophylls, carotenes, and xanthophylls. 6.6.2 Structures of the chlorophylls,
The pigment–protein complexes can be divided bacteriochlorophylls, and carotenoids
into two classes. One class, the inner antenna Chlorophylls and bacteriochlorophylls
β-carotene–Chl a–protein complex, exists as The chlorophylls are substituted tetrapyrroles
part of photosystems I and II. The second class, related to heme (Figs. 5.5 and 6.14A). The
the outer antenna pigment–protein class, forms differences are that chlorophylls have Mg2+
light-harvesting complex I (LHC I) associated in the center of the tetrapyrrole rather than
with PS I, and light-harvesting complex II (LHC Fe2+(3+), and a fifth ring (ring V) is present. The
www.ebook777.com
1996.29 It is present in an acidophilic pho-

tosynthetic bacterium called Acidiphilium
rubrum.
Carotenoids
Carotenoids are long isoprenoids, usually
C40 tetraterpenoids made of eight isoprene
units (Fig. 6.13). (For other isoprenoids, see
note 30.) Carotenoids have a system of conju-
gated double bonds, which accounts for their
absorption spectrum in the visible range at
wavelengths poorly absorbed by chlorophylls
and makes them valuable as light-harvesting
pigments. Light energy absorbed by carote-
noids is transferred to the reaction center chlo-
rophyll (Section 6.7). Carotenoid pigments
with six-membered rings at both ends of the
Fig. 6.12 Structure of phycoerythrobilin and phy-
molecule are called carotenes (e.g., β-carotene,
cocyanobilin. In ring A, the protein is covalently
Fig. 6.13). Modification of carotenes (e.g.,
bonded via a thioether linkage between one of its
cysteine residues and C2 of CH3–CH=. by hydroxylation) produces xanthophylls,
which comprise a major portion of the light-
harvesting pigments in algae. Carotenoids
different substitutions on the pyrrole rings perform a function in addition to serving as
and the number of double bonds in ring II a light-harvesting pigment. As described in
distinguish the chlorophylls from each other. Section 6.2.2 and note 14, they protect against
A Zn–bacteriochlorophyl a was reported in photooxidation.
Fig. 6.13 Structure of carotenoids. An isoprene molecule is shown at the top. One end can be called the head (h)
and the other the tail (t). All carotenoids are made of isoprene subunits with a “tail-to-tail” connection in the
middle. Carotenoids in which the ends have circularized into rings are called carotenes. Carotenes that have been
modified (e.g., by hydroxylation) are called xanthophylls. Lycopene is the carotenoid responsible for the red color
in tomatoes. β-Carotene is part of the inner antenna complex in all plants and algae. Lutein is a xanthophyll found
in algae. Isorenieratene is a carotenoid found in green sulfur bacteria.
photosynthesis 193
Fig. 6.14 (A) Structure of bacteriochlorophyll. There is no double bond in ring II of bacteriochlorophylls a,
b, and g between C3 and C4. (Nitrogen is position 1.) (B) Table identifying the R groups distinguishing the
compounds schematized in (A).
6.7 The Transfer of Energy from the resonance transfer, electrons do not travel
Light-Harvesting Pigments to the between complexes, but energy is transferred.
Essentially, when the electron in the excited
Reaction Center
molecule drops back to its ground state, the
6.7.1 Mechanism of energy transfer energy is transferred to a molecule close by,
When light is absorbed in one region of the pig- resulting an electron in that molecule being
ment system, excitation energy is transferred raised to the excited singlet state. This can
to other regions and eventually to the reaction occur if there is an overlap between the fluo-
center. The mechanism that probably accounts rescence spectrum of the donor molecule and
for the transfer of excitation energy from one the absorption spectrum of the acceptor, and
pigment complex to another throughout the if the molecules are in the proper orientation
pigment system is called inductive resonance to each other. That is the reason for the critical
transfer. importance of the steric relationships of the
light-harvesting pigments to each other and to
Inductive resonance transfer the reaction center. It must be emphasized that
In an unexcited molecule, all the electrons transfer by this mechanism does not occur as a
occupy molecular orbitals having the lowest result of emission and reabsorption of light.
available energy (the ground state). When
a photon of light is absorbed, an electron Delocalized excitons
becomes energized and occupies a higher A second method of excitation transfer occurs
energy orbital that had been empty. This usu- over very short distances (i.e., <2 nm). When
ally occurs without a change in the spin quan- the excited electron is raised to the singlet state,
tum number (Fig. 6.15). The electron is now it is said to leave a positive “hole” in the ground
in the excited singlet state. The excitation can state (Fig. 6.15). The combination of the
be transferred from one pigment complex excited electron and the positive hole is called
to another via inductive resonance. During an exciton. If two or more similar molecules are
www.ebook777.com
Fig. 6.15 Light raises an electron to a higher energy level in an unoccupied orbital. The electron is then said
to be in the excited singlet state. As the diagram shows, the electron may drop down to a lower excited singlet
state. The electron may return to the ground state, releasing energy as light (fluorescence) or heat. Under
appropriate conditions, in the process called inductive resonance, the energy made available when the elec-
tron returns to the ground state is not released but instead is used to raise an electron in a nearby pigment to an
excited singlet state, thereby transferring the energy. In the reaction center, electrons raised to a higher energy
orbital are transferred to a nearby acceptor molecule, creating a redox gradient.
very close to each other (<2 nm), the exciton 6.8 The Structure of Photosynthetic
migrates over the molecular orbitals belonging Membranes in Bacteria
to both molecules; that is, the exciton becomes
The anoxygenic phototrophic bacteria have
delocalized. Thus, the excitation energy is actu-
photosynthetic membrane structures of a vari-
ally shared by the group of interacting mol-
ety of types, depending upon the organism
ecules. However, delocalized excitons in the
(Fig. 6.16). For example, the purple photo-
pigment complexes extend only over very short
synthetic bacteria have numerous intracellular
distances and cannot connect the outer regions
photosynthetic membranes in which the light-
of the light-harvesting complex with the reac-
harvesting pigments and the reaction centers
tion center. On the other hand, the sharing of
are closely associated. The green bacteria sepa-
excitation energy by delocalized excitons is
rate most of the light-harvesting pigments into
more rapid than inductive resonance transfer
chlorosomes.
and is therefore expected to be an important
factor in the behavior of molecules at small
intermolecular distances. 6.9 Summary
When the energy from a photon of light reaches
6.7.2 Evidence that energy absorbed reaction center chlorophyll, an electron in the
by the light-harvesting pigments is chlorophyll becomes energized and is trans-
transferred to the reaction center ferred to a primary acceptor molecule within
When isolated chloroplasts are irradiated, the reaction center. This leaves the chlorophyll
fluorescence from reaction center chlorophyll in an oxidized form. The fate of the electron lost
(chlorophyll a) can be measured, but fluores- by the primary donor differentiates two differ-
cence from the light-harvesting pigments (chlo- ent types of reaction center. One type of reac-
rophyll b, carotenoids) cannot. This means that tion center reduces quinone. Quinone-reducing
the energy absorbed by the light-harvesting pig- reaction centers are the reaction centers in pur-
ments is efficiently transferred to the reaction ple photosynthetic bacteria, and reaction cen-
center chlorophyll, whereupon an electron in ters II in chloroplasts and cyanobacteria. In the
the reaction center chlorophyll becomes excited photosynthetic bacteria, the quinone is ubiqui-
to a higher energy level. In the absence of fluo- none. In chloroplasts and cyanobacteria, it is
rescence, the energized electron can reduce the plastoquinone. A second type of reaction cen-
primary electron acceptor. It has been suggested ter reduces an FeS protein. The latter reaction
that the initial redox reaction is initiated in the centers are reaction center I in chloroplasts and
reaction center chlorophyll by an electron in the cyanobacteria, and the reaction center in green
lowest excited singlet state.31 sulfur bacteria and heliobacteria.
photosynthesis 195
Fig. 6.16 Photosynthetic membranes in anoxygenic phototrophic bacteria. The green bacteria (Chlorobium,
Chloroflexus, Prosthecochloris) contain chlorosomes attached to the cytoplasmic surface of the cell mem-
brane. Most of the purple photosynthetic bacteria have extensive invaginations of the cell membrane that
can take the form of stacked lamellae (Ectothiorhodospira sp., Rhodopseudomonas palustris, R. viridis),
vesicles (Rhodospirillum sp., Rhodopseudomonas sp.), or tubules (Thiocapsa pfennigii). A few photosyn-
thetic bacteria do not have internal membranes or chlorosomes (Rhodospirillum tenue, Heliobacterium
chlorum). Source: Sprague, S. G., and A. R. Varga. 1986. Membrane architecture of anoxygenic photo-
synthetic bacteria, pp. 603–619. In: Photosynthesis III: Photosynthetic Membranes and Light-Harvesting
Systems. L. A. Staehelin and C. J. Arntzen (Eds.). Springer-Verlag, Berlin. With kind permission from Springer
Science+Business Media.
In the purple photosynthetic bacteria the In cyanobacteria and chloroplasts, the

reduced quinone transfers the electrons to a bc1 reduced quinone leaves reaction center II and
complex. The bc1 complex reduces cytochrome reduces the b6f complex, which is similar to the
c2, which returns the electron to the reaction bc1 complex. The b6f complex reduces plasto-
center. This is called cyclic electron flow. A ∆p cyanin, which transfers the electron to reaction
is generated by the bc1 complex. center I. A ∆p is established by the b6f complex.
www.ebook777.com
Having lost an electron, reaction center II 3. When bacteriochlorophyll absorbs a pho-

accepts an electron from water, thus producing ton of 870 nm light (about 1.4 eV of energy),
oxygen as a by-product of the light reactions in an electron travels out of the reaction cen-
the cyanobacteria and in chloroplasts. Reaction ter, through a bc1 complex, and back to
center I is also energized by light and trans- the reaction center (cyclic flow) with the
fers the electron to ferredoxin. The ferredoxin extrusion of two protons to the outer mem-
reduces NADP+. The combination of reaction brane surface at the site of the bc1 complex.
centers I and II is called noncyclic electron flow. Protons return to the cytoplasm through
In noncyclic flow, both ATP and NADPH are the ATP synthase with 3 as the H+/ATP
synthesized. Cyclic flow is also possible as ratio. Assuming that the ∆Gp is 45 kJ, what
the electron returns to reaction center I from is the efficiency of photophosphorylation?
reduced ferredoxin.
4. For all anoxygenic photosynthetic bacteria
The green sulfur photosynthetic bacteria
except the green sulfur bacteria, electrons
have a reaction center similar to reaction center
from the reductant do not pass through
I which reduces ferredoxin and NAD+. The elec-
the reaction center. This means that light
tron donors are inorganic sulfur compounds
does not stimulate electron flow from the
(e.g., sulfide, elemental sulfur). The electron can
reductant to NAD+. But many of the reduc-
return to the reaction center via a quinone and a
tants have reduction potentials more posi-
bc1 complex in cyclic flow rather than reducing
tive than that of the NAD+/NADH couple.
NAD+.
What energizes electron flow to NAD+? Use
Most of the light energy used in photosynthe-
ionophores to devise an experiment that
sis is not absorbed by reaction centers directly
would support your conclusion.
but is instead absorbed by light-harvesting pig-
ments organized either within the membrane 5. Energy from light absorbed by the acces-
and closely positioned to the reaction centers, in sory pigments is transferred to the reaction
special inclusions called chlorosomes attached center. What is the evidence for that?
to the membrane, or in granules called phyco-
6. Explain how the absorption of light by
bilisomes attached to the membrane. There is
photosynthetic membranes creates a ∆E.
considerable variability in the composition and
Explain how the ∆E is converted into a ∆p.
structure of the light-harvesting complexes.
Explain how the ∆p is used to drive ATP
However, they all consist of pigment–protein
synthesis.
complexes that transfer energy to the reac-
tion center. The mechanism of energy transfer 7. Why are two light reactions required for
between the pigment molecules demands that oxygenic photosynthesis but only one light
they be positioned very close to each other in reaction for anoxygenic photosynthesis?
a precise orientation. The proteins in the light-
harvesting complexes and in the reaction cen-
ters fulfill this function. REFERENCES AND NOTES
1. Fleischman, D. E., W. R. Evans, and I. M. Miller.

1995. In: Anoxygenic Photosynthetic Bacteria, pp.
Study Questions 123–136. R. E. Blankenship, M. T. Madigan, and
C. E. Bauer (Eds.). Kluwer Academic Publishers,
1. Measuring oxygen production per quantum Dordrecht, Netherlands.
of absorbed light reveals that the efficiency 2. Kramer, D. M., A. Kanazawa, and D. Fleischman.
falls when monochromatic light of 700 nm 1997. Oxygen dependence of photosynthetic elec-
is used. The efficiency can be restored if the tron transport in a bacteriochlorophyll-containing
700 nm light is supplemented with low-in- rhizobium. FEBS Lett. 417:275–278.
tensity light of shorter wavelength (e.g., 600 3. Reviewed in: The Molecular Biology of
nm). Explain these results in terms of the Cyanobacteria. 1995. D. A. Bryant (Ed.). Kluwer
Academic Publishers, Boston.
light reactions of oxygenic photosynthesis.
4. These groupings do not reflect the complex tax-
2. Describe the similarities and differences onomy of the photosynthetic bacteria as revealed by
between reaction centers I and II. ribosomal RNA sequencing. There are five distinct
photosynthesis 197
evolutionary lines of photosynthetic prokaryotes center from Rhodobacter sphaeroides R-26 and
(see Fig. 1.1 and Table 1.1). The purple photosyn- 2.4.1: symmetry relations and sequence compari-
thetic bacteria are a heterogeneous assemblage that sons between different species. Proc. Natl. Acad. Sci.
are part of the division of prokaryotes known as USA 85:9012–9016.
the Proteobacteria (also called the purple bacteria).
The Proteobacteria are subdivided into four sub- 12. Katona, G., U. Andreasson, E. Landau, L.-E.
divisions, the α, β, γ, and δ groups. Photosynthetic Andreasson, and R. Neutze. 2003. Lipidic cubic
bacteria, along with nonphotosynthetic bacteria, are phase crystal structure of the photosynthetic reac-
in the α, β, and γ subdivisions, whereas the δ subdi- tion centre from Rhodobacter sphaeroides at 2.35 Å
vision contains only nonphotosynthetic bacteria. resolution. J. Mol. Biol. 331:681–692.
The purple nonsulfur photosynthetic bacteria are in 13. Cogdell, R. J. 1978. Carotenoids in photosyn-
the α and β subdivisions, whereas the purple sulfur thesis. Philos. Trans. R. Soc. Lond. B 284:569–579.
bacteria are in the γ subdivision. In the α group are
Rhodospirillum, Rhodopseudomonas, Rhodobacter, 14. During photosynthesis under high light intensi-
Rhodomicrobium, and Rhodopila, all of which are ties, an electron in bacteriochlorophyll can go from
nonsulfur purple bacteria. Another nonsulfur pur- the excited singlet state to the lower energy triplet
ple bacterium, Rhodocyclus, is in the β group. The state (Section 6.7). In the triplet state, the spin of the
γ group contains purple sulfur bacteria, including excited electron has changed, so that there are now
Chromatium and Thiospirillum. The green sulfur bac- two unpaired electrons as opposed to paired electrons
teria, including Chlorobium and Chloroherpeton, are in the singlet state. (Phosphorescence occurs from the
only distantly related to the purple bacteria and are triplet state.) When triplet-state bacteriochlorophyll
in a distinctly separate evolutionary line. The green reacts with oxgyen, the oxygen becomes energized
nonsulfur bacteria (the Chloroflexus group) are in a to its first excited state, singlet oxygen, 1O2. Singlet
separate evolutionary line, phylogenetically distinct oxygen is very reactive and combines with cell mol-
from the purple photosynthetic bacteria and the green ecules such as unsaturated fatty acids, amino acids,
sulfur bacteria. The cyanobacteria occupy a fourth and purines, thus causing oxidative damage to vari-
evolutionary line. The photosynthetic heliobacteria, ous cell components, including lipids, enzymes, and
which consist of Heliobacterium, Heliospirillum, and nucleic acids. The carotenoid in the reaction center
Heliobacillus, are grouped in the evolutionary line quenches the triplet state in bacteriochlorophyll.
that consists of the gram-positive bacteria. Thus, carotenoids prevent photooxidation under
high light intensities. When the carotenoid absorbs
5. Bullerjahn, G. S., and A. F. Post. 1993. The the energy from the triplet bacteriochlorophyll, the
prochlorophytes: are they more than just chlorophyll carotenoid itself becomes energized to the triplet state
a/b-containing cyanobacteria? Crit. Rev. Microbiol. (triplet–triplet energy transfer). However, the energy
19:43–59. in triplet-state carotenoid is dissipated harmlessly as
heat to the medium. Because the triplet-state energy of
6. Olson, J. M. 1998. Chlorophyll organization carotenoids is below the singlet-state energy of oxy-
and function in green photosynthetic bacteria. gen, carotenoids will also quench singlet oxygen.
Photochem. Photobiol. 67:61–75.
15. Carotenoids are isoprenoid membrane pigments.
7. Dutton, P. L. 1986. Energy transduction in They are useful in detecting changes in membrane
anoxygenic photosynthesis, pp. 197–237. In: potential. The energy levels of electrons in the conju-
Photosynthesis III: Photosynthetic Membranes and gated-double-bond system of the carotenoids is altered
Light Harvesting Systems. L. A. Staehelin and C. J. by the electric field of the membrane potential. When
Arntzen (Eds.). Springer-Verlag, Berlin. the membrane potential changes, the carotenoid under-
goes a rapid shift in its absorption spectrum peaks. This
8. Mathis, P. 1990. Compared structure of plant can be followed by using split-beam spectroscopy simi-
and bacterial photosynthetic reaction centers. lar to that discussed for examining spectral changes of
Evolutionary implications. Biochim. Biophys. Acta cytochromes during oxidation–reduction reactions.
1018:163–167.
16. Nitschke, W., U. Feiler, and A. W. Rutherford.
9. Green nonsulfur photosynthetic bacteria have a 1990. Photosynthetic reaction center of green
reaction center resembling that of the purple photo- sulfur bacteria studied by EPR. Biochemistry 29:
synthetic bacteria. The primary donor of the electron, 3834–3842.
called P865, is a dimer of Bchl a. The initial electron
acceptor is bacteriopheophytin (Bpheo). The two 17. Miller, M., X. Liu, S. W. Snyder, M. C.
quinones (QA and QB) are menaquinones rather than Thurnauer, and J. Biggins. 1992. Photosynthetic
ubiquinones. electron-transfer reactions in the green sulfur bacte-
rium Chlorobium vibrioforme: evidence for the func-
10. Deisenhofer, J., H. Michel, and R. Huber. 1985. tional involvement of iron-sulfur redox centers on
The structural basis of photosynthetic light reactions the acceptor side of the reaction center. Biochemistry
in bacteria. Trends Biochem. Sci. 10:243–248. 31:4354–4363.
11. Komiya, H., T. O. Yeates, D. C. Rees, J. P. 18. Meent, van de E. J., M. Kobayashi, C. Erkelens,
Allen, and G. Feher. 1988. Structure of the reaction P. A. van Veelen, S. C. M. Otte, K. Inoue, T. Watanabe,
www.ebook777.com
and J. Amesz. 1992. The nature of the primary elec- structure. This is discussed in Kurisu, G., H. Zhang,
tron acceptor in green sulfur bacteria. Biochim. J. L. Smith, and W. A. Cramer. 2003. Structure of
Biophys. Acta 1102:371–378. the cytochrome b6f complex of oxygenic photosyn-
thesis: Tuning the cavity. Science. 302:1009–1014.
19. The reaction centers of the green sulfur bacte-
ria, heliobacteria, and photosystem I of chloroplasts 25. Hirschberg, J., and D. Chamovitz.
differ from reaction center II and the reaction center 1995. Carotenoids in cyanobacteria, pp. 559–579.
in purple photosynthetic bacteria in producing low- In: The Molecular Biology of Cyanobacteria. D. A.
potential, reduced-FeS proteins, rather than reducing Bryant (Ed.). Kluwer Academic Publishers, Boston.
quinone. 26. Morgan-Kiss, R. M., L.-K. Chan, S. Modla,
20. Reviewed in: ref. 7. T. S. Weber, M. Warner, K. J. Czymmek, and T. E.
Hanson. 2009. Photosynthetic Res. 99:11–21.
21. Emerson, R., and C. M. Lewis. 1943. The depen-
dence of the quantum yield of Chlorella photosynthe- 27. Grossman, A. R., M. R. Schaefer, G. G. Chiang,
sis on wavelength of light. Am. J. Bot. 30:165–178. and J. L. Collier. 1993. The phycobilisome, a light-
harvesting complex responsive to environmental
22. Emerson, R., R. Chalmers, and C. Cederstrand. conditions. Microbiol. Rev. 57:725–749.
1957. Some factors influencing the long-wave limit
of photosynthesis. Proc. Natl. Acad. Sci. USA 43: 28. Gantt, E. 1986. Phycobilisomes, pp. 260–268.
133–143. In: Photosynthesis III: Photosynthetic Membranes
and Light-Harvesting Systems. L. A. Staehelin and
23. Kurisa, G., H. Zhang, J. L. Smith, and W. A. C. J. Arntzen (Eds.). Springer-Verlag, Berlin.
Cramer. 2003. Structure of the cytochrome b6f com-
plex of oxygenic photosynthesis: tuning the cavity. 29. Wakao, N., N. Yokoi, N. Isoyama, A. Hiraishi,
Science. 302:1009–1014. K. Shimada, M. Kobayashi, H. Kise, M. Iwaki, S.
Itoh, S. Takaichi, and Y. Sakurai. 1996. Discovery
24. The b6f complex is similar to the bc1 complex of natural photosynthesis using Zn-containing bacte-
and carries out a Q cycle generating a ∆p, as does riochlorophyll in an aerobic bacterium Acidiphilium
the bc1 complex during respiration. It contains four rubrum. Plant Cell Physiol. 37:889–893.
proteins: cyt b6, subunit IV, the Rieske Fe–S pro-
tein, and cyt f. There are four prosthetic groups: two 30. Isoprenoids are an important class of molecule.
b-type hemes, one c-type heme (cytochrome f), and a Other isoprenoids are the side chains of quinones,
2Fe–2S center. The b6f complex functions in respi- the phytol chain in chlorophyll, and the retinal chro-
ration as well as in photosynthetic electron transport mophore in bacteriorhodopsin, halorhodopsin, and
in the cyanobacteria. See the review by Kallas, T. rhodopsin.
1995. The cytochrome b6f complex, pp. 259–317. 31. Diner, B. A. 1986. Photosystems I and II:
In: The Molecular Biology of Cyanobacteria. D. A. structure, proteins, and cofactors, pp. 422–436. In:
Bryant (Ed.). Kluwer Academic Publishers. Boston. Photosynthesis III: Photosynthetic Membranes and
There are some differences in structure between the Light-Harvesting System. L. A. Staehelin and C. J.
bc1 complex and the b6f complex revealed by crystal Arntzen (Eds.). Springer-Verlag, Berlin.
7
The Regulation of Metabolic Pathways
Bacteria catalyze a very large number of chemi- the student should be aware of the patterns of
cal reactions (approximately 1,000–2,000) that regulation. What is meant by “patterns of regu-
are organized into interconnecting metabolic lation” will be made clear shortly. Patterns of
pathways. To avoid inefficiency, if not chaos, it regulation will vary, even within the same path-
is of the utmost importance that there be coor- way in different organisms. Nevertheless, some
dination between the pathways. Pathways are common patterns of metabolic regulation have
regulated by adjusting the rate of one or more evolved. These are described next.
regulatory enzymes that govern the overall rate
of the pathway.
7.1.1 Feedback inhibition by an end
The rates at which regulatory enzymes oper-
product of the pathway
ate are modified in two ways. One method is by
the noncovalent binding to the enzyme of cer- For biosynthetic pathways, the end product
tain biochemical intermediates of the pathways is usually a negative allosteric effector for a
(called allosteric effectors) that either stimulate branch point enzyme. For example, in Fig. 7.1,
or inhibit the regulatory enzyme, thus signaling F and J are negative effectors for the regulatory
to the regulatory enzyme whether the pathway enzymes 1 and 2, respectively. Such control,
is producing optimal amounts of intermediates which is called end-product inhibition, or feed-
(Sections 7.1.1 and 7.1.2). Regulatory enzymes back inhibition by an end product, acts to main-
are also called allosteric enzymes because in tain a steady state in which the end product is
addition to having binding sites for their sub- utilized as rapidly as it is synthesized. When the
strates, they have separate binding sites for the rates of utilization of the end product increase,
effector molecules. (The Greek prefix allo, mean- the concentration of the end product decreases.
ing “other,” indicates that allosteric enzymes The decrease in concentration relieves the inhi-
have a second, “other,” site besides the substrate bition, and the regulatory enzyme speeds up,
site.) A second method of altering enzyme activ- resulting in more end-product synthesis to meet
ity is by the covalent modification of the enzyme the demands dictated by more rapid utilization
(Section 7.4). Covalent modifications include (e.g., during rapid growth).
the attachment and removal of chemical groups There are three recognized patterns of feed-
such as phosphates and nucleotides. back inhibition in biosynthetic pathways: sim-
ple, cumulative, and concerted (Fig. 7.2). Simple
feedback inhibition describes what happens
7.1 Patterns of Regulation of when the regulatory enzyme is inhibited by a
Metabolic Pathways single end product (Fig. 7.2A). It is encountered
In addition to learning the mechanisms of enzyme in linear biosynthetic pathways. In the cumula-
regulation (allosteric or covalent modification), tive and concerted types of inhibition, seen in
199
www.ebook777.com
Fig. 7.1 Branched metabolic pathways. The pathways provide precursors to other pathways at the branch
points. Both positive and negative regulation occurs. Here F and J are negative effectors that prevent their own
overproduction by inhibiting the regulatory enzymes, 1 and 2. The positive effector K stimulates the produc-
tion of G (in a different pathway), which is needed to react with K to form L. The precursor G activates a later
reaction (enzyme 3) in its own pathway.
Fig. 7.2 Patterns of feedback inhibition. (A) Simple feedback inhibition in an unbranched pathway. (B)
Concerted or cumulative inhibition. In concerted inhibition, there is no inhibition unless both D and E bind.
In cumulative inhibition D and E independently exert partial inhibition, but the combination is less than addi-
tive. (C) Simple inhibition in a branched pathway by means of isoenzymes.
branched pathways, more than one end prod- activity of the common enzyme, thus preventing
uct inhibits the enzyme (Fig. 7.2B). In concerted the synthesis of all the end products. However,
inhibition both end products must bind to the sometimes branched biosynthetic pathways are
regulatory enzyme simultaneously to achieve regulated by simple feedback inhibition. This
any inhibition (Fig. 7.2B). In cumulative inhibi- may occur when a reaction shared by the vari-
tion the enzyme is not completely inhibited by ous branches uses enzymes that have the same
any single end product. For example, one end catalytic sites but different effector sites (isoen-
product might inhibit the enzyme by 25% and a zymes) (Fig. 7.2C). (Thus, isoenzymes catalyze
second might inhibit the enzyme by 45%. Both the same reaction but are responsive to differ-
end products together might inhibit the enzyme ent end products.) It can be seen that different
by 60% (not 70%) (Fig. 7.2B). Thus the inhibi- regulatory patterns (cumulative, concerted,
tion is cumulative, but not necessarily additive. isoenzymes) can produce similar results (i.e.,
Cumulative or concerted inhibition is necessary the control of the rate of the pathway by the lev-
because branched pathways may share a com- els of one or more of the end products).
mon regulatory enzyme prior to the point that It should not be concluded that feedback
leads to the separate branches of the pathway. inhibition applies only to biosynthetic path-
Under these circumstances it is important that a ways. Catabolic pathways, including those for
single end product not be able to shut down the the breakdown of sugars and carboxylic acids,
the regulation of metabolic pathways 201
may also be subject to feedback inhibition by an 7.2.1 Nonregulatory enzymes

end product. A plot of the rate of formation of product (or
disappearance of substrate, S) as a function of
7.1.2 Positive regulation substrate concentration for most enzyme-cat-
A metabolic pathway can also be positively alyzed reactions generates a hyperbolic curve
regulated, sometimes by an intermediate in a similar to the one shown in Fig. 7.3. As the sub-
second pathway, as shown in Fig. 7.1, where strate concentration [S] is increased, the rate
one pathway produces an intermediate, G, of the reaction approaches a maximum, Vmax,
which combines with K in a second pathway to because the enzyme becomes saturated with
produce L. If insufficient G is formed, K accu- substrate. For each enzyme there is a substrate
mulates and stimulates the enzyme that pro- concentration that gives 12 Vmax . This is called
duces G. An example of this is the activation of the Michaelis–Menten constant, or Km. The
PEP carboxylase by acetyl–CoA described in equation that describes the kinetics of enzyme
Section 9.9.1. activity is called the Michaelis–Menten equa-
In another pattern of positive regulation, tion (eq. 6.1):
sometimes called “precursor activation,” a
precursor intermediate stimulates a regulatory v = (VmaxS/(Km+ S) (7.1)
enzyme “downstream” in the same pathway. When the substrate concentration, S, is very
This ensures that the rate of the downstream small compared with Km, then the initial veloc-
reactions matches that of the upstream reac- ity, v, is proportional to S (actually to VmaxS/Km).
tions. In Fig. 7.1, G is a precursor that activates However, when S becomes much larger than
enzyme 3, which converts I to J. An example Km, then S cancels out and v approaches Vmax.
of precursor activation is the activation of There are two units for Vmax (i.e., specific activ-
pyruvate kinase by fructose-1,6-bisphosphate ity and turnover number). The units of specific
described in Section 9.1.2. We will encounter activity are micromoles of substrate converted
other examples of both positive and negative per minute per milligram of protein, and the
regulation by multiple effectors when we dis- units for turnover number are micromoles of
cuss the regulation of the enzymes of central substrate converted per minute per micromole
metabolism in Chapter 9. of enzyme.
7.1.3 Regulatory enzymes catalyze Derivation of the Michaelis–Menten

irreversible reactions at branch points equation
There are certain generalizations that one In 1913 L. Michaelis and M. L. Menten formu-
can make about regulatory enzymes and the lated a theory of enzyme action that explained
reactions that they catalyze. The reactions the kinetics shown in Fig. 7.3. The following
catalyzed by regulatory enzymes are usually derivation of the Michaelis–Menten equation
at a metabolic branch point. For example, in was developed later by Briggs and Haldane.
Fig. 7.1, the intermediates B, D, and I are at Consider a substrate (S) being converted to
branch points. Also, regulatory enzymes often a product (P) in an enzyme-catalyzed reac-
catalyze reactions that are physiologically irre- tion. During the reaction, S combines with the
versible (i.e., far from equilibrium). Thus they enzyme (E) at the substrate site on the enzyme
are poised to accelerate in one direction when to form the enzyme–substrate complex (ES),
stimulated. which then breaks down to enzyme and prod-
uct. For the derivation of Briggs and Haldane,
7.2 Kinetics of Regulatory and it is necessary to assume a steady state in which
Nonregulatory Enzymes the enzyme is present in catalytic amounts: S >>
total enzyme. Under these conditions, the rate
To appreciate how effector molecules alter
of formation and breakdown of ES are equal,
the activities of regulatory enzymes, one must
which results in a steady state level of ES. The
understand enzyme kinetics and the enzyme
situation can be summarized as follows:
kinetic constants, Km and Vmax. This is best done k1 k3
by beginning with the kinetics of nonregulatory S + E(free) U(ES) U E + P
enzymes. k2 k4
www.ebook777.com
Equation 7.4 describes the instantaneous rate

of a bimolecular reaction between the substrate
whose concentration is S, and the free enzyme
whose concentration is Et – ES. Now we must
consider the rate of breakdown of ES. This is
given by eq. 7.5:
−d(ES) / dt = k2 (ES) + k3 (ES) (7.5)
Let us assume the reaction reaches a steady state

at the time of measurement. This is ensured by
using substrate concentrations that are in large
excess over the total amount of enzyme. In the
Fig. 7.3 Michaelis–Menten kinetics. When substrate steady state, the rate of formation of ES is equal
concentrations [S] are plotted against initial velocity to its rate of breakdown, therefore,
(v), a hyperbolic curve is obtained that approaches
a maximum rate (Vmax). The substrate concentration
k1 (Et − ES)(S) = k2 (ES) + k3 (ES) (7.6)
that yields 12 Vmax is a constant for each enzyme and
is called the Michaelis–Menten constant or Km.
which on rearrangement gives
Notice that there are four rate constants, k1, (S)(Et − ES) /(ES) = (k2 + k3 ) / k1 = Km (7.7)
k2, k3, and k4. If one were to measure the ini-
tial velocity (v), then the rate of the reaction where Km is the Michaelis–Menten constant.
(e.g., the rate of formation of P or disappear- Note that if k3 is small compared with k2 (i.e.,
ance of S) would be proportional to k3(ES). In if the rate of product formation is small with
this treatment we are ignoring the formation of respect to the rate that the substrate dissociates
(ES) from E and P, and k4. This amount would from the enzyme), then the Km is approximately
be small, in any event, if the initial rates were equal to k2/k1, that is, the dissociation con-
measured before P had accumulated. If all of the stant for the enzyme (KD). This is true for some
enzyme were bound to S, then all of E would be enzyme–substrate combinations. It is not always
in the form of (ES), and the rate would be the true, however, and it is a mistake to assume (as
maximum rate. These relationships are shown is often done) that 1/Km is a direct measure of the
in eqs. 7.2 and 7.3 affinity of a substrate for an enzyme. It is best to
refer to the Km as being equal to the substrate
v = dP/dt = k3(ES) (7.2) concentration that gives half maximal velocity,
as described in Fig. 7.3.
and, substituting total enzyme, Et, for E, when
Now we can solve for (ES):
all of the enzyme is saturated with substrate
(eq. 7.3),
(ES) = (Et )(S) /(Km + S) (7.8)
Vmax = k3 (Et S) (7.3)
where (Et) is total enzyme. Since the initial
(For enzymes that may have complex reaction rate (v) is proportional to k3(ES), we can write
pathways that depend upon more rate con- eq. 7.9:
stants than simply k3, the symbol kcat is used for
the maximal catalytic rate: i.e., Vmax = kcat(EtS). ν = k3 (Et )(S) /(Km + S) (7.9)
See note 1.
Since Vmax = k3(ES) when (ES) = Et, it is also true
The instantaneous rate of formation of (ES) is
that Vmax = k3(Et). Equation 7.9 then becomes
given by eq. 7.4:
eq. 7.10:
d(ES) / dt = k1 (Et − ES)(S) (7.4) ν = (Vmax S) /(Km + S) (7.10)
The concentration of free enzyme (E) is Et – (ES), Equation 7.10 is the Michaelis–Menten equa-
where S represents the substrate concentration. tion that describes the kinetics in Fig. 7.3.
Typical Km values range from 10–4 M (100 μM) or –1/Km from the intercept on the y axis, and 1/
to lower molarities. Vmax from the intercept on the x axis.
The Km is the substrate concentration that 7.2.2 Regulatory enzymes

gives 12 Vmax Regulatory enzymes seldom follow simple
If one substitutes 12 Vmax for v in the Michaelis– Michaelis–Menten kinetics. Instead, they
Menten equation and solves for S, then S is equal typically show sigmoidal kinetics (Fig. 7.5).
to Km. That is, the Km is equal to the substrate An explanation for substrate-dependent sig-
concentration that gives 12 Vmax . This is shown moidal kinetics is that the binding of one
in eq. 7.11: substrate molecule increases the affinity of
the enzyme for a second substrate molecule
1
2
Vmax = Vmax (S) /(Km + S) (7.11) or increases the rate of formation of product
from sites already occupied. This is called
Divide both sides by Vmax: positive cooperativity. Positive cooperativity
makes sense for a regulatory enzyme because
1
2
V = S /(Km + S) (7.12) it makes enzyme catalysis very sensitive to
small changes in substrate level concentra-
Rearrange: tions when the substrate concentrations are
very small in comparison to the Km. (Notice in
Km + S = 2 S (7.13) Fig. 7.5 how the enzyme rate rapidly changes
with small changes of substrate at critical
and, concentrations.)
Allosteric effectors (positive and negative)
Km = S (7.14) can also bind cooperatively to enzymes, mak-
ing the enzyme more sensitive to small changes
Because the Vm and therefore the Km can only be in effector concentration. Even though sigmoi-
approximated from the plot shown in Fig. 7.3, dal kinetics is not explained by the Michaelis–
eq. 7.10 is frequently written as the reciprocal: Menten equation, one can still measure the Km
and Vmax because these are defined operation-
1 / v = 1 / Vmax + (Km / Vmax )(1 / S) (7.15) ally. When one measures these “constants”
for regulatory enzymes, one finds that they are
and 1/v is plotted against 1/[S] to give a dou- subject to change, depending upon the presence
ble reciprocal, or Lineweaver–Burke, plot or absence of effector molecules. Some of the
(Fig. 7.4). One can find Km/Vmax from the slope,
Fig. 7.5 Sigmoidal kinetics of regulatory enzymes.

Initial velocities (v) are plotted against substrate con-
centrations [S] in the presence of a positive effector
(curve 1), no effector (curve 2), or a negative effector
(curve 3). The positive effector decreases the Km as
well as decreasing the sigmoidicity of the curve. The
negative effector increases the Km and the sigmoid-
icity of the curve. In other cases, the effectors may
Fig. 7.4 A Lineweaver–Burke plot. change the Vmax.
www.ebook777.com
effectors raise the Km and some lower the Km. are oligomeric, or multimeric (i.e., they have
An effector that raises the Km will decrease the multiple subunits). However, for illustrative
velocity of the reaction, and one that lowers the purposes we will first consider a monomeric
Km will increase the velocity of a reaction, pro- polypeptide (Fig. 7.6). Assume that the poly-
vided the enzyme is not saturated with substrate. peptide has three binding sites: one for the
This is shown in Fig. 7.5. Notice that the nega- substrate (substrate site), a second for a posi-
tive effector (curve 3) increases the sigmoidicity tive effector, and a third for a negative effector.
of the curve and also increases the Km. The posi- Further assume that the enzyme exists in two
tive effector (curve 1) decreases the sigmoidicity states, A and B. When the enzyme is in confor-
of the curve and lowers the Km. Effectors can mation A, it binds the substrate and also binds
also change the Vmax. However, in Fig. 7.5 the the positive effector, which locks the enzyme
Vmax is not changed. in state A. You might say that state A has a low
Km. State B has a high Km and also binds the
7.3 Conformational Changes in positive effector poorly. But state B does bind
the negative effector, which locks it into state
Regulatory Enzymes B. Thus, the positive effector lowers the Km and
When the effector binds to the allosteric site the negative effector raises the Km. In the cell,
(the effector site), the protein undergoes a con- there is an interplay of positive and negative
formational change, and this changes its kinetic effectors that adjust the ratio of active and less
constants. Many of the regulatory enzymes active enzyme.
Regulatory subunits
Many regulatory enzymes are multimeric.
Some consist of a regulatory subunit and one or
more catalytic subunits. The catalytic subunits
bear the active sites and bind the substrate. The
regulatory subunits bind the effectors. When
the regulatory subunits bind the effector,
the polypeptide undergoes a conformational
change induced by the binding. The regulatory
subunit then either inhibits or activates the cata-
lytic subunit, resulting in changes in the kinetic
Fig. 7.6 Conformational changes in a monomeric
regulatory enzyme with three binding sites: one for constants.
the substrate (S), one for the positive effector (+),
and one for the negative effector (–). (A) When the 7.4 Regulation by Covalent
enzyme binds the positive effector, a conforma- Modification
tional change occurs that lowers the Km. (B) When
the enzyme binds to the negative effector, there is Although most regulatory proteins appear to be
an increase in the Km and a loss of affinity for the regulated by conformational changes induced
positive effector. In the presence of a sufficient con- by the binding of allosteric effectors as just
centration of positive effector, essentially all of the described, there are many important instances
enzyme will be in conformation A and the enzyme in both prokaryotes and eukaryotes of regula-
will be turned “on.” On the other hand, if suffi- tion by covalent modification of the protein.
cient negative effector is present, the enzyme will The enzyme may also be regulated by allosteric
be mostly in conformation B and turned “off.” The interactions. Covalent modification occurs by
fraction of enzyme in the “on” or “off” conforma- the reversible attachment of chemical groups
tion will depend upon the relative concentrations
such as acetyl groups, phosphate groups, methyl
of substrate, positive effector, and negative effec-
groups, adenyl groups, and uridyl groups. The
tor. Multimeric enzymes are regulated in a similar
fashion except that the effector-binding sites can covalent attachment of a chemical group can
be on a subunit (regulatory subunit) separate from activate or inhibit the protein. Table 7.1 sum-
the substrate-binding site (the catalytic subunit). marizes some examples of covalent modifica-
Conformational changes in one subunit induce tion of proteins, several of which are discussed
conformational changes in the attached subunit. in later chapters.
Table 7.1 Covalent modifications of bacterial enzymes and other proteins

Enzyme Organism Modification
Glutamine synthetase E. coli and others Adenylylation

Isocitrate lyase E. coli and others Phosphorylation
Isocitrate dehydrogenase E. coli and others Phosphorylation
Chemotaxis proteins E. coli and others Methylation
PII E. coli and others Uridylylation
Ribosomal protein L7 E. coli and others Acetylation
Citrate lyase Rhodopseudomonas gelatinosa Acetylation
Histidine protein kinase Many bacteria Phosphorylation
Phosphorylated response regulators Many bacteria Phosphorylation
Source: Neidhardt, F. C., J. L. Ingraham, and M. Schaechter. 1990. Physiology of the Bacterial Cell. Sinauer Associates,
Sunderland, MA.
7.5 Summary consisting of catalytic subunits and regulatory

Metabolism is regulated by key regulatory subunits. The allosteric effector binds to the
enzymes that help keep metabolism in a steady regulatory subunit and effects a conformational
state in which concentrations of intermedi- change in the catalytic subunit that alters its
ates do not change. Generally, the regulatory kinetic constants.
enzymes catalyze physiologically irreversible
reactions at metabolic branch points. The Study Questions
enzymes can be regulated by negative effectors,
by positive effectors, by negative and positive 1. Plot the following data and derive a Km and
effectors together, and by covalent modifica- a Vmax.
tion. In biosynthetic pathways, the negative
effectors are the end products. This is called Time Product (arbitrary units) for the
feedback or end-product inhibition. Positive (min) formation of substrates (M × 10–6)
regulation by a precursor intermediate can 16 24 48 144
speed up a subsequent reaction in the same
metabolic pathway to avoid buildup of the pre- 0.4 0.55 0.70 0.70 1.00
cursor. Also, an intermediate in one pathway 0.8 1.10 1.30 1.35 1.70
may exert positive regulation on a reaction in 1.2 1.65 1.85 2.00 3.00
a second pathway if the second pathway pro- 1.6 2.05 2.35 2.55 3.15
vides a precursor metabolite for the first path- 2.0 2.50 2.85 3.10 3.80
way. Enzymes that are regulated by effector
molecules that bind to effector sites are called 2. What is a rationale for having regulatory
allosteric enzymes, and the effector molecules enzymes catalyze physiologically irrevers-
are called allosteric effectors. ible reactions at branch points?
Nonregulatory enzymes display simple satu-
3. What are three patterns of feedback inhibi-
ration kinetics called Michaelis–Menten kinet-
tion for the regulation of a branched bio-
ics. Two kinetic constants are the Vmax and the
synthetic pathway?
Km. The Vmax is the velocity of the enzyme when
it is saturated with substrate. The Km is the sub- 4. Under what circumstances might you
strate concentration yielding 12 Vmax . expect to see positive regulation of enzyme
Regulatory enzymes usually show sigmoi- activity?
dal kinetics. Nevertheless a Vmax and a Km can
5. What is meant by positive cooperativity?
be measured. A positive effector can raise the
What is the physiological advantage to it?
Vmax and lower the Km, but not necessarily both.
A negative effector does just the opposite. 6. How does the cell maintain a “steady state”
Regulatory enzymes are frequently multimeric, whereby the end product of a synthetic
www.ebook777.com
pathway is synthesized at the same rate as REFERENCES AND NOTES

it is utilized?
1. The turnover number is equal to kcat (k3) and is
7. What is meant by “regulatory subunit” of related to the concentration of active sites Et [i.e.,
an enzyme, and what binds to a regulatory Vmax = kcat(Et)]. The turnover number is measured
subunit? What is the advantage to having a with pure enzyme whose molecular weight is
known.
regulatory subunit in terms of broad meta-
bolic regulation between pathways?
8
Bioenergetics in the Cytosol
Two different kinds of energy drive all cellular 8.1 High-Energy Molecules and
reactions. One of these is electrochemical Group Transfer Potential
energy, which refers to ion gradients (in bacte-
High-energy molecules such as ATP have bonds
ria primarily the proton gradient) across the cell
that have a high free energy of hydrolysis.
membrane (Chapter 4). Electrochemical energy
These bonds, which are sometimes depicted by
energizes solute transport, flagella rotation,
a “squiggle” (~), are sometimes called “high-
ATP synthesis, and other membrane activities.
energy” bonds; as will be discussed shortly, how-
The second type of energy is chemical energy in
ever this is a misnomer because the bonds have
the form of high-energy molecules (e.g., ATP)
normal bond energy. The important point is that
in the soluble part of the cell. High-energy mol-
the chemical group attached to the “squiggle”
ecules are important because they drive biosyn-
is transferred with a large free energy release to
thesis in the cytoplasm, including the synthesis
acceptor molecules. Therefore, the high-energy
of nucleic acids, proteins, lipids, and polysac-
molecules are said to have a high group transfer
charides. Additionally, the uptake into the cell
potential. When the chemical groups are trans-
of certain solutes is driven by high-energy mol-
ferred, new linkages between molecules (e.g.,
ecules rather than by electrochemical energy.
ester linkages, amide linkages, glycosidic link-
This chapter introduces the various high-
ages, ether linkages) are made. This results in
energy molecules used by the cell and explains
the synthesis of the different small molecules in
why the adjective “high-energy” is applied to
the cell (e.g., complex lipids, nucleotides, etc.),
them, how they are made, and how they are used.
as well as polymers such as nucleic acids, pro-
The major high-energy molecules include ATP
teins, polysaccharides, and fatty acids. We will
as well as other nucleotide derivatives, phospho-
first consider group transfer reactions in gen-
enolpyruvate, acyl phosphates, and acyl–CoA
eral, and then explain why certain molecules
derivatives. The chapter begins with a discus-
have a high group transfer potential.1 Finally,
sion of the chemistry of the major high-energy
we will present some examples of the use of
molecules (Section 8.1), followed by explana-
group transfer reactions in biosynthesis.
tions of how high-energy molecules are used to
drive biosynthetic reactions (Section 8.2) and
how the high-energy molecules are synthesized 8.1.1 Group transfer potential
(Section 8.3). The information in this chapter A common chemical group that is transferred
will be referred to in later chapters, which con- between molecules is the phosphoryl group,
sider in more detail the metabolic roles of the and phosphoryl group transfer will serve as an
high-energy molecules. A historical perspective example of a group transfer reaction. Let us
in given in Box 8.1. consider a generic phosphoryl group transfer
207
www.ebook777.com
BOX 8.1 HISTORICAL PERSPECTIVE: ENERGY

TRANSFER IN THE CYTOSOL
In 1927 Eggleton and Eggleton reported sin, which was known to form the basis of
that the contraction of vertebrate muscle muscle contraction, is an ATPase.
was associated with an organic phospho- Between 1939 and 1941 Fritz Lipmann
rus compound that they named “phospha- made a long-lasting contribution to our
gen.” When muscle contracted, the levels understanding of how energy transfer takes
of phosphagen decreased, and the process place in the cytosol. It was Lipmann who
was accompanied by a rise in inorganic introduced the term “energy-rich phos-
phosphate. When oxygen was present, phate bond” to describe the phosphate
this sequence was reversed. In 1927 Fiske linkage in creatine phosphate, ATP, phos-
and Subbarow identified phosphagen as phoenolpyruvate, 1,3-diphosphoglyceric
creatine phosphate. In 1929 Lohmann, as acid, and acetylphosphate, whose hydro-
well as Fiske and Subbarow, isolated from lysis is associated with a large negative
striated muscle a compound that could free-energy change (kilocalories or kilo-
be hydrolyzed to adenylic acid and pyro- joules per mole). Lipmann also intro-
phosphate. This was ATP, and its struc- duced the term “group potential” to
ture was elucidated subsequently. The heat describe the ability of these compounds
released when creatine phosphate or ATP to donate the phosphoryl group, or other
was hydrolyzed was measured, and it was groups such as the acetyl group, to other
clear that either of these compounds could molecules.
provide the energy for muscle contraction. For additional background on these
However, it was not known at that time developments, see refs. 1 and 2.
which of the two did so.
It is now known that creatine phosphate REFERENCES
is an energy-storage compound in muscle
1. Hoffmann-Ostenhof, O. (Ed.). 1987. Inter-
and is synthesized via ATP, which is the mediary Metabolism. Van Nostrand Reinhold,
molecule that drives muscle contraction. New York.
This realization was due to in large part to 2. Fruton, J. S. 1972. Molecules and Life:
a paper published in 1939 by Engelhardt Historical Essays on the Interplay of Chemistry
and Lyubimova. They showed that myo- and Biology. John Wiley & Sons, New York.
reaction (shown in Fig. 8.1). We will examine not simply phosphoryl group transfers. That is,
both the chemical mechanism of transfer and the a nucleophile bonds to an electropositive center
thermodynamics. Notice that the phosphorus in and displaces a leaving group with its bonding
all phosphate groups carries a positive charge, electrons. The reactions are called nucleophilic
(i.e., the P=O bond is drawn as the semipolar displacements or SN2 (substitution nucleophilic
P+–O– bond). This is because phosphorus forms bimolecular) reactions. As we shall see later,
double bonds poorly, and the electrons in the various molecules such as ATP, acyl phosphates,
bond are shifted toward the electron-attracting and phosphoenolpyruvate have a high phospho-
oxygen. During the phosphoryl group transfer ryl group transfer potential and undergo similar
reaction, the phosphorus atom is attacked by a nucleophilic displacement reactions.
nucleophile (an attacking atom with a pair of But, how can one compare the phosphoryl
electrons seeking a positive center) shown in Fig. group transfer potential of all these molecules,
8.1. The chemical group Y is displaced with its since the acceptors (the attacking nucleophiles)
bonding electrons, and the phosphoryl group is differ? A scale is used in which the standard
transferred to the hydroxyl, forming ROP. This nucleophile is the hydroxyl group of water and
is a general scheme for group transfer reactions, the phosphoryl donors are all compared with
bioenergetics in the cytosol 209
respect to the tendency to donate the phospho- “high-energy” bonds, although in fact they have
ryl group to water. The group transfer potential normal bond energy. The group transfer poten-
is thus defined as the negative of the standard tial is not really a potential in an electrical sense,
free energy of hydrolysis at pH 7. It is a quanti- but a free energy change per mole of substrate
tative assessment of the tendency of a molecule hydrolyzed. However, the word “potential” is
to donate the chemical group to a nucleophile. widely used in this context, and the convention
For example, suppose the standard free energy will be followed here. The molecules with high
of hydrolysis of the phosphate ester bond in YOP phosphoryl group transfer potentials that we will
is –29,000 J/mol. Then its phosphoryl group revisit in the ensuing chapters are listed in Table
transfer potential is the negative of this number, 8.1. Also listed is glucose-6-phosphate, which
or +29,000 J/mol. Bonds that have a standard has a low phosphoryl group transfer potential.
free energy of hydrolysis at pH 7 equal to or Group transfer potentials are a convenient
greater than –29,000 J/mol are usually called way of estimating the direction in which a reac-
tion will proceed. For example, the phosphoryl
group transfer potential of ATP at pH 7 is 35
kJ/mol and for glucose-6-phosphate it is only 14
kJ/mol. This means that ATP is a more energetic
donor of the phosphoryl group than is glucose-6-
phosphate. It also means that ATP will transfer
the phosphoryl group to glucose to form glucose-
6-phosphate with the release of 35 minus 14 or
21 kJ/mol under standard conditions, pH 7.
ATP + glucose glucose-6-phosphate + ADP

ΔG90 = –21 kJ/mol
This can also be seen by summing the hydroly-

sis reactions, since their sum equals the transfer
of the phosphoryl group from ATP to glucose.
This can be done for thermodynamic calcula-
tions, because the overall energy change is
Fig. 8.1 Phosphoryl group transfer reaction. The phos- independent of the path of the reaction, even
phate group is shown as ionized. Because phosphorus though glucose-6-phosphate is not synthesized
is a poor double-bond former, the phosphorus–oxygen by hydrolysis reactions as written:
bond exists as a semipolar bond. The positively charged
phosphorus is attacked by the electronegative oxygen ATP + H2O ADP + Pi ΔG90 = –35 kJ/
in the hydroxyl. The leaving group is YOH. If ROH is mol
water, then the reaction is a hydrolysis and the product Glucose + Pi glucose-
is inorganic phosphate and YOH. One can compare the 6-phosphate ΔG9 = +14
tendency of different molecules to donate phosphoryl + H2 O
0
groups by comparing the free energies released when

kJ/mol
the acceptor is water (i.e., the free energy of hydrolysis). ATP + glucose glucose-
The group transfer potential is the negative of the free 6-phosphate ΔG90 = –21 kJ/
energy of hydrolysis. mol
+ ADP
Table 8.1 Group transfer potentials

Compound ΔG90,hyd (kJ/mol) Phosphoryl group transfer potential (kJ/mol)
PEP + H2O → pyruvate + Pi –62 +62

1,3-BPGA + H2O → 3-PGA + Pi –49 +49
Acetyl–P + H2O → acetate + Pi –47.7 +47.7
ATP + H2O → ADP + Pi –35 +35
Glucose-6-P + H2O → glucose + Pi –14 +14
www.ebook777.com
The release of 21 kJ/mol means that the equi-

librium lies far in the direction of glucose-6-
phosphate. Because ΔG90 = –RT ln K9eq = –5.80
log 10K9eq kJ (at 30 oC), the equilibrium con-
stant is 4.2 × 103 in favor of glucose-6-phos-
phate and ADP. However, whether a particular
reaction will proceed in the direction written
depends upon the actual free energy change, not
the equilibrium constant. The actual free energy
change at pH 7 is ΔG9, which is a function of the
physiological concentrations of products and
reactants and in this case is
Fig. 8.2 Structure of ATP. ATP has three phosphate
ΔG9= ΔG90 + RT ln(glucose-6-phosphate) groups. (A) Un-ionized. (B) Although all the phos-
(ADP)/(ATP)(glucose) phate groups are shown ionized, giving them a net
negative charge, the actual number at pH 7 is 3 or 4
Reactions proceed only in the direction of a
for most of the ATP. Note that the phosphate–oxy-
negative ΔG9. In the cell, the preceding reaction
gen double bond is drawn as a semipolar bond, which
proceeds only in the direction of glucose-6- takes into account the electronegativity of the oxygen
phosphate and ADP. This is because the ratio of and the low propensity of phosphorus to form dou-
glucose-6-phosphate)(ADP) to (ATP)(glucose) ble bonds. The structure predicts strong electrostatic
would have to be greater than 4.2 × 103 to repulsion between the phosphate groups. The elec-
change the sign of the ΔG9 to cause the reaction trostatic repulsion favors the transfer of phosphate
to proceed in the direction of ATP and glucose. to a nucleophile (e.g., water). (C) ATP drawn with
This does not occur, and therefore under physi- squiggles to show the bonds with high free energy of
ological conditions, the direction of phospho- hydrolysis. The phosphates are drawn as phosphoryl
ryl flow is always from ATP to glucose. We will groups as illustrated in Fig. 8.1.
now consider why ATP and the other molecules
in Table 8.1 have such high group transfer
potentials (high free energies of hydrolysis). the appropriate enzyme is present to catalyze
the reaction. In this way, ATP can phospho-
rylate many different compounds. Enzymes
8.1.2 Adenosine triphosphate (ATP) that catalyze phosphoryl group transfer reac-
As seen in Table 8.1, the standard free energy of tions are called kinases. In summary, then, the
hydrolysis of the phosphate ester bond in ATP at high free energy of hydrolysis of ATP is a good
pH 7 is –35 kJ/mol. To understand why so much predictor for the tendency of ATP to donate a
energy is released during the hydrolysis reaction, phosphoryl group to nucleophiles such as the
we must consider the structure of the ATP mole- hydroxyl groups in sugars. Much of the bio-
cule (Fig. 8.2). Notice that at pH 7 the phosphate chemistry of ATP is directly related to this ten-
groups are ionized. (Actually, at pH 7 most of the dency. We will return to this point later, but first
ATP is a mixture of ATP3– and ATP4–.) This pro- we must further discuss the “squiggle” and the
duces electrostatic repulsion between the nega- “high-energy bond.”
tively charged phosphates, which accounts for
much of the free energy of hydrolysis. Reactions The “squiggle”
during which phosphate is removed from ATP As mentioned, the “squiggle” symbol (~) denotes
will be favored because the electrostatic repul- a high negative free energy of hydrolysis, and
sion is decreased as a result of the hydrolysis. thus a greater group transfer potential. ATP is
usually drawn with two squiggles because there
Transfer of a phosphoryl group to acceptors are two phosphate ester bonds with a high free
other than water energy of hydrolysis (Fig. 8.3).
Any group that is electronegative (e.g., the Note that the squiggle does not refer to the
hydroxyl groups in sugars) can attack the elec- energy in the phosphate bond but, rather, to the
tropositive phosphorus shown in Fig. 8.2 and free energy of hydrolysis. To make the distinc-
result in phosphoryl group transfer, provided tion clear, consider the definition of bond energy.
Bond energy is the energy required to break a energetic phosphoryl donor than ATP and
bond. It is not the energy released when a bond will donate the phosphoryl group to ADP to
is broken. In fact, the P–O bond energy is about make ATP, with release of 62 – 35 or 27 kJ/mol
+413 kJ (100 kcal). Compare this with the (under standard conditions, pH 7) (Table 8.1).
–35 kJ of hydrolysis energy, as shown in Fig 8.4. As discussed in Section 7.3.4, this is an impor-
tant source of ATP. PEP also donates the
8.1.3 Phosphoenolpyruvic acid phosphoryl group to sugars during sugar trans-
Another high-energy phosphoryl donor is phos- port in the phosphotransferase (PTS) system
phoenolpyruvate (PEP). In fact, PEP is a more (Chapter 18).
Why does PEP have a high phosphoryl

group transfer potential?
The reason for PEP’s high phosphoryl group
transfer potential is different from that for ATP.
Consider the following reaction, which is illus-
trated in Fig. 8.5:
PEP + H2O pyruvic acid + Pi
Fig. 8.3 ATP drawn with squiggles to denote a high ΔG90 = –62 kJ/mol
free energy of hydrolysis (i.e., a high group transfer
The hydrolysis removes the phosphate and
potential). In ATP, two of the phosphate ester bonds
allows the enol form of pyruvic acid to tau-
have a high free energy of hydrolysis owing to elec-
tomerize into the keto form. Energy is released
trostatic repulsion. The phosphates are labeled α, β,
because the keto form is more stable than the
and γ, starting with the one nearest to the ribose. If
enol form. One can account for the energy
the α phosphate is attacked, then AMP is transferred
and pyrophosphate (PPi) is displaced. The subscript released during hydrolysis by the difference in
i refers to inorganic: that is, there are no attached bond energies between the keto and enol forms
carbon atoms. If the γ phosphate is attacked, then of pyruvic acid. A summation of bond ener-
the phosphoryl group is transferred and ADP is dis- gies reveals that the keto form has 76 kJ/mol
placed. Less frequently, the β phosphate is attacked (18 kcal/mol) more bond energy than the enol
and the pyrophosphoryl group is transferred, displac- form. (Recall that bond energy is the energy
ing the AMP. The reaction that takes place depends required to break a bond and therefore is equal
upon the specificity of the enzyme. to the energy released when the bond is formed.
BOND ENERGY
ADP O
O
P
homolysis
O– ADP O . . +
O
P O– +413 kJ/mol
O– O–
HYDROLYSIS ENERGY
O O
. OH
ADP O P O– + . 2 ADP OH + HO P O– –35kJ/mol
O– O–
Fig. 8.4 Schematic representation of (A) bond energy and (B) hydrolysis energy.
www.ebook777.com
Fig. 8.5 Conversion of PEP to pyruvate. This is

an enol–keto tautomerization. The removal of the
phosphoryl group allows the electrons to shift into
the keto form. Energy is released because the keto Fig. 8.6 Structures of acyl group, acyl–CoA, and acyl
form is more stable than the enol form. The keto phosphates. Notice that the carbonyl group is polar-
form of pyruvic acid is more stable than the enol ized, and the carbon is subject to nucleophilic attack
form by 42 to 50 kJ/mol (10–12 kcal/mol). The fol- during acyl group transfer reactions. The phosphate
lowing tabulation of the difference in bond energies ester and thioester bonds in the acyl phosphates and
between pyruvic acid and enolpyruvic acid shows acyl–CoAs have high free energy of hydrolysis.
that pyruvic acid has 76 kJ more bond energy than
enolpyruvic acid:
Bond energies (kJ)

Enolpyruvic acid Pyruvic acid
C–O 293 C–C 247
O–H 460 C–H 364
C=C 418 C=O 636 Fig. 8.7 (A)Resonance of an oxygen ester. Note that
1,171 kJ 1,247 kJ electrons shift from the oxygen in the C–OR9 to form
a double bond. This is unlikely in phosphate esters
(B) because the phosphorus atom bears a positive
Hence the formation of a molecule with higher charge and prevents electrons from shifting in from
bond energy will result in the release of energy.) the oxygen atom to form the double bond. Resonance
Thus the hydrolysis of the phosphate ester bond is made less likely in thioesters because the sulfur
results in the release of energy because it pro- atom does not form double bonds.
motes the enol–keto tautomerization.
oxygen acquires a positive charge. In acyl–CoA
8.1.4 Acyl derivatives of phosphate and and acyl phosphate derivatives, the oxygen
coenzyme A attached to R′ (Fig. 8.7B) is replaced by a phos-
Acyl derivatives of phosphate and coenzyme A phorus or sulfur atom. Resonance is hindered in
also have a high free energy of hydrolysis. An phosphate esters because the positive charge on
acyl group is a derivative of a carboxylic acid the phosphorus atom prevents electrons from
and has the structure shown in Fig. 8.6, where R shifting in from the oxygen to form a double
may be an alkyl or aryl group. The acyl deriva- bond, which would leave two adjacent positive
tives of coenzyme A (CoA) and of phosphate are centers. Thioesters also do not resonate as well
also shown in Fig. 8.6. as normal esters. The reason for this is that sul-
fur forms double bonds with carbon poorly. The
Why do acyl derivatives have high group poor resonance of the phosphate and thioesters
transfer potentials? is in sharp contrast to the high resonance of the
The acyl derivatives of phosphate esters and free carboxylate group that forms when the
thioesters have high group transfer potential phosphoryl group or CoA is transferred to an
because they do not resonate well. Consider the acceptor molecule. Thus, the hydrolysis of the
resonance forms of a normal ester (Fig. 8.7A) acyl derivatives of phosphate and coenzyme A
and a phosphate ester (Fig. 8.7B). In a normal is energetically favored because it leads to prod-
ester, when two electrons shift from the oxy- ucts stabilized by resonance with respect to the
gen to form a double bond during resonance, reactants.
The importance of acyl derivatives in group example, this occurs during protein synthe-
transfer reactions sis, discussed in Section 8.2.2. Some enzymes
Acyl derivatives are very versatile. Depending catalyze the transfer of the pyrophosphoryl
on the specificity of the enzyme catalyzing the group when the β phosphate is attacked. For
reaction, they donate acyl groups, CoA groups, example, enzymes that synthesize phosphoe-
or phosphoryl groups. For example, fatty acids nolpyruvate from pyruvate make an enzyme–
and proteins are synthesized via acyl group pyrophosphate derivative by transferring
transfer reactions, CoA transfer takes place dur- the pyrophosphoryl group from ATP to the
ing fermentative reactions in bacteria, and acyl enzyme (Section 9.13.2).
phosphates donate their phosphoryl groups to Another example is the synthesis of phospho-
ADP to form ATP. Enzymes that catalyze the ribosylpyrophosphate (PRPP), during which
transfer of acyl groups are called transacy- the pyrophosphoryl group is transferred from
lases. Those that transfer CoA or phosphoryl ATP to the C1 of ribose-5-phosphate (Sect-
groups are called CoA transferases and kinases, ion 10.2.2).
respectively. In a subsequent reaction, the ribose-5-phos-
phate is transferred from PRPP to an appropri-
8.2 The Central Role of Group ate acceptor molecule in the synthesis of purines
and pyrimidines. If the γ phosphate is attacked,
Transfer Reactions in Biosynthesis the phosphoryl group is transferred, forming
Group transfer reactions are central to all of phosphorylated derivatives and ADP as the
metabolism because biological molecules such leaving group. This is a very common reaction
as proteins, lipids, carbohydrates, and nucleic in metabolism (e.g., the synthesis of glucose-6-
acids are synthesized as a result of group trans- phosphate).
fer reactions by means of high-energy donors As stated, all the foregoing reactions can
such as ATP or other nucleotide derivatives, occur. The reaction that takes place depends
acyl–coenzyme As, and acyl–phosphates. upon which enzyme is the catalyst, since it is
the enzyme that determines the specificity of the
8.2.1 How ATP can be used to form amide attack. As examples, we will consider the cases
linkages, thioester bonds, and ester bonds in which ATP is used to form amide linkages,
ATP as a donor of AMP, phosphoryl groups, thioesters, and esters. In all these reactions, a
or pyrophosphoryl groups carboxyl group accepts either AMP or phos-
ATP can donate other parts of the molecule phate from ATP in a group transfer reaction
(e.g., AMP, PPi) besides the phosphoryl group. to form a high-energy intermediate. The high-
This is very important for the synthesis of many energy intermediate is an acyl derivative with a
polymers (e.g., proteins, polysaccharides, high group transfer potential. The acyl–AMP or
nucleic acids) as well as other biochemical reac- acyl phosphate can donate the acyl group in a
tions. That is because when these groups are subsequent reaction, forming an amide, ester,
transferred to an acceptor molecule, the accep- or thioester bond, depending upon whether
tor molecule itself becomes a high-energy donor the attacking nucleophile is N:, O:, or S:,
for subsequent biosynthetic reactions. In other respectively.
words, the energy from ATP can be transferred Consider the reactions in Fig. 8.8. Notice the
to other molecules, which can then drive bio- sequence of reactions. First there is a displace-
synthetic reactions (i.e., the formation of new ment on either the α or γ phosphate of ATP to
covalent bonds). form the acyl derivative. Then there is a dis-
Let us look again at the structure of ATP placement on the carbonyl carbon to transfer
and examine some of the group transfer the acyl moiety to the nucleophile to form the
reactions that it can undergo (Fig. 8.3). The ester or amide bond. Notice that a pyrophos-
phosphates are labeled α, β, and γ, count- phatase is associated with reactions in which
ing from the ribose moiety, with α being pyrophosphate is displaced. Because of the high
the phosphate closest to the ribose. If the α free energy of hydrolysis of the pyrophosphate
phosphate is attacked, then AMP is the group bond, the group transfer reaction is driven to
transferred and PPi is the leaving group. For completion.
www.ebook777.com
Fig. 8.8 ATP provides the energy to make ester and amide linkages. (A) A carboxyl attacks the α phosphate
of ATP, displacing pyrophosphate, and forming the AMP derivative. The hydrolysis of pyrophosphate cata-
lyzed by pyrophosphatase drives the reaction to completion. The AMP is then displaced by an attack on the
carbonyl carbon by a hydroxyl or amino group, resulting in the transfer of the acyl group to form the ester or
the substituted amide. (B) Similar to (A) except that the attack is on the γ phosphate of ATP, displacing ADP,
and forming the acyl phosphate.
8.2.2 How ATP is used to form peptide The aminoacyl–RNA is attacked by the nucleo-
bonds during protein synthesis philic nitrogen of an amino group from another
As an example of the principles described thus amino acid, and the acyl portion is transferred
far in Section 8.2, we will consider the forma- to the amino group, forming the peptide bond.
tion of peptide bonds during protein synthe- This last reaction (Fig. 8.9C) takes place on
sis. The acyl donor is made by derivatizing the the ribosome. The reaction goes to completion
α-carboxyl group on the amino acid with AMP, because of the large difference in group transfer
using ATP as the AMP donor (Fig. 8.9A). The potential between the aminoacyl–RNA and the
acyl–AMP is the high-energy intermediate. In peptide that is formed.
the next reaction (Fig. 8.9B) the acyl group is The reactions shown in Fig. 8.9 exemplify
transferred to transfer RNA (tRNA) so that the principle that the energy to make covalent
the carboxyl group becomes derivatized with bonds (in this case a peptide bond) derives from
tRNA. The acyl–tRNA also has a high group a series of group transfer reactions starting with
transfer potential. Thus, energy has flowed from ATP, in which ATP provides the energy to make
ATP to aminoacyl–MP to aminoacyl–RNA. high-energy intermediates that serve as group
Fig. 8.9 Formation of a peptide bond (an amide linkage between two amino acids). Peptide bonds are formed
as a result of a series of group transfer reactions. (A) Transfer of AMP from ATP to the carboxyl group of the
amino acid. In this reaction the α phosphorus of ATP is attacked by the OH in the carboxyl group. Recall that
the P=O bonds in ATP are semipolar and that the phosphorus is an electropositive center. Most of the group
transfer potential of ATP is trapped in the product, and the reaction is freely reversible. However, the reaction
is driven to completion by the hydrolysis of the pyrophosphate (not shown). (B) The displacement of the AMP
by tRNA. This is not done for energetic reasons but, rather, because the tRNA is an adapter molecule that aids
in placing the amino acid in the correct position with respect to the mRNA on the ribosome. The synthesis of
the aminoacyl–RNA is reversible, indicating that the group transfer potential of the aminoacyl–RNA is simi-
lar to that of the aminoacyl–MP. (C) The displacement of the tRNA by the amino group of a second amino
acid, resulting in the synthesis of a peptide bond, takes place on the ribosome. This reaction proceeds with the
release of a relatively large amount of free energy and is irreversible. On the ribosome, it is the amino group
of the incoming aminoacyl–RNA at the “a” site that attacks the carbonyl of the resident aminoacyl–RNA at
the “p” site. The polypeptide is thus transferred from the “p” site to the “a” site as the tRNA at the “p” site is
released. The tRNA of the incoming amino acid at the “a” site is represented by R1, and the incoming amino
acid is glycine.
donors. In this way, all the large complex mol- ΔEh volts provide the energy to establish the
ecules (proteins, nucleic acids, polysaccharides, Δp (nΔEh = yΔp, where n is the number of
lipids, etc.) are synthesized. electrons transferred and y is the number of
protons extruded). ATP in the soluble part of
8.3 ATP Synthesis by Substrate-Level the cell is made by phosphorylating ADP in a
Phosphorylation process called substrate-level phosphoryla-
tion. We can define a substrate-level phospho-
We have seen how ATP can drive the synthesis
rylation as the phosphorylation of ADP in the
of biological molecules via a coupled series of
soluble part of the cell by means of a high-en-
group transfer reactions. But how is ATP itself
ergy phosphoryl donor. Substrate-level phos-
made? The answer depends upon whether the
phorylations are catalyzed by enzymes called
ATP is synthesized in the membranes or in the
kinases:
cytosol.
In the membranes, the phosphorylation of Y~P + ADP ⎯⎯ → Y + ATP
kinase
ADP is coupled to oxidation–reduction reactions

via the generation of an electrochemical gradient During a substrate-level phosphorylation, an
of protons (Δp), which is then used to drive the oxygen in the β phosphate of ADP acts as a
phosphorylation of ADP via the membrane ATP nucleophile and bonds to the phosphate phos-
synthase. That process, called oxidative phosphorus in the high-energy donor. The phosphoryl
phorylation or electron transport phosphoryla- group is transferred to ADP, making ATP. (This
tion, was discussed in Sections 4.6.2 and 4.7.1. is a phosphoryl group transfer reaction similar
In electron transport phosphorylation, elec- to that shown in Fig. 8.1.) Consider the phos-
trons traveling over a potential difference of phoryl group transfer from an acyl phosphate
www.ebook777.com
to ADP (Fig. 8.10). Phosphoryl donors for ATP The single exception is the synthesis of phos-
synthesis during substrate-level phosphoryla- phoenopyruvate, which results from a dehydra-
tions include 1,3-bisphosphoglycerate (BPGA), tion. During the oxidation–reduction reaction,
phosphoenolpyruvate (PEP), acetyl phosphate, –nFΔE joule is used to create a molecule with
and succinyl–CoA plus inorganic phosphate; a high phosphoryl group transfer potential.
for these high-energy phosphoryl donors, see The synthesis of the phosphoryl donors and the
Table 8.1. The four major substrate-level phos- substrate-level phosphorylations of ADP are
phorylations, listed in Table 8.2, occur in the described next.
following metabolic pathways:
1. The substrate-level phosphorylations that use 8.3.1 1,3-Bisphosphoglycerate
BPGA and PEP take place during glycolysis. During the degradation of sugars in the meta-
2. The succinyl–CoA reaction is part of the cit- bolic pathway called glycolysis, the 6-carbon
ric acid cycle. sugar glucose is cleaved into two 3-carbon frag-
3. Acetyl phosphate is formed from acetyl– ments called phosphoglyceraldehyde (PGALD)
CoA, itself formed from pyruvate. This is an (Chapter 9). The phosphoglyceraldehyde is
important source of ATP in anaerobically then oxidized to 1,3-bisphosphoglycerate,
growing bacteria. using inorganic phosphate as the source of
phosphate and NAD+ as the electron acceptor
The synthesis of ATP via substrate-level phos- (Fig. 8.11). The reaction is catalyzed by phos-
phorylation first requires the synthesis of one phoglyceraldehyde dehydrogenase. The energy
of the high-energy phosphoryl donors listed in that would normally be released as heat from
Table 8.2. All but one of the reactions that syn- the oxidation (–2FΔE90 = –44,000 J) is used to
thesize a high-energy molecule are oxidations. drive the synthesis of 1,3-bisphosphoglycerate,
which has a phosphoryl group transfer potential
of 49,000 (–ΔC90,hyd). (These are standard free
energy changes. The actual free energy changes
depend upon the ratios of concentrations of prod-
ucts to reactants.) The 1,3-bisphosphoglycerate
Fig. 8.10 A substrate-level phosphorylation. This is then donates a phosphoryl group to ADP, in a
an example of an acyl phosphate donating a phospho- substrate-level phosphorylation, to form ATP
ryl group to ADP, specifically, the phosphorylation and 3-phosphoglycerate in a reaction catalyzed
of ADP by 1,3-bisphosphoglycerate. The carboxylic by phosphoglycerate kinase (Fig. 8.12). Thus,
acid is displaced. the oxidation of phosphoglyceraldehyde by
NAD+ drives ATP synthesis.
Table 8.2 Four substrate-level phosphorylations
8.3.2 Acetyl phosphate
1,3-BPGA + ADP → 3-PGA + ATP Acetyl phosphate, another high-energy phos-
PEP + ADP → pyruvic acid + ATP phoryl donor, can be made from pyruvate via
Acetyl–P + ADP → acetic acid + ATP
Succinyl–CoA + Pi + ADP → succinic acid + ATP + acetyl–oA. The sequence is
CoASH
Pyruvate → acetyl – CoA → acetyl phosphate
Fig. 8.11 Oxidation of phosphoglyceraldehyde (PGALD). The incorporation of inorganic phosphate (Pi) into
a high-energy phosphoryl donor occurs during the oxidation of PGALD. The product is an acyl phosphate,
1,3-bisphosphoglycerate (BPGA). Energy that would normally be released as heat is trapped in the BPGA
because inorganic phosphate, rather than water, is the nucleophile. As a consequence, an acyl phosphate,
rather than a free carboxylic acid, is formed.
Fig. 8.12 The phosphoglycerate kinase reaction.

ADP carries out a nucleophilic attack on the phos-
phoryl group of 1,3-bisphosphoglycerate (BPGA),
displacing the free carboxylic acid (3-phosphoglyc-
erate, 3-PGA).
Fig. 8.13 Three enzyme systems that decarboxy-
The acetyl phosphate donates the phosphoryl
late pyruvic acid to acetyl–CoA. Aerobically grow-
group to ADP (to form ATP), and the acetate that
ing bacteria and mitochondria use the pyruvate
is produced is excreted into the medium. These dehydrogenase complex. The acetyl–CoA that is
reactions are extremely important for ferment- produced is oxidized to carbon dioxide in the citric
ing bacteria, and they account for the acetic acid acid cycle. Some anaerobic bacteria may also have
that is produced during certain fermentations. a pyruvate dehydrogenase. Anaerobically grow-
The oxidation of pyruvate to acetyl–CoA will ing bacteria generally use pyruvate–ferredoxin
be discussed first (the pyruvate dehydrogenase, oxidoreductase or pyruvate–formate lyase instead
pyruvate–ferredoxin reductase, and pyruvate– of pyruvate dehydrogenase. The acetyl–CoA that
formate lyase reactions). This will be followed is produced anaerobically (during fermentations)
by a description of the conversion of acetyl–CoA is converted to acetyl phosphate via the phospho-
transacetylase reaction. The acetyl phosphate
to acetyl phosphate (the phosphotransacety-
serves as a phosphoryl donor for ATP synthesis (the
lase reaction). Then the synthesis of ATP and
acetate kinase reaction), and the product acetate is
acetate from acetyl phosphate and ADP will be excreted.
described (the acetate kinase reaction).
Formation of acetyl–CoA from pyruvate found in any of the archaae thus far examined.2)
The pyruvate–ferredoxin oxidoreductase reac-
Acetyl–CoA is usually made by the oxidative
tion is an oxidative decarboxylation of pyruvate
decarboxylation of pyruvate, which is a key
to acetyl–CoA in which the electron acceptor
intermediate in the breakdown of sugars. There
is ferredoxin. The pyruvate–formate lyase is
are three well-characterized enzyme systems
an oxidative decarboxylation of pyruvate to
in the bacteria that decarboxylate pyruvate
acetyl–CoA in which the electrons remain with
to acetyl–CoA. One is found in aerobic bacte-
the carboxyl group (rather than being trans-
ria (and mitochondria) and is called pyruvate
ferred to an acceptor such as NAD+ or ferre-
dehydrogenase. It is usually not present in
doxin), which is released as formate.
anaerobically growing bacteria. The pyruvate
An important difference between pyruvate
dehydrogenase reaction is an oxidative decar-
dehydrogenase and the other two enzymes is
boxylation of pyruvate to acetyl–CoA in which
that the pyruvate dehydrogenase reaction pro-
the electron acceptor is NAD+. Acetyl–CoA
duces NADH. This result can be disadvanta-
formed aerobically via pyruvate dehydrogenase
geous to fermenting bacteria because there is
is not a source of acetyl phosphate but usually
often no externally provided electron acceptor
enters the citric acid cycle, where it is oxidized
to reoxidize the NADH. This may be why pyru-
to CO2 (Chapter 9).
vate dehydrogenase is usually not found in fer-
The other two enzyme systems that oxi-
menting bacteria.
dize pyruvate to acetyl–CoA are found only
in bacteria growing anaerobically. These 1. The pyruvate dehydrogenase reaction
are pyruvate–ferredoxin oxidoreductase and Bacteria that are respiring aerobically use
pyruvate–formate lyase (Fig. 8.13). (Pyruvate– pyruvate dehydrogenase to decarboxylate
ferredoxin oxidoreductase has been found in pyruvic acid to acetyl–CoA. This enzyme
several archaea, but pyruvate–formate lyase reaction (Fig. 8.14) is also found in mito-
and pyruvate dehydrogenase have not been chondria. A more detailed description of the
www.ebook777.com
Fig. 8.14 The pyruvate dehydrogenase reaction. In this reaction, the two electrons that bond the carboxyl
group to the rest of the molecule are transferred by the enzyme to NAD+. At the same time, coenzyme A
attaches to the carbonyl group to form the acylated coenzyme A derivative. If the oxidation were to take place
by means of the :OH from water instead of the :SH from CoASH to supply the fourth bond to the carbonyl
carbon, the product would be acetic acid and a great deal of energy would be lost as heat. But the thioester
of the carboxyl group cannot resonate as well as the free carboxyl group and, for this reason, the energy that
normally would have been released during the oxidation is “trapped” in the acetyl–SCoA, a molecule with a
high group transfer potential.
Fig. 8.15 The pyruvate–ferredoxin oxidoreductase reaction and the hydrogenase. The electrons travel from
the ferredoxin (Fd) to protons via the enzyme hydrogenase. Enzymes: 1, pyruvate–ferredoxin oxidoreductase;
2, hydrogenase.
pyruvate dehydrogenase reaction can be found is reversible and is used for autotrophic CO2
in Section 9.7. fixation in certain anaerobic bacteria (Sections
14.1.3 and 14.1.4).
2. Pyruvate–ferredoxin oxidoreductase
Most anaerobically growing bacteria do not 3. Pyruvate–formate lyase
use pyruvate dehydrogenase to oxidize pyru- Pyruvate–formate lyase is an enzyme found
vate to acetyl–CoA and CO2. Instead, they in some fermenting bacteria (e.g., the enteric
use pyruvate–ferredoxin oxidoreductase or bacteria and certain lactic acid bacteria.). In
pyruvate–formate lyase. Pyruvate–ferredoxin the reaction, the electrons stay with the car-
oxidoreductase is found in the clostridia, boxyl that is removed, and therefore formate is
sulfate-reducing bacteria, and some other formed instead of carbon dioxide. Among the
anaerobes. The enzyme catalyzes a reaction advantages of using pyruvate–formate lyase are
similar to that of pyruvate dehydrogenase that neither reduced ferredoxin nor NADH is
except that the electron acceptor is not NAD+. produced and the electrons are disposed of as
Instead, it is an iron–sulfur protein called ferre- part of the formate. The reaction is illustrated
doxin (Fig. 8.15). in Fig. 8.16.
An important feature of the pyruvate–
ferredoxin oxidoreductase in fermenting bac- Formation of acetyl phosphate from
teria is that the enzyme is coupled to a second acetyl–CoA
enzyme called hydrogenase. The hydroge- The acetyl–CoA that is made by using
nase catalyzes the transfer of electrons from either pyruvate–ferredoxin oxidoreductase or
reduced ferredoxin to H+ to form hydrogen gas, pyruvate–formate lyase is converted to acetyl
accounting for much of the hydrogen gas pro- phosphate by the displacement of the CoASH
duced during fermentations. The importance of by inorganic phosphate in a reaction catalyzed
the hydrogenase reaction is that it reoxidizes the by phosphotransacetylase. Sometimes referred
reduced ferredoxin, thus allowing the continued to as the PTA enzyme, phosphotransacetylase:
oxidation of pyruvate. The ferredoxin-linked
decarboxylation of pyruvate to acetyl–CoA Acetyl–CoA + Pi ↔ acetyl-P + CoASH
(Acetyl-P can also be made directly from pyru- Acetate + ATP + CoA
vate and inorganic phosphate by using pyru-
←⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
AMP-dependent acetyl–CoA synthetase
→
vate oxidase. This is a flavoprotein enzyme
found in certain Lactobacillus species. See acetyl–CoA + AMP + PPi
Section 15.9.)
An ADP-dependent acetyl–CoA synthetase
Formation of ATP from acetyl phosphate
Although essentially all acetyl–CoA synthetases
The acetyl phosphate then donates the phospho-
that have been studied generate AMP and PPi
ryl group to ADP in a substrate-level phospho-
(the AMP-dependent acetyl–CoA synthetases)
rylation catalyzed by acetate kinase, sometimes
and have been isolated from bacteria, methano-
referred to as the ACK enzyme. The route to
genic archaea, and eukaryotes, there also exist
the formation of acetate from acetyl–CoA via
the so-called ADP-dependent acetyl–CoA syn-
ACK and PTA is referred to as the ACK–PTA
thetases, which generate ADP and Pi. The lat-
pathway. Fermenting bacteria oxidize pyruvate
ter type of enzyme has thus far not been found
to acetate by using both the phosphotransacety-
to be widely distributed and has been reported
lase and the acetate kinase, and they derive an
in Entamoeba histolytica and Giardia lamblia,
ATP from the process while excreting acetate
two eukaryotic microorganisms, and in cer-
into the medium. We will return to this subject
tain archaea, including the hypothermophilic
in Chapter 15, where fermentations are dis-
archaeon Pyrococcus furiosus, which ferments
cussed in more detail.
carbohydrates and peptides and grows at tem-
Acetyl-P + ADP ←⎯⎯⎯ → acetate +ATP
acetate kinase peratures as high as 105 °C.3 The latter three
organisms excrete acetate into the medium and
Making acetyl–CoA during appear to use the ADP-dependent acetyl–CoA
growth on acetate synthetase as a means of generating ATP. Note
that the AMP-dependent acetyl–CoA synthetase
Note that acetate kinase and phosphotrans-
is generally used for acetate utilization, whereas
acetylase catalyze reversible reactions and
the ADP-dependent acetyl–CoA synthetase
can be used during growth on acetate to make
may be involved primarily with acetate produc-
acetyl–CoA, which is then incorporated into cell
tion (and ATP synthesis), and can be viewed as
material or oxidized to CO2. Another means of
an alternative to the ACK–PTA pathway.
making acetyl–CoA when bacteria are growing
on acetate is by use of the enzyme acetyl–CoA Acetate + ATP + CoA
synthetase (the ACS enzyme). ADP-dependent acetyl −CoA synthetase
←⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →
acetyl − CoA + ADP + Pi
8.3.3 Succinyl–Coenzyme A
Succinyl–CoA is made by the oxidative decar-
Fig. 8.16 The pyruvate–formate lyase reaction. Part boxylation of α-ketoglutarate, a reaction that
of the molecule becomes oxidized and part becomes occurs in the citric acid cycle (Fig. 8.17). It is
reduced. The part that becomes reduced is the car- strictly analogous in its mechanism and cofac-
boxyl group that leaves as formic acid. tor requirements to the oxidative decarboxyla-
tion of pyruvate by pyruvate dehydrogenase.
Fig. 8.17 The α-ketoglutarate dehydrogenase reaction. Notice that the substrate molecule resembles pyru-
vic acid. The difference is that in pyruvic acid the R group is H, whereas in α-ketoglutaric acid, it is CH2–
COOH.
www.ebook777.com
The enzyme that carries out the oxidation of can be metabolized.) However, these reactions
α-ketoglutarate is called α-ketoglutarate dehy- cannot account for the synthesis of net ATP
drogenase. Succinyl–CoA then drives the syn- from ADP and inorganic phosphate, since the
thesis of ATP from inorganic phosphate and phosphate in the PEP originated from ATP (or
ADP in a reaction catalyzed by the citric acid PEP), rather than from inorganic phosphate.
cycle enzyme succinate thiokinase. The reaction Phosphoenolpyruvate is also necessary for the
is as follows: synthesis of muramic acid and certain amino
acids (Chapters 10 and 12).
Succinyl–CoA + ADP + Pi →
succinate + CoASH +ATP
8.4 Summary
Succinyl phosphate is not a free intermedi- Biochemical reactions in the cytosol are driven
ate. Perhaps the CoASH is transferred from by high-energy molecules, so called because they
succinyl–CoA to the enzyme, where it is displaced have a bond with a high free energy of hydrolysis.
by phosphate. The phosphorylated enzyme The reasons for this depend upon the structure of
would then be the phosphoryl donor for ADP the whole molecule, not on any particular bond.
in ATP synthesis. It should be pointed out that High free energies of hydrolysis can be due to elec-
whereas bacteria and plants produce ATP from trostatic repulsion between adjacent phosphate
succinyl–CoA, animals produce GTP instead. groups and/or diminished resonance. There are
Succinyl–CoA is important not only for ATP syn- several high-energy molecules (e.g., ATP, BPGA,
thesis but also as a precursor for heme synthesis. PEP, acetyl-P, acetyl–CoA, succinyl–CoA).
The term “high-energy bond” is a misno-
8.3.4 Phosphoenolpyruvate mer because the free energy of hydrolysis is not
Cells that are using the glycolytic pathway to bond energy. The term “group transfer poten-
grow on sugars make phosphoenolpyruvate tial” refers to the negative of the free energy of
from 2-phosphoglycerate (2-PGA), an interme- hydrolysis and is a useful concept when one is
diate in the breakdown of the sugars. The re- comparing the tendency of chemical groups to
action is a dehydration and is catalyzed by the be transferred to attacking nucleophiles. Thus,
enzyme enolase (Fig. 8.18). Phosphoenolpyruvate ATP has a high phosphoryl group transfer
then phosphorylates ADP in a reaction cata- potential (around 35 kJ under standard con-
lyzed by pyruvate kinase: ditions, pH 7) whereas glucose-6-phosphate
has a low phosphoryl group transfer potential,
PEP + ADP → pyruvate + ATP (around 14 kJ). Hence, ATP will transfer the
phosphoryl group to glucose with the release of
The enolase and pyruvate kinase reactions are 35 – 14 = 21 kJ of energy.
important in energy metabolism because they Group transfer reactions can occur with
serve to regenerate the ATP that is used to phos- conservation of energy to form high-energy
phorylate the sugars during the initial stages intermediates, which themselves can be group
of sugar catabolism. (With rare exceptions, donors in coupled reactions. This explains how
sugars must be phosphorylated before they ATP can provide the energy to drive a series of
coupled chemical reactions that result in the
synthesis of nucleic acids, proteins, polysac-
charides, lipids, and so on.
The formation of a high-energy molecule
usually involves an oxidation of an aldehyde
(3-phosphoglyceraldehyde) or the oxidative
decarboxylation of a β-keto carboxylic acid
(pyruvate or succinate). In substrate-level phos-
Fig. 8.18 The enolase reaction, in which 2-phospho-
glycerate (2-PEA) is dehydrated to phosphoenolpyru- phorylation, the energy of the redox reaction is
vate. Isotope exchange studies suggest that the first trapped in an acyl phosphate or an acyl–CoA deriv-
step is the removal of a proton from C2 to form a car- ative. This is in contrast to respiratory phosphory-
banion intermediate, which loses the hydroxyl and lation, where the energy from the redox reaction
becomes phosphoenolpyruvate (PEP). is trapped in a Δp, which drives ATP synthesis.
Phosphoenolpyruvate is not formed as a 4. What features do the synthesis of BPGA,

result of an oxidation–reduction reaction but acetyl–CoA, and succinyl–SCoA have in
from a dehydration. However, the phosphate common with each other but not with PEP?
in phosphoenolpyruvate was already present
5. What do the syntheses of acetyl–CoA and
in 2-phosphoglycerate, having been donated by
succinyl–CoA have in common?
ATP or phosphenolpyruvate during the sugar
phosphorylations. Therefore, synthesis of ATP 6. Write a series of reactions that result in the
from phosphenolpyruvate does not represent synthesis of a substituted amide or an ester
the formation of ATP from ADP and inorganic from a carboxylic acid. Use ATP as the
phosphate but, rather, the regeneration of source of energy. How is protein synthesis
ATP that had been used to phosphorylate the a modification of this reaction?
sugars.
Study Questions 1. For a discussion of high-energy molecules, see:
Ingraham, L. L. 1962. Biochemical Mechanisms.
1. What is the definition of group transfer John Wiley & Sons, New York.
potential? Why do ATP, acyl phosphates, 2. Selig, M., and P. Schonheit. 1994. Oxidation
and PEP have high phosphoryl group of organic compounds to CO2 with sulfur or thio-
transfer potentials? sulfate as electron acceptor in the anaerobic hyper-
thermophilic archaea Thermoproteus tenax and
2. Write a series of hypothetical reactions in Pyrobaculum islandicum proceeds via the citric acid
which PEP drives the synthesis of A–B from cycle. Arch. Microbiol. 162:286–294.
A, B, and ADP. 3. Mail, X., and M. W. W. Adams. 1996. Purification
and characterization of two reversible and ADP-
3. What are two features that distinguish dependent acetyl–coenzyme A synthetases from the
substrate-level phosphorylations from hyperthermophilic archaeon Pyrococcus furiosus. J.
electron transport phosphorylation? Bacteriol. 178:5897–5903.
www.ebook777.com
9
Central Metabolic Pathways
The central metabolic pathways provide the phosphoryl group to ADP in a substrate-level
precursor metabolites to all the other pathways. phosphorylation. (See Section 8.3.1.)
They are the pathways for the metabolism of The fate of the pyruvate that is formed dur-
carbohydrates and carboxylic acids, such as C4 ing the catabolism of carbohydrates depends on
dicarboxylic acids and acetic acid. The major whether the cells are respiring. If the organisms
carbohydrate pathways are the Embden– are respiring, then the pyruvate that is formed
Meyerhof–Parnas pathway (also called the EMP by the carbohydrate catabolic pathways is oxi-
pathway or glycolysis), the pentose phosphate dized to acetyl–CoA, which is subsequently oxi-
pathway (PPP), and the Entner–Doudoroff dized to carbon dioxide in the citric acid cycle.
pathway (ED). The Entner–Doudoroff pathway The latter generally operates only during aero-
has been found to be restricted almost entirely to bic respiration. (However, see Section 8.8.4.) If
the prokaryotes, being present in gram-negative fermentation rather than respiration is taking
and gram-positive bacteria, as well as in archaea. place, then the pyruvate is converted to fermen-
(It has been reported in the amoeba Entamoeba tation end products such as alcohols, organic
histolytica and in two fungi, Aspergillus niger acids, and solvents, rather than oxidized in the
and Penicillium notatum.1) The three pathways citric acid cycle. Fermentations are discussed in
differ in many ways, but two generalizations Chapter 15.
can be made: Figure 9.1 presents an overview of the carbo-
hydrate catabolic pathways and their relation to
1. All three pathways convert glucose to
one another and to the citric acid cycle. Several
phosphoglyceraldehyde, albeit by different
points can be made about this figure. Notice
routes.
that there are three substrate-level phosphory-
2. The phosphoglyceraldehyde is oxidized to
lations, two during carbohydrate catabolism
pyruvate via reactions that are the same in
and one in the citric acid cycle. Furthermore,
all three pathways.
there are six oxidation reactions, one in glycoly-
From an energetic point of view, the reactions sis, one in the pyruvate dehydrogenase reaction,
that convert phosphoglyceraldehyde to pyru- and four in the citric acid cycle. These oxida-
vate are extremely important because they gen- tions produce NADH (primarily) and FADH2.
erate ATP from inorganic phosphate and ADP. The NADH and FADH2 must be reoxidized to
This is because there is an oxidation in which regenerate the NAD+ and FAD that are required
inorganic phosphate is incorporated into an for the oxidations.
acyl phosphate (i.e., the oxidation of phospho- The route of reoxidation and the energy
glyceraldehyde to 1,3-bisphosphoglycerate). yield depend upon whether the organism is
The 1,3-bisphosphoglycerate then donates the respiring or fermenting. During respiration,
222
central metabolic pathways 223
the NADH and FADH2 are reoxidized via elec- reoxidized in the cytosol by an organic accep-
tron transport with the formation of a Δp. (The tor, but ATP is not made. (However, see note
Δp is used for ATP synthesis via respiratory 2.) The different pathways for the reoxidation
phosphorylation as explained in Chapter 4.) of NADH in fermenting bacteria are discussed
In fermenting cells, most of the NADH is in Chapter 15.
Fig. 9.1 Relationships between the major carbohydrate pathways and the citric acid cycle. The pathway
from glucose-6-phosphate to pyruvate is the Embden–Meyerhof–Parnas pathway (glycolysis). The pentose
phosphate pathway (PPP) and the Entner–Doudoroff pathway (ED) branch from 6-phosphogluconate. Both
these pathways intersect with the glycolytic pathway at phosphoglyceraldehyde. All the carbohydrate path-
ways produce pyruvate, which is oxidized to acetyl–CoA. In aerobically growing organisms, the acetyl–CoA
is oxidized to CO2 in the citric acid cycle. The electrons from NAD(P)H and FADH2 are transferred to the
electron transport chain in respiring organisms, with the formation of ATP. In fermenting cells, the NADH
is reoxidized by an organic acceptor (B) that is generated during catabolism. The citric acid cycle does not
operate as an oxidative pathway during fermentative growth. Abbreviations: G6P, glucose-6-phosphate;
F6P, fructose-6-phosphate; FDP, fructose-1,6-bisphosphate; PGALD, 3-phosphoglyceraldehyde; PEP, phos-
phoenolpyruvate; PPP, pentose phosphate pathway; ED, Entner–Doudoroff pathway; FAD, flavin adenine
dinucleotide; FADH2, reduced FAD.
www.ebook777.com
The student will notice that the citric acid cycle which began with the demonstration of cell-free
generates a great deal of NADH and FADH2. fermentation in yeast extracts. Some of the high-
The reoxidation of the NADH and FADH2 lights of this period are summarized in Box 9.1.
requires adequate amounts of electron acceptor, It is best to think of glycolysis as occurring in
such as is provided to respiring organisms. In two stages:
fact, the oxidative citric acid cycle as illustrated
Stage 1. This stage catalyzes the splitting of the
in Fig. 9.1 is coupled to respiration, and during
glucose molecule (C6) into two phosphoglyc-
fermentative growth it becomes modified into a
eraldehyde (C3) molecules. It consists of four
reductive pathway (Section 9.10).
consecutive reactions. Two ATPs are used
per glucose metabolized, and these donate the
9.1 Glycolysis phosphoryl groups that become the phosphates
The history of biochemical research had its ori- in phosphoglyceraldehyde. (The phosphoglyc-
gins in the elucidation of the glycolytic pathway, eraldehyde becomes phosphoenolpyruvate
BOX 9.1. HISTORICAL PERSPECTIVE: CELL-FREE YEAST

FERMENTATION AND THE BEGINNINGS OF BIOCHEMISTRY1–3
The discovery of cell-free yeast in the accidental discovery of cell-free fer-

fermentation mentation by yeast extracts.
Earlier, in 1893, Eduard Büchner had
During the 1890s two German broth- been working in the laboratory of Adolf
ers, Hans Büchner, a bacteriologist, and von Baeyer at Munich University on a
Eduard Büchner, a chemist, were engaged method to break yeast cells open by grind-
in developing methods to extract proteins ing yeast in a mortar with fine sand. Lack
from microbial cells. Hans Büchner was the of funds, however, had put a stop to the
director of the Munich Institute of Hygiene, research. Eduard later became a professor
and Eduard Büchner was a professor of of analytical and pharmaceutical chemistry
analytical and pharmaceutical chemistry at at Tübingen University.
Tübingen University. Meanwhile, in 1894 Hans Büchner
Hans Büchner wanted to break open bac- became chair of the Department of Hygiene
terial cells to test the idea that pathogenic in Munich, and in 1896 he resumed his
bacteria contained in their protoplasm attempts to break open bacterial cells to
both protein toxins and antitoxins and that isolate the protoplasmic proteins. Knowing
immunized animals were protected from of his brother’s work with sand grinding of
disease by those antitoxins in their sera. At yeast, Hans suggested that his assistant try
the time it was known that animals could be the method. The assistant, Martin Hahn,
made immune to diphtheria or tetanus by ground yeast cells with quartz sand and
injecting them with sera from animals that added water to produce a paste, which was
had survived the disease. wrapped in cloth and pressed in a hydraulic
Hans Büchner was wrong in his suppo- press. A yellowish juice was produced. The
sition that there were protein antitoxins in plan was to study the immunological charac-
bacterial protoplasm that accounted for teristics of the yellow yeast juice. To preserve
immunization. (It was later discovered that the yeast juice while Hahn went on vaca-
polysaccharides in the bacterial cell enve- tion, Hans Büchner suggested the addition
lope act as antigens and elicit an immune of 40% glucose, since sugar was a known
response.) However, his research did result preservative, added to fruit preserves.
Meanwhile, Eduard Büchner was vis- Harden and William John Young. While
iting from Tübingen, using his vacation studying fermentation by yeast extracts,
time to work on pressed yeast juice in his they learned that the rate of fermentation
brother’s laboratory. He prepared yeast could be stimulated by adding inorganic
juice with 40% glucose, and on returning phosphate to the medium. The phosphate
later to the laboratory, noticed bubbles they added rapidly disappeared and was
emanating from the juice. He immediately replaced by a form of phosphate that was not
realized that a cell-free fermentation was precipitable by magnesium citrate mixture.
taking place. Eduard Büchner continued This led the investigators to search for phos-
the research, published his conclusions in phorylated intermediates and resulted in
1897, and in 1907 won the Nobel Prize in the discovery of fructose-1,6-bisphosphate
Chemistry for his work. (the “Harden–Young ester”). When the
Eduard Büchner thought that the fer- fructose-1,6-bisphosphate was added back
mentation was carried out by a single to yeast extracts, the mixture fermented,
soluble intracellular enzyme that he called proving that the ester was indeed an inter-
“zymase.” We now know that sugar fer- mediate in the fermentation pathway.
mentation requires a mixture of enzymes Harden and Young also discovered that
that take part in the glycolytic pathway. dialysis of the yeast extracts destroys the
fermentative activity but that the addition
of boiled yeast extract (not simply inor-
The elucidation of the glycolytic ganic phosphate) restores fermentation.
pathway This led to the discovery of NAD+. Arthur
Harden was knighted in 1926 and in
Between Büchner’s paper in 1897 and the 1929 shared the Nobel Prize in Chemistry
1940s, the chemical reactions of the entire with the Swedish chemist Hans von Enler-
glycolytic pathway were elucidated, as well Chelpin.
as the enzymes catalyzing these reactions.
Chemical analysis identified the metabolic REFERENCES
intermediates, which were made to accu-
mulate in various ways, including the use 1. Fruton, J. S. 1972. Molecules and Life:
Historical Essays on the Interplay of Chemistry
of specific enzyme inhibitors, or by the and Biology. John Wiley & Sons, New York.
removal of specific coenzymes (e.g., by
2. Florkin, M. (Ed.). Comprehensive Bio-
dialysis). chemistry, Vol. 31, A History of Biochemistry,
In addition to cell-free yeast preparations, Part III. 1975. Elsevier Scientific Publishers,
much research was done with cell-free mus- Amsterdam.
cle preparations that convert glucose to lac- 3. Hoffmann-Ostenhof, O. (Ed.). 1987. Inter-
tic acid. An important discovery was made mediary Metabolism. Van Nostrand Reinhold,
in 1905 by the English scientists Arthur New York.
in stage 2, and the phosphate that originated pathway and the Entner–Doudoroff path-
from ATP is returned to ATP in the pyruvate way, accounting for ATP synthesis in these
kinase step, thus regenerating the ATP.) pathways.
Stage 2. This stage catalyzes the oxidation Stage 1: Glucose + 2 ATP → 2 PGALD
of phosphoglyceraldehyde to pyruvate. It + 2 ADP
consists of five consecutive reactions. Stage
Stage 2: 2 PGALD + 2 Pi + 4 ADP + 2 NAD+
2 generates four ATPs per glucose metabo-
→ 2 pyruvate + 4 ATP
lized, hence the net yield of ATP is 2. Stage 2
+ 2 NADH + 2H+
reactions are not unique to glycolysis and also
occur when pyruvate is formed from phos- Sum: Glucose + 2 ADP + 2 Pi + 2 NAD+ →
phoglyceraldehyde in the pentose phosphate 2 pyruvate + 2 ATP + 2 NADH + 2H+
www.ebook777.com
Fig. 9.2 Glycolysis. The following enzymes serve as catalysts: 1, hexokinase; 2, isomerase; 3, phosphofruc-
tokinase; 4, fructose-1,6-bisphosphate aldolase; 5, triosephosphate isomerase; 6, triosephosphate dehydro-
genase; 7, phosphoglycerate kinase; 8, mutase; 9, enolase; 10, pyruvate kinase.
The reactions are summarized in Fig. 9.2. The intramolecular redox reaction. The C2 alcohol
pathway begins with the phosphorylation of becomes oxidized to a ketone as two electrons
glucose to form glucose-6-phosphate (reac- shift to the C1 aldehyde of glucose-6-phosphate
tion 1). The phosphoryl donor is ATP in a and reduce it to an alcohol (Section 9.1.3).
reaction catalyzed by hexokinase. The ATP is The fructose-6-phosphate is phosphorylated
regenerated from phosphoenolpyruvate in stage at the expense of ATP to fructose-1, 6-bis-
2. Some bacteria phosphorylate glucose during phosphate (FBP) by the enzyme fructose-6-
transport into the cell via the phosphotrans- phosphate kinase (reaction 3). The ATP used
ferase (PTS) system, in which case the phos- to phosphorylate fructose-6-phosphate is
phoryl donor is phosphoenolpyruvate. (See also regenerated from phosphoenolpyruvate
Section 17.3.4 for a discussion of the phospho- in stage 2. The fructose-1,6-bisphosphate is
transferase system.) The glucose-6-phosphate split into phosphoglyceraldehyde (PGALD)
(G6P) isomerizes to fructose-6-phosphate (F6P) and dihydroxyacetone phosphate (DHAP) by
in a reaction catalyzed by the enzyme isomerase fructose-1,6-bisphosphate aldolase (reaction 4).
(reaction 2). The isomerization is formally an The splitting of fructose-1,6-bisphosphate is
facilitated by the electron-attracting keto group phosphoenolpyruvate serves as the phosphoryl

at C2, thus rationalizing the isomerization of donor in a second-site, substrate-level phospho-
glucose-6-phosphate to fructose-6-phosphate rylation to form two more moles of ATP and
(Section 9.1.3). The dihydroxyacetone phos- two moles of pyruvate (reaction 10). Notice
phate is isomerized to phosphoglyceraldehyde that the phosphate in the 2-phosphoglycerate
(reaction 5), in a reaction similar to the earlier originated from ATP during the phosphoryla-
isomerase reaction (reaction 2). Thus, stage 1 tions in stage 1 (reactions 1 and 3). Thus, the net
produces two moles of phosphoglyceraldehyde synthesis of ATP from ADP and inorganic phos-
per mole of glucose. phate in glycolysis is coupled to the oxidation of
In stage 2, both moles of phosphoglyceral- phosphoglyceraldehyde to 3-phosphoglycerate
dehyde are oxidized to 1,3-bisphosphoglycer- (reactions 6 and 7). For a more detailed discus-
ate (also called disphosphoglycerate, DPGA) sion of substrate-level phosphorylations, review
(reaction 6). The bisphosphoglycerate serves Sections 8.3.1 through 8.3.4.
as the phosphoryl donor for a substrate-level
phosphorylation catalyzed by the enzyme 9.1.1 Glycolysis as an anabolic pathway
phosphoglycerate kinase (reaction 7). At this The glycolytic pathway serves not only to oxidize
point, two ATPs are made, one from each of carbohydrate to pyruvate and to phosphorylate
the two bisphosphoglycerates. The product ADP, but it also provides precursor metabolites
of the phosphoglycerate kinase reaction is for many other pathways. Figure 9.3 summa-
3-phosphoglycerate (3-PGA). The two moles rizes the glycolytic reactions and points out only
of 3-phosphoglycerate are converted to two a few of the branch points to other pathways.
moles of 2-phosphoglycerate (2-PGA) (reac- For example, glucose-6-phosphate is a precur-
tion 8), which are dehydrated to two moles of sor to polysaccharides, pentose phosphates, and
phosphoenolpyruvate (PEP) (reaction 9). The aromatic amino acids; fructose-6-phosphate
Fig. 9.3 Glycolysis as an anabolic pathway and its regulation in E. coli. The rationale for the pattern of regula-
tion is that when the ADP and AMP levels are high, the ATP levels are low and therefore glycolysis is stimu-
lated. Steady state levels of intermediates are maintained by positive and negative feedback inhibition.
www.ebook777.com
is a precursor to amino sugars (e.g., muramic in regulating the directionality of carbon flow
acid and glucosamine found in the cell wall); are phosphofructokinase (reaction 1) and fruc-
dihydroxyacetone phosphate is a precursor to tose-1,6-bisphosphate phosphatase (reaction
phospholipids; 3-phosphoglycerate is a pre- 2), which catalyze physiologically irreversible
cursor to the amino acids glycine, serine, and steps. The kinase catalyzes the phosphorylation
cysteine; and phosphoenolpyruvate is a precur- of fructose-6-phosphate to fructose-1,6-bis-
sor to aromatic amino acids and to the lactyl phosphate, whereas the phosphatase catalyzes
portion of muramic acid. Note that the thin the dephosphorylation of fructose-1,6-bis-
dark arrows denote reversible steps, whereas phosphate to fructose-6-phosphate. Models for
the large arrows denote irreversible steps that the regulation of glycolysis are based primarily
are highly regulated on in vitro studies of the allosteric properties of
When the organisms are not growing on car- the enzymes. Important effector molecules are
bohydrate, they must synthesize these glyco- AMP and ADP. When these are high, ATP is
lytic intermediates from other carbon sources. low, since both are derived from ATP. That is,
Figure 9.3 shows that some of the carbon from
amino acids, carboxylic acids (organic acids), ATP → ADP + Pi
and lipids are converted to phosphoenolpyru- ATP → AMP + PPi
vate, from which the glycolytic intermediates
can be synthesized. Also, pyruvate can serve as a Thus, high ADP and AMP concentrations are
carbon source and therefore must be converted a signal that the ATP levels are low. (Allosteric
to glycolytic intermediates. However, it can be activation and inhibition are discussed in
seen that the glycolytic pathway can be reversed Chapter 7.) Since glycolysis produces ATP, it
from phosphoenolpyruvate only to fructose- makes sense to stimulate glycolysis when the
1,6-bisphosphate (FBP), and not at all from pyru- ATP levels are low. E. coli accomplishes this
vate. This is because the high free energy in the by allosterically activating the phosphofruc-
phosphoryl donors with respect to the phospho- tokinase with ADP, which, as mentioned, is at
rylated products renders the pyruvate kinase and a higher concentration when the ATP levels are
phosphofructokinase reactions physiologically low. At the same time that glycolysis is stimu-
irreversible. Therefore, to reverse glycolysis, the lated by ADP, the reversal of glycolysis is slowed
kinase reactions are bypassed. The conversion by AMP, which is also at a higher concentra-
of fructose-1,6-bisphosphate to fructose-6- tion when the ATP levels are low. These trends
phosphate requires fructose-1,6-bisphosphate occur simultaneously because AMP inhibits
phosphatase (Fig. 9.4). Alternative ways to con- the fructose-1,6-bisphosphate phosphatase
vert pyruvate directly to phosphoenolpyruvate reaction. The student may notice that the sum
without using pyruvate kinase are discussed in of the phosphofructokinase and fructose-1,6-
Section 9.13. bisphosphatase reaction is the hydrolysis of
ATP (i.e., ATPase activity). The stimulation of
9.1.2 Regulation of glycolysis the phosphofructokinase by ADP and the inhi-
Figure 9.3, in addition to presenting glycolysis bition of the phosphatase by AMP prevent the
as an anabolic pathway, illustrates the regula- unnecessary hydrolysis of ATP when ATP levels
tion of the process in E. coli. Two key enzymes are low.
Fig. 9.4 The conversion of fructose-1,6-bisphosphate, via fructose-1,6-bisphosphatase, to fructose-6-

phosphate.
In E. coli, glycolysis is regulated not only by an electron-attracting keto group at C2 of the

AMP and ADP but also by phosphoenolpyru- sugar, and the electron-attracting keto group
vate and fructose-6-phosphate. As indicated in is necessary to break the bond between C3 and
Fig. 9.3, the phosphofructokinase is feedback C4 in the aldolase reaction. These reactions are
inhibited by phosphoenolpyruvate. This can shown in Fig. 9.5.
be considered to be an example of end-product The isomerization can be viewed as the oxi-
inhibition. The pyruvate kinase, another physi- dation of C2 by C1, because two electrons shift
ologically irreversible reaction, is positively from C2 to C1. This happens in two steps. A pro-
regulated by fructose-1,6-bisphosphate, which ton dissociates from C2, and two electrons shift
is an example of a precursor metabolite activat- in to form the cis-enediol. Then, the proton in the
ing a later step in the pathway. (Feedback inhi- C2 hydroxyl dissociates and two electrons shift
bition and precursor activation are discussed in in, forcing the two electrons in the double bond
Sections 7.1.1 and 7.1.2.) to go to the C1, which recaptures the proton that
dissociated from C2. The result is fructose-6-
9.1.3 The chemical bases for the phosphate. The fructose-6-phosphate becomes
isomerization and aldol cleavage phosphorylated to fructose-1,6-bisphosphate.
reactions in glycolysis The fructose-1,6-bisphosphate is split by the
It is important to understand the chemistry of aldolase when the keto group on C2 pulls elec-
metabolic reactions as well as to learn the path- trons away from the C–C bond between C3 and
ways and their physiological roles. To this end, C4, as two electrons shift in from the hydroxyl
the isomerization and aldol cleavage reactions on C4 (Fig. 9.5). The products of the split are
will be explained because they are common phosphoglyceraldehyde and dihydroxyacetone
reactions that we will see later in other path- phosphate. A second isomerization converts
ways. Consider the isomerization of glucose- the dihydroxyacetone phosphate to phospho-
6-phosphate to fructose-6-phosphate. The rat- glyceraldehyde via the same mechanism as the
ionale for this isomerization is that it creates isomerization between glucose-6-phosphate
Fig. 9.5 Making two phosphoglyceraldehydes from glucose-6-phosphate. Glucose-6-phosphate itself cannot
be split because there is no electron-attracting group to withdraw the electrons from the C–C bond between
C3 and C4. An electron-withdrawing keto group is created on C2 when glucose-6-phosphate is isomerized to
fructose-6-phosphate.
www.ebook777.com
and fructose-6-phosphate. In this way, two coupling site according to the chemiosmotic
phosphoglyceraldehydes can be formed from theory.)
glucose-6-phosphate. Fermentation (anaerobic)
9.1.4 Why are the glycolytic NADH + H + + B(organic) → NAD+ + BH 2

intermediates phosphorylated? Hydrogenase (anaerobic)
In glycolysis, the phosphorylation of ADP by
inorganic phosphate is due to two reactions that NADH + H + → H 2 + NAD+
take place in stage 2: the oxidation of the C1
aldehyde of 3-phosphoglyceraldehyde to the Aerobic and anaerobic respiration are dis-
acyl phosphate and the subsequent transfer of cussed in Chapter 5. In the absence of respira-
the phosphoryl group to ADP. The first reaction tion, NADH can be reoxidized in the cytosol
is catalyzed by triosephosphate dehydrogenase, via fermentation (discussed in Chapter 15). A
and the second reaction is catalyzed by phos- third way to reoxidize NADH is via the enzyme
phoglycerate kinase. Given that these are the hydrogenase in the cytosol. Hydrogenases that
steps in which net ATP is made from ADP and use NADH as the electron donor are found in
inorganic phosphate, one can ask why the other fermenting bacteria. However, the oxidation
intermediates are phosphorylated. of NADH with the production of hydrogen gas
Phosphorylation of the intermediates requires generally proceeds only when the hydrogen gas
the use in stage 1 of two ATPs, which are simply concentration is kept low (e.g., during growth
regenerated in the pyruvate kinase step. There is with hydrogen gas utilizers). This is because
probably more than one reason for the phospho- the equilibrium favors the reduction of NAD+.
rylation of all the intermediates. One such rea- Interspecies hydrogen transfer is discussed in
son may be that the kinase reactions in stage 1 Section 15.4.1.
are irreversible and therefore drive the reactions
rapidly in the direction of pyruvate. Another 9.3 Why Write NAD+ instead of NAD,
possible reason has to do with the physiological and NADH instead of NADH2?
role of the pathway. Glycolysis is not simply a
pathway for the oxidation of glucose and the Oxidized nicotinamide adenine dinucleotide
provision of ATP. Very importantly, the gly- is written as NAD+, and the reduced form is
colytic pathway also provides phosphorylated written NADH, not NADH2. To understand
precursors to many other pathways. In subse- why, we must examine the structures (Fig. 9.6).
quent chapters, we will study these interconnec- Notice that the molecule can accept two elec-
tions with other pathways. trons but only one hydrogen. That is why it is
written as NADH + H+. The oxidized molecule
9.2 The Fate of NADH

If the NADH were not reoxidized to NAD+, then
all pathways (including glycolysis) that require
NAD+ would stop. Clearly, glycolysis must be
coupled to pathways that reoxidize NADH
back to NAD+. Bacteria have three ways of
reoxidizing NADH: respiration, fermentation,
and the hydrogenase reaction.
Respiration (aerobic or anaerobic)
Fig. 9.6 The structures of NAD+, NADH, and nico-
+ tinamide. NAD+ is a derivative of nicotinamide, to
NADH + H + B + yADP + yPi
which ADP–ribose is attached by the nitrogen of nic-
→ NAD+ + BH 2 + yATP otinamide. In the oxidized form, the nitrogen has four
bonds, hence carries a positive charge. NAD+ accepts
Here y is approximately equal to the number two electrons but only one hydrogen (hydride ion) to
of coupling sites and B is the terminal electron become NADH. The second hydrogen removed from
acceptor. (See Section 5.5.2 for a discussion the electron donor (the reductant) is released into the
of the number of ATP molecules formed per medium as a proton.
is written NAD+ because the nitrogen carries a maltose glucose

ADP
formal positive charge. hexokinase
(ADP)
AMP
glucose -6-phosphate
9.4 A Modified EMP Pathway in
the Hyperthermophilic Archaeon fructose-6-phosphate
ADP
phosphofructokinase
Pyrococcus furiosus (ADP) AMP
fructose-1,6-bisphosphate
In comparison to the long history of research on
bacteria, the study of archaeal metabolism is a
glyceraldehyde-3-phosphate dihydroxyacetone phosphate
relatively recent focus of attention, and there CoASH
GAPOR 2e-
have been many rewarding findings that sug- Fd FNOR NADP H2–ase
3-phosphoglycerate
gest the presence of several metabolic features +
2H
distinct from those of the bacteria.3,4 Examples H2
include the novel ether-linked lipids described 2-phosphoglycerate
in Chapter 1 and their biosynthesis, dis- enolase H2O
cussed in Chapter 10. Additional features of phosphoenolpyruvate

ADP
archaeal metabolism that appear to be unique ATP
to these microorganisms include the synthesis pyruvate
ferredoxin
pyruvate
2e -
of methane, and the presence of novel coen- oxidoreductase
Fd FNOR NADP H2–ase
zymes for acetate and methane metabolism acetyl-CoA + CO2
+
ADP + Pi 2H
discussed in Chapter 14. A modified Embden– acetyl-Coa
synthetase ATP + CoaSH H2
Meyerhof–Parnas pathway has been proposed acetate
for Pyrococcus furiosus.5 This organism has a
growth temperature optimum of 100 ºC and Fig. 9.7 Proposed pathway for the fermentation
ferments carbohydrates and peptides to acetate, of maltose, from glucose, to acetate, CO2, and H2
H2, and CO2. Monosaccharides such as glucose in Pyrococcus furiosus: GAPOR, glyceraldehyde-
and fructose do not support growth, but other 3-phosphate ferredoxin oxidoreductase; Fd, ferre-
doxin; FNOR, ferredoxin NADP oxidoreductase;
carbohydrates such as maltose (a disaccharide
H2ase, hydrogenase. Source: Adapted from Mukund,
of glucose) are transported into the cell and con-
S., and M. W. W. Adams. 1995. Glyceraldehyde-3-
verted to glucose. If Sº is present in the medium, phosphate ferredoxin oxidoreductase, a novel tung-
it is used as an electron sink and reduced to H2S. sten-containing enzyme with a potential glycolytic
Figure 9.7 shows pathway proposed for the role in the hyperthermophilic archaeon Pyrococcus
catabolism of glucose to acetate.6 The pos- furiosus. J. Biol. Chem. 270:8389–8392.
tulated pathway resembles the classic EMP
pathway but differs in several respects. Very
been detected in extracts of P. furiosus, and the
interestingly, the phosphoryl donor in the
pathway drawn in Fig. 9.7 is supported by the
hexokinase and fructokinase reactions appears
results of in vivo NMR studies of the products
to be not ATP but, rather, ADP. Another dif-
formed from [13C]glucose.3
ference is that the enzyme that oxidizes glycer-
aldehyde-3-phosphate to 3-phosphoglycerate
is suggested to be a ferredoxin-linked enzyme 9.5 The Pentose Phosphate Pathway
rather than an NAD+-linked enzyme, and Another important pathway for carbohydrate
it appears that 1,3-bisphosphoglycerate is metabolism is the pentose phosphate pathway.
not an intermediate. Pyruvate is oxidized to The pentose phosphate pathway is important
acetyl–CoA and CO2 by pyruvate:ferredoxin first because it produces the pentose phos-
oxidoreductase. (See Section 8.3.2 for a phates, which are the precursors to the ribose
description of this reaction.) Finally, an ADP- and deoxyribose in the nucleic acids, and sec-
dependent acetyl–CoA synthetase catalyzes ond because it provides erythrose phosphate,
a reaction in which ADP is phosphorylated which is the precursor to the aromatic amino
and acetate is formed. See Section 8.3.2 for a acids phenylalanine, tyrosine, and tryptophan.
description of the ADP-dependent acetyl–CoA Also, the NADPH produced in the pentose phos-
synthetase and its distribution. All the enzymes phate pathway is a major source of electrons for
proposed for the modified EMP pathway have biosynthesis in most of the pathways in which
www.ebook777.com
reductions occur. (See note 7 for an alternative Stage 1: Oxidation–decarboxylation reactions

means of generating NADPH.) The oxidation–decarboxylation reactions are
The pentose phosphate pathway is important shown in Fig. 9.8. These reactions oxidize the
to learn for yet one more reason. Several of its C1 in glucose-6-phosphate to a carboxyl group
reactions are the same as the reactions in the and remove it as carbon dioxide. Glucose actu-
Calvin cycle, which is used by many autotrophic ally exists as a ring structure, which forms
organisms to incorporate CO2 into organic car- because the aldehyde group at C1 reacts with
bon (Chapter 14). the C5 hydroxyl group, forming a hemiacetal.
The overall reaction of the pentose phosphate The C1 is therefore not a typical aldehyde in that
pathway is it does not react in the Schiff test and does not
form a bisulfite addition product. Nevertheless,
G6P + 6 NADP + → 3CO2 + 1PGALD it is easily oxidized. The glucose-6-phosphate
+ 6 NADPH + 6H + (9.1) is oxidized by NADP+ to 6-P-gluconolactone
by glucose-6-phosphate dehydrogenase (reac-
9.5.1 The reactions of the pentose tion 1). The lactone is then hydrolyzed to 6-P-
phosphate pathway gluconate by gluconolactonase (reaction 2).
The pentose phosphate pathway is complex In the oxidation of glucose-6-phosphate to
and can be best learned by dividing the reac- 6-phosphogluconate, water contributes the sec-
tions into three stages. Stage I, which consists ond oxygen in the carboxyl group and an acyl
of oxidation–decarboxylation reactions, pro- phosphate intermediate is not formed. This
duces CO2 and NADPH. Stage 2 consists of means that the energy from the oxidation is lost
isomerization reactions that make the precur- as heat, and the reaction is physiologically irre-
sors for stage 3. The stage 3 reactions are sugar versible, as is often the case for the first reaction
rearrangements. The phosphoglyceraldehyde is in a metabolic pathway. (Recall that during the
produced in stage 3. oxidation of phosphoglyceraldehyde, inorganic
Fig. 9.8 The oxidation–decarboxylation reactions. Enzymes: 1, glucose-6-phosphate dehydrogenase; 2, glu-

conolactonase; 3 and 4, 6-phosphogluconate dehydrogenase.
phosphate is added and 1,3-bisphosphoglycer- different structural formulas; that is, their parts
ate is formed. The energy of oxidation is trapped have been switched around. For example, the
in the acyl phosphate rather than being lost as chemical formula for ribulose-5-phosphate is
heat. A subsequent substrate-level phosphoryla- C5H11O8P. Ribose-5-phosphate and xylulose-
tion recovers the energy of oxidation in the form 5-phosphate have the same chemical formula.
of ATP.) The product of the oxidation, 6-P- However, their structures are different (Fig. 9.9).
gluconate, is then oxidized on the C3 to gener- One of the isomerases is an epimerase, RuMP
ate a keto group β to the carboxyl (reaction 3). epimerase, which catalyzes a movement of the
A β-decarboxylation then occurs, generating hydroxyl group from one side of the C3 in rib-
ribulose-5-phosphate (reaction 4). (The mech- ulose-5-phosphate to the other. The product is
anism of β-decarboxylations is described in xylulose-5-phosphate (the epimer8 of ribulose-
Section 9.11.2.) Therefore, the products of the 5-phosphate). The other isomerase converts
three stages are as follows: 1, carbon dioxide; 2, ribulose-5-phosphate to ribose-5-phosphate.
NADPH, and 3, a five-carbon sugar phosphate, Stage 3: The sugar rearrangement reactions
ribulose-5-phosphate (RuMP). The rest of the Stage 3 of the pentose phosphate pathway
pathway continues with ribulose-5-phosphate. involves sugar rearrangement reactions. There
Stage 2: The isomerization reactions are two basic types. One kind transfers a two-
During the second stage, some of the ribulose- carbon fragment from a ketose to an aldose. The
5-phosphate is isomerized to ribose-5-phosphate enzyme that catalyzes the transfer of the two-
and to xylulose-5-phosphate. Isomers are mol- carbon fragment is called a transketolase (TK).
ecules having the same chemical formula but A second kind of reaction transfers a three-
Fig. 9.9 The isomerization reactions of the pentose phosphate pathway. Enzymes: 1, RuMP epimerase; 2,
ribose-5-phosphate isomerase.
Fig. 9.10 The transketolase (TK) and transaldolase (TA) reactions. The donor is always a ketose with the keto
group on C2 and the hydroxyl on C3 on the “left.” In the transketolase reaction, a C2 unit is transferred with its
bonding electrons to the carbonyl group on an aldehyde acceptor. The transaldolase transfers a three-carbon
fragment. In the transketolase reaction, the newly formed alcohol group is on the “left,” which means that the
products of both the transketolase and transaldolase reactions can act as donors in a subsequent transfer.
www.ebook777.com
carbon fragment from a ketose to an aldose. The and fructose-6-phosphate. You will notice that
enzyme that catalyzes the transfer of a three- the sequence of reactions is
carbon fragment is called a transaldolase (TA).
The rule is that the donor is always a ketose Transketolase → transaldolase
(with the OH group of the third carbon “on the → transketolase
left,” as in xyulose-5-phosphate) and the accep-
tor is always an aldose. This rule is important The result is that three moles of glucose-
to learn because we shall see other transketolase 6-phosphate are converted to two moles of
and transaldolase reactions later. Knowing the fructose-6-phosphate and one mole of phos-
requirements will make it easier to remember phoglyceraldehyde. The two moles of fructose-
the reactions. The transketolase and transaldo- 6-phosphate become glucose-6-phosphate by
lase reactions are summarized in Fig. 9.10. isomerization, and the net result is the conver-
sion of one mole of glucose-6-phosphate to one
Summarizing the pentose phosphate mole of phosphoglyceraldehyde, three moles of
pathway carbon dioxide, and six moles of NADPH. This
Figure 9.11 summarizes the pentose phosphate is shown in the carbon balance below.
pathway. Reactions 1 through 3 comprise the
oxidative decarboxylation reactions of stage 1, Carbon balance for the pentose
which produces three moles of ribulose-5-phos- phosphate pathway
phate. Reactions 4 and 5 are the isomerization Later we will study the Calvin cycle, a pathway
reactions of stage 2, in which the three moles of by which many organisms can grow on CO2 as
ribulose-5-phosphate are converted to one mole the sole source of carbon. In the Calvin cycle,
of ribose-5-phosphate and two moles of xylulose- CO2 is first reduced to phosphoglyceraldehyde.
5-phosphate. Reactions 6, 7, and 8 comprise stage The phosphoglyceraldehyde is then converted
3. Reaction 6 is a transketolase reaction in which via sedoheptulose-7-phosphate to pentose
a xylulose-5-phosphate (C5) transfers a two- phosphates, essentially by means of a reversal of
carbon moiety to ribose-5-phosphate (C5), with the pentose phosphate pathway. However, the
the formation of sedoheptulose-7-phosphate Calvin cycle has no transaldolase and synthe-
(C7) and phosphoglyceraldehyde (C3). The two- sizes sedoheptulose-7-phosphate by an alternate
carbon moiety is highlighted as a boxed area. route, which runs irreversibly in the direction of
Reaction 7 is a transaldolase reaction in which pentose phosphates (Section 14.1.1).
the sedoheptulose-7-phosphate transfers a three- The pentose phosphate pathway serves
carbon moiety to the phosphoglyceraldehyde to important biosynthetic functions. Notice that
form erythrose-4-phosphate (C4) and fructose- stage 1 (the oxidative decarboxylation reac-
6-phosphate (C6). The three-carbon moiety is tions) and stage 2 (the isomerization reactions)
highlighted as a dashed box. Reaction 8 is a trans- generate the pentose phosphates required for
ketolase reaction in which xylulose-5-phosphate nucleic acid synthesis. Stage 1 also produces
transfers a two-carbon moiety to the erythrose- NADPH, which is used as a reductant in bio-
4-phosphate forming phosphoglyceraldehyde synthetic pathways. Stage 3 generates the
Oxidative decarboxylation 3 Glucose-6-P → 3 ribulose-5-P + 3CO2

3C6 3C5 3C1
Isomerizations 3 Ribulose-5-P → 2 xylulose-5-P + ribose-5-P
3C5 2C5 C5
Transketolase Xylulose-5-P + ribose-5-P → sedoheptulose-7-P + phosphoglyceraldehyde
C5 C5 C7 C3
Transaldolase Sedoheptulose-7-P + phosphoglyceraldehyde → fructose-6-P + erythrose-4-P
C7 C3 C6 C4
Transketolase Xylulose-5-P + erythrose-4-P → fructose-6-P + phosphoglyceraldehyde
C5 C4 C6 C3
Sum: Glucose-6-P → phosphoglyceraldehyde + 3CO2

C6 C3 3C1
Fig. 9.11 The pentose phosphate pathway. Enzymes: 1, glucose-6-phosphate dehydrogenase; 2, lactonase; 3,
6-phosphogluconate dehydrogenase; 4, ribulose-5-phosphate epimerase; 5, ribose-5-phosphate isomerase;
6 and 8, transketolase; 7, transaldolase. The two-carbon moiety transferred by the transketolase is shown in
the boxed area. The three-carbon fragment transferred by the transaldolase is shown in the dashed box. The
product of the glucose-6-phosphate dehydrogenase reaction is the lactone, which is unstable and hydrolyzes
spontaneously to the free acid. However, there exists a specific lactonase that catalyzes the reaction.
erythrose-4-phosphate necessary for aromatic phosphoglyceraldehyde via the pentose phos-

amino acid biosynthesis. phate pathway. The phosphoglyceraldehyde is
then oxidized to pyruvate via reactions that are
Some bacteria rely completely on the pentose the same as stage 2 of the EMP pathway; then the
phosphate pathway for sugar catabolism pyruvate is oxidized to CO2 via the citric acid cycle.
Thiobacillus novellus and Brucella abortus
lack both stage 1 of the Embden–Meyerhof– Relationship of the pentose phosphate
Parnas pathway and the enzymes of the Entner– pathway to glycolysis
Doudoroff pathway. These organisms use only The pentose phosphate pathway and glyco-
an oxidative pentose phosphate pathway to lysis interconnect at phosphoglyceraldehyde
grow on glucose. They oxidize the glucose to and fructose-6-phosphate (Fig. 9.1). Thus
www.ebook777.com
organisms growing on pentoses can make Glucose + NADP + + NAD+ + ADP + Pi

hexose phosphates. Furthermore, because
stages 2 and 3 of the pentose phosphate path- → 2 pyruvic acid + NADPH + 2H +
way are reversible, it is possible to synthesize + NADH + ATP
pentose phosphates from phosphoglyceral-
dehyde and avoid the oxidative decarboxyla- It can be seen that the pathway catalyzes
tion reactions of stage 1. Such reactions would the same overall reaction as the Embden–
uncouple pentose phosphate synthesis from Meyerhof–Parnas pathway (i.e., the oxidation
NADPH production, possibly conferring an of one mole of glucose to two moles of pyruvic
advantageous metabolic flexibility on the cells. acid), except that only one ATP is made, and
one NADPH and one NADH are made instead
of two NADHs. Only one ATP is made because
9.6 The Entner–Doudoroff Pathway
only one phosphoglyceraldehyde is made from
Many prokaryotes have another pathway for glucose (Fig. 9.12).
the degradation of carbohydrates called the
Entner–Doudoroff (ED) pathway. The other
pathways that we have been studying are com- 9.6.1 The reactions of the Entner–
mon to all cells, whether they be prokaryotes or Doudoroff pathway
eukaryotes. But the Entner–Doudoroff path- The first oxidation is the oxidation of the C1
way has been found almost entirely among the in glucose-6-phosphate to the carboxyl in 6-P-
prokaryotes, including bacteria and archaea. gluconate (Fig. 9.12, reaction 2). These are the
(Reviewed in ref. 1.) The pathway is wide- same enzymatic reactions that oxidize glucose-
spread, particularly among the aerobic gram- 6-phosphate in the pentose phosphate pathway,
negative bacteria. It is usually not found among and they proceed through the gluconolactone.
anaerobic bacteria, perhaps because of the low The pathway diverges from the pentose phos-
ATP yields, as we shall discuss shortly. phate pathway at this point because some of the
Most bacteria degrade sugars via the Embden– 6-P-gluconate is dehydrated to 2-keto-3-deoxy-
Meyerhof–Parnas pathway, but when grown on 6-P-gluconate (KDPG), rather than being oxi-
certain compounds (e.g., gluconic acid), they use dized to ribulose-5-phosphate (reaction 3). The
the Entner–Doudoroff pathway. However, some KDPG is split by KDPG aldolase to pyruvate
strictly aerobic bacteria cannot carry out stage 1 and phosphoglyceraldehyde (reaction 4). The
of the Embden–Meyerhof–Parnas pathway and phosphoglyceraldehyde is oxidized to pyruvate
rely entirely on the Entner–Doudoroff pathway in a sequence of reactions (5–9) identical to
for sugar degradation (Table 9.1). The overall those in stage 2 of the EMP pathway.
reaction for the Entner–Doudoroff pathway is Reaction 3, which is the dehydration of
6-phosphogluconate to KDPG, takes place
via an enol intermediate that tautomerizes to
Table 9.1 Distribution of the EMP and ED pathways KDPG. This is illustrated in Fig. 9.13. In this
in certain bacteriaa
way, it is similar to the dehydration of 2-phos-
Bacterium EMP ED phoglycerate to phosphoenolpyruvate by eno-
lase described in Section 8.3.4. However, in
Arthrobacter species + –
Azotobacter chromococcus + – phosphoenolpyruvate, the enol derivative is
Alcaligenes eutrophus – + stabilized by the phosphate group and the tau-
Bacillus species + –
Escherichia coli and other enteric + – tomerization does not take place until the phos-
bacteriab phoryl group has been transferred.
Pseudomonas species – +
Rhizobium species – +
Xanthomonas species – + 9.6.2 Physiological role for the Entner–
Thiobacillus species – + Doudoroff pathway
a
The absence of the EMP pathway means that the bacteria Since the Entner–Doudoroff pathway produces
rely exclusively on the ED pathway to grow on glucose.
b
When organisms such as E. coli are growing on gluconate,
only one ATP, one can ask why the pathway is
they synthesize the enzymes of the ED pathway. so common in the bacteria. Whereas hexoses
Source: Gottschalk, G. 1986. Bacterial Metabolism. are readily degraded by the Embden–Meyerhof–
Springer-Verlag, Berlin. Parnas pathway, such aldonic acids (aldoses
Fig. 9.12 The Entner–Doudoroff pathway. Because there is only one phosphoglyceraldehyde formed, there
is only one ATP made. The enzymes unique to this pathway are the 6-phosphogluconate dehydratase (reac-
tion 3) and the KDPG aldolase (reaction 4). The other enzymes are present in the pentose phosphate pathway
and the glycolytic pathway. Enzymes: 1, hexokinase; 2, glucose-6-phosphate dehydrogenase; 3, 6-phospho-
gluconate dehydratase; 4, KDPG aldolase; 5, triosephosphate dehydrogenase; 6, PGA kinase; 7, mutase; 8,
enolase; 9, pyruvate kinase.
Fig. 9.13 Dehydration of a carboxylic acid with hydroxyl groups in the α and β positions. The dehydra-
tion of a carboxylic acid with hydroxyl groups in both the α and the β positions leads to the formation of
an enol, which tautomerizes to the keto compound. That is because the hydroxyl on the C3 leaves with
its bonding electrons, and the electrons bonded to the hydrogen on the C2 shift in to form the double
bond. This happens when 6-phosphogluconate is dehydrated to 2-keto-3-deoxy-6-phosphogluconate in
the Entner–Doudoroff pathway and when 2-phosphoglycerate is dehydrated to phosphoenolpyruvate dur-
ing glycolysis. The phosphoenolpyruvate tautomerizes to pyruvate when the phosphate is removed during
the kinase reaction.
www.ebook777.com
oxidized at the aldehydic carbon) as glucon- kinase to form KDPG, which is metabolized to
ate are not; they can, however, be degraded via pyruvate by means of the ordinary ED reactions.
the Entner–Doudoroff pathway. (Aldonic acids Bacteria that have a modified ED pathway can
occur in nature and are sometimes important use it for the catabolism of gluconate, or glucose
nutrients.) if the glucose dehydrogenase is present.
An example of degradation of aldonic
acids occurs when E. coli is transferred from
a medium containing glucose as the carbon
9.7 The Oxidation of Pyruvate
source to one in which gluconate is the source to Acetyl–CoA: The Pyruvate
of carbon. Growth on gluconate results in the Dehydrogenase Reaction
induction of three enzymes: a gluconokinase Pyruvate is the common product of sugar
that makes 6-P-gluconate from the gluconate catabolism in all the major carbohydrate cata-
(at the expense of ATP), 6-P-gluconate dehy- bolic pathways (Fig. 9.1). We now examine the
dratase, and KDPG aldolase. Thus, E. coli uses metabolic fate of pyruvate. One of the fates of
the Entner–Doudoroff pathway to grow on pyruvate is to be oxidized to acetyl–CoA. As
gluconate and the Embden–Meyerhof–Parnas described in Section 8.3.2, this is catalyzed in
pathway to grow on glucose. anaerobically growing bacteria by either pyru-
vate–ferredoxin oxidoreductase or pyruvate–
Some bacteria do not have a complete formate lyase. The acetyl–CoA thus formed is
Embden–Meyerhof–Parnas pathway converted to acetyl phosphate via the enzyme
and rely on the Entner–Doudoroff phosphotransacetylase, and the acetyl phos-
pathway for hexose degradation phate donates the phosphoryl group to ADP
Several prokaryotes (e.g., pseudomonads) do in a substrate-level phosphorylation catalyzed
not make phosphofructokinase or fructose by acetate kinase. The acetate thus formed is
bisphosphate aldolase. Hence, they cannot excreted. (However, certain anaerobic archaea
carry out glucose oxidation via the Embden– oxidize pyruvate to acetyl–CoA by using pyru-
Meyerhof–Parnas pathway. Instead, they use vate–ferredoxin oxidoreductase and then oxi-
the Entner–Doudoroff pathway. (See Table dize the acetyl–CoA to CO2 via the citric acid
9.1 for the distribution of the EMP and ED path- cycle.9) The oxidation of pyruvate to acetyl–
ways.) Some of the pseudomonads even oxidize CoA during aerobic growth is carried out by
glucose to gluconate before degrading it via the the enzyme complex pyruvate dehydrogenase,
Entner–Doudoroff pathway, instead of making which is widespread in both prokaryotes and
glucose-6-phosphate. The oxidation of glucose eukaryotes. (The acetyl–CoA formed during
to gluconate may confer a competitive advan- aerobic growth is oxidized to CO2 in the citric
tage, since it removes glucose, which is more acid cycle.) The overall reaction for pyruvate
readily utilizable by other microorganisms. dehydrogenase is
CH 3COCOOH + NAD+ + CoASH
9.6.3 A partly nonphosphorylated
→ CH 3COSCoA + CO2 + NADH + H +
Entner–Doudoroff pathway
A modified ED pathway has been found in The pyruvate dehydrogenase complex is a
the archaeon Halobacterium saccharovorum very large enzyme complex (in E. coli, about
and in several bacteria, including members 1.7 times the size of the ribosome) located in the
of the genera Clostridium, Alcaligenes, and mitochondria of eukaryotic cells and in the cyto-
Achromobacter, and in Rhodopseudomonas sol of prokaryotes. The pyruvate dehydrogenase
sphaeroides. The pathway is characterized as from E. coli consists of 24 molecules of enzyme
having nonphosphorylated intermediates prior E1 (pyruvate dehydrogenase), 24 molecules of
to 2-keto-3-deoxygluconate. The first reaction enzyme E2 (dihydrolipoate transacetylase), and
is the oxidation of glucose to gluconate via an 12 molecules of enzyme E3 (dihydrolipoate
NAD+-dependentdehydrogenase.Thegluconate dehydrogenase). Several very important cofac-
is then dehydrated to 2-keto-3-deoxygluconate tors are involved. The cofactors are thiamine
by a gluconate dehydratase. The 2-keto-3–de- pyrophosphate (TPP), derived from the vitamin
oxygluconate is phosphorylated by a special thiamine; flavin adenine dinucleotide (FAD),
derived from the vitamin riboflavin; lipoic Step 1. Pyruvate is decarboxylated to form
acid (RS2); nicotinamide adenine dinucleotide “active acetaldehyde” bound to TPP (Fig.
(NAD+), derived from the vitamin nicotin- 9.14). The reaction is catalyzed by pyruvate
amide; and coenzyme A, derived from the vita- dehydrogenase (E1).
min pantothenic acid. (See Box 9.2 for a more Step 2. The “active acetaldehyde” is oxidized
complete discussion of vitamins.) The large to the level of carboxyl by the disulfide in
size of the complex is presumably designed to lipoic acid. The disulfide of the lipoic acid
process the heavy stream of pyruvate that is is reduced to a sulfhydryl. During the reac-
generated during the catabolism of sugars and tion, TPP is displaced and the acetyl group is
other compounds. As described next, the pyru- transferred to the lipoic acid. The reaction is
vate dehydrogenase complex catalyzes a short also catalyzed by pyruvate dehydrogenase.
metabolic pathway rather than simply a single Step 3. A transacetylation occurs in which lipoic
reaction. The individual reactions carried out acid is displaced by CoASH, forming acetyl–
by the pyruvate dehydrogenase complex are as CoA and reduced lipoic acid. The reaction is
follows (Fig. 9.14). catalyzed by dihydrolipoate transacetylase, E2.
BOX 9.2. VITAMINS

Vitamins are growth factors that cannot (retinol) is an important component of the
be made by animals. During the early years light-sensitive pigment in the rod cells of
of vitamin research, vitamins were divided the eye.
into two groups: the fat-soluble vitamins In addition to not being able to make
and the water-soluble vitamins (B complex vitamins, animals cannot make polyunsatu-
and vitamin C). Vitamins are generally rated fatty acids (“essential fatty acids”). In
assayed according to specific diseases that this case, the reason is lack of the enzyme to
they cure in animals fed a vitamin-deficient desaturate monounsaturated fatty acids. A
diet. We now know that the water-soluble diet deficient in polyunsaturated fatty acids
vitamin most responsible for stimulating leads to poor growth, skin lesions, impaired
growth in rats is riboflavin (B2). Vitamin B6 fertility, and kidney damage. Lipoic acid is
(pyridoxine) prevents rat facial dermatitis. synthesized by animals and is therefore not
Pantothenic acid cures chick dermatitis. a vitamin. However, certain bacteria and
Nicotinic acid (a precursor to nicotinamide) other microorganisms require lipoic acid
cures human pellagra. (The symptoms of for growth. It was soon recognized that
pellagra, which can be fatal, are weakness, lipoic acid was required for the oxidation
dermatitis, diarrhea, and mental disorder.) of pyruvic acid. The chemical identifica-
Vitamin C (ascorbic acid) prevents scurvy. tion of lipoic acid followed the purification
Folic acid and vitamin B12 (cobalamin) pre- of 30 mg from 10 tons (!) of water-soluble
vent anemia. Vitamin D, or calciferol (fat residue of liver. This feat was accomplished
soluble), prevents rickets. Vitamin E, or by Lester Reed at the University of Texas
tocopherol (fat soluble), is required for rats in 1949, with collaboration from scientists
for full-term pregnancy and prevents ste- at Eli Lilly and Company. Lipoic acid is a
rility in male rats. Vitamin K (fat soluble) growth factor for some bacteria. It is an
is necessary for normal blood clotting. A eight-carbon saturated fatty acid in which
deficiency in vitamin A (fat soluble) leads C6 and C8 are joined by a disulfide bond to
to dry skin, conjunctivitis of the eyes, night form a ring when the molecule is oxidized.
blindness, and retardation of growth. Male The reduced molecule has two sulfhydryl
rats fed a diet deficient in vitamin A do not groups. The coenzyme functions in the oxi-
form sperm and become blind. Vitamin A dative decarboxylation of α-keto acids.
www.ebook777.com
Fig. 9.14 The pyruvate dehydrogenase complex (PDH). (1), Pyruvate is decarboxylated to “active acetalde-
hyde.” The decarboxylation requires thiamine pyrophosphate (TPP). (2) The “active acetaldehyde” is oxi-
dized to an acylthioester with a high acyl group transfer potential. The oxidant is reduced lipoic acid (R-S2).
(3) The lipoylacylthioester transfers the acetyl group to coenzyme A (CoASH) to form acetyl–CoA. (4) The
reduced lipoic acid is reoxidized by FAD, which in turn is reoxidized by NAD+ (5) The products of the reaction
are acetyl–CoA, CO2, and NADH. Enzymes: E1, pyruvate dehydrogenase; E2, dihydrolipoate transacetylase;
E3, dihydrolipoate dehydrogenase.
Step 4. The lipoic acid is oxidized by dihydroli-

poate dehydrogenase, E3–FAD.
Step 5. The E3–FADH2 transfers the electrons
to NAD+.
All the intermediates remain bound to the
complex and are passed from one active site to
another. Presumably, this offers the advantage
inherent in all multienzyme complexes (i.e., there
is no dilution of intermediates in the cytosol,
and side reactions are minimized). The student Fig. 9.15 Regulation of pyruvate dehydrogenase in
should refer to Section 1.2.6 for a discussion of E. coli. The activity of the enzyme in vitro is modified
multienzyme complexes in the cytoplasm. by several effector molecules. NADH and acetyl–CoA
are negative effectors, and PEP and AMP are positive
9.7.1 Physiological control effectors.
The pyruvate dehydrogenase reaction, which is
physiologically irreversible, is under metabolic and NADH. This can be rationalized as ensuring
control by several allosteric effectors (Fig. 9.15). that the enzyme produces only as much acetyl–
The E. coli pyruvate dehydrogenase is feedback CoA and NADH as can be used immediately. The
inhibited by the products it forms, acetyl–CoA enzyme is also stimulated by phosphoenolpyruvate
(the precursor to pyruvate), presumably signal- The acetyl–CoA that is formed by pyruvate
ing the dehydrogenase that more pyruvate is on dehydrogenase is oxidized to CO2 in the citric
the way. It is also stimulated by AMP, which sig- acid cycle (Fig. 9.1). The overall reaction is:
nals low ATP. The stimulation by AMP probably
reflects the fact that the oxidation of the product Acetyl–CoA + 2H 2O + ADP + Pi + FAD +
acetyl–CoA in the citric acid cycle is a major source
NADP + + 2 NAD+ → 2CO2 + ATP + FADH 2 +
of ATP (via respiratory phosphorylation).
NADPH + 2 NADH + 3H + + CoASH
9.8 The Citric Acid Cycle
See Box 9.3 for a historical perspective on the Notice that there are four oxidations per acetyl–
research that led to the elucidation of the citric CoA, producing two NADHs, one NADPH,
acid cycle. and one FADH2, plus one substrate-level
BOX 9.3. HISTORICAL PERSPECTIVE: THE CITRIC ACID CYCLE
During the first decades of the twentieth carbohydrate, presumably glycogen or a

century, there was considerable effort by product derived from glycogen, in the mus-
biochemists to understand how animal cle tissue. Szent-Györgyi also showed that
tissues oxidized carbohydrates to carbon malonate, previously shown to be an inhibi-
dioxide and water. Between 1910 and tor of succinate dehydrogenase, inhibits the
1920, it was learned that anaerobic sus- respiration of muscle suspensions.
pensions of minced animal tissue trans-
fer hydrogen from the dicarboxylic acids
succinate, fumarate, malate, and the tri- Krebs
carboxylic acid citrate to methylene blue,
reducing the dye to a colorless form. The In 1937 H. A. Krebs and W. A. Johnson
enzymes catalyzing these reactions were published their classic paper on the citric
named “dehydrogenases.” acid cycle. They postulated the cycle on the
bases of results published earlier by others,
including Szent-Györgyi, and their own
Szent-Györgyi experimental results with minced pigeon
breast muscle, especially the finding that cit-
What people wanted to understand was rate catalytically stimulates respiration, that
the link between carbohydrate degradation citrate is converted first to α-ketoglutarate
and the electron transport chain, called the and then to succinate, and that oxaloacetate
cytochrome–cytochrome oxidase system. It is converted to citrate by the addition of two
was suspected that the carboxylic acids pro- carbons from an unidentified donor termed
vided the link. In 1935 the Hungarian-born “triose” (refer to Fig. 9.1). They proposed
scientist Albert Szent-Györgyi discovered the following “citric acid cycle”:
that small amounts of fumarate, malate, or
succinate caused oxygen uptake by minced CO2
↑
pigeon breast muscle far in excess of that
which would be required to completely oxi- Citrate → isocitrate → oxalosuccinate →
dize the carboxylic acids to carbon dioxide CO2
↑
and water. (He determined the ratio of CO2
α -ketoglutarate → succinate → fumarate
produced to O2 consumed, called the respi-
ratory quotient.) He concluded that the car- → malate → oxaloacetate → citrate
boxylic acids were acting catalytically and ↑
2C from "triose"
stimulating the oxidation of an endogenous
www.ebook777.com
In their summary they conclude: combines with CoA to form acetyl–CoA,

which is an active acetyl donor. In a paper
Citric acid catalytically promotes oxi- published in 1947, Lipmann and his col-
dations in muscle, especially in the pres- leagues determined that coenzyme A is a
ence of carbohydrate. The rate of the derivative of pantothenic acid.2
oxidative removal of citric acid from Fritz Lipmann was born in 1899 in
muscle was measured. The maximum Germany. He earned an M.D. degree
figure for Qcitrate observed was –16.9.
in 1924 at Berlin. He studied and did
α-Ketoglutaric acid and succinic acid research in Germany, the Netherlands, and
were found as products of the oxida-
tion of citric acid. These experiments
Denmark. In 1939 he came to the United
confirm Martius and Knoop’s results States, where he held teaching and research
obtained with liver citric dehydrogenase. positions at prestigious institutions, shar-
Oxaloacetic acid, if added to muscle, ing the 1953 Nobel Prize with Hans Krebs.
condenses with an unknown substance Fritz Lipmann died in 1986.
to form citric acid. The unknown sub-
stance is in all probability a derivative of
carbohydrate. The catalytic effect of cit-
Ochoa
rate as well as the similar effects of succi-
nate, fumarate, malate and oxaloacetate Severo Ochoa and some colleagues studied
described by Szent-Györgyi and by Stare the enzyme that converted oxaloacetate
and Baumann are explained by the series to citrate, and in 1952 they reported that
of reactions summarized [elsewhere]. The acetyl–CoA condenses with oxaloacetate
quantitative data suggest that the “citric to form citrate.3
acid cycle” is the preferential pathway Severo Ochoa was born in Spain in
through which carbohydrate is oxidized 1905. He did research and studied in Spain,
in animal tissues.1 Germany, England, and the United States,
becoming an American citizen in 1956.
Hans Krebs was born in Germany in Ochoa shared the Nobel Prize in Physiology
1900 and worked there until 1933, when or Medicine with Arthur Kornberg in 1959.
he emigrated to England. He shared He died in 1993.
the 1953 Nobel Prize in Physiology or
Medicine with Fritz Lipmann for contribu-
REFERENCES
tions toward the understanding of cellular
metabolism. Hans Krebs was knighted in 1. Krebs, H. A., and W. A. Johnson. 1937. The
1958 and died in 1981. role of citric acid in intermediate metabolism in
animal tissue. Enzymologia 4:148–156.
2. Lipmann, F., N. O. Kaplan, G. D. Novelli, L.
Lipmann C. Tuttle, and B. M. Guirard. 1947. Coenzyme
for acetylation, a pantothenic acid derivative. J.
Between 1947 and 1950, Fritz Lipmann Biol. Chem. 167:869–870.
and coworkers discovered and character- 3. Stern, J. R., S. Ochoa, and F. Lynen. 1952.
ized coenzyme A, the so-called coenzyme Enzymatic synthesis of citric acid. V. Reaction of
of acetylation. They learned that acetate acetyl coenzyme A. J. Biol. Chem. 198:313–321.
phosphorylation producing ATP. The cycle 9.8.1 The reactions of the citric acid cycle
usually operates in conjunction with respira- The citric acid cycle is outlined in Fig. 9.16.
tion that reoxidizes the NAD(P)H and FADH2. Reaction 1 is the addition of the acetyl group
Other names for this pathway are the tricar- from acetyl–CoA to oxaloacetate to form cit-
boxylic acid (TCA) cycle and the Krebs cycle. rate. In this reaction the methyl group is depro-
The latter name honors Sir Hans Krebs, who tonated and the resulting anion in which the
did much of the pioneering work and proposed carbon has an unshared pair of electrons acts as
the cycle in 1937. a nucleophile that forms a bond with the carbon
Fig. 9.16 The citric acid cycle. Enzymes: 1, citrate synthase; 2 and 3, aconitase; 4 and 5, isocitrate dehydroge-
nase; 6, α-ketoglutarate dehydrogenase; 7, succinate thiokinase; 8, succinate dehydrogenase; 9, fumarase; 10,
malate dehydrogenase. cis-Aconitate is bracketed because it is an enzyme-bound intermediate.
in the keto group of oxaloacetate. The reaction citric acid cycle, the C3 and C5 carbons will
is driven to completion by the hydrolysis of the be removed as CO2, and carbons C1 to C5
thioester bond of acetyl–CoA, which has a high will regenerate the oxaloacetate. In reaction
free energy of hydrolysis. Reaction 1 is catalyzed 2, catalyzed by aconitase, the citrate is dehy-
by citrate synthase. This enzyme operates irre- drated to cis-aconitate, which remains bound
versibly in the direction of citrate. The oxalo- to the enzyme. Reaction 3 (also catalyzed by
acetate acts catalytically in the cycle; if it is not aconitase) is the rehydration of cis-aconitate
regenerated or replenished, the pathway stops. to form isocitrate, an isomer of citrate. In reac-
Examine the structure of citrate shown in tion 4 (isocitrate dehydrogenase) the isocitrate
Fig. 9.16. In the subsequent reactions of the is oxidized to oxalosuccinate. This oxidation
www.ebook777.com
creates a keto group β to the carboxyl group. 9.8.2 Regulation of the citric acid cycle
The creation of the keto group is necessary The citric acid cycle is feedback inhibited by
for the decarboxylation that takes place in the several intermediates that can be viewed as
next reaction. (β-Keto decarboxylations are end products of the pathway. In gram-negative
explained in Section 9.11.2.) bacteria, the citrate synthase is allosterically
Reaction 5 (isocitrate dehydrogenase) is inhibited by NADH, and in facultative anaer-
the decarboxylation of oxalosuccinate to obes such as E. coli, α-ketoglutarate is an addi-
α-ketoglutarate. Reaction 6 (α-ketoglutarate tional inhibitor. The inhibition of the citrate
dehydrogenase) is the oxidative decarboxy- synthase by NADH may be a way to prevent
lation of α-ketoglutarate to succinyl–CoA. oversynthesis of NADH. The inhibition by
This is an α-decarboxylation, in contrast to a α-ketoglutarate can also be viewed as an exam-
β-decarboxylation. It is a complex reaction and ple of end-product inhibition, in this case to
requires the same cofactors as does the decar- prevent overproduction of the amino acid gluta-
boxylation of pyruvate to acetyl–CoA. mate, which is derived from α-ketoglutarate.10
Reaction 7 (succinate thiokinase) is a substrate- The citrate synthase from gram-positive bac-
level phosphorylation resulting in the formation teria and eukaryotes is not sensitive to NADH
of ATP from ADP and inorganic phosphate. This and α-ketoglutarate but is inhibited by ATP,
reaction was described in Section 8.3.3. Reaction another end product of the citric acid pathway.
8 (succinate dehydrogenase) is the oxidation Recall the discussion in Chapter 7 emphasizing
of succinate to fumarate, catalyzed by a flavin that the pattern of regulation of a particular
enzyme. Succinate dehydrogenase is the only cit- pathway need not be the same from bacterium
ric acid cycle enzyme that is membrane bound. In to bacterium.
bacteria it is part of the cell membrane and trans-
fers electrons directly to quinone in the respiratory
chain. (See the description of the electron trans- 9.8.3 The citric acid cycle as an
port chain in Section 5.3.) Reaction 9 (fumarase) anabolic pathway
is the hydration of fumarate to malate. Finally, The citric acid cycle reactions provide pre-
the oxaloacetate is regenerated by oxidizing the cursors to 10 of the 20 amino acids found in
malate to oxaloacetate in reaction 10 (malate proteins. It is therefore a multifunctional path-
dehydrogenase). way and is not used simply for the oxidation
The citric acid cycle proceeds in the direc- of acetyl–CoA. Succinyl–CoA is necessary for
tion of acetyl–CoA oxidation because of two the synthesis of the amino acids L-lysine and
irreversible steps: the citrate synthase and the L-methionine. Succinyl–CoA is also a precursor
α-ketoglutarate dehydrogenase reactions. to tetrapyrroles, which are the prosthetic group
in several proteins, including cytochromes and
Summing up the citric acid cycle chlorophylls. Oxaloacetate is a precursor to
When one examines the reactions in the citric the amino acid aspartate, which itself is the pre-
acid cycle, it can be seen that there is no net syn- cursor to five other amino acids. In some bac-
thesis. In other words, all the carbon that enters teria, fumarate is also a precursor to aspartate.
the cycle exits as CO2. This is made clear by writ- α-Ketoglutarate is the precursor to the amino
ing a carbon balance. In the carbon balance that acid glutamate, which itself is the precursor
follows, C2 represents the two-carbon molecule to three other amino acids. The biosynthesis
acetyl–CoA, C6 represents citrate or isocitrate, of amino acids is described in Chapter 10. As
C5 represents α-ketoglutarate, and C4 represents noted earlier in Section 9.8, however, since
either succinate fumarate, malate, or oxaloac- the citric acid cycle requires a constant level
etate. Of course, C1 represents carbon dioxide. of oxaloacetate to function, net synthesis of
these molecules requires replenishment of the
Carbon balance for citric acid cycle oxaloacetate.
C2 + C4 → C6
C6 → C5 + C1
9.8.4 Distribution of the citric acid cycle
C5 → C4 + C1
The citric acid cycle is present in most
Sum: C2 → 2C1 heterotrophic bacteria growing aerobically.
Fig. 9.17 Carboxylation reactions that replenish the supply of oxaloacetate. Enzymes: 1, pyruvate carbox-
ylase; 2, PEP carboxylase. Bacteria may have one or the other.
However, not all aerobic bacteria have a com-

plete citric acid cycle. For example, organ-
isms that grow on C1 compounds (methane,
methanol, etc., described in Chapter 14) lack
α-ketoglutarate dehydrogenase and carry
out a reductive pathway as described shortly
(Section 9.9). An oxidative citric acid cycle is
not necessary for these organisms, since acetyl–
CoA is not an intermediate in the oxidation of
the C1 compounds.
Although the oxidative citric acid cycle is a
pathway usually associated with aerobic bacte-
ria, it is also present in certain anaerobes that
oxidize organic compounds completely to CO2.
These include the group II sulfate reducers (dis-
cussed in Section 13.2.2) and certain archaea.
The latter are anaerobic hyperthermophilic
archaea that use sulfur or thiosulfate as the ter-
minal electron acceptor.9 Fig. 9.18 The regulation of PEP carboxylase in
E. coli. The carboxylase is positively regulated by
acetyl–CoA and negatively regulated by aspartate.
9.9 Carboxylations That Replenish
Oxaloacetate: The Pyruvate and
Phosphoenolpyruvate Carboxylases the oxaloacetate (Fig. 9.17). Two enzymes
Because the citric acid cycle intermediates are that carry out the carboxylation of phosphoe-
constantly being removed to provide precur- nolpyruvate and pyruvate are PEP carboxylase
sors for biosynthesis, they must be replaced and pyruvate carboxylase, which are wide-
(Section 9.8.3). Failure to do this would spread among the bacteria. A bacterium will
decrease the level of oxaloacetate that is con- have one or the other. PEP carboxylase is not
tributed to the citrate synthase reaction, and found in animal tissues or in fungi.
thus would prevent the continuation of the
cycle. If the organism is growing on amino acids 9.9.1 Regulation of PEP carboxylase
or organic acids (e.g., malate), replenishment In E. coli, PEP carboxylase is an allosteric
of oxaloacetate is not a problem, since these enzyme that is positively regulated by acetyl–
molecules are easily converted to oxaloacetate CoA and negatively regulated by aspartate
(see later: Fig. 9.27). If the carbon source is a (Fig. 9.18). Presumably, if the oxaloacetate lev-
sugar (e.g., glucose), then the carboxylation of els drop, acetyl–CoA will accumulate and result
pyruvate or phosphoenolpyruvate replenishes in the activation of the PEP carboxylase. This
www.ebook777.com
should produce more oxaloacetate. Aspartate Thus, the pathway is blocked between
can slow down its own synthesis by negatively α-ketoglutarate and succinyl–CoA, and cannot
regulating PEP carboxylase. operate in the oxidative direction (Fig. 9.19).
Succinyl–CoA is made by reversing the reac-
tions between oxaloacetate and succinyl–CoA;
9.10 Modification of the Citric Acid
the enzyme fumarate reductase is used instead
Cycle into a Reductive (Incomplete) of succinate dehydrogenase. The latter enzyme
Cycle during Fermentative Growth is replaced by fumarate reductase under anaer-
In the presence of air, the citric acid cycle obic conditions. These reactions consume 4H.
operates as an oxidative pathway coupled to If one includes the reactions from citrate to
aerobic respiration in respiratory organisms. α-ketoglutarate that produce 2H, the net result
Since fermenting organisms are not carry- is that the reductive pathway consumes 2H.
ing out aerobic respiration, it seems best not The reductive citric acid pathway is found
to have an oxidative pathway that produces not only in fermenting bacteria, but also in some
so much NADH and FADH2. On the other other bacteria (including the enteric bacteria)
hand, the reactions that make oxaloacetate, that are carrying out anaerobic respiration with
succinyl–CoA, and α-ketoglutarate are neces- nitrate as the electron acceptor.11 The reason
sary because these molecules are required for for this is that oxygen induces the synthesis of
the biosynthesis of amino acids and tetrapyr- α-ketoglutarate dehydrogenase in certain fac-
roles. (See Sections 9.7.3 and 9.15, especially ultative anaerobes, and under anaerobic condi-
Fig. 9.27.) tions, the enzyme levels are very low. (However,
The solution to the problem is to convert the some nitrate respirers, e.g., Pseudomonas
citric acid cycle from an oxidative into a reduc- stutzeri grown under denitrifying conditions,
tive pathway. The reductive pathway is also do have an oxidative citric acid cycle.12)
referred to as an incomplete citric acid cycle. One of the consequences of an incomplete cit-
Fermenting bacteria have little or no activity ric acid cycle is that acetate is excreted as a by-
for the enzyme α-ketoglutarate dehydrogenase. product of sugar metabolism during anaerobic
Fig. 9.19 The reductive citric acid pathway in fermenting bacteria. There are two “arms.” One route oxidizes
citrate to α-ketoglutarate. A second route reduces oxaloacetate to succinyl–SCoA. The enzyme α-ketoglutarate
dehydrogenase is missing. The enzyme fumarase replaces succinate dehydrogenase. Enzymes: 1, citrate syn-
thase; 2 and 3, aconitase; 4 and 5, isocitrate dehydrogenase; 6, succinate thiokinase; 7, fumarate reductase; 8,
fumarase; 9, malate dehydrogenase; 10, PEP carboxylase; 11, pyruvate carboxylase.
growth. That is because some of the acetyl–CoA chemistry of the polarized carbonyl group. The
is converted to acetate concomitant with the for- electrons in the carbonyl group are not shared
mation of an ATP. These reactions, which are equally by the carbon and oxygen; rather, the
an important source of ATP, are discussed in oxygen is much more electronegative than the
Chapter 15. It should be pointed out that some carbon and pulls the electrons in the double
strict anaerobes (e.g., the green photosynthetic bond closer to itself. That is, the C=O group is
sulfur bacteria) have a reductive citric acid path- polarized, making the carbon slightly positive
way that is “complete” in that it reduces oxaloac- (Fig. 9.20).
etate to citrate. The pathway, called the reductive Because the oxygen in the carbonyl group is
carboxylic acid pathway, is a CO2 fixation path- electron attracting, there is a tendency to pull
way used for autotrophic growth and differs in electrons away from the C–H bond in the car-
some key enzymological reactions from the path- bon adjacent to the carbonyl. This results in
ways discussed here. The reductive carboxylic the formation of an enolate ion, which acts as
acid pathway is described in Section 14.1.9. a nucleophile (Fig. 9.21). The enolate anion
seeks electrophilic centers (e.g., the carbon
9.11 Chemistry of Some of the atoms in carbonyl groups). So acetyl–CoA can
be a nucleophile at its methyl end, attacking
Reactions in the Citric Acid Cycle other carbonyl groups, even other acetyl–CoA
This section presents a rational basis for under- molecules. In the formation of citric acid, the
standing the chemistry of some of the key reac- methyl group of acetyl–CoA attacks a carbonyl
tions in the citric acid cycle. Similar reactions group in oxaloacetate to form citric acid. At the
are seen in other pathways. same time, the thioester linkage to coenzyme A
is hydrolyzed, driving the reaction to comple-
9.11.1 Acetyl–CoA condensation tion. Later, we will examine other pathways in
reactions which the methyl carbon of acetyl–CoA attacks
Acetyl–CoA is a precursor for the biosynthesis of a carbonyl.
many different molecules besides citrate. These Another result of the polarization of the car-
include lipids (Chapter 10) and various fermen- bonyl group is that the carbon in the carbonyl
tation end products (Chapter 15). Acetyl–CoA is electron deficient and subject to attack by
is so versatile because it undergoes condensa- nucleophiles that seek a positive center (i.e., it
tions at both the methyl and carboxyl ends of is electrophilic). Thus, acetyl–CoA undergoes
the molecule. Condensations at both ends of condensations at the carboxyl end in reac-
acetyl–CoA can be understood in terms of the tions in which the CoASH is displaced during a
nucleophilic displacement and the acetyl portion
is transferred to the nucleophile (Fig. 9.22). For
example, this occurs during fatty acid synthesis
and during butanol fermentations (Chapters 10
Fig. 9.20 The carbonyl group in acetyl–CoA conden- and 15), as well as in several other pathways.
sation reactions is polarized. Electrons are attracted Because of the reactivity of acetyl–CoA at both
by the oxygen, leaving a partial positive charge on the methyl and carboxyl ends, the molecule is
the carbon. widely used in building larger molecules.
Fig. 9.21 The methylene carbon of acetyl–CoA can act as a nucleophile. Because of the electron-attracting ability
of the carbonyl group, a hydrogen dissociates from the methylene carbon, forming an enolate anion that resonates.
Electrons can shift to the methylene carbon, which then seeks a positive center (e.g., a carbonyl group). Because of
this, acetyl–CoA will form covalent bonds to the carbon of carbonyl groups in condensation reactions.
www.ebook777.com
Fig. 9.22 Acetyl group transfer. Because the carbonyl group in acetyl–CoA is polarized, it is subject to nucleo-
philic attack. As a result, the acetyl group is transferred to the nucleophile and CoASH is displaced.
9.11.2 Decarboxylation reactions

Oxalosuccinate is a β-ketocarboxylic acid (i.e.,
the carboxyl group is β to a keto group). The
decarboxylation of β-ketocarboxylic acids
occurs throughout metabolism. Another β-
decarboxylation occurs in the pentose phos-
phate pathway. When 6-phosphogluconate is
oxidized by NADP+, a 3-keto intermediate is Fig. 9.23 The decarboxylation of a β-ketocarboxylic
formed (Fig. 9.8). acid. The keto group attracts electrons, causing an
The decarboxylation of β-ketocarboxylic acids electron shift and the breakage of the C–C bond
is relatively straightforward. A single enzyme is holding the carboxyl group to the molecule. The
required, and the cofactor requirements are met decarboxylation of oxalosuccinate is physiologically
by Mn2+ or Mg2+. A β-decarboxylation is shown reversible. Notice the resemblance to the aldol cleav-
age shown in Fig. 9.5.
in Fig. 9.23. The β-keto group attracts electrons,
facilitating the breakage of the bond holding the
carboxyl group to the molecule. Note the simi- The glyoxylate cycle is required by aerobic bac-
larity to the aldolase reaction, where the keto teria to grow on fatty acids and acetate. (Plants
group facilitates the breakage of a C–C bond and protozoa also have the glyoxylate cycle.
that is β to the keto group (Fig. 9.5). However, it is absent in animals.) The glyoxy-
late cycle resembles the citric acid cycle except
Decarboxylation of α-ketoglutarate that it bypasses the two decarboxylations in the
α-Ketoglutarate is an α-ketocarboxylic acid, citric acid cycle. For this reason the acetyl–CoA
and its decarboxylation is more complex than is not oxidized to CO2.
that of a β-ketocarboxylic acid. The mecha- Examine the summary of the glyoxylate
nism is the same as for the decarboxylation of cycle shown in Fig. 9.24. The glyoxylate cycle
pyruvate, another α-ketocarboxylic acid, and shares with the citric acid cycle the reactions
is described in more detail in Section the sub- that synthesize isocitrate from acetyl–CoA.
section of 15.10.2 entitled. How thiamine The two pathways diverge at isocitrate. In the
pyrophosphate catalyzes the decarboxylation glyoxylate cycle the isocitrate is cleaved to suc-
of 2-ketocarboxylic acids. Pyruvate and α- cinate and glyoxylate by the enzyme isocitrate
ketoglutarate dehydrogenases are similar in lyase (reaction 1). The glyoxylate condenses
that they can be separated into three compo- with acetyl–CoA to form malate, in a reac-
nents: the TPP-containing decarboxylase, the tion catalyzed by malate synthase (reaction 2).
FAD-containing dihydrolipoyl dehydrogenase, The malate synthase reaction is of the same
and the dihydrolipoyl transacetylase. Other type as the citrate synthase (i.e., the methylene
α-ketocarboxylic acid dehydrogenases exist carbon of acetyl–CoA attacks the carbonyl
(e.g., for the catabolism of α-keto acids derived group in glyoxylate). The reaction is driven
from the degradation of the branched-chain to completion by the hydrolysis of the coen-
amino acids leucine, isoleucine, and valine). zyme A–thioester bond just as during citrate
synthesis. The malate replenishes the oxaloac-
etate, leaving one succinate and NADH as the
9.12 The Glyoxylate Cycle products. The cells incorporate the succinate
We now come to a second pathway central to the into cell material by first oxidizing it to oxalo-
metabolism of acetyl–CoA: the glyoxylate cycle, acetate, from which phosphoenolpyruvate is
also called the glyoxylate bypass (Fig. 9.24). synthesized.
Fig. 9.24 The glyoxylate cycle. Enzymes: 1, isocitrate lyase; 2, malate synthase. The dashed arrows represent
reactions of the citric acid cycle that are bypassed.
The net result of the glyoxylate cycle is the isocitrate lyase is high relative to the concentra-
condensation of two molecules of acetyl–CoA to tions of isocitrate. This means that the isocitrate
form succinate. The carbon balance is as follows: lyase requires a high intracellular concentration
of isocitrate. It is presumed that the partial inac-
Carbon balance for the glyoxylate pathway tivation of isocitrate dehydrogenase increases
C2 + C4 → C6 the concentration of isocitrate, resulting in an
C6 → C4 + C2 increase in flux through the glyoxylate cycle.
C2 + C2 → C4
C2 + C2 → C4 9.13 Formation of
Phosphoenolpyruvate
9.12.1 Regulation of the glyoxylate cycle Organic acids such as lactate, pyruvate, acetate,
As Fig. 9.24 illustrates, isocitrate is at a branch succinate, malate, and the amino acids can be
point for both the citric acid cycle and the gly- used for growth because pathways exist to con-
oxylate cycle. That is, it is a substrate for both vert them to phosphoenolpyruvate, which is a
the isocitrate lyase and the isocitrate dehydro- precursor to the glycolytic intermediates. The
genase. What regulates the fate of the isocitrate? phosphoenolpyruvate is generally made in two
In E. coli, the isocitrate dehydrogenase activ- ways: via the decarboxylation of oxaloacetate
ity is partially inactivated by phosphorylation or via the phosphorylation of pyruvate. These
when cells are grown on acetate.13 (The regula- two reactions are described next.
tion of enzyme activity by covalent modification
is discussed in Chapter 7, and other examples 9.13.1 Formation of phosphoenopyruvate
are listed in Table 7.1.) Acetate also induces the from oxaloacetate
enzymes of the glyoxylate cycle. Therefore, in The ATP-dependent decarboxylation catalyzed
the presence of acetate, isocitrate lyase activity by PEP carboxykinase, shown in Fig. 9.25,
increases while the isocitrate dehydrogenase is results in the formation of phosphoenopyruvate
partially inactivated. However, the Km for the from oxaloacetate. The enzyme is widespread
www.ebook777.com
PEP synthetase and pyruvate–phosphate

dikinase reactions
Prokaryotes that convert pyruvate directly to
phosphoenolpyruvate use one of two enzymes:
PEP synthetase or pyruvate–phosphate diki-
Fig. 9.25 The PEP carboxykinase reaction. The PEP nase. These enzymes are found in prokaryotes
carboxykinase generally operates in the direction and plants, but not in animals. The following
of PEP synthesis, although during fermentation in reactions are illustrated in Fig. 9.26:
anaerobes it can work in the direction of oxaloac-
etate, which is subsequently reduced to the fermenta- PEP synthetase (reaction 1)
tion end product, succinate.
Pyruvate + H2O + ATP → PEP + AMP + Pi
Pyruvate–phosphate dikinase (reaction 2)
and accounts for phosphoenolpyruvate syn-
thesis in both eukaryotes and prokaryotes. Pyruvate + ATP + Pi → PEP + AMP + PPi
This enzyme, along with malic enzyme, is Pyrophosphatase (reaction 3)
important for growth on succinate and malate
(Section 9.14). PPi + H2O → 2Pi
Although PEP carboxykinase is an important In the PEP synthetase reaction, a pyrophospho-
enzyme for the synthesis of phosphoenolpyru- ryl group is transferred from ATP to the enzyme.
vate from oxaloacetate, it catalyzes the synthe- (See Fig. 8.3 for a description of pyrophospho-
sis of oxaloacetate from phosphoenolpyruvate ryl group transfer reactions.) One phosphate is
in some anaerobic bacteria. For example, removed by hydrolysis, leaving a phosphory-
anaerobic bacteria that ferment glucose to lated enzyme. The phosphorylated enzyme then
succinate may use this enzyme to carboxylate donates the phosphoryl group to pyruvate to
phosphoenolpyruvate to oxaloacetate and then form PEP. The reaction catalyzed by the pyru-
in other reactions reduce the oxaloacetate to vate–phosphate dikinase is similar, except that
succinate.14 instead of being hydrolytically removed from
the pyrophosphorylated enzyme, the phos-
9.13.2 Formation of phate is transferred to inorganic phosphate to
phosphoenolpyruvate from pyruvate form pyrophosphate. The pyrophosphate is
Prokaryotes are able to synthesize phospho- hydrolyzed by a pyrophosphatase, pulling the
enolpyruvate by phosphorylating pyruvate reaction to completion. The net result from
instead of converting the pyruvate to oxalo- both reactions is the same (i.e., the sum of the
acetate and then decarboxylating the oxaloac- pyruvate–phosphate dikinase and pyrophos-
etate. This is necessary for growth on pyruvate phatase reactions is the same as the PEP syn-
for bacteria that cannot synthesize oxaloac- thetase reaction). In either case, the synthesis of
etate from pyruvate. Prokaryotes that fall into phosphoenolpyruvate from pyruvate requires
the latter category do not have a glyoxylate the hydrolysis of two phosphodiester bonds
pathway and also lack pyruvate carboxylase. with a high free energy of hydrolysis.
For example, E. coli does not have a glyoxylate
cycle unless growing on acetate, and it has PEP
carboxylase instead of pyruvate carboxylase.
Furthermore, there are strict anaerobes (e.g.,
methanogens and green sulfur photosynthetic
bacteria) that grow autotrophically, convert-
ing CO2 to acetyl–CoA, which is carboxylated
to pyruvate (Sections 14.1.2 and 14.1.3).
These microorganisms do not have a glyoxy-
late cycle and must phosphorylate the pyruvate
to form phosphoenolpyruvate. The phospho-
rylation of pyruvate to phosphoenolpyruvate Fig. 9.26 The PEP synthetase (1), pyruvate–phos-
is described next. phate dikinase (2), and pyrophosphatase (3) reactions.
9.14 Formation of Pyruvate from or malate, although strains lacking only one of
Malate these enzymes will grow.
Bacteria and mitochondria can convert malate
directly to pyruvate by using the malic enzyme: 9.15 Summary of the Relationships
Malic enzyme between the Pathways
ZX Figure 9.27 illustrates the relationship between
L-Malate + NAD+ YZ pyruvate + NADH + CO2 the different pathways discussed in this chapter.
Bacteria and mitochondria actually possess two Reactions 1 and 2 are key enzymes in the glyoxy-
malic enzymes: one specific for NAD+ and the late cycle. Reaction 1 is isocitrate lyase. Reaction
other for NADP+. The latter provides NADPH 2 is malate synthase. Fatty acids and acetate are
for reductions that occur during biosynthesis converted to acetyl–CoA, which enters the citric
(e.g., the biosynthesis of fatty acids, as discussed acid cycle and the glyoxylate cycle. Dicarboxylic
in Chapter 10). Malic enzyme, along with PEP acids and amino acids are eventually degraded
carboxykinase (Section 9.13.1), is an important to citric acid cycle intermediates. Sugars can be
enzyme for growth on citric acid cycle inter- catabolized via the Embden–Meyerhof–Parnas
mediates such as succinate or malate. E. coli pathway, the pentose phosphate pathway, or
mutants that lack both PEP carboxykinase the Entner–Doudoroff pathway. All the sugar
and malic enzyme will not grow on succinate pathways intersect at phosphoglyceraldehyde.
Fig. 9.27 Relationship between the glyoxylate cycle, the citric acid cycle, and the major carbohydrate path-
ways. Enzymes: 1, isocitrate lyase; 2, malate synthase; 3, pyruvate carboxylase; 4, PEP carboxylase; 5, PEP
carboxykinase; 6, PEP synthetase or pyruvate phosphodikinase; 7, phosphofructokinase; 8, fructose-1,6-
bisphosphatase.
www.ebook777.com
9.16 Summary bisphosphate aldolase, key enzymes in the first

The nearly ubiquitous pathway for glucose stage of the EMP pathway, and so rely on the ED
degradation is the Embden–Meyerhof–Parnas pathway for growth on glucose. Interestingly,
pathway, also called the glycolytic pathway. modifications of the ED pathway exist in a few
The pathway oxidizes one mole of glucose to bacteria and archaea in which some of the inter-
two moles of pyruvate, and produces two moles mediates are not phosphorylated.
of NADH and two moles of ATP. There is only Among the many pathways that intersect
one oxidation, and that is the oxidation of with the second stage of glycolysis is the pen-
phosphoglyceraldehyde to phosphoglycerate. tose phosphate pathway. The pentose phos-
An intermediate, 1,3-bisphosphoglycerate, is phate pathway oxidizes glucose-6-phosphate
formed, which donates a phosphoryl group to to pentose phosphates. The pentose phosphates
ADP in a substrate-level phosphorylation. Since can be used for the synthesis of nucleic acids,
two phosphoglyceraldehydes are formed from or they can be converted via the transaldolase
one glucose, two ATPs are made. and transketolase reactions to phosphoglycer-
The pathway can be considered as two aldehyde. The pathway also produces NADPH
stages. Stage 1 generates two phosphoglyc- for biosynthetic reductions and erythrose-4-
eraldehydes. Stage 2 is the oxidation of phos- phosphate for aromatic amino acid biosynthe-
phoglyceraldehyde to pyruvate. Not only does sis. The pentose phosphate pathway is also used
glycolysis provide ATP, NADH, and pyruvate, when pentoses are the source of carbon and
but its intermediates are used for biosynthesis in energy.
other pathways; this will become more evident In aerobic prokaryotes, pyruvate is oxida-
in subsequent chapters when we examine other tively decarboxylated via pyruvate dehydro-
metabolic pathways. The pathway cannot be genase to acetyl–CoA, CO2, and NADH. The
reversed from pyruvate to glucose-6-phosphate acetyl–CoA enters the citric acid cycle, where
because of two irreversible reactions, the pyru- it is oxidized completely to CO2. The electrons
vate kinase reaction (phosphoenolpyruvate to generated during the oxidations are transferred
pyruvate) and the phosphofructokinase reac- to the electron transport chain, where a Δp is
tion (fructose-6-phosphate to fructose-1,6- created.
bisphosphate). Both these reactions proceed so The citric acid cycle is also an anabolic path-
far in the direction of products that they are not way. Several biosynthetic pathways draw citric
physiologically reversible. However, the path- acid cycle intermediates out of the pools. For
way can be reversed from phosphoenolpyru- example, oxaloacetate and α-ketoglutarate are
vate, provided fructose-1,6-bisphosphate used for amino acid biosynthesis, and succinyl–
phosphatase is present to convert fructose- CoA is used for the synthesis of tetrapyrrole,
1,6-bisphosphate to fructose-6-phosphate. lysine, and methionine. Oxaloacetate is also
A modification of the EMP pathway exists in used to synthesize phosphoenolpyruvate for
the archaea, where ADP serves as the phos- glucogenesis. Since oxaloacetate is used catalyt-
phoryl donor in the kinase reactions and ferre- ically in the cycle, it must be replenished during
doxin is the electron acceptor in the oxidation growth. When organisms are growing on car-
reactions. bohydrates, the oxaloacetate is replenished by
A second sugar-catabolizing pathway is the carboxylating pyruvate or phosphoenolpyru-
Entner–Doudoroff (ED) pathway. The ED path- vate. Growth on proteins or amino acids poses
way is widespread among prokaryotes, where it no problem, since these substances are degraded
can be used for growth on gluconate. An inter- to citric acid cycle intermediates.
mediate of this pathway is phosphoglyceral- Pyruvate must be converted to phospho-
dehyde, which is oxidized to pyruvate via the enolpyruvate before it can be used as a source
same reactions that occur in stage 2 of glycoly- of cell carbon. The reason for this is that
sis. Because only one phosphoglyceraldehyde is phosphoenopyruvate is a precursor to sev-
produced from glucose-6-phosphate in the ED eral metabolites, including the intermediates
pathway, only one ATP is made. Several bacte- of glycolysis. One route is the ATP-dependent
ria lack phosphofructokinase and fructose-1,6- carboxylation of pyruvate to oxaloacetate
(pyruvate carboxylase) combined with the ent. What might be the metabolic defect in
ATP-dependent decarboxylation to phosphoe- the mutant?
nolpyruvate via the enzyme PEP carboxykinase.
Two ATPs are therefore used to convert one 2. There are two physiologically irreversible
pyruvate to phosphoenolpyruvate. A second reactions in the EMP pathway, starting
route is the direct phosphorylation of pyruvate with glucose-6-phosphate and ending with
to phosphoenolpyruvate via either PEP syn- pyruvate. Which ones are they? How are
thetase or pyruvate–phosphate dikinase com- they regulated in E. coli?
bined with a pyrophosphatase. The products 3. Do you think that the glyoxylate cycle and
are phosphoenolpyruvate, AMP, and Pi. Thus, the citric acid cycle can operate at the same
regardless of the pathway used, two ATPs are time in an organism growing on acetic acid?
required to synthesize phosphoenolpyruvate How might this arrangement be regulated
from pyruvate. in E. coli?
Bacteria growing aerobically on fatty acids or
acetate use the glyoxylate cycle for net incorpo- 4. Fermenting bacteria have little or no
ration of C2 units into cell material. This path- α-ketoglutarate dehydrogenase activity
way is a modification of the citric acid cycle, in and therefore have a block in the citric acid
which the two decarboxylation reactions are cycle. Under these conditions, the pathway
bypassed. The decarboxylation reactions are operates reductively instead of oxidatively.
bypassed by using isocitrate lyase and malate Draw a reductive citric acid pathway show-
synthetase, two enzymes unique to the glyoxy- ing how these organisms make oxaloac-
late cycle. The net result is that acetyl–CoA is etate, succinyl–CoA, and α-ketoglutarate.
converted to succinate rather than to carbon Why is it important to be able to synthe-
dioxide. size these compounds under all growth
The citric acid cycle is usually present only dur- conditions?
ing aerobic growth. Under anaerobic conditions,
5. Which nonglycolytic enzyme is necessary
prokaryotes have a modified pathway called the
to reverse glycolysis from PEP to G6P?
reductive citric acid pathway. However, some
sulfate reducers, which are anaerobic bacteria 6. Some bacteria (e.g., Brucella abortus) lack
that use sulfate as an electron acceptor, have an key enzymes in the EMP and ED pathways
oxidative citric acid cycle. In the reductive citric but can still grow on glucose. Brucella
acid pathway there are two major changes. The abortus is an animal pathogen that causes
α-ketoglutarate dehydrogenase activity is low spontaneous abortion in cattle and can also
or missing, and fumarate reductase replaces infect humans, causing fever, headache,
succinate dehydrogenase. The result is that and joint pains. It has a citric acid cycle.
oxaloacetate is reduced to succinyl–CoA, and Write a series of reactions showing how the
citrate is oxidized to α-ketoglutarate. Thus, pentose phosphate pathway can result in
the intermediates necessary for biosynthesis of glucose oxidation completely to CO2 with-
amino acids and tetrapyrroles are made, but the out the involvement of stage 1 of either the
number of reductions exceeds the number of EMP or the ED pathway.
oxidations.
7. The ED pathway is only 50% as efficient
as the EMP pathway in making ATP from
Study Questions glucose. Why is that?
1. Suppose you isolated a mutant that did 8. E. coli mutants lacking glucose-6-P dehy-
not grow on glucose as the sole source of drogenase can be grown on glucose as
carbon. However, the mutant did grow on the sole source of carbon. Since this is the
the glucose when it was supplemented with enzyme that catalyzes the entrance of G6P
succinate, fumarate, or malate. An exami- into the pentose phosphate pathway, how
nation of broken cell extracts revealed that can these mutants make NADPH or pen-
all the citric acid cycle enzymes were pres- tose phosphates?
www.ebook777.com
9. Show how the pentose phosphate path- 3. Danson, M. J. 1988. Archaebacteria: the com-
way might oxidize glucose completely to parative enzymology of their central metabolic
pathways, pp. 166–231. In: Advances in Microbial
CO2 without using the citric acid cycle.
Physiology, Vol. 29. A. H. Rose and D. W. Tempest
10. The combination of phosphofructoki- (Eds.). Academic Press, New York.
nase (fructose-6-phosphate kinase) and 4. Verhees, C. H., S. W. M. Kengen, J. E. Tuininga,
fructose-1,6-bisphosphatase has ATPase G. J. Schut, M. W. W. Adams, W. M. de Vos, and
activity. Write these reactions and sum V. der Oost. 2003. The unique features of glycolytic
pathways in Archaea. Biochem. J. 375:231–246.
them to satisfy yourself that this is so.
How does E. coli ensure that these reac- 5. Kengen, S. W. M., F. A. M. de Bok, N.-D. van
Loo, C. Dijkema, A. J. M. Stams, and W. M. de Vos.
tions do not use up all the ATP in the
1994. Evidence for the operation of a novel Embden–
cell? Meyerhof pathway that involves ADP-dependent
kinases during sugar fermentation by Pyrococcus
11. When E. coli is grown on pyruvate as its
furiosus. J. Biol. Chem. 269:17537–17541.
sole source of carbon, it does not synthe-
size the glyoxylate cycle enzymes. How is 6. Mukund, S., and M. W. W. Adams. 1995.
Glyceraldehyde-3-phosphate ferredoxin oxidoredu-
the pyruvate converted to PEP? [Hint: For ctase, a novel tungsten-containing enzyme with a
Questions 11–14, start by examining Fig. potential glycolytic role in the hyperthermophilic
9.27.] archaeon Pyrococcus furiosus. J. Biol. Chem. 270:
8389–8392.
12. Suppose you are growing a bacterial cul-
7. However, some bacteria (e.g., E. coli) also contain
ture on ribose as the only source of car-
an enzyme called transhydrogenase, which catalyzes
bon. Write a series of reactions that will the reduction of NADP+ by NADH. Therefore, for
convert ribose to glucose-6-phosphate. those bacteria the pentose phosphate pathway is not
Why is it important to be able to synthe- essential for NADPH synthesis.
size glucose-6-phosphate? 8. Sugars that differ in the configuration at a single
asymmetric carbon are called epimers. For example,
13. Suppose you are growing a culture on xylulose-5-phosphate and ribulose-5-phosphate are
gluconate. How would the cells convert epimers of each other at C3.
the gluconate to phosphoglyceraldehyde
9. Selig, M., and P. Schonhet. 1994. Oxidation of
if they lacked 6-phosphogluconate dehy- organic compounds to CO2 with sulfur or thiosul-
drogenase. How might they make ribose- fate as electron acceptor in the anaerobic hyper-
5-phosphate and erythrose-4-phosphate? thermophilic archaea Thermoproteus tenax and
Why is it necessary to have a supply of the Pyrobaculum islandicum proceeds via the citric acid
cycle. Arch. Microbiol. 162:286–294.
latter two compounds?
10. Danson, M. J., S. Harford, and P. D. J. Weitzman.
14. Suppose you are growing a bacterial cul- 1979. Studies on a mutant form of Escherichia coli
ture on succinate as the source of carbon. citrate synthase desensitized to allosteric effectors.
Write the reactions by which succinate is Eur. J. Biochem. 101:515–521.
converted to PEP. Why is it important to 11. Stewart, V. 1988. Nitrate respiration in rela-
be able to synthesize PEP? tion to facultative metabolism in Enterobacteria.
Microbiol. Rev. 52:190–232.
REFERENCES AND NOTES 12. Spangler, W. J., and C. M. Gilmour. 1966.

Biochemistry of nitrate respiration in Pseudomonas
stutzeri. Aerobic and nitrate respiration routes of
1. Reviewed in: Conway, T. 1992. The Entner–
carbohydrate catabolism. J. Bacteriol. 91:245–250.
Doudoroff pathway: history, physiology, and molec-
ular biology. FEMS Microbiol. Rev. 103:1–28. 13. LaPorte, D. C. 1993. The isocitrate dehydroge-
nase phosphorylation cycle: regulation and enzymol-
2. During some fermentations, a portion of the
ogy. J. Cell. Biochem. 51:14–18.
NADH is reoxidized by fumarate via membrane-
bound electron carriers and a Δp is created. This is 14. Podkovyrov, S. M., and J. G. Zeikus. 1993.
called fumarate respiration. Fumarate respiration Purification and characterization of phosphoe-
occurs during mixed-acid fermentation discussed in nolpyruvate carboxykinase, a catabolic CO2-fixing
Section 15.10; propionate fermentation is discussed enzyme, from Anaerobiospirillum succiniciprod-
in Section 15.7. ucens. J. Gen. Microbiol. 139:223–228.
10
Metabolism of Lipids, Nucleotides,
Amino Acids, and Hydrocarbons
The central pathways described in Chapter 9 Table 10.1. Fatty acids differ in the number of
(i.e., glycolysis, the pentose phosphate path- carbon atoms, in the number of double bonds
way, the Entner–Doudoroff pathway, and the they contain, in where the double bonds are
citric acid cycle) provide precursors to all the placed in the molecule, and in whether the mol-
other metabolic pathways. This chapter focuses ecule is branched (e.g., tuberculostearic acid) or
on some of those metabolic pathways that feed whether it contains cyclopropane (e.g., lactoba-
off the central pathways and are necessary for cillic acid). However, some generalizations can
lipid, protein, and nucleic acid metabolism. be made about the fatty acids. They are usually
16 or 18 carbons long and either are saturated
or have one double bond. Generally, gram-pos-
10.1 Lipids
itive bacteria are richer in branched-chain fatty
Lipids are a structurally heterogeneous group acids than are gram-negative bacteria.
of substances that share the common property
of being highly soluble in nonpolar solvents The role of fatty acids
(e.g., methanol, chloroform) and relatively Fatty acids do not occur free in bacteria but are
insoluble in water. Essentially all of the lipids in covalently attached to other molecules. Most of
prokaryotes are in the membranes. The major the fatty acids are esterified to glycerolphosphate
membrane lipids in bacteria and eukaryotes, derivatives to make phosphoglycerides, which
which are phospholipids consisting of fatty are an important structural component of mem-
acids esterified to glycerol phosphate deriva- branes. Fatty acids may also be esterified to car-
tives, are called phosphoglycerides. Archaea bohydrate. For example, the lipid A portion of
can also have phosphoglycerides, but their lipopolysaccharide consists of fatty acids esteri-
structure is different, as is their mode of synthe- fied to glucosamine (see later: Section 12.2.1).
sis. Section 1.2.5 summarized the structures of Fatty acids can also be esterified to protein. For
archaeal lipids; their synthesis is the subject of example, gram-negative bacteria have a lipopro-
Section 10.1.3. The metabolism of fatty acids is tein that is covalently attached to the peptido-
discussed first. glycan and protrudes into the outer membrane.
The lipoprotein apparently is important for the
10.1.1 Fatty acids stability of the outer membrane (Section 1.2.3).
Types of fatty acid found in bacteria Because fatty acids are nonpolar, they anchor
Fatty acids are chains of methylene carbons the molecules to which they are attached to the
with a carboxyl group at one end. They can membrane. Furthermore, whether a fatty acid is
be branched (methyl), saturated, unsaturated, unsaturated, saturated, or branched has impor-
or hydroxylated. Some examples are shown in tant significance for the physical properties of
255
www.ebook777.com
Table 10.1 Some fatty acids
Fatty acid Number of carbon atoms
Palmitic CH3(CH2)14COOH 16
Stearic CH3(CH2)16COOH 18
Oleic CH3(CH2)7CH=CH(CH2)7COOH 18
Linoleic CH3(CH2)4CH=CHCH2CH=CH(CH2)7COOH 18
Lactobacillic CH3(CH2)5CH – CH(CH2)9COOH 19
CH2
Tuberculostearic CH3(CH2)7CH(CH2)8COOH 19
CH3
the membrane. Unsaturated fatty acids and in the citric acid cycle (Fig. 9.16). The carbo-
branched-chain fatty acids make the membrane nyl of the β-ketoacyl–CoA is then attacked by
more fluid, which is a necessary condition for CoASH, displacing an acetyl–CoA (reaction 5)
membrane function. catalyzed by β-ketothiolase. Usually the bacte-
rium is growing aerobically and the acetyl–CoA
β-Oxidation of fatty acids is oxidized to carbon dioxide in the citric acid
Many bacteria can grow on long-chain fatty cycle. The acetyl–CoA can also be a source of
acids (e.g., pseudomonads, various bacilli, E. cell carbon when it is assimilated via the gly-
coli). The fatty acids are oxidized to acetyl–CoA oxylate cycle (Section 9.12). The displacement
via a pathway called β-oxidation (Fig. 10.1). of the acetyl–CoA results in the generation of a
Before such oxidation can take place, the fatty acid acyl–CoA that is recycled through the
fatty acid must be converted to the acyl–CoA β-oxidation pathway.
derivative in a reaction catalyzed by acyl–CoA If the fatty acid has an even number of carbon
synthetase. This is a two-step process. First, atoms, the entire chain is degraded to acetyl–
pyrophosphate is displaced from ATP to make CoA. However, if the fatty acid is an odd-chain
the AMP derivative of the carboxyl group (acyl fatty acid, then the last fragment is propio-
adenylate), which remains tightly bound to the nyl–CoA rather than acetyl–CoA. Two of the
enzyme (Fig. 10.1, reaction 1). Then the AMP various possible pathways for the oxidation
is displaced by CoASH to make the fatty acyl– of propionyl–CoA to acetyl–CoA are shown
CoA. The activation of a carboxyl group by in Figs. 10.2 and 10.3. E. coli forms pyruvate
formation of an acyl adenylate is common in from propionate via methylcitrate (Fig. 10.3).1,2
metabolism and is discussed in Section 8.2.1. Review the glyoxylate cycle in Section 9.12 and
For example, amino acids are activated for pro- Fig. 9.24. Notice the similarity to the methylci-
tein synthesis in this way. As with other reac- trate pathway.
tions in which pyrophosphate is displaced from E. coli has two β-oxidation pathways, one
ATP, the reaction is driven to completion by that operates only under aerobic conditions and
hydrolysis of the pyrophosphate. A keto group a second that operates anaerobically.3 Thus, the
is then generated β to the carboxyl. The fol- bacteria can be grown anaerobically on fatty
lowing sequence of reactions leads to the keto acids if a terminal electron acceptor, such as
group: (1) an oxidation to form a double bond nitrate, is available.
(reaction 2), catalyzed by acyl–CoA dehydro-
genase; (2) hydration of the double bond to The synthesis of fatty acids
form the hydroxyl (reaction 3), catalyzed by An examination of metabolic pathways reveals
3-hydroxyacyl–CoA hydrolyase; and (3) oxi- that catabolic pathways are usually differ-
dation of the hydroxyl to form the keto group ent from synthetic pathways. In other words,
(reaction 4), catalyzed by L-3-hydroxyacyl– biosynthesis is not due simply to reversing the
CoA dehydrogenase. The chemistry is the same reactions of catabolism. This is true because
as the conversion of succinate to oxaloacetate there is usually at least one irreversible step in
metabolism of lipids, nucleotides, amino acids, and hydrocarbons 257
the catabolic pathway. For example, glycolysis pathway differs from the β-oxidation pathway
has two irreversible reactions, the phospho- in the following ways.
fructokinase and pyruvate kinase reactions,
1. The reductant is NADPH, rather than
and these must be bypassed to reverse glycoly-
NADH.
sis. Often, an entirely different set of reactions
2. Biosynthesis requires CO2 and proceeds via
is used for biosynthesis. This is seen in fatty acid
malonyl–CoA, a carboxylated derivative of
metabolism, where the biosynthetic pathway
acetyl–CoA.
consists of reactions completely different from
3. The acyl carrier is not CoA, as it is in the deg-
the β-oxidation pathway.4 The biosynthetic
radative pathway, but a protein called the
acyl carrier protein (ACP). The acyl carrier
protein is a small protein (MW 10 kDa in
E. coli), having a residue similar to CoASH
attached to one end. (See note 5 for compari-
son of the chemical structures of CoASH and
ACP.)
In eukaryotic cells, fatty acid biosynthesis takes
place in the cytosol, and the degradation takes
place in the matrix of the mitochondria. Since
prokaryotes do not have similar organelles,
both synthesis and degradation take place in
the same compartment, the cytosol. The fatty
acid biosynthetic enzymes exist as multien-
zyme complexes in eukaryotes, including yeast.
However, in E. coli they are present in the cyto-
sol as separate enzymes.
The sequence of reactions begins with
the carboxylation of acetyl~CoA to make
malonyl~CoA (Fig. 10.4A, reaction 1). The
carboxylation, which requires ATP and a vita-
min, biotin, is catalyzed by acetyl–CoA car-
boxylase. (See note 6 for more information
on acetyl–CoA carboxylase.) Then the CoA
is displaced by the acyl carrier protein, ACP,
to form malonyl–ACP in a reaction catalyzed
by malonyl–CoA:ACP transacetylase (reac-
tion 2). A similar reaction, catalyzed by acetyl–
CoA:ACP transacetylase, displaces CoA from
another molecule of acetyl–CoA to form
the acetyl–ACP derivative (reaction 3). (See
Fig. 10.1 β-Oxidation of fatty acids. Enzymes: 1, note 7.) The methylene carbon in malonyl–
acyl–CoA synthetase; 2, fatty acyl–CoA dehydroge- ACP acts as a nucleophile and displaces the
nase; 3, 3-hydroxyacyl–CoA hydrolyase; 4, L-3-hy- ACP from acetyl–ACP to form the β-ketoacyl–
droxyacyl–CoA dehydrogenase; 5, β-ketothiolase. ACP derivative in a reaction catalyzed by
Fig. 10.2 Oxidation of propionyl–CoA to pyruvate via acrylyl–CoA.
www.ebook777.com
Fig. 10.3 Oxidation of propionyl–CoA to pyruvate via the methylcitrate pathway. This pathway seems to
operate in E. coli. Enzymes: a, 2-methylcitrate synthase; b and c, 2-methylaconitase; d, 2-methylisocitrate
lyase. Source: Adapted from Textor, S., V. F. Wendisch, A. A. De Graff, U. Müller, M. I. Linder, D. Linder,
and W. Buckel. 1997. Propionate oxidation in Escherichia coli: evidence for operation of a methylcitrate cycle
in bacteria. Arch. Microbiol. 168:428–436.
β-ketoacyl–ACP synthase (reaction 4). (See of identical reactions initiated by the attack of
Section 9.1.1 for a discussion of condensation malonyl–ACP on the carboxyl end of the grow-
reactions of these types.) At the same time, the ing acyl–ACP chain, displacing the ACP. Fatty
newly added carboxyl is removed, driving the acid synthesis must be regulated because long-
reaction to completion. (In E. coli, since there chain acyl–ACPs do not accumulate. The mech-
is more than one condensing enzyme, either anism of regulation has not been elucidated,
acetyl–CoA or acetyl–ACP can be used for the but it is reasonable to suggest that it involves
condensation with malonyl–ACP. See note 8 feedback inhibition of a regulatory fatty acid
for more information on fatty acid synthesis in biosynthetic enzyme by long-chain acyl–ACPs.
E. coli.) The β-ketoacyl–ACP is then reduced (See Section 7.1.1 for a discussion of feedback
to the hydroxy derivative by an NADPH- inhibition.)
dependent β-ketoacyl–ACP reductase (reac- When the acyl–ACP chain is completed, the
tion 5), and dehydrated to the α, β derivative acyl portion is immediately transferred to mem-
by the β-hydroxylacyl–ACP dehydrase to form brane phospholipids by the glycerol phosphate
the unsaturated acyl–ACP derivative (reaction acyltransferase reactions described shortly
6). (The double bond is between C2, the carbon in Section 10.1.2. Of course, not all the fatty
α to the carboxyl carbon, and C3, the carbon β acids are incorporated into phospholipids. For
to the carboxyl carbon.) The unsaturated acyl– example, some acyl–ACPs are used for lipid A
ACP derivative is then reduced by enoyl–ACP biosynthesis (Section 12.2.2). These include
reductase to the saturated acyl–ACP (reaction β-hydroxymyristoyl–ACP, lauroyl–ACP, and
7). The acyl–ACP chain is elongated by a series myristoyl–ACP.
Fig. 10.4 Biosynthesis of fatty acids. Enyzmes: 1, acetyl–CoA carboxylase; 2, malonyl transacetylase; 3, an
acetyl transacetylase; 4, β-ketoacyl–ACP synthase; 5, β-ketoacyl–ACP reductase; 6, β-hydroxyacyl–ACP
dehydrase; 7, enoyl–ACP reductase; 8, β-hydroxydecenoyl–ACP dehydrase.
Synthesis of unsaturated fatty acids polyunsaturated acids are required in the diets of
If the fatty acid is unsaturated, it is almost animals.) Depending upon the organism, unsat-
always monounsaturated (one double bond). urated fatty acids are formed in two different
Polyunsaturated fatty acids are more typical of ways: the anaerobic pathway and the aerobic
eukaryotic organisms. (As explained in note 9, pathway. The anaerobic pathway is restricted
www.ebook777.com
to prokaryotes. It is widespread in bacteria, Because the phosphate groups are ionized,

being found in Clostridium, Lactobacillus, phospholipids always have negatively charged
Escherichia, Pseudomonas, the cyanobacteria, groups. There may also be positively charged
and the photosynthetic bacteria. The aerobic groups (e.g., the protonated amino group in
pathway is found in Bacillus, Mycobacterium, phosphatidylethanolamine).
Corynebacterium, and Micrococcus, and in There are several different types of phos-
eukaryotes. pholipid. However, the ones that concern us
In the anaerobic pathway, a special dehydrase here are the major phospholipids in bacteria.
desaturates the C10 hydroxyacyl–ACP interme- These are the phosphoglycerides, phospho-
diate to the trans-α,β-decenoyl–ACP, which is lipids that contain fatty acids and phosphate
isomerized while bound to the enzyme to cis- esterified to glycerol, and usually some other
β,γ-decenoyl–ACP (Fig. 10.4B, reaction 8). In molecule (e.g., an amino acid, an amine, or
the β,γ derivative, the double bond is between a sugar covalently bound to the phosphate).
C3 and C4, whereas in the α,β derivative the The structure of a typical phosphoglyceride is
double bond is between C2 and C3. Very impor- shown in Fig. 10.5. The fatty acids at positions
tantly, the cis-β,γ does not serve as a substrate C1 and C2 of the glycerol molecule need not
for the enoyl reductase. As a consequence, the be the same.
cis-β,γ derivative is not reduced. Instead, it is
elongated, leading to a fatty acid with a double Phospholipids and the structure of the
bond. Two common monounsaturated fatty cell membrane
acids in bacteria synthesized in this way are A major structural feature of phospholipids is
palmitoleic acid, which has 16 carbons (cis-Δ9- that they are amphibolic. That is, one part of the
hexadecenoic acid), and cis-vaccenic acid, which molecule is hydrophobic (apolar) and another
has 18 carbons (cis-Δ11-octadecenoic acid). (See part is hydrophilic (polar). The hydrophobic
note 10 for a description of cis and trans fatty area (tail) is the part where the fatty acids are
acids.) In palmitoleic acid, the double bond is located, and the hydrophilic area (head) is the
between C9 and C10 in the final product. This end containing the phosphate and its attached
is because six carbons are added to the C1 of group.
the C10 β,γ derivative. Since vaccenic acid is The amphibolic nature of the phospholip-
two carbons longer, the double bond is between ids explains their orientation in membranes.
C11 and C12. Phospholipids are oriented in membranes as a
The aerobic pathway makes use of special bilayer with the charged head groups pointing
desaturases (oxidases) that introduce a double out into the aqueous phase and the hydropho-
bond into the completed fatty acyl–CoA or bic fatty acid tails interacting with each other
fatty acyl–ACP derivative. The enzyme system in a hydrophobic interior. The cell membrane
requires molecular oxygen and NADPH. In a is completed by proteins, as shown earlier
multienzyme reaction sequence, two electrons (Fig. 1.15). Some properties of phospholipid
are removed from the fatty acyl derivative membranes that should be remembered are as
forming the double bond, and two electrons are follows.
removed from NADH. The four electrons are
1. They are permeable only to water, gases, and
transferred to O2, forming two H2O molecules.
small hydrophobic molecules.
Bacteria that use the aerobic pathway make
2. They have a low ionic conductance and are
oleic acid from its C18 saturated progenitor,
capable of doing work when ions (especially
stearic acid, rather than cis-vaccenic. Oleic acid
protons and sodium ions) are transported
has a double bond between C9 and C10 and is
through them via special carriers along an
called cis-Δ9-octadecenoic acid.
electrochemical gradient.
3. For a phospholipid membrane to function,
10.1.2 Phospholipid synthesis its lipid portion must remain fluid.
in bacteria 4. The fluidity of the membrane is due to
Phospholipids are lipids with covalently the presence of unsaturated fatty acids or
attached phosphate groups. They are an branched-chain (methyl) fatty acids in the
important constituent of cell membranes. phospholipids.
Fig. 10.5 Some common phosphoglycerides in bacteria; R represents a fatty acyl moiety esterified to the
glycerol phosphate. Note that phospholipids have an apolar and a polar end. The configuration consist-
ing of the glycerol-3-phosphate and the glycerol-3-phosphate moiety of phosphoglycerides belongs to the
L-stereochemical series and is called sn-glycerol-3-phosphate, where sn indicates stereospecific numbering.
Horizontal bonds are projected to the front and vertical bonds behind the plane of the page. Thus, the H and
OH are in front of the plane of the page, and C1 and C3 are behind.
The synthesis of the phosphoglycerides by CDP–diglyceride synthase. The formation

Phosphoglyceride synthesis starts with dihy- of CDP–diacylglycerol is driven to comple-
droxyacetone phosphate (DHAP), an inter- tion by the hydrolysis of the PPi catalyzed
mediate in glycolysis (Fig. 10.6). Step 1 is the by a pyrophosphatase. Step 4 is the displace-
reduction of dihydroxyacetone phosphate to ment of CMP by serine, catalyzed by phos-
glycerol phosphate by the enzyme glycerol phatidylserine synthase (PS synthase). Step 5
phosphate dehydrogenase. Step 2 is the trans- is the decarboxylation of phosphatidylserine
fer to glycerol phosphate of fatty acids from to yield phosphatidylethanolamine, which
fatty acyl–S–ACP. The reactions are catalyzed is the major phospholipid in several bacte-
by membrane-bound enzymes called G3P acyl ria. Bacteria make other phospholipids from
transferases. (Some bacteria are also capable of CDP–diglyceride. The displacement of CDP by
using acyl–SCoA derivatives as the acyl donor.) α-glycerol phosphate yields phosphatidylglyc-
The first phospholipid made is phosphatidic erol phosphate (step 6). Phosphatidylglycerol
acid, which is glycerol phosphate esterified to phosphate is dephosphorylated to yield phos-
two fatty acids. The other phospholipids are phatidylglycerol (step 7). The latter displaces
made from phosphatidic acid. glycerol from a second molecule of phosphati-
Step 3 is a reaction in which the phosphate dylglycerol to form diphosphatidylglycerol
on the phosphatidic acid reacts with cytidine (cardiolipin) (step 8). Note that modifica-
triphosphate (CTP) and displaces PPi to form tion of the head groups in all cases occurs via
CDP–diacylglycerol. The reaction is catalyzed a nucleophilic attack by a hydroxyl on the
www.ebook777.com
Fig. 10.6 Phosphoglyceride synthesis. The biosynthetic pathway for phosphoglycerides branches off glycoly-
sis at DHAP. The G3P dehydrogenase is a soluble enzyme, but the G3P acyl transacylase and all subsequent
reactions take place in the cell membrane. The fatty acyl–ACP derivatives are made in the cytosol, diffuse
to the membrane, and transfer the acyl portion to the phospholipid on the inner surface of the membrane.
Abbreviations: DHAP, dihydroxyacetone phosphate; G3P, glycerol phosphate; PA, phosphatidic acid; PS,
phosphatidylserine; PGP, phosphatidylglycerol phosphate; PG, phosphatidylglycerol; CL, cardiolipin
(diphosphatidylglycerol); CDL, cardiodilipid. Enzymes: 1, G3P dehydrogenase; 2, G3P acyltransferase and
1-acyl–G3P acyltransferase; 3, CDP–diglyceride synthase; 4, phosphatidylserine synthase; 5, phosphatidyl-
serine decarboxylase; 6, PGP synthase; 7, PGP phosphatase; 8, CL synthase.
electropositive phosphorus in the phosphate an accumulation of intracellular (p)ppGpp

group. and is accompanied by an inhibition of phos-
pholipid synthesis, as well as stable RNA syn-
The inhibition of phospholipid biosynthesis thesis. During the stringent response imposed
during the stringent response by amino acid starvation in E. coli, long-
Recall that in Section 2.2.2 it was pointed out chain acyl–ACPs accumulate, suggesting that
that the stringent response, which is a response the acyltransferase is inhibited, which could
to starvation for a required amino acid or a account for the inhibition of phospholipid bio-
carbon and energy source, is correlated with synthesis.11 How (p)ppGpp might be involved
in the inhibition of the acyltransferase is not 2. The linkage to glycerol is via an ether bond
known. rather than an ester bond.
10.1.3 Synthesis of archaeal lipids The metabolic pathway for the synthesis of the
glycerol ethers in the archaea is not well under-
The student should review the structure of
stood. Figure 10.7 illustrates a proposal for
archaeal lipids and membranes, as described in
Halobacterium halobium published in 1993.
Section 1.2.5. The lipids of archaea differ in the
The glycerol backbone is synthesized from
following two ways from those of bacteria.
either glycerol-3-phosphate or dihydroxyacetone
1. Instead of fatty acids, archaeal lipids have phosphate. The alcohol is believed to be derived
long-chain alcohols called isoprenyl alcohols. from geranylgeranyl pyrophosphate, which is
Fig. 10.7 Postulated pathway for the synthesis of glycerol ether lipids in archaea. Two pathways are depicted,
one for Halobacterium and one for Methanobacterium and Sulfolobus. It is proposed that a nucleophilic dis-
placement takes place on geranylgeranyl-PP displacing the pyrophosphate and forming the ether linkage. The
molecule that condenses with the geranylgeranyl-PP might be glycerol-3-phosphate, glycerol-1-phosphate,
or dihydroxyacetone phosphate. The rest of the pathway is unknown. Source: Koga et al. 1993. Ether polar
lipids of methanogenic bacteria: structures, comparative aspects, and biosynthesis. Microbiol. Rev. 57:164–
182. Reproduced with permission from American Society for Microbiology.
www.ebook777.com
synthesized from acetyl–CoA via mevalonic acid triphosphate. There are even separate names for
in a well-known pathway that is widespread the nucleoside monophosphates. For example,
among the bacteria and higher organisms. (The adenosine monophosphate is also called ade-
alcohol group is esterified to the pyrophosphate.) nylic acid. Guanosine monophosphate is gua-
The C1 hydroxyl on glycerol-3-phosphate or the nylic acid. We also have cytidylic acid, uridylic
dihydroxyacetone phosphate makes a nucleo- acid, and thymidylic acid. The structures of the
philic attack on the geranylgeranyl pyrophos- purines, pyrimidines, and sugars are shown in
phate, displacing pyrophosphate and forming Fig. 10.8. Notice that deoxyribose differs from
the ether linkage. If dihydroxyacetone phos- ribose in having a hydrogen substituted for the
phate is used, then 3-monoalkenyl–DHAP is hydroxyl group on C2 of the sugar.
formed. If glycerol-3-phosphate is used, then the
monoalkenylglycerol-1-phosphate is formed.
10.2.2 The synthesis of the pyrimidine
The monoalkenylglycerol-1-phosphate, which is
nucleotides
formed either directly or in two steps, reacts with
The pyrimidine part of the nucleotide is made
another geranylgeranyl pyrophosphate to form
from aspartic acid, ammonia, and carbon diox-
2,3-dialkenylglycerol-1-phosphate.
ide (Fig. 10.9). Step 1 is the biotin-dependent
It is proposed that in Methanobacterium
synthesis of carbamoyl phosphate from HCO3−,
and Sulfolobus, glycerol-1-phosphate, rather
glutamine, and ATP (Fig. 10.10). The enzyme
than dihydroxyacetone phosphate or glycerol-
that catalyzes the reaction is called carbamoyl
3-phosphate, attacks the geranylgeranyl pyro-
phosphate synthetase. (See note 12.) It requires
phosphate. The rest of the pathway involves a
two ATPs. One ATP is used to form a carboxy-
reduction of the double bonds in the hydrocar-
lated biotin intermediate, and the second ATP
bons, the attachment of the polar head groups,
is used to phosphorylate the carboxyl group.
and the formation of the tetraether lipids.
Notice that carbamoyl phosphate is an acyl
phosphate and therefore has high group trans-
10.2 Nucleotides fer potential.
10.2.1 Nomenclature and structures In step 2, the carbamoyl group is transferred
A nucleotide is a molecule containing a pyrimi- to aspartate as the phosphate is displaced by
dine or purine, a sugar (ribose or deoxyribose), the nitrogen atom of aspartate. The product
and phosphate. If there were no phosphate, the is N-carbamoylaspartate. The enzyme that
molecule would be called a nucleoside. Thus, catalyzes step 2 is aspartate transcarbamylase
nucleotides are nucleoside phosphates. If one (ATCase), which is feedback inhibited by the
phosphate is present, the molecule is called a end product, CTP. Step 3 is the cyclization
nucleoside monophosphate. If two phosphates of N-carbamoylaspartate with loss of water
are present, then it is called a nucleoside diphos- to form the first pyrimidine, dihydroorotate.
phate; We shall encounter nucleoside triphos- Step 4 is the oxidation of dihydroorotate to
phate (NTP) in Chapter 11. orotate.
The three major pyrimidines are cytosine, In step 5 the orotate displaces PPi from phos-
thymine, and uracil. The corresponding nucle- phoribosylpyrophosphate (PPRP) to form the
osides are called cytidine, thymidine, and uri- first nucleotide, orotidine-5′-phosphate, also
dine, respectively. The cytidine nucleotides are called orotidylate. The reaction is driven to
called cytidine monophosphate (CMP), cytidine completion by the hydrolysis of PPi catalyzed by
diphosphate (CDP), and cytidine triphosphate a pyrophosphatase. Note that the PRPP itself is
(CTP). A similar naming system is used for the synthesized from ribose-5-phosphate and ATP
uridine nucleotides (UMP, UDP, UTP) and the via a phosphoryl group transfer reaction from
thymidine nucleotides (TMP, TDP, TTP). If the ATP to the C1 carbon of ribose-5-phosphate.
sugar is deoxyribose rather than ribose, then the Orotodine-5′-phosphate is decarboxlated
nucleotide is written dCTP, and so on. in step 6 to uridine monophosphate (UMP).
The two major purines are adenine and gua- Phosphorylation of UMP, using ATP as the
nine. The nucleosides are adenosine and guanos- donor, yields UDP and UTP (steps 7 and 8). CTP
ine. The nucleotides are adenosine mono,- di-, is synthesized from UTP by an ATP-dependent
and triphosphate and guanosine mono-, di-, and amination via NH+4 (step 9). (See note 13.)
Fig. 10.8 The structures of the nucleotides and their subunits. The bases are attached to the C1 of the sugars
and the phosphates to the C5. Not shown are the nucleoside di- and triphosphates.
The CTP can be dephosphorylated to CDP,

which is the precursor to the deoxyribonucle-
otides dCTP and dTTP. The sequence of reac-
tions is
CDP → dCDP → dCTP → dUTP → dUMP
dTMP → dTDP → dTTP
Fig. 10.9 Origins of atoms in the pyrimidine ring: C2
Also, UDP can be reduced to dUDP, which is derived from CO2, and N3 comes from ammonia,
can be converted to dUTP. Thus, there are two via carbamoyl phosphate. The rest of the atoms are
routes to dUTP, one from CDP and the other (a derived from aspartate.
minor route) from UDP.
the amide nitrogen of glutamine displaces the
10.2.3 The synthesis of the purine nucle- pyrophosphate in a nucleophilic displacement.
otides The pyrophosphate that is released has a high
A purine is drawn in Fig. 10.11, showing the free energy of hydrolysis, and the amination is
origin of the atoms. The synthesis of the purine driven to completion by the hydrolysis of the
ring requires as precursors glutamine and pyrophosphate by pyrophosphatase.
aspartic acid (to donate amino groups), glycine, In step 2, glycine is added to the amino group
carbon dioxide, and a C1 unit at the oxidation of phosphoribosylamine to form 5-phosphori-
state of formic acid. The latter three donate all bosylglycineamide in an ATP-dependent step.
of the carbon atoms in the purine. The C1 unit In this reaction the carboxyl group of glycine
at the oxidation level of formic acid can come is added to the amino group of phosphoribo-
from formic acid itself, or from serine. sylamine to make the amide bond. The ATP first
phosphorylates the carboxyl group, forming an
Enzymatic reactions acyl phosphate. Then the nitrogen in the amino
The biosynthetic pathway for purines is shown group displaces the phosphate forming the
in Fig. 10.12. The purine molecule is synthe- amide bond. The student should review these
sized in stages while attached to ribose phos- kinds of reaction in Section 8.2.1.
phate, which in turn comes from PRPP. Step 1 Step 3 is the formylation of the amino group
is the attachment of an amino group to PRPP, on the glycine moiety. This requires a molecule
donated by glutamine. The product is 5-phos- to donate the formyl group (i.e., a molecule
phoribosylamine. This is a reaction in which with high formyl group transfer potential).
www.ebook777.com
Fig. 10.10 The biosynthesis of pyrimidine nucleotides. Enzymes: 1, carbamoyl phosphate synthetase; 2,
aspartate transcarbamoylase; 3, dihydroorotase; 4, dihydroorotate dehydrogenase; 5, orotate phosphoribo-
syltransferase; 6, orotidine-5-phosphate decarboxylase; 7, nucleoside monophosphate kinase; 8, nucleoside
diphosphate kinase; 9, CTP synthetase; 10, PRPP synthetase; 11, pyrophosphatase.
The donor of the formyl group is formyl–tetra- step 5 the 5-phosphoribosyl-N-formylglycineami-

hydrofolic acid (formyl–THF). The product is dine cyclizes to 5-phosphoribosyl-5-aminoimida-
5-phosphoribosyl-N-formylglycineamide. zole in a reaction that requires ATP. Probably the
In step 4 the amide group is changed into an carbonyl is phosphorylated and the phosphate is
amidine, the nitrogen being donated by glutamine displaced by the nitrogen to form the C–N bond.
in an ATP-dependent reaction. The product is The cyclized product tautomerizes to form the
5-phosphoribosyl-N-formylglycineamidine. In amino group on aminoimidazole ribonucleotide.
used, examine Fig. 10.14. The C1 units can be

donated to THF from serine, which donates its
β carbon at the oxidation level of formaldehyde
to THF acid to form methylene–THF and gly-
cine. (This is also the pathway for glycine bio-
synthesis, to be described in Section 10.3.1.)
Methylene–THF is then oxidized to formyl–
THF, which donates the formyl group in
purine biosynthesis.
Bacteria can also use formate for purine bio-
synthesis. The formate is attached to THF acid
Fig. 10.11 Metabolic origins of atoms in purines.
to make formyl–THF in an ATP-dependent
reaction. THF is also an important methyl car-
rier. Instead of methylene–THF being oxidized
Reaction 6 is the carboxylation of 5-phosphori-
to formyl–THF, it is reduced to methyl–THF.
bosyl-5-aminoimidazole to form 5-phosphoribo-
Methyl–THF donates the methyl group in vari-
syl-5-aminoimidazole-4-carboxylic acid. In step
ous biosynthetic reactions, including the bio-
7 aspartic acid combines with 5-phosphoribo-
synthesis of methionine.
syl-5-aminoimidazole-4-carboxylic acid to form
Sulfanilamide and its derivatives, called sul-
5-phosphoribosyl-4-(N-succinocarboxyamide)-
fonamides (the “sulfa drugs”), inhibit the for-
5-aminoimidazole in an ATP-dependent step. In
mation of folic acid and therefore inhibit purine
step 8 fumarate is removed, leaving the nitrogen,
biosynthesis in bacteria. Sulfonamides resemble
to form 5-phosphoribosyl-4-carboxamide-5-
p-aminobenzoic acid, an intermediate in folic
aminoimidazole, which is formylated in step 9
acid biosynthesis, and the sulfonamides inhibit
to form 5-phosphoribosyl-4-carboxamide-5-
the enzyme that utilizes p-aminobenzoic acid
formaminoimidazole. The latter is cyclized in
for folic acid synthesis. Because animals obtain
step 10 to inosinic acid (IMP). The purine itself is
their folic acid from their diet, whereas most
called hypoxanthine. Inosinic acid is the precur-
bacteria synthesize folic acid, the sulfonamides
sor to all the purine nucleotides.
are toxic to bacteria but not to animals.
Biosynthesis of AMP and GMP from IMP
IMP is converted to AMP by substitution of the 10.2.5 Synthesis of deoxyribonucleotides
carbonyl oxygen at C6 with an amino group The deoxyribonucleotides are synthesized from
(Fig. 10.13). The amino donor is aspartate, the ribonucleoside diphosphates by a reduc-
and the reaction requires GTP. AMP can then tive dehydration catalyzed by ribonucleoside
be phosphorylated to ADP by using ATP as the diphosphate reductase (Fig. 10.15). The elec-
phosphoryl donor. The ADP can be reduced to tron donor, a sulfhydryl protein called thiore-
dADP or phosphorylated to ATP via substrate- doxin, obtains its electrons from NADPH.
level phosphorylation or respiratory phospho-
rylation. GMP is synthesized by an oxidation 10.3 Amino Acids
of IMP at C2 followed by an amination at that There are 20 different amino acids in proteins.
carbon. The oxidation produces a carbonyl Inspection of Table 10.2 reveals that six amino
group, which becomes aminated at the expense acids are synthesized from oxaloacetate and
of ATP. four from α-ketoglutarate, two intermediates
of the citric acid cycle. In addition, succinyl–
10.2.4 The role of tetrahydrofolic acid CoA donates a succinyl group in the forma-
Derivatives of tetrahydrofolic acid (THF; tion of intermediates in the biosynthesis of
synthesized from the B vitamin folic acid) are lysine, methionine, and diaminopimelic acid.
coenzymes that carry single-carbon groups (The succinate is removed at a later step and
and are very important in metabolism. For does not appear in the product.) This again
example, the C1 units used in purine biosyn- emphasizes that the citric acid cycle is impor-
thesis are carried by derivatives of tetrahydro- tant not only for energy generation, but also
folic acid. For an explanation of how THF is for biosynthesis.
www.ebook777.com
Fig. 10.12 Biosynthesis of purine nucleotides. Enzymes: 1, PRPP amidotransferase; 2, phosphoribosylglycine-

amide synthetase; 3, phosphoribosylglycineamide formyltransferase; 4, phosphoribosyl–formylglycineamidine
synthetase; 5, phosphoribosyl–aminoimidazole synthetase; 6, phosphoribosylaminoimidazole carboxylase;
7, phosphoribosylaminoimidazole succinocarboxamide synthetase; 8, adenylosuccinate lyase; 9, phospho-
ribosylaminoimidazole carboxamide formyltransferase; 10, IMP cyclohydrolase. Abbreviations: gln, glu-
tamine; glu, glutamate; asp, aspartate; fum, fumarate; THF, tetrahydrofolate.
Table 10.2 also points out that three glyco- erythrose-4-P, which is a precursor to aromatic
lytic intermediates [pyruvate, 3-phosphoglyc- amino acids. (It is actually used with PEP in the
eric acid (3-PGA), and phosphoenolpyruvate synthesis of the aromatic amino acids.) Finally,
(PEP)] are precursors to nine more amino acids. 5-phosphoribosyl-1-pyrophosphate (PRPP),
The pentose phosphate pathway provides which donates the sugar for nucleotide synthesis,
Fig. 10.13 Synthesis of AMP and GMP from IMP. Enzymes: 1, adenylosuccinate synthetase and adenylosuc-
cinate lyase; 2, IMP dehydrogenase; 3, GMP synthetase. Abbreviations: IMP, inosinic acid; AMP, adenylic
acid; XMP, xanthylic acid; GMP, guanylic acid.
is also the precursor to the amino acid histidine. donor for the synthesis of purines, pyrimidines,
Therefore, the three central pathways—glyco- amino sugars, histidine, tryptophan, asparag-
lysis, the pentose phosphate pathway, and the ine, NAD+, and p-aminobenzoate.
citric acid cycle—provide the precursors to all Glutamate is synthesized by two alternate
the amino acids. To illuminate some principles routes. One requires the enzyme glutamate
of amino acid metabolism, we will now exam- dehydrogenase, and the second requires two
ine the biosynthesis and catabolism of a select enzymes: glutamine synthetase (GS) and gluta-
group of amino acids. mate synthase, which is also called glutamine
oxoglutarate aminotransferase, or the GOGAT
10.3.1 Synthesis enzyme.
Glutamate and glutamine synthesis Glutamate dehydrogenase catalyzes the
The ability to synthesize glutamate and glu- reductive amination of α-ketoglutarate (Fig.
tamine is of extreme importance because this 10.16). However, it should be pointed out that
is the only route for incorporation of inorganic because of its high Km for ammonia, the glu-
nitrogen into cell material. All inorganic nitro- tamate dehydrogenase reaction can be used
gen must first be converted to ammonia, which for glutamate synthesis only when ammonia
is then incorporated as an amino group into concentrations are high (>1 mM), otherwise
glutamate and glutamine. (The synthesis of ammonia is incorporated into glutamate via
ammonia from nitrate and from nitrogen gas glutamine. This is the situation in many natural
is described in Chapter 13). The amino group environments.
is then donated from these amino acids to all Under conditions of low ammonia concentra-
the other nitrogen-containing compounds in tion, bacteria use two enzymes in combination
the cell. Glutamate is the amino donor for most for ammonia incorporation. One is L-glutamine
of the amino acids, and glutamine is the amino synthetase (Fig. 10.17, reaction 1). This enzyme
www.ebook777.com
Fig. 10.14 Tetrahydrofolic acid (THF) as a C1 carrier. Tetrahydrofolic acid is the reduced form of the vita-
min folic acid. Tetrahydrofolic acid and its derivatives are important coenzymes that function in many path-
ways to carry one-carbon units. The one-carbon units can be at the oxidation level of formic acid (HCOOH),
formaldehyde (HCHO), or methyl (CH3–). This arrangement is essential for purine biosynthesis because it
transfers groups at the oxidation level of formic acid in two separate steps. The most common precursor of
the C1 unit is the amino acid serine, which donates the C1 unit as formaldehyde to THF. The product can
then be oxidized to the level of formate to form 10-formyl–THF. Some bacteria can use formic acid itself
as the source of the C1 unit. Methylene–THF can also be reduced to methyl–THF, which serves as a methyl
donor.
incorporates ammonia into glutamate to form incorporated ammonia from glutamine to

glutamine. The glutamine synthetase reaction α-ketoglutarate to form glutamate. Notice in
uses ATP, which drives the reaction to comple- Fig. 10.17 that glutamine is used catalytically.
tion. The other enzyme is glutamate synthase This is also seen in the following reactions,
(reaction 2). This enzyme transfers the newly driven to completion by the use of ATP.
Fig. 10.15 Formation of deoxynucleotides by ribonucleoside diphosphate reductase. The deoxynucleotides

are made from the nucleoside diphosphates. The OH on C2 of the ribose leaves with its bonding electrons
to form water. The OH is replaced by a hydride ion (H:–) donated by reduced thioredoxin, a dithiopro-
tein. Thioredoxin, R–(SH)2, is re-reduced by thioredoxin reductase, a flavoprotein. Thioredoxin reductase,
in turn, accepts electrons from NADPH. Enzymes: 1, ribonucleotide diphosphate reductase; 2, thioredoxin
reductase.
Table 10.2 Precursors for amino acid biosynthesis Transamination reactions from glutamate
are required for the synthesis of the other
Precursor Amino acid amino acids
Pyruvic acid L-Alanine, L-valine, L-leucine The synthesis of the other amino acids requires
Oxaloacetic acid L-Aspartate, L-asparagine, that glutamate donate an amino group to
L-methionine, L-lysine, an α-ketocarboxylic acid. The enzyme that
L-threonine, L-isoleucine
catalyzes the amino group transfer is called a
α-Ketoglutaric acid L-Glutamate, L-glutamine,
L-arginine, L-proline
transaminase. A generalized transamination
3-PGA L-Serine, L-glycine, L-cysteine reaction is shown in Fig. 10.18A. Note that after
PEP and erythrose-4-P L-Phenylalanine, L-tyrosine, the glutamate has donated its amino group, it
L-tryptophan becomes α-ketoglutarate again, which then is
PRPP and ATP L-Histidine
aminated once more to form another molecule
of glutamate. Thus, glutamate can be thought
of as a conduit through which ammonia passes
The glutamine synthetase and glutamate syn- into other amino acids. Transaminases use a
thase reactions coenzyme called pyridoxal phosphate (derived
from vitamin B6) to carry the amino group from
Glutamate + ATP + NH3 glutamate to the α-ketocarboxylic acids.
→ glutamine + ADP + Pi
Synthesis of aspartate and alanine
Glutamine + α-ketoglutarate + NADPH + H+ Aspartate is synthesized via a transamination
→ 2 glutamate + NADP+ from glutamate to oxaloacetate (Fig. 10.18B),
and alanine is made via a transamination from
α-Ketoglutarate + ATP + NH3 + NADPH + H+ glutamate to pyruvate (Fig. 10.18C).
→ glutamate + ADP + Pi + NADP+
Synthesis of serine, glycine, and cysteine
As mentioned, glutamine is an amino donor Serine is synthesized from 3-phosphoglycerate
for many compounds, including pyrimidines (Fig. 10.19). The phosphoglycerate is first oxi-
and purines. Therefore, glutamine synthetase dized to the α-keto acid, phosphohydroxypyru-
is an indispensable enzyme unless the medium vate. A transamination, using glutamate as the
is supplemented with glutamine. Glutamate donor, converts the phosphohydroxypyruvate
dehydrogenase, however, is not necessary for to phosphoserine. Then the phosphate is hydro-
growth, since the organism can synthesize glu- lytically removed to form L-serine. The hydro-
tamate from glutamine and α-ketoglutarate by lysis of the phosphate ester bond drives the
using glutamate synthase. reaction to completion.
www.ebook777.com
Fig. 10.16 The reductive amination of α-ketoglutarate, catalyzed by glutamate dehydrogenase.
Fig. 10.17 The GS and GOGAT reactions. Enzymes: 1, L-glutamine synthetase; 2, glutamine: α-oxoglutarate
aminotransferase, also called the GOGAT enzyme, or glutamate synthase.
Serine is a precursor to both glycine and cycle. These seven intermediates are pyruvate,
cysteine. Glycine is formed by transfer of the acetyl–CoA, acetoacetyl–CoA, α-ketoglutarate,
β carbon of serine to tetrahydrofolic acid succinyl–CoA, fumarate, and oxaloacetate.
(Fig. 10.14). The route to cysteine is initiated Acetoacetyl–CoA itself is a precursor to
when serine accepts an acetyl group from acetyl– acetyl–CoA.
CoA to become O-acetylated. Sulfide then dis-
places the acetyl group from O-acetylserine, Removal of the amino group
forming cysteine (see later: Fig. 13.2). There are several ways of removing amino
groups from amino acids during their catab-
10.3.2 Catabolism olism (Fig. 10.21). Usually amino acids are
A scheme showing the overall pattern of carbon oxidatively deaminated to their correspond-
flow during amino acid catabolism is shown ing keto acid. The oxidation may be cata-
in Fig. 10.20. The first reaction is always the lyzed by a nonspecific flavoprotein oxidase
removal of the amino group to generate the that oxidizes any one of a number of amino
α-keto acid, which eventually enters the citric acids and feeds the electrons directly into the
acid cycle. All 20 amino acids are degraded to electron transport chain (Fig. 10.21A). These
seven intermediates that enter the citric acid oxidases can be D-amino oxidases as well as
Fig. 10.18 The transamination reaction. (A) Glutamate donates an amino group to an α-ketocarboxylic acid
to form an α-amino acid. The glutamate becomes α-ketoglutarate, which is aminated via either the gluta-
mate dehydrogenase or the GS and GOGAT reactions. (B) Formation of L-aspartate from oxaloacetate. (C)
Formation of L-alanine from pyruvate.
L-amino oxidases. The D-amino acid oxidases as ammonia, and pyruvate is regenerated by
are useful because several biological molecules L-alanine dehydrogenase. α-Ketoglutarate and
(e.g., peptidoglycan, certain antibiotics) have its dehydrogenase serve the same function in
D-amino acids. amino acid catabolism.
There are also NAD(P)+-linked dehydroge- Certain amino acids are deaminated by spe-
nases (Fig. 10.21B), which are more specific cific deaminases (Fig. 10.21D). In these cases,
than the flavoprotein oxidases. The dehy- a redox reaction is not involved. These amino
drogenases catalyze the reversible reductive acids include serine, threonine, aspartate, and
amination of the keto acid. A transaminase histidine. The deamination of serine and of thre-
coupled to a dehydrogenase can lead to the onine proceeds by the elimination of water.
deamination of amino acids. In the example
shown in Fig. 10.21C, a nonspecific transami- 10.4 Aliphatic Hydrocarbons
nase transfers an amino group to pyruvate- Many bacteria can grow on long-chain hydro-
forming L-alanine. The amino group is released carbons, for example, alkanes (C10–C18). A
www.ebook777.com
Fig. 10.19 The synthesis of serine, glycine, and cysteine from 3-PGA. Enzymes: 1, phosphoglycerate dehydro-
genase; 2, phosphoserine aminotransferase; 3, phosphoserine phosphatase; 4, serine hydroxymethyltrans-
ferase; 5, serine transacetylase; 6, O-acetylserine sulfhydrylase.
few bacteria (mycobacteria, flavobacteria, of molecular oxygen, in a reaction catalyzed by

Nocardia) grow on short-chain hydrocarbons a monooxygenase. Membrane-bound enzymes
(C2–C8). The hydrocarbons exist as droplets then oxidize the long-chain alcohol to the car-
of oil outside the cell, and a major problem in boxylic acid (Fig. 10.22A). The carboxylic acid
using these water-insoluble molecules entails is derivatized with CoASH in an ATP-dependent
transferring them from the oil layer across the reaction and oxidized via β-oxidation in the
cell wall to the cell membrane, where they can cytoplasm.
be metabolized. Some bacteria have cell walls Since aliphatic hydrocarbons can be oxi-
containing glycolipids in which the hydro- dized to CO2 by certain sulfate-reducing bac-
carbons can dissolve and be transported to teria in the absence of oxygen, there must exist
the membrane. Acinetobacter strains secrete an alternative mechanism that does not require
particles resembling the outer membrane in oxygen.14 However, the mechanism is not
which the hydrocarbons dissolve. The parti- known. In the sequence shown in Fig. 10.22A,
cles with the dissolved hydrocarbon fuse with a short electron transport chain carries elec-
the outer membrane, and the hydrocarbons trons from NADH via flavoprotein and an FeS
are then transferred to the cell membrane. protein to a b-type cytochrome called P450.15
Once in the cell membrane, the degradation The P450 is in a complex with the hydrocar-
of the alkanes requires their hydroxylation, bon substrate and is called a cosubstrate. The
via molecular oxygen and an enzyme called reduced P450 binds O2 and activates it, pre-
monooxygenase. sumably converting it to bound O2– (superox-
ide) or O22– (peroxide), which attacks the C1
10.4.1 Degradative pathways carbon on the hydrocarbon. One of the oxygen
Hydroxylation at one end atoms replaces a hydrogen on the hydrocar-
After the hydrocarbon dissolves in the mem- bon and becomes a hydroxyl group, while the
brane, it is hydroxylated at one end by means other oxygen atom is reduced to H2O. Some
Fig. 10.20 Fates of the carbon skeletons of amino acids. Amino acid carbon can be used to synthesize the cell
components, or it can be oxidized to CO2. In the absence of the glyoxylate cycle, carbon entering at acetyl–
CoA cannot be used for net glucogenesis except in some strict anaerobic bacteria that can carboxylate acetyl–
CoA to pyruvate (Chapter 14).
bacteria use an n-alkane monooxygenase is the β-oxidative pathway. First, the fatty acid
instead of P450 to hydroxylate the hydrocar- is converted to the acyl–CoA derivative in an
bon. The iron protein rubredoxin (Fe2+ or Fe3+) ATP-dependent reaction catalyzed by acyl–CoA
is reduced by NADH and is a cosubstrate in the synthetase. The first reaction is a displacement
monooxygenase reaction. of pyrophosphate from ATP to form the acyl–
AMP, which remains bound to the enzyme.
Hydroxylation on the penultimate carbon Then the AMP is displaced by CoASH to form
An alternative degradative pathway is found in the acyl–CoA. The reaction is driven to comple-
Nocardia species (Fig. 10.22B). The hydroxylation by hydrolysis of the pyrophosphate, cata-
tion takes place on the second carbon, forming lyzed by pyrophosphatase. Thus, the activation
a secondary alcohol, which is then oxidized to of fatty acids requires the equivalent of two ATP
form a ketone. Then, in an unusual reaction, molecules. A keto group is then introduced β to
a second monooxygenase reaction creates an the carboxyl via an oxidation, hydration, and
acetyl ester, which is subsequently hydrolyzed oxidation to form the β-ketoacyl–CoA deriva-
to acetate and the long-chain alcohol. The alco- tive. Finally, the β-ketoacyl–CoA derivative is
hol is oxidized to the carboxylic acid, which is cleaved with CoASH to yield acetyl–CoA.
degraded via the β-oxidation pathway. The process is repeated while two-carbon
fragments are sequentially removed as acetyl–
CoA from the fatty acid. If the fatty acid is an
10.5 Summary odd-chain fatty acid, then the last fragment
Many aerobic bacteria can grow on long-chain is propionyl–CoA, which can be oxidized to
fatty acids. The major pathway for degradation pyruvate. Because fatty acids are degraded to
www.ebook777.com
Fig. 10.21 The removal of amino groups from amino acids during catabolism. (A) Amino acid oxidases
are flavoproteins that are specific for L- or D-amino acids, but generally not for the particular amino acid.
Electrons are transferred from the flavoprotein to the electron transport chain. (B) An amino acid dehydro-
genase. These molecules are NAD(P) linked and more specific for the amino acid than are the flavoprotein
oxidases. (C) Transamination catalyzed by a nonspecific transaminase that uses pyruvate as the amino acid
acceptor. The amino group is removed from the alanine by alanine dehydrogenase. α-Ketoglutarate can also
accept amino groups and be deaminated by α-ketoglutarate dehydrogenase. (D) The deamination of serine
by serine dehydratase.
acetyl–CoA, growth on fatty acids uses the same pathway, which is found in eukaryotes and cer-
metabolic pathways as growth on acetyl–CoA. tain aerobic bacteria, unsaturated fatty acids
Fatty acid synthesis occurs via an entirely are synthesized in a pathway requiring oxygen
different set of reactions. Acetyl–CoA is car- and NADPH. The double bond is introduced
boxylated to malonyl–CoA. A transacetylase into the completed fatty acyl–ACP or –CoA
then converts malonyl–CoA to malonyl–ACP. derivative by special desaturases (oxidases) that
Acetyl–CoA is also transacylated to acetyl-ACP. require O2 and NAD(P)H as substrates. During
Malonyl–ACP then condenses with acetyl–ACP the introduction of the double bond, four elec-
(or acetyl–CoA, depending upon which synthase trons are transferred to O2 to form two H2O
is used) to form a β-ketoacyl–ACP. The con- molecules. Two electrons are contributed by
densation with malonyl–ACP is accompanied NAD(P)H, and two electrons are derived from
by a decarboxylation reaction that drives the the fatty acid when the double bond is formed.
condensation. The β-ketoacyl–ACP is reduced Bacteria that use this pathway make oleic acid
via NADPH, dehydrated, and reduced again to rather than cis-vaccenic acid, which is made in
form the saturated acyl–ACP. The acyl–ACP is the anaerobic pathway. They are both C18 car-
elongated by condensation with malonyl–ACP. boxylic acids, but in oleic acid the double bond
There are two systems found among the is between C9 and C10, whereas in cis-vaccenic
bacteria for synthesizing unsaturated fatty acid it is between C11 and C12.
acids: an anaerobic and an aerobic pathway. The phosphoglycerides are synthesized
In the anaerobic pathway the C10 hydroxyacyl– from the fatty acyl–ACP derivatives and glyc-
ACP is desaturated and elongated rather than erol phosphate, the latter being the reduced
being desaturated and reduced. In the aerobic product of dihydroxyacetone phosphate. The
Fig. 10.22 Oxidation of an aliphatic hydrocarbon to a carboxylic acid. (A) Pathway in yeast and
Corynebacterium species. The hydrocarbon is hydroxylated by oxygen, which is activated by cytochrome
P450. The second oxygen atom is reduced to water. Other bacteria (e.g., Pseudomonas spp.) use a monooxy-
genase instead of P450. In the latter case the electron transport is from NADH to rubredoxin:NADH oxi-
doreductase, to rubredoxin, which is a cosubstrate in the monooxygenase reaction, rather than P450. The
primary alcohol is oxidized to an aldehyde, which is then oxidized to a carboxylic acid. The carboxylic acid
is oxidized via the β-oxidation pathway. (B) Pathway in Nocardia species. The C2 carbon is hydroxylated,
forming a secondary alcohol, which is then oxidized to a ketone. A second monooxygenase reaction converts
the ketone into an acetyl ester. The acetyl ester is hydrolyzed to acetate and a long-chain alcohol, which is
oxidized as in (A). Enzymes: 1, 4, 6, monooxygenase; 2, 5, alcohol dehydrogenase; 3, 6, aldehyde dehydroge-
nase; 7, acetyl esterase.
phosphoglycerides differ with respect to the Diphosphatidylglycerol is made from two phos-
group that is substituted on the phosphate. The phatidylglycerol molecules. The transfer of the
major phospholipid in cell membranes from fatty acyl groups to the glycerol and all subse-
several bacteria is phosphatidylethanolamine. quent steps in phospholipid synthesis take place
Substitution on the phosphate begins with the in the cell membrane. Archaeal phospholipids
displacement of pyrophosphate from CTP by are synthesized from either dihydroxyacetone
the phosphate on phosphatidic acid, the product phosphate or glycerol phosphate. The precur-
being CDP–diglyceride. The pyrophosphate is sor to the alcohol portion is geranylgeranyl
hydrolyzed by a pyrophosphatase, which makes pyrophosphate, which forms an ether linkage
the pathway irreversible. CMP is then displaced to the glycerol backbone.
by serine to form phosphatidylserine. A decar- The ribose and deoxyribose moieties of the
boxylation leads to phosphatidylethanolamine. nucleotides are derived from 5-phosphoribo-
Phosphatidylglycerol phosphate is formed by syl-1-pyrophosphate (PPRP), which is syn-
displacing CMP from CDP–diacylglyceride thesized from ribose-5-phosphate and ATP.
with glycerol phosphate. A subsequent dephos- In this reaction, the OH on the C1 of ribose-5
phorylation produces phosphatidylglycerol. -phosphate displaces AMP from ATP to form
www.ebook777.com
the pyrophosphate derivative. It is therefore a synthesized by the reductive amination of

pyrophosphoryl group transfer rather than the α-ketoglutarate to glutamate by glutamate
usual phosphoryl or AMP group transfer. The dehydrogenase, a reaction that requires high
pyrophosphate itself is displaced from PPRP by concentrations of ammonia.
orotic acid during pyrimidine synthesis or by an The degradation of amino acids occurs by
amino group from glutamine during purine bio- pathways different from the biosynthetic ones.
synthesis. A pyrophosphatase hydrolyzes the The first step is the removal of the amino group,
pyrophosphate to inorganic phosphate, thus by an amino acid oxidase, an amino acid dehy-
driving nucleotide synthesis. drogenase, or a deaminase. The carbon skeleton
Whenever pyrophosphate is hydrolyzed to eventually enters the citric acid cycle.
inorganic phosphate, the equivalent of two Hydrocarbon catabolism begins with a
ATPs is necessary to restore both the phos- hydroxylation to form the alcohol, which is
phates to ATP. Pyrimidine and purine biosyn- oxidized to the carboxylic acid. The carboxylic
thesis differ in that the pyrimidines are made acid is degraded via the β-oxidative pathway.
separately and then attached to the ribose phos- Some bacteria use an alternative pathway to
phate, whereas the purine ring is built piece by initiate hydrocarbon degradation, in which the
piece while attached to the ribose phosphate. hydrocarbon is oxidized to a ketone, which is
The deoxyribonucleotides are formed by reduc- eventually hydrolyzed to acetate and a long-
tion of the ribonucleotide diphosphates. chain alcohol.
All the phosphates in the nucleotides are
donated by ATP. The α phosphate is derived
from ribose-5-phosphate, which in turn is syn- Study Questions
thesized from glucose-6-phosphate. In some
bacteria (e.g., E. coli), the phosphate in glu- 1. Fatty acid synthesis is not simply the reverse
cose-6-phosphate may come from PEP during of oxidation. What features distinguish the
transport into the cell via the phosphotrans- two pathways from each other?
ferase system, whereas in other bacteria it is
2. What is it about the structure of phospho-
transferred to glucose from ATP via hexoki-
lipids that causes them to form bilayers
nase or glucokinase. The β and γ phosphates
spontaneously?
are derived from kinase reactions in which
ATP is the donor. The inorganic phosphate 3. Glycerol can be incorporated into phospho-
is incorporated into ATP via substrate-level lipids. Write a pathway showing the syn-
phosphorylation or electron transport phos- thesis of phosphatidic acid from glycerol.
phorylation. Since the various nucleotide
4. The C3 of glycerol becomes the C2 and C8
triphosphates provide the energy for the syn-
of purines. Write a sequence of reactions
thesis of nucleic acids, protein, lipids, and
showing how this can occur.
polysaccharides, as well as other reactions, it
is clear that ATP fuels all the biochemical reac- 5. Show how three carbons of succinic acid
tions in the cytosol. can become C4, C5, and C6 of pyrimidines.
To incorporate ammonia into cell material, What happens to the fourth carbon from
cells must synthesize both glutamate and glu- succinate?
tamine. The glutamate donates amino groups
6. What drives the condensation reaction in
to the amino acids, and the glutamine donates
fatty acid synthesis?
amino groups to purines, pyrimidines, amino
sugars, and some amino acids. Glutamine is 7. Write a reaction sequence by which bacte-
synthesized by the ATP-dependent amination of ria incorporate the nitrogen from ammonia
glutamate to form glutamine. This is catalyzed into glutamate and glutamine when ammo-
by glutamine synthetase. The glutamine then nia concentrations are low. How might this
donates the amino group to α-ketoglutarate occur when ammonia concentrations are
to form glutamate. The latter reaction is cata- high? What is the fate of the nitrogen incor-
lyzed by glutamate synthase, also known as porated into glutamate, and that incorpo-
the GOGAT enzyme. Glutamate can also be rated into glutamine?
8. ATP drives the synthesis of phospholipids. 8. There are actually three synthases and three pos-
Write the reactions showing the incorpo- sible routes to acetoacetyl–ACP in E. coli. In one
pathway, β-ketoacyl–ACP synthase III catalyzes
ration of glycerol into phosphatidylserine.
the condensation of acetyl–CoA with malonyl–
Focus on the steps that require a high-en- ACP. A defect in synthase III leads to overproduc-
ergy donor. How is ATP involved? (You tion of C18 fatty acids, whereas overproduction of
must account for the synthesis of CTP.) synthase III leads to a decrease in the average chain
lengths of the fatty acids synthesized. The decrease
9. Bacteria that utilize aliphatic hydrocarbons in the fatty acid chain lengths in strains overproduc-
as a carbon and energy source are usually ing synthase III has been rationalized by assuming
aerobes. What is the explanation for the that synthase III is primarily active in the initial
condensation of acetyl–CoA and malonyl–ACP
requirement for oxygen?
and that the increased levels of synthase III stimu-
late the initial condensation reaction and divert
malonyl–ACP from the terminal elongation reac-
REFERENCES AND NOTES tions. In a second initiation pathway, acetyl-ACP is
a substrate for β-ketoacetyl–ACP synthase I or II. In
1. Textor, S., V. F. Wendisch, A. A. De Graff, U. a third pathway, which is not thought to be physio-
Müller, M. I. Linder, D. Linder, and W. Buckel. logically significant under most growth conditions,
1997. Propionate oxidation in Escherichia coli: evi- malonyl–ACP is decarboxylated by synthase I and
dence for operation of a methylcitrate cycle in bacte- the resultant acetyl–ACP condenses with malonyl–
ria. Arch. Microbiol. 168:428–436. ACP. The different synthases appear to be involved
in determining the types of fatty acid made. In
2. Gerike, U., D. W. Hough, N. J. Russell, M. L. particular, synthases I and II, which catalyze con-
Dyall-Smith, and M. J. Danson. 1998. Citrate syn- densations in both saturated and unsaturated fatty
thase and 2-methylcitrate synthase: structural, func- acid synthesis, appear to have specific roles in the
tional and evolutionary relationships. Microbiology synthesis of unsaturated fatty acids. For example,
144:929–935. mutants of E. coli that lack synthase I do not make
3. Campbell, J. W., R. M. Morgan-Kiss, and J. any unsaturated fatty acids, suggesting that only
E. Cronan Jr. 2003. A new Escherichia coli meta- synthase I is capable of catalyzing the elongation of
bolic competency: growth on fatty acids by a novel cis-3-decenoyl–ACP. Mutations in synthase II lead
anaerobic beta-oxidation pathway. Mol. Microbiol. to an inability to synthesize cis-vaccenate, in agree-
47:793–805. ment with the finding that synthase II can elongate
palmitoleoyl–ACP but synthase I cannot.
4. Magnuson, K., S. Jackowski, C. O. Rock, and J. E.
Cronan Jr. 1993. Regulation of fatty acid biosynthe- 9. Mammals require linoleate (18:2 cis-∆9, ∆12)
sis in Escherichia coli. Microbiol. Rev. 57:522–542. and linolenate (18:3 cis-∆9, ∆12, ∆15). Hence these
are essential fatty acids and must be supplied in the
5. CoASH = P–AMP-pantothenic acid–β-mercapto- diet of mammals. Mammals cannot synthesize these
ethylamine–SH. Acyl carrier protein = protein–pan- fatty acids because they do not have the enzymes to
tothenicacid–β-mercaptoethylamine–SH. Coenzyme introduce double bonds in fatty acids longer than
A has a phosphorylated derivative of AMP (AMP- C9. Arachidonate, a C20 fatty acid with four double
3′-phosphate) attached via a pyrophosphate link- bonds, can be synthesized by mammals from lino-
age to the vitamin pantothenic acid (a B2 vitamin), lenate. Arachidonate, in turn, is a precursor to vari-
which is covalently bound to β-mercaptoethylamine ous signaling molecules such as prostaglandins.
via an amide linkage. The β-mercaptoethylamine 10. Naturally occurring fatty acids are mostly cis
provides the SH group at the end of the molecule. with respect to the configuration of the double bond:
In the acyl carrier protein, the AMP is missing and
the pantothenic acid is bound directly to the pro-
tein. Therefore, the functional end of the acyl car-
rier protein is identical to CoASH, but the end that
binds to the enzymes is different.
6. Acetyl–CoA carboxylase consists of four subunit
proteins. These are biotin carboxylase, biotin car- This cis configuration produces a bend of about
boxyl carrier protein, and two proteins that carry out 30° in the chain, whereas the trans configuration is a
the transcarboxylation of the carboxy group from straight chain, as is the saturated.
biotin to acetyl–CoA.
7. Whether there is a separate acetyl–CoA:ACP
transacetylase is not certain. Reaction 3 can be cata-
lyzed by β-ketoacyl–ACP synthase III (acetoacetyl–
ACP synthase), which has transacetylase activity.
www.ebook777.com
11. Heath, R. J., S. Jackowski, and C. O. Rock. 1994. enzyme can use either NH+4 or glutamine. In both
Guanosine tetraphosphate inhibition of fatty acid and cases the requirement for ATP can be explained by
phospholipid synthesis in Escherichia coli is relieved the formation of a phosphate ester with the carbo-
by overexpression of glycerol-3-phosphate acyltrans- nyl oxygen. The amino group then displaces the
ferase (plsB). J. Biol. Chem. 266:26584–26590. phosphate.
12. In addition to the carbamoyl synthetase described 14. Aeckersberg, F., F. Bak, and F. Widdel. 1991.
here, vertebrates have a carbamoyl synthetase that Anaerobic oxidation of saturated hydrocarbons to
combines NH24, CO2, 2 ATP, and H2O to form car- CO2 by a new type of sulfate-reducing bacterium.
bamoyl phosphate that is used to convert NH24, to Arch. Microbiol. 156:5–14.
urea in the urea cycle.
15. Munro, A. W., and J. G. Lindsay. 1996. Bacterial
13. In mammals the amino group is donated cytochromes P-450. Mol. Microbiol. 20:1115–
from the amide group of glutamine. In E. coli the 1125.
11
RNA and Protein Synthesis
This chapter will focus on the synthesis of ribo- are condensed. These components are either
nucleic acids (RNAs) and proteins. However, the the products of central biosynthetic pathways
synthesis of virtually all macromolecules shares described in other chapters (ribose in Chapter 9
a similar molecular logic. The student should and nucleobases in Chapter 10), or they are
seek to apply the principles outlined below to taken up from the medium (phosphate). In the
the biosynthesis of DNA (Chapter 3), carbohy- case of proteins, the synthesis involves conden-
drates (Chapter 12), and lipids (Chapter 10). sation of amino acids, activated as esters with
The three fundamental principles of macro- a tRNA specific to its cognate amino acid. It is
molecular synthesis, discussed here in turn are crucial to recognize that the long biopolymers
that (1) synthesis is modular; (2) assembly of of cells are built from small molecules that most
components relies on dehydration reactions; prokaryotes are readily capable of synthesizing.
and (3) the chemical reactions are energetically The synthesis of macromolecules is built on
unfavorable. In simpler terms, macromolecules dehydration reactions. The mechanistic details
such as RNA, proteins, DNA, lipids, and carbo- of RNA and protein synthesis are described in
hydrates are built from small, easily synthesized detail in the sections that follow. An important
components. Assembly of these components principle that is easily obscured by the detailed
is usually based on a dehydration reaction mechanisms is that when a sugar and phosphate
between an acid, either carboxylic or phos- condense, or when two amino acids are linked
phoric, and a nucleophile, either an alcohol or to become a peptide, the reaction involves the
amine. Such condensation reactions require the elimination of a water molecule (Fig. 11.1). The
input of energy. sequence of events in macromolecular assembly
The synthesis of macromolecules is modu- is inherently similar. It involves the activation
lar. In the case of RNA, the modular nature of of an acid (carboxylic or phosphoric), followed
assembly is evident on two levels. RNAs, such as by the attack of a nucleophile (an alcohol or
mRNA, rRNA, or tRNAs, are synthesized as a amine) on the activated center.
repeating polymer comprising a sugar (ribose), The synthesis of macromolecules is energeti-
phosphate, and one of four heterocyclic bases. cally unfavorable. The elimination of water in
RNA synthesis, catalyzed by an RNA polyan aqueous environment is an energetically
merase, catalyzes the formation between an unfavorable reaction. This means that for
alcohol (the 3′-OH of ribose) and an activated each dehydration reaction, at least one mol-
phosphoric acid. On a deeper level, each of the ecule with a high group transfer potential must
four RNA nucleotides (UMP, GMP, AMP, be expended. (See Section 8.2 for a descrip-
and CMP) is built from smaller components. tion of group transfer reactions). Hydrolysis
One sugar, one phosphate, and one nucleobase of at least two phosphate–phosphate bonds is
281
www.ebook777.com
O O
..
A (5') Ribose–CH 2–O–P OH + HO Ribose (3') (5') Ribose–CH2–O–P–O–Ribo
O– H2O O–
B H H H R H R
R R
.. OH .. OH
H2N H2N H2N N
H
O O H2O O O
Fig. 11.1 Synthesis of macromolecules involves dehydration. (A) Nucleic acid polymers are built by forming
phosphoric acid esters between ribose and phosphate. The nucleobase is not shown, nor is the activation step.
(B) Peptides are polymers of amino acids built by forming amides between successive amino acids. R describes
an amino acid side chain. With both nucleic acid and amino acid polymers, the acid must first be activated
using a molecule with a high group transfer potential such as ATP (not shown).
ultimately required to incorporate each mono- referred to as the plus (+) strand. The RNA is
mer into a growing polymer of RNA or protein. synthesized in opposite strand polarity to the
Macromolecular synthesis is therefore inher- template DNA strand; that is, synthesis occurs
ently expensive from an energetic perspective. in the 5′-to-3′ direction, and the 3′ end of the
The converse of this rule implies an important RNA (the growing end) extends toward the 5′
feature of biological macromolecules, namely, end of the DNA. For transcription to occur, the
that degradation (hydrolysis) is energetically DNA must unwind at the site of transcription
favorable. Only a catalyst is required to facili- so that DNA-dependent RNA polymerase, the
tate hydrolysis, such as a ribonuclease for RNA enzyme that synthesizes RNA, has access to the
or a protease for proteins. While some nucleases single-stranded template DNA. There are three
and proteases use ATP, the requirement is not formal stages in transcription: initiation, chain
for the chemical step of hydrolysis. Breakdown elongation, and termination.
of macromolecules yields the same building
blocks that may later be repurposed for synthe- 11.1.1 RNA holoenzyme structure
sis or degraded to yield ATP.
The core RNA polymerase (RNAP) in bacteria
This chapter discusses important features of
usually has five subunits (α2β′β′ω). For tran-
the biosynthesis of RNA and protein, includ-
scription to be initiated, a sixth subunit, called
ing metabolic regulation of the pathways.
sigma (σ) factor, must attach to the core enzyme
Structural studies involving the RNA poly-
to form the RNAP holoenzyme. Sigma factors
merase holoenzyme or the ribosome complexed
are responsible for recognition of initiation
with substrates and regulatory molecules are
sequences called promoters and are described
also described, and these provide a powerful
later. Crystallographic structures for the core
context for interpreting decades of genetic and
polymerase structures have been solved for a
biochemical observations.
number of bacteria, and the overall topology is
highly conserved (Fig. 11.2). The overall isoelec-
11.1 RNA Synthesis tric point is generally low (pI ~5.1), consistent
RNA synthesis is called transcription and with an overall negative surface charge. Each
involves accessing one of the strands of DNA structure has a characteristic cleft (~25 Å deep)
as a template to make an RNA copy. While lined with positive charges that interacts with the
the information is preserved, RNA replaces DNA substrate. The structure is often compared
the thymine and deoxyribose from DNA with to the claw of a crab (Fig. 11.2A), much as DNA
uracil and ribose. The template strand is some- polymerases are likened to a hand, with fingers,
times called the coding strand or the minus (–) palm, and thumb to grasp DNA. The active
strand. The nontemplate strand is sometimes site is buried deep in the primary cleft formed
rna and protein synthesis 283
– – – 11.1.2 Initiation, chain elongation,

– – – – – and termination
– – Initiation begins at the promoter region
Initiation of transcription is sequence specific
+ – and takes place at the promoter site, which is
° ++
+ ++ – a region of DNA at the beginning of the gene
A + where the σ subunit of the RNA polymerase
– binds (Fig. 11.3).1 The promoter region of most
– ′ – genes is identifiable by conserved nucleotides
– centered approximately –10 and –35 base pairs
– – – – – upstream from the site at which transcription is
– – – actually initiated, that is, where the first nucle-
otide is incorporated. This is also called the
+1 position, or the start site. The polymerase
holoenzyme complex spans the entire promoter,
not simply the –10 and –35 regions. However,
the –10 and –35 regions are critical because
° ′ mutations that affect promoter function usually
B occur in one of these sequences. Furthermore,
GreB as discussed later, the frequency of initiation
binding of transcription is specific for promoters and
reflects variations in the sequences in the –35
and –10 regions.
Fig. 11.2 The bacterial RNA polymerase core struc- In E. coli, the most common sequence in most
ture. (A) The core comprises five subunits, α, α, β, β′, of the genes at –10, sometimes called the Pribnow
and ω, and is often described as a claw. The enzyme
box, is 5′-TATAAT-3′ and the sequence at –35
binds RNA in the deep, positively charged channel
is 5′-TTGACA-3′. These are called the consen-
that cuts between the β and β′ subunits. The active
site Mg2+ (black sphere) is positioned at the base of the sus sequences, and although they are not identi-
channel. (B) A side view of the core structure reveals cal in all promoters, there is sufficient similarity
a secondary channel that provides access to the back to permit the investigator [and the major RNA
of the active site. This allows access for nucleotides to polymerase holoenzyme, called the sigma 70
the active site in the closed elongation complex and for (σ70) polymerase] to recognize these sequences.
cleavage factors such as GreA (not shown) or GreB; Promoter sequences other than the consen-
see Marr, M. T., and J. W. Roberts. 2000. Function sus –35 and –10 sequences exist, but these are
of transcription cleavage factors GreA and GreB at not recognized by the major RNA polymerase
a regulatory pause site. Mol. Cell 6(6):1275–1285. holoenzyme (core plus sigma factor). Instead,
These proteins can induce transcript cleavage when
they are recognized by RNA polymerases com-
the polymerase stalls, apparently by activating an
plexed with other sigma factors. (See the discus-
endonucleolytic activity inherent to the core poly-
merase. Source: Adapted from Darst, S. A. 2001. sion of the sigma subunit later.) The binding of
Bacterial polymerase response. Curr. Opin. Struct. the RNA polymerase to these sequences and
Biol. 11:155–162. to the rest of the promoter region leads to the
localized unwinding of the DNA strands, which
allows the RNA polymerase to move down the
template strand, making an RNA copy begin-
by the ββ′ subunits. This is where the catalytic ning at the start site. In contrast to DNA replica-
magnesium is located, which identifies the site tion, a helicase is not required.
of nucleotide condensation. Because the active
site becomes buried when the substrate DNA is Steps in initiation and chain elongation
captured and the cleft is closed, the nucleoside A structural view of the stages of RNA tran-
triphosphates (NTPs) must enter through a sec- scription is presented in Fig. 11.4. The sigma (σ)
ondary “back door” channel in the β′ subunit factor (dark gray structure) associates with the
(Fig. 11.2B). core polymerase to form the holoenzyme. The
www.ebook777.com
Fig. 11.3 Consensus promoter sequences for –10 and –35 regions in noncoding strand. Centered about 10
base pairs upstream of the mRNA start site (at position +1) is a 6-base region that is usually 5′-TATAAT-3′ in
the noncoding strand or some minor variation thereof. It is called the –10 sequence or Pribnow box. A second
consensus sequence is centered approximately 35 base pairs upstream of the transcription start site. It is gener-
ally referred to as the –35 sequence and is 5′-TTGACA-3′. The RNA polymerase binds to the entire promoter,
not simply to the –35 and –10 regions. Unwinding of the DNA to initiate transcription occurs at the –10 region
and extends for about 20 base pairs past the mRNA transcription start site. Interactions between the RNA
polymerase and single-stranded template DNA help maintain the open complex.
binding is tight, with a dissociation constant of itself enforces the open complex throughout
~1 nM, built from multiple weak interactions the process of elongation.
between the core and the four σ factor domains.
It is an extended protein structure comprising 3. Binding of initiating ribonucleotides
four subdomains that span the –35 and –10 Transcription is usually initiated at the start site
regions. The sigma group subdomain (σ2) binds (+1) with ATP or GTP, whose ribose provides
the –10 region, and the group 4 subdomain (σ4) the free 3′-hydroxyl that attacks the α phos-
binds the –35 region. phate in the incoming nucleotide, displacing
pyrophosphate. The newly synthesized RNA
1. Formation of the closed complex therefore has a triphosphate at its 5′ end. As
The RNA polymerase holoenzyme bound shown in Figs. 11.2B and 11.4E, the incoming
to the promoter region is referred to as the NTPs enter the active site through a secondary
“closed” complex (Fig. 11.4A). At this stage channel because the elongation complex closes
the DNA exists as a double helix, and the DNA on the DNA substrate, preventing access via the
is recognized via specific contacts with the σ cleft.
subunit.
4. Elongation
2. Formation of open complex RNA synthesis is said to be in its elongation
The DNA duplex unwinds at the promoter stage after a few nucleotides (about 12) have
region catalyzed by the σ factor to form the been added to the growing chain of RNA. The
open complex. There is no requirement for emerging RNA chain is thought to cause a con-
a helicase or a nucleotide triphosphate such formational change that weakens the interac-
as ATP. Instead, it appears that a small set of tion between the σ4 domain and the –35 region
aromatic residues within the σ2 domain are of the promoter. The sigma subunit thus disso-
perfectly positioned to capture the A/T-rich ciates from the polymerase, and the polymerase
–10 region as it “breathes,” or transiently moves forward from the promoter region as it
melts (Fig. 11.4B). A tryptophan residue (W) synthesizes the RNA. In the DNA duplex, there
is highly conserved in RNA polymerase struc- is an opening about 18 bases long, called a tran-
tures and base stacks at the –12 site, defining scription bubble, in which the elongating RNA
the upstream end of the bubble. As the open forms an RNA/DNA hybrid with the template
complex initially forms, a long extension of 90 DNA strand. The RNA/DNA hybrid helps keep
amino acids (the σ1 domain) swings out of the the RNA polymerase attached to the DNA. The
active site cleft in preparation for DNA load- average rate of elongation of messenger RNA
ing. Formation of the fully opened complex for E. coli is between 40 and 50 nucleotides per
(Fig. 11.4C) is associated with the complete second, versus approximately 1,000 nucleotides
capture of the template strand in an enclosed, per second for elongation of DNA. The rate of
positively charged channel between the β and elongation of mRNA matches the rate of pro-
β′ subunits. Thus the RNA polymerase core tein synthesis because approximately 16 amino
DNA melting
–35/ 4 –10/ 2
WY
F
4 3
A 2
Mg2+
Mg2+ 1
s1 B
5′ lid zipper ++
+ +
rudder
RNA
Mg2+
Mg2+ 4
2
++
+ +
E NTP entry C
Mg2+
Fig. 11.4 Initiation of transcription. The sigma subunit (darkest gray) initially interacts with the template
DNA duplex (helical strands of gray and black that lie on top of the structure). The catalytic Mg2+ defines the
active site (dark sphere). The view is through the β′ subunit, which is partially cut away to reveal the active
site cleft (lighter gray). The nascent RNA strand appears as a heavier gray single line that initiates adjacent
to the central Mg2+. (A) The RNA polymerase, with its attached sigma subunit, binds to the DNA along the
length of the promoter region. This is called the closed complex. (B) Interaction between the σ subunit (σ2)
and the –10 region initiate the opening of the transcription bubble (the open complex). (C) The extended
σ1 subdomain clears the cleft and the open complex is extended into the cleft of the core polymerase. The
negatively charged template DNA strand is fed through a positively charged channel that enforces the open
complex. RNA synthesis is initiated at the start site with either ATP or GTP. (D) The σ4 subdomain loses
contact with the –35 region of the promoter, and the σ subunit is expelled. The final structure (E) is called the
TEC, or Transcription Elongation Complex, and is highly processive. Small arrows identify the direction of
DNA movement through the core polymerase. During elongation, an opening in the DNA duplex of about 18
bases called a transcription bubble is maintained in which the elongating RNA forms an RNA/DNA hybrid.
Adapted from Murakami, K. S., and S. A. Darst. 2003. Bacterial polymerases: the whole story. Curr. Opin.
Struct. Biol. 13:31–39.
acids per second are added to a growing poly- in fast-growing cells is synthesized at a rate of
peptide chain. For this to occur, the messenger about 90 nucleotides per second.
RNA must move through the ribosome at a rate
of 48 nucleotides (i.e., 16 codons), per second, Termination
or approximately the same rate as its synthe- Upon chain termination the RNA and RNA
sis. See the discussion of coupled transcription/ polymerase are released from the DNA,
translation in Section 11.2.8. Ribosomal RNA and the DNA duplex re-forms at that site.2,3
www.ebook777.com
Transcription is terminated by a DNA sequence to the DNA template (Fig. 11.5B). The loop is
called a terminator. There are two patterns of followed by a short RNA–DNA duplex consist-
termination. One requires an additional protein ing of a string of A–U base pairs that is unstable.
factor and is called factor-dependent termina- (The interactions between A and U are weaker
tion. The best-studied termination factor is a than the other base pair interactions.) The con-
protein called Rho. We begin, however, with a sequence is that the RNA spontaneously disso-
type of termination that does not require a pro- ciates from the template along the A–U region
tein factor: factor-independent termination. and from the polymerase. The RNA polymerase
also dissociates from the DNA and the DNA
1. Factor-independent type
duplex reforms.
Factor-independent termination occurs when
the RNA polymerase transcribes a self-comple- 2. Rho-dependent type
mentary sequence of bases (Fig. 11.5A) that pro- Other terminator regions require protein fac-
motes the localized hybridization of the RNA to tors for termination and do not rely on RNA–
itself (formation of a hairpin loop) rather than RNA hairpin loops to disrupt the RNA/DNA
Fig. 11.5 Rho-independent termination. (A) At the 3′ terminus of the mRNA there is an inverted repeat sepa-
rating two nonrepeating sequences. (B) For example, the sequence in the E. coli trp mRNA (encoding enzymes
to synthesize tryptophan) is AAAGGCUCC-UUUU-GGAGCCUUU, which causes the mRNA to form a stem-
and-loop (hairpin) structure near the 3′ end of the mRNA. Near the loop or within the stem there is a sequence
rich in G + C base pairs. There are also 6 to 8 uracils at the 3′ end of the mRNA (corresponding to a string of
adenines in the template DNA). The uracils are necessary for termination, as shown by mutants in which some
of the A–T base pairs in this region of the DNA have been removed. According to the model, when the RNA
hairpin forms, the RNA hybridizes to itself rather than to the DNA and thus the RNA–DNA duplex is dis-
rupted in the transcription complex. In addition, the U–A duplex following the hairpin is relatively unstable.
This leads to the dissociation of the RNA from the DNA.
hybrid in the transcription bubble. The nucle- regions near the promoter and either stimulate
otide sequences in the terminator regions are RNA polymerase activity (e.g., by enhancing its
not closely related and are not always easy to binding or the formation of the open complex,
recognize. There are three termination factors or inhibit the activity of the polymerase). These
in E. coli. They are Rho, Tau, and NusA. The transcription factors are extremely important
best studied is Rho, which is an RNA-dependent for the regulation of the timing of gene expres-
ATPase and an ATP-dependent RNA/DNA sion. See Section 11.1.2 for a discussion of the
helicase. As we shall see, these two activities of regulation of RNA transcription.
Rho are necessary for termination.
Rho works by binding to particular sequences, 11.1.4 Role of topoisomerases
called rut (rho utilization sites) in the RNA Just as during DNA replication, the unwind-
directly behind the RNA polymerase. The rut ing of the duplex produces positive supercoils
sequences are characterized as rich in cytosine downstream of the unwinding region. If nothing
over guanosine, which is essential for Rho bind- were done with these supercoils, DNA would
ing. In general, the Rho protein is excluded from stop unwinding and transcription would stop.
the transcript if a translating ribosome is pres- Topoisomerases change the positive supercoils
ent. If, however, translation is terminated and to negative supercoils within which the DNA
rut sequences are exposed, they serve as assem- strands are more readily unwound.
bly sites for a hexamer of Rho that encircles the
transcript. The pore in the Rho hexamer is only 11.1.5 The sigma subunit
large enough for single-stranded RNA, and thus Sigma factors are responsible for recognition of
duplexes are excluded. The Rho protein acts as the promoter sequences (for reviews, see refs. 4
an ATP-dependent RNA helicase, rapidly mov- and 5). They are extended structures that must
ing along the transcript in the 5′-to-3′ direction, first associate with a core RNA polymerase to
trailing and then overtaking the core poly- function. All sigma factors share the function
merase. The action of Rho is thought to physi- of catalyzing the formation of an open DNA
cally pull apart the RNA/DNA hybrid duplex. complex (see Section 11.1.2) in the active site
This results in the disengagement of the RNA cleft of the core polymerase following rec-
polymerase from the DNA and the re-forma- ognition of its cognate promoter sequence.
tion of the DNA–DNA duplex. Thus, when the Different sigma factors recognize different
ribosome stops translating at the translational consensus sequences in the promoter region.
stop site in the mRNA, Rho can bind at the next Thus, sigma factors determine whether a gene
accessible rut site and transcription will stop at will be transcribed at any point in time. This is
the next Rho-dependent termination site. Rho- aided by a large number of transcription fac-
dependent termination can also explain polar- tors that activate and repress the transcription
ity, as discussed in Section 11.2.10. of specific genes. Clearly this is not a simple
task: a bacterium such as Escherichia coli has
11.1.3 Frequency of initiation approximately 4.6 million base pairs and 2,000
As stated, promoter regions are not identical, genes. After initiation, sigma factor dissociates
and they do differ in the extent to which they from the polymerase; that is, it is not needed for
influence the frequency of initiation of transcrip- chain elongation.
tion (“promoter strength”). Strong promoters A bacterium can have more than one sigma
cause initiation every few seconds and weak factor, and the core polymerase can exchange
promoters every few minutes. This discrepancy one for another, thus contributing toward the
is related to the consensus sequences at the –10 timing and specificity of gene transcription. For
and –35 regions because strong promoters have example, the main sigma factor in E. coli is σ70
sequences that are more similar to the consen- (having a molecular weight of 70 kDa), which
sus sequences, whereas weak promoters have recognizes the consensus sequence in most of
base substitutions in these regions. the promoters and is termed the “housekeep-
Various transcription factors, both positive ing” sigma. However, several alternative sigma
and negative, also influence the frequency of factors may be present, adding specificity to
transcription. The transcription factors bind to which genes are transcribed.
www.ebook777.com
Sequence analysis has revealed that all the The factor σs was discovered because of its role
sigma factors fall into two families: σ70 and σ54. in transcribing genes during stationary phase.
Most bacteria will have several different sigma However, activation of σs is also triggered by
factors of the σ70 family but only one repre- high osmolarity and by stress conditions such as
sentative of the σ54 family. Bacteria vary in the oxidative stress, UV radiation, and heat shock.
number of sigma factors. Two spore-forming Indeed, σs has been called a master regulator of
bacteria, Bacillus subtilis and Streptomyces the general stress response. The regulation of
coelicolor, have 18 and 63 members, respec- σs is complex and occurs at the transcriptional,
tively, of the σ70 family. Not all bacteria have a translational, and post-translational levels. (See
σ54 gene. For example, high-GC, gram-positive Section 2.2.2.)
bacteria and cyanobacteria, are not known to When subjected to heat stress, E. coli increases
possess one. the amount of another sigma factor in the σ70
Sigma factors belonging to the σ70 family can family, σ32, which stimulates the transcription
initiate transcription without any additional of genes encoding heat-shock proteins that
energy source or other proteins. Members of fold or degrade cytoplasmic proteins damaged
the σ54 family are different. They require an by stresses such as heat shock (Section 16.3).
activator protein that binds about 100 to 150 Another sigma factor in E. coli that responds
base pairs upstream of the promoter region at to stress is σE. It is induced by unfolded proteins
“enhancer sites” to ensure that an open com- in the cell envelope and stimulates the expres-
plex forms and transcription is initiated at the sion of genes that repair the envelope.6 The
promoter region. The activator proteins are genes stimulated by σE include those that encode
ATP dependent. As an example, read the discus- chaperones, as well as proteases that refold or
sion in Section 19.4.1 of the use of σ54 by E. coli degrade misfolded proteins in the envelope.
to transcribe genes in the Ntr regulon when the Still another sigma factor in E. coli that
organism is growing under limiting ammonia responds to stress, σFecI, stimulates the transcrip-
conditions, and examine Fig. 19.8. tion of the fecABCDE operon that encodes a
E. coli has six sigma factors in the σ70 family: transport system for bringing iron citrate into
σ (RpoD), σS (RpoS), σ32 (RpoH), σF (FliA), σE
70
the cell. The stress signal for this sigma factor is
(RpoE), and σFecI, and they differ in the speci- iron starvation. The sigma factor σF (FliA), also
ficity of the genes that they transcribe. As men- called σ28, initiates transcription of genes neces-
tioned previously, σ70 is the “housekeeping” sary for flagellum biosynthesis and chemotaxis.
sigma, and the alternative sigma factors are used These include genes that encode the flagellar fil-
for the transcription of genes activated by stress ament protein (HAPS protein) and components
conditions, growth transitions, and morpho- of the chemotaxis system.
logical changes such as flagellar biogenesis. For
example, when E. coli enters stationary phase Anti-sigma factors
as a result of starvation, it increases the amount How do bacteria regulate the activities of the
of σs (also called RpoS or σ38), which recognizes sigma factors? One strategy for control is based
promoters for genes that are expressed during on engaging anti-sigma factors that bind to them
stationary phase (Section 2.2.2). The amounts and inhibit their activity. For a description of an
of σs also increase during growth in high osmo- anti-sigma factor that is part of the regulatory
larity, and this reflects, at least in part, the role system for sporulation genes in Bacillus subtilis,
of σs in transcribing the genes encoding mech- see the discussion of SpoIIAB in Section 2.3.3.
anosensitive channel proteins discussed in An anti-sigma factor in Escherichia coli is
Section 16.2.5. The mechanosensitive channels anti-σE, called RseA. (For a review of this factor,
protect the cells from hypo-osmotic shock when whose short form stands for regulators of sigma
they are diluted into low osmolarity media. The E, read refs. 4 and 5). RseA is a transmembrane
promoter sequence recognized by σs is actu- protein with a cytoplasmic domain that binds to
ally the same as the core consensus promoter σE and keeps it inactive. Recall that σE is used to
sequence recognized by σ70. However, there transcribe genes that are important for repair-
exist regulatory factors that enhance selectivity ing the cell envelope when its proteins become
of σs. These are reviewed in Gruber and Gross.6 damaged. Under these circumstances, two cell
membrane proteases, DegS and Yael, sequen- signal transduction pathways important for
tially degrade RseA, releasing σE into the cyto- these responses are described in Chapters 16
plasm, where it binds to core RNA polymerase and 19 through 23.
and stimulates transcription of the genes in the
σE regulon. Operons
As mentioned earlier, the genes stimulated Frequently bacterial genes are arranged in oper-
by σE include those that refold or degrade mis- ons. An operon consists of two or more genes
folded envelope and periplasmic proteins. A that are cotranscribed into a polycistronic
potent signal for the proteolysis of RseA is the mRNA from a single promoter at the beginning
presence of unfolded porin proteins in the outer of the first gene. All the genes in the operon are
membrane. It has been suggested that when coordinately regulated from a single promoter.
the porin, which is a trimer, unfolds as a con-
sequence of membrane stress, part of the pro- Transcription factors
tein that is normally buried in the folded porin Transcription is commonly regulated at the ini-
trimer is exposed and binds to the membrane tiation step by proteins that affect in some way
protease DegS. This binding activates the pro- the binding of RNA polymerase to the promoter,
teolytic activity of DegS, leading to partial pro- the “melting” of the DNA to form the transcrip-
teolysis of RseA. The partially degraded RseA, tion bubble, or the movement of the polymerase
in turn is a substrate for the membrane protease along the DNA. The transcription factors bind
Yael, which completes the degradation of RseA, to specific sequences in or near the promoter
in turn resulting in the release of σE. For a more region or, if the transcription factor stimulates
complete description of this model, see Ades. 5 gene expression, in so-called enhancer regions,
The anti-sigma factors comprise a wide range upstream from the promoter.
of structures and are controlled by diverse bio- Although most of the known transcription
logical signals (reviewed in ref. 4). The regulatory factors freely diffuse in the cytoplasm, two tran-
schemes, however, do not require transcription scription factors that activate the ToxT pro-
to respond to an environmental cue. Like the moter in Vibrio cholerae, ToxR and TcpP, are
σE example just cited, the σ factor/anti-σ factor bound to the cell membrane. This is described
pairs are always present in the cell. This condi- in Section 19.11.1. The transcription factors
tion allows a rapid transcriptional response that generally have a helix–turn–helix motif, as
is not dependent upon new protein synthesis. described later. The majority of transcription
factors that activate transcription bind to the
11.1.6 Regulation of transcription RNA polymerase and recruit the polymerase
Bacteria generally regulate protein synthesis at to the promoter. An example of this type of
the transcriptional level (see ref. 7 for a review.) transcription factor is the cyclic AMP recep-
Since mRNA in bacteria is generally unstable tor protein (CRP) discussed in Section 17.3.4.
(i.e., is rapidly degraded enzymatically), the Some transcription factors work by binding
inhibition of transcription of a particular gene to the DNA, altering its structure, and in this
generally means that the synthesis of the protein way positively influencing the initiation of tran-
encoded by that gene ceases quickly. Conversely, scription. An example is IHF (integration host
stimulation of transcription of a particular gene factor), discussed in Section 19.4.1. IHF binds
usually results in a rapid synthesis of the pro- to DNA and bends it so that a second bound
tein encoded by that gene. In this way, bacteria transcription activator is brought to the pro-
are able to modulate the mix of enzymes and moter and activates transcription. IHF can also
other proteins in response to environmental repress transcription by binding at or near the
challenges such as the presence or absence of promoter region of certain genes, as described
specific carbon and energy sources, or inor- in Section 19.7.2.
ganic compounds such as inorganic phosphate E. coli has about 350 transcription factors.6
or ammonium ion, and environmental stress The regulation of RNA synthesis by transcrip-
such as starvation, changes in pH, changes in tion factors is especially important when the
osmolarity, and changes in temperature. The levels of the sigma factor for the particular gene
responses to environmental challenges and the are not changed. Under these circumstances it is
www.ebook777.com
only the transcription factors that determine the Because helix 2 lies in the major groove, the
rates of transcription of specific genes. Global amino acid side chains that protrude from the
transcription factors regulate the transcription surface of the protein can contact and form
of genes in several different operons. All the noncovalent bonds (e.g., hydrogen bonds) with
operons regulated by a single transcription fac- base pairs in the DNA, and therefore can bind
tor are collectively called a regulon. to specific nucleotide sequences (Fig. 11.6C).
Hydrogen bonds can form between the carbonyl
Helix–turn–helix motif and amino groups in the protruding side chains
Many prokaryotic DNA-binding proteins, of certain amino acids in the DNA-binding
including activators and repressors of tran- region of the protein, and amino groups, ring
scription, have a helix-turn-helix motif in part nitrogen, and carbonyl groups in specific bases.
of the protein (Fig. 11.6A).8 In one part of the The amino acids that are involved in the bond-
protein approximately 7 to 9 amino acids form ing to the bases are generally asparagine, glu-
an α-helix, which is called helix 1. Helix 1 is tamine, glutamate, lysine, and arginine. For
connected by about 4 amino acids to a second example, glutamine can bind to A–T base pairs
α-helix of about the same size. The helices are because the carbonyl group and amino group in
approximately at right angles to each other. glutamine can hydrogen-bond with the amino
Helix 1 lies across the DNA and helix 2 (the rec- group on C6 and the N7 in adenine, respectively
ognition helix) lies in the major groove of the (Fig. 11.6C). This does not interfere with the
DNA (Fig. 11.6B). hydrogen bonding of the bases to each other.
Fig. 11.6 Helix–turn–helix motif and DNA binding. (A) A DNA-binding protein typically has in one portion
two α-helices connected by a short segment of about four amino acids. (B) Helix 2 lies in the major groove of
the DNA duplex, and helix 1 lies at approximately right angles to helix 2. (C) Because helix 2 lies in the major
groove, it makes contact with the bases. Certain amino acids in helix 2 with amino or carbonyl groups in their
side chains can hydrogen-bond to specific bases. Here glutamine (–CH2–CH2–CONH2) is hydrogen bonded to
adenine in an A:T base pair. Other amino acids, for example, asparagine (–CH2–CONH2), glutamate (–CH2–
CH2–COOH), lysine (–CH2–CH2–CH2–CH2–NH3+), and arginine (–CH2–CH2–CH2–NH–C(NH2)=N+H2),
can also hydrogen-bond to specific base pairs, and in this way the DNA-binding protein can bind to specific
nucleotide sequences. DNA-binding proteins are frequently dimers and bind to inverted repeats in the DNA.
How transcription factors work 1. Positive regulators

Activators bind to specific sites in the DNA Positive transcription regulators bind to the
upstream of the RNA polymerase and make promoter region or to enhancer sites, which
contact with the RNA polymerase while are sequences upstream from the promoter
enhancing its activity. Repressors bind down- region. They may make contact with RNA
stream of the RNA polymerase (in the operator polymerase and promote its binding to the
region) and block progression of the RNA poly- promoter region, and some may facilitate the
merase. Figure 11.7 illustrates general models “melting” of DNA to form the transcription
of how transcription activators and repressors bubble. Consider the cyclic AMP receptor pro-
are believed to work. Some specific examples of tein. When it binds to cAMP, the CRP can then
positive and negative transcription factors are bind to specific sites in the DNA and stimulate
described next. transcription.
Fig. 11.7 Models for how transcription activators and repressors work. Although not shown, several tran-
scription factors are known to function as dimers. (A) Activator proteins can bind at specific sites, called
activator sites, upstream of the promoter, where they make contact with the RNA polymerase, facilitating its
binding to the promoter. Some activators must bind to a small effector molecule (inducer) before they can bind
to the DNA. (B) If the activator binds a distance away, upstream of the promoter, then the DNA must bend to
bring the activator to the polymerase. (C) Sometimes a second protein (e.g., integration host factor, IHF), is
required to bend the DNA to bring the activator site to the RNA polymerase. Here IHF binds to a site between
the promoter region and the activator site and brings the activator to the RNA polymerase. (D) Repressor
proteins bind to the operator region, which may or may not be in the promoter region. The repressors block
progression of the RNA polymerase. Some repressor proteins (called aporepressors) must bind to an effector
molecule (corepressor) before they can bind to the DNA.
www.ebook777.com
For example, consider the stimulation by the transcription of the nitrate reductase gene
cAMP–CRP of the lac operon, which encodes (Section 19.3). According to the model, IHF
proteins required for the uptake and metabolism bends the DNA to bring the bound NarL–P to
of lactose. cAMP–CRP binds to specific nucle- the promoter, where it stimulates transcrip-
otide sequences upstream of the lac promoter as tion (Fig. 11.7). IHF can also inhibit transcrip-
well as to an incoming RNA polymerase. The tion if it binds to the promoter region. (See
binding of cAMP–CRP to the RNA polymerase Section 19.7.2.) Also, see the discussion of the
enhances its binding to the promoter and thus nucleoid in Section 1.2.6 for a description of
stimulates transcription. IHF and other DNA binding proteins that bind
Many unrelated genes are positively regu- to bent DNA.
lated by cAMP–CRP and are said to be part of Many operons are specifically activated by a
the CRP regulon. They are called cAMP-depen- protein (activator) only when the protein binds
dent genes. cAMP-dependent genes may encode an inducer. Examples include the activation
proteins required for the catabolism of several of the L-ara operon, which encodes proteins
different carbon and energy sources, such as required for the conversion of L-arabinose into
lactose, galactose, arabinose, and maltose. In xylulose-5′-phosphate, an intermediate in the
such bacteria, mutants unable to synthesize pentose phosphate pathway. The activator pro-
cAMP or CRP are unable to grow on these car- tein is AraC, which must bind to L-arabinose
bon sources. before it can bind to the DNA and activate
Because cAMP–CRP stimulates the transcrip- transcription of the operon. This requirement is
tion of many unrelated genes, it is called a global believed to be due to a conformational change
regulator. The levels of cAMP are lowered by in the activator increasing its affinity for the
glucose uptake in many bacteria. This lowers DNA sequence upon binding of the positive
the transcription of cAMP-dependent genes, effector. As described later, some repressor
explaining glucose repression of genes required molecules must also bind to small effector mol-
for the catabolism of carbon sources other than ecules before they bind to DNA.
glucose. Glucose repression mediated by cAMP
is discussed in Section 17.3.4. (Refer to Sections 2. Negative regulators
19.10.1 and 19.10.2, respectively, for glucose Negative regulators are called repressors. They
repression that does not involve cAMP and for a bind to nucleotide sequences in the DNA called
general discussion of catabolite repression.) operator regions and inhibit transcription
Another example of a global transcription (Fig. 11.7D). For some genes, the operator is
activator is NRI–P, which is part of a two-com- within the promoter region; but for others it is
ponent regulatory system that activates the ntr outside the promoter and may even be upstream
regulon as discussed in Section 19.4.1. NRI–P of the start site. An important negative repres-
binds to enhancer sites that are 100 to 130 base sor is LexA, which is inactivated by proteolytic
pairs upstream from the promoter region of the cleavage during the SOS response, as discussed
target genes and interacts with the RNA poly- in Chapter 16. The SOS response is activated
merase to form the open complex, thus stimu- when DNA is damaged (e.g., by UV radiation).
lating transcription. For this to happen, the Single-stranded DNA that results from such
DNA must bend enough to bring NRI–P to the damage activates RecA, which then activates
polymerase (Fig. 11.7). As shown in Fig. 11.7C, the cleavage of LexA. As a consequence of the
integration host factor (IHF) promotes DNA cleavage of LexA, many genes important for
bending. NRI–P helps the RNA polymerase (σ54 DNA repair are induced.
polymerase) to separate the strands of DNA Another well-known repressor is the lac
around the promoter so that the polymerase repressor, which binds to the operator region
can gain access to the transcription start site. that overlaps the start site of the first gene in the
Several other positive transcriptional activators lac operon. Transcription is inhibited when the
are described in Chapter 19. repressor is bound because the repressor blocks
Another example of a positive transcription RNA polymerase from proceeding into the first
factor that requires the IHF DNA binding pro- gene of the operon. The repressor is inactivated
tein for activity is NarL–P, which stimulates when it binds an isomeric product of lactose
metabolism (allolactose), accounting for the a region called the leader region (trpL), which
induction of the lactose operon by lactose. (The lies between the promoter and the first gene of
allolactose binds to the repressor, causing a the trp operon (trpE). When the tryptophan
conformational change in the repressor protein, concentrations rise, this results in an increased
which lowers its affinity for the DNA. As a con- amount of aminoacylated tRNATrp, which ter-
sequence, the repressor comes off the DNA and minates transcription in trpL. As a consequence,
the lac operon is induced.) Often investigators the operon is not transcribed. How this occurs
use isopropyl-β-D-thiogalactoside (IPTG) as an is described next.
inducer of the lac operon because IPTG is not The model for attenuation of the trp operon
metabolized. is diagrammed in Fig. 11.8. The four regions
Sometimes a repressor must bind to a small in the trpL RNA, numbered 1, 2, 3, and 4, can
molecule before it can bind to the operator. form hairpin loops with one another (i.e., 1:2,
This is the case for biosynthetic operons, oper- 2:3, and 3:4). There are two adjacent trypto-
ons that encode enzymes for the biosynthesis phan codons in trpL, and these are in region
of amino acids and other small molecules. As 1. When the ribosome reaches these codons, it
an example, consider the Trp repressor in E. will stall if there is insufficient aminoacylated
coli, which represses the trp operon. (The trp tRNATrp. If the ribosome stalls at the trp codons
operon encodes the enzymes required to syn- in region 1, then a hairpin loop forms between
thesize tryptophan from chorismic acid.) The the newly synthesized regions 2 and 3. (Region
repressor can bind to the operator only if it 4 has not yet been synthesized.) If, on the other
first binds to tryptophan, which is a corepres- hand, there is sufficient tryptophan to prevent
sor. (In the absence of the corepressor, the inac- the ribosome from stalling in region 1, it moves
tive repressor is called an aporepressor.) In this to the stop codon in region 2. This frees region
way, tryptophan regulates the synthesis of its 3 to pair with region 4 of the RNA when it is
biosynthetic enzymes by feedback repression of synthesized. If regions 3 and 4 form a hairpin
transcription. loop, transcription stops because this hairpin is
a factor-independent termination signal. (See
Some transcription regulators can be Section 11.1.2 for a discussion of factor-inde-
activators or repressors pendent termination.)
Some transcription regulators activate certain
genes and repress others. Several examples
are discussed in Chapter 19. For example, see 11.1.7 Processing of ribosomal and
Section 19.10.1, which discusses the catabolite transfer RNAs
repressor/activator (Cra) system. Ribosomal RNA
To synthesize rRNA, bacteria first make a 30S
Attenuation of the trp operon preribosomal transcript from a single gene. The
Whereas positive and negative regulators of 30S transcript contains the 16S, 23S, and 5S
transcription act at the promoter region to rRNAs, as well as one or more tRNAs, and is
influence the initiation of transcription, attenu- processed according to the diagram in Fig. 11.9.
ation refers to regulation of transcription after There are actually seven sets of rRNA genes in
initiation. In attenuation, transcription is ter- E. coli that yield the 30S preribosomal tran-
minated before the first gene in the operon has script. The genes all code for the same rRNAs
been transcribed. Operons can be regulated by but differ with respect to the number and type
attenuation as well as by positive and negative of tRNAs encoded. Eukaryotes also process
regulation of initiation of transcription. Several rRNA transcripts, as described in note 9.
operons that encode enzymes required for
the biosynthesis of amino acids are known to Transfer RNA
be regulated by attenuation, including the trp Transfer RNAs are extensively processed after
operon in E. coli. (As suggested earlier, regu- transcription in both bacteria and eukaryotes
lation of the trp operon occurs via feedback (Fig. 11.10).10 The initial transcript has seg-
repression by tryptophan.) Downstream of the ments at both the 3′ and 5′ ends that are enzy-
promoter region in the trp operon there exists matically removed. The removal of the 5′ end
www.ebook777.com
Fig. 11.8 Model for attenuation of the trp operon. There are four regions in the leader RNA that precedes
the transcript for the first gene in the trp operon. Region 1 has two trp codons (T). (A) The RNA polymerase
transcribes trpL, just past region 2 and pauses briefly as regions 1 and 2 form a hairpin loop. The translational
start and stop sites are marked, as are the tryptophan codons. (B) While the polymerase is paused, a ribo-
some begins translating the mRNA at the 5′ end. As the ribosome moves, region 1 of the mRNA is drawn
into the ribosome. As translation proceeds, the polymerase begins to move again and transcribes region 3.
(B) Region 1 has been drawn into the ribosome and is being translated. This frees region 2 to pair with the
newly transcribed region 3. If the ribosome stalls at region 1 because of an insufficiency of tryptophan, region
2 remains paired with region 3. This prevents a termination loop between regions 3 and 4 from forming, and
transcription of the trp operon continues. If the ribosome does not stall at region 1, it will move to the end of
trpL, where it stops at the UGA codon in region 2, which is now in the ribosome. While the ribosome occupies
the UGA codon in region 2, it prevents region 2 from pairing with region 3. (C) Meanwhile, the polymerase
continues to move and transcribes region 4. A termination loop forms between regions 3 and 4. Thus, in the
presence of excess tryptophan, transcription of the trp operon is terminated.
segment is catalyzed by RNase P, an endonu- remaining nucleotides to generate the mature

clease, which is present in both prokaryotes and 3′ end. In addition to exo- and endonucleolytic
eukaryotes. RNAse P is a ribonucleoprotein in cleavage, processing may occur at the 3′ end. All
which the RNA has catalytic activity. It is there- eukaryotic tRNAs and some bacterial tRNAs
fore an example of a ribozyme. (The peptidyl lack the CCA-3′ terminus, and this must be
transferase is another example of a ribozyme; added to the molecule after transcription by the
see note 11.) The 3′ end is processed by an endo- enzyme tRNA nucleotidyltransferase. In addi-
nuclease that removes a terminal piece, followed tion, all tRNAs have bases at specific sites that
by RNase D, an exonuclease, which trims the are modified after transcription (Fig. 11.11).
Fig. 11.9 Processing of preribosomal RNA transcripts. E. coli has seven genes that code for a 30S transcript
containing the 16S, 23S, and 5S rRNAs and one or more tRNAs. For example, one of the E. coli 30S tran-
scripts is 5′-16S-tRNAIle-tRNAAla-23S-5S-tRNAAsp-tRNATrp-3′. After methylation of specific bases, the 30S
transcript is cleaved into 17S, 25S, and 5S rRNA transcripts and the tRNAs. Specific nucleases then generate
the 16S rRNA from the 17S transcript and the 23S rRNA from the 25S transcript. Recall that each ribosome
has one 23S, one 16S, and one 5S rRNA. Cleaving the three rRNA transcripts from a single transcript rather
than having the three genes transcribed separately ensures that the rRNA molecules are produced in a 1:1:1
ratio. Not all tRNA genes are present within the rRNA transcript; several of the tRNA genes are encoded
outside the rRNA genes.
removed by exo- amino acids esters

and endonucleases formed at the 3′ end
3′
}
RNAse P
cuts A 3′ A
C C
C C
5′ 5′
A B
arm, present in
some tRNAs
• •• •• •
anticodon anticodon
Fig. 11.10 Processing of transfer RNA. (A) Transfer RNA is made as a larger precursor and processed at both
the 3′ and 5′ ends. An exonuclease called RNase D removes the nucleotides at the 3′ end up to the CCA tri-
nucleotide. Some tRNAs do not have CCA in the original transcript. Thus, after the 3′ end has been trimmed,
tRNA nucleotidyltransferase is used to add CCA-3′-OH. The 5′ end is processed by a ribozyme called RNase
P, which trims the 5′ end to produce the 5′ terminus in mature tRNAs. After the 3′ and 5′ ends have been
trimmed, certain of the bases are enzymatically modified. (See text.) (B) Processed transfer RNA molecules
have between 73 and 93 nucleotides. They consist of a 3′ end that ends in CCA. The amino acid is esterified to
a hydroxyl in the ribose portion of the terminal AMP. Often a short fifth arm exists between two of the arms.
The anticodon specifies incorporation of the appropriate amino acid by pairing with the mRNA.
www.ebook777.com
Fig. 11.11 Some post-transcriptionally modified nucleosides in tRNAs. Inosine differs from adenosine in
having hypoxanthine as its base rather than adenine. The difference is that adenine has an amino group
attached to C6, whereas hypoxanthine has an oxygen. Pseudouridine differs from uridine in having the
ribose attached to C5 rather than N1. Thiouridine differs from uridine in having sulfur attached to C4
rather than oxygen. Dihydroxyuridine differs from uridine in having the ring reduced between C5 and C6.
N6-Isopentenyladenosine differs from adenosine in having an isopentenyl group attached to the amino group
at the C6 position. Methylguanosine differs from guanosine in having a methyl group attached to N1.
11.1.8 Some antibiotic and other chemical eukaryotic RNA polymerase II (synthesizes
inhibitors of transcription mRNA) but not bacterial RNA polymerase.
Transcription in both prokaryotes and eukary-
otes is inhibited by the antibiotic actinomycin
D, which inserts into the DNA between base 11.2 Protein Synthesis
pairs and deforms the DNA. The result is that 11.2.1 Overview of translation
the RNA polymerase cannot continue to move Protein synthesis is called translation because
down the template, and elongation is inhibited. information from one language, based on the
Another inhibitor of elongation is acridine, four-letter alphabet of RNA, is translated into
which behaves in a similar way. A commonly an entirely different language based on an
used antibiotic to inhibit transcription in alphabet of 20 amino acids. More precisely, a
prokaryotes is rifamycin, or the chemically macromolecular machine called the ribosome
modified derivative rifampicin, which is more (Fig. 11.12) systematically moves along a mes-
effective. Rifamycin (and rifampicin) binds to senger RNA and reads the RNA alphabet in 3
a subunit in bacterial RNA polymerases (the β base units (a codon). The ribosome uses this
subunit) and inhibits the initiation of transcrip- information to direct the synthesis of a specific
tion. It is not an inhibitor of RNA synthesis in protein. The process of translation is usually
eukaryotic nuclei. A related antibiotic, which broken down into three formal stages: (1) initia-
also binds to the β subunit of bacterial RNA tion, (2) elongation, and (3) termination.12 Each
polymerase, is streptolydigin. Amanitin inhibits stage represents a very different mechanistic
ribosomal movements. The tRNA entry, pep-

tide transfer reaction, and exiting steps occur
PTC at specialized sites within the ribosome called
E P A GAC
50S the A, P, and E sites, described shortly. The pro-
cess may be subdivided into three major events
that are repeatedly carried out: (1) recruitment
30S of an appropriately charged tRNA bound to
5′ an elongation factor (EF-Tu) to the aminoacyl
3′ (A) tRNA binding site; (2) a chemical reaction
in which a peptide bond is formed between the
mRNA
newly arrived amino group in the A site and the
activated ester in the P site (see Figs. 8.9 and
Fig. 11.12 An assembled ribosome comprises a 70S
particle (see text) built from two smaller subunits of
11.16, step 3); and (3) a ratcheting movement in
30S and 50S. The mRNA is held between the two sub- which the two ribosomal subunits rotate relative
units and passes through a cleft on the 30S subunit to one another. This process is called transloca-
(not seen). The unshaded areas represent the binding tion. It moves the mRNA by one codon length
sites for the exit (E), peptide (P), and aminoacyl (A) toward the 3′ end, slides the spent tRNA into a
tRNAs. The peptide formation reaction occurs at the third “exit” (E) site on the ribosome, and moves
peptidyl transferase center (PTC) and GTP hydroly- the polypeptide bearing tRNA into the P site.
sis reactions occur at the GAC (GTPase activating The A site is thus cleared to repeat the cycle.
center). Source: Adapted from the crystal structure of
Schmeing, T. M., and V. Ramakrishnan. 2009. What
Termination
recent ribosome structures have revealed about the
mechanism of translation. Nature 461:1234–1242. When a stop codon on the mRNA arrives in-
frame at the A site, a protein called a termination
factor enters the site instead of a tRNA. Water
problem in protein synthesis. A summary of enters the site where peptide bond formation
each stage is given, followed by a mechanistic normally occurs, whereupon the polypeptide
description. is hydrolyzed from the final tRNA, releasing it
from the ribosome.
Initiation
The process of initiating translation is one of Removal of the formyl group and of
component assembly. The goal is to construct methionine
a functional ribosome from two large subunits Most completed proteins do not have formyl
that precisely capture an mRNA template at a methionine, or even methionine, at the N ter-
start site near the 5′ end. This assembly must minus. Two enzymes remove these moieties.
also capture an initiator tRNA (tRNAf) carry- Peptide deformylase removes the formyl group,
ing formylmethionine within the P, for peptidyl, and methionine aminopeptidase removes
site on the ribosome. It is important to recog- methionine.
nize that amino acids do not selectively bind to
mRNA, and it is the tRNA that is responsible for Elongation rate
bringing an activated amino acid into reactive For E. coli growing at 37 °C, the polypeptide
site of the ribosome. The charged tRNAf must elongation rate is about 16 amino acids per
be positioned to bind the start site (AUG) near second. The average protein (MW 40,000) has
the 5′ end of the mRNA. It must also position around 364 amino acids. This means that it
the activated formylmethionine in the active takes only about 20 seconds to make an average
site of the ribosome for protein synthesis. The size protein.
process is facilitated by a small set of proteins
called initiation factors, or IFs. Comparison to cytosolic eukaryotic
protein synthesis
Elongation Overall, protein synthesis by cytosolic eukary-
Elongation describes a cycle of charged otic ribosomes is quite similar to protein synthe-
tRNA binding, peptide bond formation, and sis by bacterial ribosomes. The two processes
www.ebook777.com
differ in certain details, however, as described peptidyl transferase center, or PTC. Many of
in note 13. the proteins that facilitate translation also
bind GTP. These cofactors all bind to a simi-
11.2.2 Ribosomes lar region of the ribosome. Hydrolysis of GTP
Ribosomes are ribonucleoprotein particles that always occurs at the GTPase activating center
function as elaborate enzymes that catalyze (GAC). The mRNA passes through a cleft in the
protein synthesis (Section 1.2.6). Rapidly grow- 30S subunit, and each peptide bond synthesis is
ing bacteria contain about 70,000 ribosomes associated with large conformational changes
per cell, all of which are engaged in the synthe- that enforce directional movement of the ribo-
sis of protein. Each ribosome has a diameter of some along the mRNA toward the 3′ end.
approximately 20 nm (200 Å) and a sedimenta-
tion coefficient of 70S. The bacterial ribosome 11.2.3 Charging of the tRNA
has two subunits identified by their sedimenta- (making the aminoacyl–tRNA)
tion coefficients. These are the 50S subunit and Overview
the 30S subunit. (See Section 1.1.1 and Table Prior to becoming involved in protein synthe-
1.2 for a comparison with archaeal and eukary- sis on the ribosome, each amino acid must be
otic ribosomes.) Together, the 50S and 30S sub- covalently linked to a tRNA. The tRNA esters
units form the 70S ribosome, which makes the of amino acids have a high group transfer
proteins. potential with respect to amino acid hydroly-
The 30S subunit has 21 different proteins and sis or transfer, and thus charging the tRNAs is
one 16S rRNA molecule; its proteins are named equivalent to activation. Each of the 20 differ-
S1, S2, and so on, with S standing for “small ent amino acids must be activated in this way,
subunit.” The 16S rRNA helps to align the and the process is catalyzed by 20 different
mRNA so that it is positioned correctly on the enzymes called aminoacyl–tRNA synthetases.14
ribosome for the initiation of protein synthesis. The number of tRNAs can be more than 20
The 50S subunit has 31 different proteins, one because the code is degenerate; that is, there
23S rRNA molecule, and one 5S rRNA mole- exists more than one codon for most amino
cule. Its proteins (L1, L2, etc.) are named for the acids. For example, E. coli can have from 1 to 6
large subunit. The 23S rRNA serves as a struc- codons for each of its amino acids, resulting in
tural scaffold for peptide bond formation and it 61 codons. (However, there are fewer than 61
interacts with the tRNA substrates. different tRNAs in E. coli. The reason for this is
explained in note 15).
Key sites within the ribosome The aminoacyl–tRNA synthetases recognize
Recent crystallographic structures and cryo- both the amino acid and certain structural fea-
electron microscopic reconstructions of ribo- tures of the tRNA. Some synthetases recognize
somal subunits and the assembled ribosome the anticodon, whereas others recognize nucle-
provide a powerful structural framework to otide sequences in other parts of the tRNA. The
help understand decades of genetic and bio- process of attaching the amino acid to the tRNA
chemical studies on the ribosome. Figure 11.12 is called charging the tRNA; it takes place in the
depicts a ribosome drawn roughly to scale that cytosol, and the product is called aminoacyl–
identifies the most important locations. A clear tRNA. Of particular importance is the require-
view of this layout will help in constructing an ment that the amino acid be attached to the right
understanding of the mechanism of protein tRNA: it is the tRNA that recognizes the codon,
synthesis directed by coded information inher- and if the wrong amino acid is attached, then a
ent to the mRNA substrate. Note that three protein with the wrong sequence of amino acids
distinct binding sites for tRNAs have been will be made.
identified, labeled E (exit), P (peptidyl), and A The 20 tRNA synthetases may be divided into
(aminoacyl). The base of each site is adjacent two classes, each based on a distinct protein
to the mRNA to facilitate continuous contact fold. The class I synthetases are distinguished by
between each tRNA and its cognate mRNA three common features: they initially acylate the
during translation. The reaction center where 2′-OH group of their cognate tRNAs, they tend
peptide bond formation occurs is called the to be monomeric structures, and they all bind
to the same face of the generic tRNA structure. specificity has little to do with the anticodon but
The amino acids they react with tend to be the is determined by other parts of the tRNA mol-
larger, more hydrophobic structures, includ- ecule. The synthesis of the aminoacyl–tRNA
ing val, leu, ile, met, trp, tyr, glu, gln, cys, and takes place via a route similar to the synthesis of
arg. The amino acids are subsequently isomer- acyl–CoA derivatives, that is, via an acyl–AMP
ized to the 3′-OH by the synthetase. The class II intermediate. The first reaction is an attack of
enzymes directly acylate the 3′-hydroxyl of the an oxygen atom of the carboxyl group of the
terminal ribose and they bind to a distinct face amino acid on the α-phosphate of ATP displac-
of the tRNA structure opposite that occupied ing pyrophosphate from the ATP and forming
by the class I synthetases. Class II enzymes are an aminoacyl–AMP derivative. Then an oxygen
multimeric (dimeric or tetrameric) in structure atom of a hydroxyl group on the ribose of the
and charge tRNAs with the smaller, generally tRNA attacks the aminoacyl–AMP, displacing
less hydrophobic amino acids (gly, ser, thr, asn, the AMP and forming the aminoacyl–tRNA
asp, his, lys, pro, ala, and phe). derivative. The bond in the final product is an
ester linkage between the carboxyl group of
Transfer RNA the amino acid and the 3′-hydroxyl group on
Transfer RNAs are small single-stranded RNAs the ribose portion of the tRNA.14 (See note 19
that are folded with three hairpin loops resem- for a further description of this reaction.) As
bling a cloverleaf (Fig. 11.10). The amino acid explained in Chapter 8, the reaction is highly
is ultimately esterified via its carboxyl group to favorable energetically because pyrophos-
the 3′-OH in the ribose part of the adenylyic acid phatase hydrolyzes the pyrophosphate prod-
residue at the 3′ end, which is 5′-CCA-OH-3′. uct to two phosphate molecules. (See Fig. 8.9.)
One of the loops presents the anticodon triplet, Therefore, the equivalent of 2 ATPs is required
which hybridizes to the codon in the RNA tem- to make one aminoacyl–tRNA.
plate and thus brings the correct amino acid to
the growing polypeptide chain. All tRNAs have Proofreading mischarged tRNAs
bases that are modified after transcription dur- Most aminoacyl tRNA synthetases are highly
ing the processing of the tRNA, and these bases effective at matching the correct tRNA with
appear in or near the hairpin loops. The struc- the cognate amino acid. The error rate for mis-
tures of some of these modified bases are shown incorporation is on the order of 10−4 to 10−5.
in Fig. 11.11. However, some enzymes have difficulty dis-
criminating among amino acids that are closely
Attaching the amino acid to the tRNA related structurally. For example, threonyl
(charging the tRNA) tRNA synthetase must recognize threonine
The charging of the tRNA takes place in two and discriminate against the closely related
steps, both of which are catalyzed by the same amino acids serine and valine. Serine lacks the
enzyme (Fig. 11.13). The class of enzymes is β-methyl group of threonine, and valine has a
called aminoacyl–tRNA synthetases, and each methyl group in place of the hydroxyl of threo-
individual enzyme recognizes a specific amino nine. Serine is incorporated at a rate of 10−2 to
acid and its corresponding tRNA. (There are 10−3. To remove such mischarging, the threonyl
some examples of the same aminoacyl syn- tRNA synthetase enzyme possesses an editing
thetase charging two different tRNAs, followed site approximately 20 Å from the charging site.
by the appropriate modification of the amino The flexible 3′ end of the tRNA is capable of
acid in one of the tRNAs.16 There are also reaching both the active site and editing site by
examples of two different aminoacyl–tRNA a conformational change. Thus, without tRNA
synthetases activating the same amino acid and dissociation from the enzyme, the charged
charging the same tRNA.17 All these cases are amino acid may swing to the editing site. The
summarized in note 18.) second, editing, site discriminates against thre-
Some aminoacyl–tRNA synthetases recog- onine and acts as a hydrolase to remove the mis-
nize simply the anticodon region of the tRNA, charged amino acids. The tRNA remains bound
which determines the specificity of the synthetase to the synthetase, and the charging attempt is
for the tRNA, whereas for other synthetases the repeated.
www.ebook777.com
Fig. 11.13 The aminoacyl–synthetase reaction. (A) The carboxyl group of the amino acid attacks the
α-phosphate in ATP and displaces pyrophosphate to form the AMP derivative of the amino acid. (B) A hydroxyl
on the ribose in the terminal adenylate residue attacks the carbonyl group of the amino acid and displaces
the AMP. (C) The aminoacyl~tRNA. The amino acid is attached to the 3′-hydroxyl of the terminal adenylate
residue in the tRNA prior to formation of the peptide bond on the ribosome. (D) Transesterification reaction.
Some synthetases catalyze an attack by the 2′-hydroxyl of the terminal adenylate residue in the tRNA. If this
occurs, the synthetase transfers the amino acid by a transesterification reaction to the 3′-hydroxyl. If the syn-
thetase catalyzes an attack by the 3′-hydroxyl, a transesterification step is not necessary.
Making fMet–tRNAfMet that attaches methionine to the tRNA (tRNAMet)

The first amino acid in the growing polypeptide responsible for inserting methionine in internal
chain in bacteria is formylmethionine (fMet), positions in the protein. Then a transformylase
and protein synthesis begins with the synthesis (tRNA methionyl transformylase) binds to the
in the cytosol of fMet–tRNAfMet. The formyl methionine on the Met–tRNAfMet and cata-
group is typically removed after the protein has lyzes the transfer of a formyl group from N10-
been synthesized. Sometimes the methionine is formyltetrahydrofolate (10-formyl-THF) to
also removed. The fMet–tRNAfMet is made in the the Met–tRNAfMet to form fMet–tRNAfMet. (See
cytosol when methionine is attached to a special Fig. 10.14 for a description of THF as a C1 car-
initiator tRNA (tRNAfMet). The reaction is cata- rier.) The transformylase recognizes not only
lyzed by the same aminoacyl–tRNA synthetase the methionine but the tRNAfMet as well, and
is able to distinguish between tRNAfMet and cooperative manner whose order is less clear.
tRNAMet. The stage is now set for the initiation The mRNA binding depends upon a rapid
of protein synthesis on the ribosome. Note that association with single-stranded RNA and a
the formylation of methionine yields an amide slower phase dependent upon the presence of
bond analogous to those present during synthe- the fMet–tRNA, the IF-2/GTP complex and the
sis so that the first amino acid may be placed in Shine–Dalgarno sequence (discussed in its own
the P (peptidyl) site in the ribosome. section). High concentrations of IF-2/GTP can
circumvent the need for the Shine–Dalgarno
sequence, supporting a role for IF-2/GTP as a
11.2.4 Initiation chaperone for the mRNA.20 Note the location
Making the preinitiation complex of IF-3 binding, which allows it a role in sensing
A preinitiation complex is formed with the the correct codon–anticodon pairing within the
30S ribosomal subunit, messenger RNA, P site. The IF-1 occupies what will become the
formylmethionyl–tRNAfMet (fMet–tRNAfMet), A site in the assembled ribosome and interacts
and initiation factors IF-1, IF-2/GTP, and IF-3 with IF-2/GTP. Together they help position the
(Fig. 11.14). First IF-3 binds to the 30S ribo- fMet–tRNAfMet in the region that will become
somal subunit, preventing premature asso- the P site. When correctly assembled, the
ciation between the 30S and 50S subunits. fMet–tRNA binds to the start codon (AUG),
The other players appear to bind in a highly via the anticodon on the fMet–tRNA (UAC).
fMet
IF2 GTP
P site IF1, IF2-GTP IF1

UAC
IF3 IF3
AUG AUG
5′ fMet-tRNA
5′
30S 3′ 30S 3′
mRNA mRNA
50S
P site
50S
fMet fMet
IF2 GTP
70S IF1, IF2-GDP IF1
UAC UAC
IF3
AUG AUG
5′ Pi 5′
3′ 30S 3′
mRNA mRNA
Fig. 11.14 Assembly of the 70S ribosome: translation initiation. The dark line indicates the mRNA to be
translated, and the 30S and 50S are designated. Assembly requires three protein initiation factors, (IF-1–IF-3).
Early events include binding of the mRNA and IF3 to the 30S subunit. This is followed by binding of the
charged fMet–tRNA in the P (peptide) site with the anticodon (AUC) interacting with the AUG initiator
within the mRNA. IF-1 and a complex of IF-2 and GTP bind to create the substrate for 50S recruitment.
Correct assembly is associated with GTP hydrolysis and loss of IF-1 and IF-2 (with GDP bound). The com-
pleted structure is primed for elongation and now has a sedimentation coefficient of 70S.
www.ebook777.com
Fig. 11.15 Binding of initiator region of mRNA to the 16S rRNA. Upstream from the initiation codon in the
mRNA there is a short sequence of nucleotides called the Shine–Dalgarno sequence. The Shine–Dalgarno
sequence hybridizes to a complementary sequence of nucleotides at the 3′ end of the 16S rRNA, and this places
the initiation codon at the P site, where the fMet–tRNAfMet binds. Shine–Dalgarno sequences vary, and they
recognize 16S rRNA with differing affinity. Some bacterial genes do not have a Shine–Dalgarno sequence;
that is, the initiation codon is at the very end of the mRNA, as revealed by sequencing the N-terminal region
of the protein and comparing it with the nucleotide sequence at the 5′ end of the mRNA.
The assembly of initiation factors, mRNA, and 11.2.5 Chain elongation

charged tRNA provides a binding surface to Start codon
engage the 50S subunit. Translation usually begins at an AUG codon
(the start codon), which codes for methionine in
Assembling the 70S initiation complex the initiator region of the mRNA. In a relatively
The 70S initiation complex is formed when the small number of cases the fMet is encoded by the
50S subunit is added to the complex. This brings valine codon (GUG) or by the leucine codons
the IF-2/GTP complex into the GTPase activat- (CUG, UUG) if they are at the start site. These
ing center (GAC, Fig. 11.12). The translation- codons encode valine or leucine, respectively, if
ready 70S complex is achieved as IF-1 IF-2, they are downstream from the start site. There
IF-3, GDP, and Pi are released following GTP may be internal (downstream) start codons, but
hydrolysis on IF-2. It is very important that the these do not serve as translation initiation sites.
mRNA bind to the 30S subunit in the correct
position to ensure that the start codon is at the Leader sequence
P site. As described next, for many mRNAs, it is RNA transcription may begin some distance
the Shine–Dalgarno sequence that positions the upstream of the translational start codon. Under
mRNA on the ribosome. these circumstances an untranslated region at
the 5′ end of the mRNA exists. It is called the
Shine–Dalgarno sequence leader sequence.
The Shine–Dalgarno sequence is a string of about
5 to 10 nucleotides approximately 6 to 10 nucle- Binding of aminoacylated tRNA to the A site
otides upstream of the start codon. It is really a Chain elongation is initiated when an incom-
family of closely related sequences whose iden- ing aminoacylated tRNA enters the A site on
tity depends on the specific mRNA. The Shine– the ribosome. The anticodon of the tRNA must
Dalgarno orientation sequences each base pair correctly base pair with the codon in the mRNA
with varying complementarity and affinity to a (Fig. 11.16, step 1). The process is chaperoned
fixed sequence close to the 3′ end of the 16S rRNA. by elongation factor Tu (EF-Tu), which binds
In many mRNAs it defines the ribosome-binding the charged tRNA and helps protect the sensi-
site (Fig. 11.15). In cooperation with the IF-2/ tive ester bond from hydrolysis. The correct
GTP complex, it helps orient the start codon of positioning of the incoming aminoacyl–tRNA
the mRNA adjacent to the P site in the ribosome. is facilitated EF-Tu–GTP, which severely bends
This is where fMet–tRNAfMet will bind. Not all the charged tRNA and prevents its productive
mRNAs have a Shine–Dalgarno sequence, and in association in the A site. The EF-Tu–GTP also
fact the start codon may be at the extreme 5′ end associates with the GTPase activating center
of the mRNA. Under these circumstances, the (GAC in Fig. 11.16), but GTP hydrolysis occurs
start codon must be free of involvement in sec- only when the correct codon–anticodon pair-
ondary structure elements, and its delivery to the ing is achieved. Hydrolysis is associated with
correct site becomes dependent upon IF-2/GTP. dissociation of EF-Tu–GDP and delivery of the
PTC
fMet fMet aa2
Pi GAC
E P An+1 EF-Tu 2 E P A
n GTP n
n+1
EF-Tu EF-Tu
n+1
GTP 5′ 5′
GDP
3′ 3′
1 mRNA
3 fMet
fMet PTC aa2
*
GAC EF-Tu n+1
GAC
E P A n+1 En P A
n
GTP
1
6
5′ EF-G
GDP 5′
3′ 4 3′
mRNA mRNA
fMet fMet
n n+1 aa2 aa2 GTP
EF-G
GAC Pi n+1 GAC
GTP
E P A GDP En P AEF-G
5
e pa trans- e pa
5′ location 5′
3′ 3′
mRNA mRNA
Fig. 11.16 The elongation phase of translation. The first cycle of peptide bond formation is shown, beginning
with the initiator tRNA–fMet positioned in the P (peptidyl) site in the ribosome (center left). The subsequent
charged tRNA (n + 1) is chaperoned to the complex by EF-Tu (step 1). GTP hydrolysis and EF-Tu release
at the GTPase activating center (GAC) follow recognition of the mRNA by the anticodon sequence of the
charged tRNA. A large conformational change then allows correct positioning of the n + 1 tRNA in the A
(aminoacyl) site (step 2). Peptide bond formation occurs in the PTC (peptidyl transferase center) and is accom-
panied by the formation of a hybrid state where the tRNA–peptide complex slides into the P site (step 3).
Translocation is initiated by the association of EF-G with GTP bound (step 4). GTP hydrolysis leads to a
conformational change that advances the mRNA one codon unit through the ribosome, depicted by arrows
(step 5). Both tRNAs remain associated with the mRNA. The uncharged tRNA now occupies the E (exit) site,
and the peptide-bearing tRNA occupies the P site. Dissociation of the EF-G/GDP complex opens the A site to
allow the next charged tRNA, chaperoned by EF-Tu–GTP, to enter the A site (step 6). The asterisk serves as a
reminder that the fMet tRNA is no longer occupying the P site and that the E site is filled.
charged tRNA fully into the A site by reliev- This reaction is analogous to the role of GEFs
ing the bend. The amino acid ester substrate is (guanine exchange factors) in more complex
thus correctly positioned in the peptidyl trans- organisms.
ferase center (PTC), ready for the bond-forming
reaction. Formation of the peptide bond
A peptide bond forms when the free α-amino
Recycling of EF-Tu–GTP group on the amino acid bound to the tRNA
After GTP is hydrolyzed and EF-Tu–GDP is in the A site displaces the tRNAf in the P site
released from the ribosome, the GDP dissoci- (Fig. 11.16, step 3). The mechanism of cataly-
ates to facilitate recharging with GTP. The sis, based on the crystallographic analysis of the
process is catalyzed by a small dimeric protein intact ribosome with a transition-state mimic
called EF-Ts that stimulates the exchange of in a catalytic site, appears to be solely entropic.
GDP for GTP. EF-Ts interaction disrupts the In other words, the ribosome simply brings
Mg2+ binding site on EF-Tu that is essential for together the substrates in close proximity and in
nucleotide binding, thus releasing EF-Tu–GDP. the correct orientation for catalysis to occur. No
www.ebook777.com
protein subunit comes within 18 Å of the cata- directional ratchet that moves back and forth
lytic center, and catalytic roles for specific bases with each amino acid addition.
within ribosomal RNAs have been disproved.
The formation of the peptide bond as the 11.2.6 Termination
ester linkage is broken between the amino acid Release factors
and the tRNA does not require any additional Chain termination is initiated when a stop codon
energy input. Recall that ATP was used to form reaches the A site (Fig. 11.17). The stop codons
the reactive ester linkage in the first place. Thus, are UAA, UGA, and UAG. There are no tRNAs
the energy to make the peptide bond is derived with anticodons for these codons, and because
from ATP. In fact, the formation of the peptide of this no aminoacyl–tRNA enters the A site
bond uses the equivalent of two ATPs, since the when a stop codon is present. Release factors
formation of the aminoacyl–tRNA is driven to (proteins) bind to the stop codons and cause the
completion by the hydrolysis of pyrophosphate. release of the completed polypeptide from the
The result is the creation of a molecule with a ribosome. There are three release factors in E.
high group transfer potential with respect to coli: RF-1, RF-2, and RF-3. The first release fac-
amino acid transfer, the aminoacyl tRNA. tor binds to UAA and UAG; RF-2 binds to UAA
and UGA, and RF-3/GTP facilitates clearing the
Translocation release factors and tRNAs from the ribosome.
The formation of the peptide bond is followed The steps in termination and ribosome dis-
by translocation (Fig. 11.16, steps 4 and 5). assembly are shown in Fig. 11.17. Binding of a
The elongation factor EF-G/GTP (translocase) release factor (RF-1 or RF-2) occurs in the A site
is required for translocation. It partially occu- and allows water access to the peptidyl trans-
pies the A site, and the bound GTP is positioned ferase center. The completed peptide is hydro-
at the GTPase activating center. Hydrolysis of lyzed from the terminal tRNA. The complex
GTP to GDP and Pi is associated with a large recruits RF-3/GTP, which interacts with RF-1/2
conformational change in EF-G itself. This and positions the GTP at the GTPase activat-
is coupled to movement of the mRNA by one ing center. GTP hydrolysis leads to dissociation
codon length through the ribosome. Carefully of RF-1 or RF-2 and the terminal tRNA. The
note that the two tRNAs present stay associated ribosome is recycled by the action of two other
with the mRNA and mark the reading frame at proteins: RRF (ribosome recycling factor) and
all times. The uncharged tRNA that was in the P EF-G/GTP. Recall that EF-G/GTP also has an
site now moves to the E (exit) site, maintaining important role in the specific step of translo-
its interaction with the corresponding codon cation. Now GTP hydrolysis, in the presence
in the mRNA. The E site accommodates only of RRF, leads to ribosome disassembly. The
uncharged tRNAs, and this has been postulated subunits are recycled into further rounds of
to help drive the reaction by capturing one of translation.
the products of the reaction. The tRNA bear-
ing the dipeptide (and later the polypeptide) is The requirement for GTP
now forced over to the P site, again maintain- Guanosine triphosphate is an essential cofac-
ing its contact with the corresponding codon tor for each of the phases of translation. In
in the mRNA. The EF-G/GDP complex now contrast to ATP and other molecules with high
dissociates. This opens the A site to allow the group transfer potentials, the role of GTP is not
next charged tRNA to enter. Note that in the to drive the formation of covalent bonds or to
following cycles of elongation, the placeholder phosphorylate proteins. Instead, it often acts as
tRNA in the E site is expelled by the action of a conformational switch or trigger. The GTP
translocation. bound form of the individual enzymes is associ-
The movement of the ribosome itself during ated with a protein conformation distinct from
translocation has been clarified by the numer- that of the GDP-bound form. Each conforma-
ous structural studies of translation (reviewed tion is associated with a specific action or event.
by Ramakrishnan21). The two ribosome sub- During translation, for example, GTP hydrolysis
units move relative to each other on an interface is catalyzed by protein occupation in the GAC,
that is largely RNA, not protein. They act as a or GTPase activating center, in the ribosome.
C-term. n n+1
GAC GTP
GAC
E P A E P A RF3 GDP
RF3
RF1/2 U U
G GTP G RF1/2
A A
5′ RF3 5′
3′ 3′
H2O mRNA mRNA
C-term.
GAC
E P A GAC E P A
U
G
A
5′ 50S GAC EF-G
GTP 5′
3′ 3′
mRNA RRF mRNA
GAC
E PRRFA GTP
EF-G
5′ 30S
5′
3′
mRNA 3′
mRNA
Fig. 11.17 Termination of translation. The ribosome at the center-left position has reached a stop codon
(UGA shown) at the A site. A deacylated tRNA lies in the E site, and the polypeptide-bearing terminal tRNA
lies in the P site. A release factor (RF-1 or 2, shown as RF-1/2) is recruited to the A site, which also allows
water into the active site. Hydrolysis of the peptidyl–tRNA structure leads to protein release (dark gray line).
The ribosome associates with release factor 3 (RF-3) with GTP bound to the GTPase activating center (GAC).
Hydrolysis of GTP is associated with a movement that drives the penultimate tRNA from the E site and the
ultimate tRNA into the E site. The 70S complex is dissociated when another release factor (RRF) binds and
recruits EF-G with GTP bound. GTP hydrolysis produces a conformational change that leads to complex
disassembly. This allows the cycle to begin anew on another mRNA.
During initiation, GTP hydrolysis occurs when on RF-3 during the termination triggers a con-
the 70S ribosome is correctly assembled, causing formational change that leads to dissociation of
a conformational change that leads to dissocia- release factors and uncharged tRNA. Thus the
tion of the initiation factors. During elongation, nucleotide bound status (GTP or GDP) deter-
EF-Tu with GTP bound chaperones the appro- mines the conformational status of the pro-
priate charged tRNA to the A site. Hydrolysis tein, enabling the protein to selectively bind or
of GTP validates entry of the correct charged release interacting molecules.
tRNA and triggers the release of EF-Tu. During
the crucial step of translocation, GTP hydroly-
sis produces a conformational change in EF-G 11.2.7 Alternative roles for
that advances the mRNA and tRNAs in the aminoacyl tRNAs
ribosome. EF-G has an analogous role later in The aminoacyl tRNAs are carboxylate esters of
ribosome disassembly. Last, hydrolysis of GFP amino acids with the 3′-hydroxyl of the terminal
www.ebook777.com
adenosyl ribose. The energy from two ATPs was 11.2.8 Polysomes: Coupled transcription
invested in charging the tRNA, and the molecules and translation
themselves have a relatively high group transfer Messenger RNAs are translated by several ribo-
potential, where the transferred species is the somes (typically around 50) in series, ensuring
amino acid. Many organisms make use of these that several copies of the protein can be made
molecules in pathways outside of protein biosyn- from a single mRNA. The complex of multiple
thesis. For example, in Staphylococcus aureus, ribosomes attached to an mRNA is called a
lysylphophatidylglycerol synthase (LysPGS) polysome. This arrangement increases the rate
transfers lysine from Lys–tRNALys to phos- of protein synthesis, since the amount of protein
phatidylglycerol (PG) in the cytoplasmic mem- made per unit time from a particular mRNA is
brane. This serves as a strategy for modulating proportional to the number of ribosomes trans-
the surface charge of the bacterial envelope. The lating the mRNA.
cationic surface is associated with resistance to Furthermore, translation and transcrip-
cationic antibacterial peptides. The presence tion are coupled in bacteria. What happens
of lysylphophatidylglycerol synthase is also is that ribosomes attach to the 5′ end of the
associated with virulence in other pathogenic mRNA even before the message is completely
organisms.22 transcribed; then they move along the mRNA
Another striking role for aminoacyl tRNAs toward the 3′ end as transcription continues
is in modulating protein stability. The identity (Fig. 11.18). Because translation is coupled to
of the amino acid at the N-terminus of a protein transcription, proteins can begin to be made
is a determinant of its half-life in the cell, which even before the mRNA is completed, thus short-
is called the end rule (see note 23). An enzyme ening the delay between the onset of transcrip-
called leucylphenylalanyltransferase transfers tion and the appearance of the protein. Coupled
the destabilizing hydrophobic amino acids transcription and translation cannot take place
leucine and phenylalanine to the N-termini of in eukaryotes, since transcription and transla-
proteins using the corresponding aminoacyl tion occur in separate cell compartments (i.e.,
tRNA molecules as substrates. This marks the the nucleus and cytoplasm, respectively).
modified proteins for ATP-dependent deg-
radation by the ClpAP–serine protease com-
plex. The system is roughly analogous to the 11.2.9 Polycistronic messages
ubiquitinylation system of marking proteins for Bacteria frequently make polycistronic mes-
degradation in eukaryotes by targeting them to sages. In this kind of messaging, which
the proteasome. eukaryotes do not engage in, several genes are
Fig. 11.18 Coupled transcription and translation. In bacteria, protein synthesis begins before the mRNA is
completed. The result is a string of ribosomes on the same mRNA. There may be up to 100 ribosomes on the
same mRNA, comprising what is called a polysome.
Fig. 11.19 Polycistronic mRNA. The polycistronic mRNA has more than one gene. At the beginning of each
gene there is a ribosome-binding site (RBS) at which initiation begins. The ribosome then begins translating
at the initiation codon and stops at the stop codon at the end of the gene. Upon termination, the ribosome and
the polypeptide are released, and initiation begins anew at the beginning of the next gene. Between the genes
there is an intercistronic spacer region, which may have as few as one or as many as 30 to 40 bases. The amino
and carboxy terminals of the polypeptide are labeled N and C, respectively.
transcribed in series into a single mRNA mol- happen if, for example, a mutation results in a
ecule. Initiation of transcription of the cotrans- nonsense codon (i.e., a codon that does not bind
cribed genes is from a single promoter upstream to a tRNA) in the upstream gene and there is a
of the gene cluster, and the cluster of genes is Rho-dependent transcription termination site
called an operon. Usually genes that encode in the upstream gene after the translational stop
proteins that function in a common pathway site. Under these circumstances the RNA poly-
are in the same operon. Operons in bacteria are merase will disengage from the DNA and tran-
the rule, and they may have as few as 2 genes or scription will stop before the polymerase has
as many as 10 or more. reached the downstream gene. For a discussion
For individual proteins to be made from a of Rho-dependent termination of transcription,
polycistronic message, there must be transla- see Section 11.1.1.
tional start sites at the beginning of each gene
transcript and translational stop sites at the 11.2.11 Coupled translation
end (Fig. 11.19). Furthermore, there must be In some instances translation can be coupled.
initiating regions (i.e., ribosome-binding sites) This means that in a polycistronic mRNA,
at the beginning of each gene transcript. Thus, translation of the immediate upstream gene
the translation initiation complex forms at the is required for translation of the downstream
beginning of each gene transcript in the poly- gene. For example, the translation of the genes
cistronic message, and the ribosome, mRNA, in the operons coding for ribosomal proteins in
tRNA, and polypeptide dissociate at the termi- E. coli is coupled. This affords the opportunity
nation site of each gene transcript in the polycis- to regulate the translation of the entire operon
tronic message. by regulating the translation of one of the genes.
(See note 22 in Chapter 2 for a further discussion
11.2.10 Polarity of the regulation of translation of the ribosomal
Polarity refers to the phenomenon in poly- protein operons.) A model for translational
cistronic mRNAs in which a block in the coupling postulates that the translational start
translation of an upstream gene can prevent region, including the start codon, for the down-
transcription of a downstream gene. This can stream gene is inside a hairpin loop in the mRNA
www.ebook777.com
so that it cannot bind to a ribosome. When the associate with the newly synthesized proteins
ribosome translating the upstream gene comes and assist in the folding.
to the stop codon of the upstream gene, the ribo- One role of chaperone proteins is to prevent
some may disrupt the secondary structure of the misfiling and aggregation. Aggregation will
mRNA, making the translational start region occur if hydrophobic regions of the newly syn-
for the downstream gene accessible to a second thesized, partially unfolded proteins remain at
ribosome (Fig. 11.20). the aqueous interface, where protein–protein
interactions can occur, rather than moving into a
11.2.12 Folding of newly synthesized hydrophobic core during folding. (Electrostatic
proteins: The role of chaperone proteins repulsion between proteins with a net charge
Chaperone proteins are proteins that temporar- decreases the propensity for aggregation.) It has
ily bind to other proteins and, depending upon been estimated that at least 50% of newly syn-
the chaperone protein, aid in proper folding of thesized proteins are aided in their folding by
newly made proteins, transport of proteins to chaperone proteins that ensure rapid folding.
and across membranes (Sections 16.3.6 and Some proteins, however, fold independently of
18.1.2), and refolding of proteins that have chaperone proteins. These are probably rapidly
become unfolded as a consequence of stress folding small proteins.
such as heat shock. For reviews, the student In addition to helping newly synthesized pro-
should consult refs 24 and 25. teins rapidly fold, there are chaperone proteins
that prevent the N-terminal portions of nascent
The role of chaperone proteins in folding polypeptides, emerging from the ribosomes,
newly synthesized proteins from misfolding, and from aggregating with
The information to fold into the active tertiary identical nascent chains being synthesized on
structure is contained in the sequence of amino neighboring ribosomes in a polysome.
acids, that is, the primary structure of the pro-
tein. (See note 26 for definitions of primary, sec- Measuring the flux of proteins through
ondary, tertiary, and quaternary structures of the chaperonins
proteins.) However, the proper folding of many One way to quantitate the flux of newly synthe-
newly synthesized proteins cannot proceed sized proteins through the chaperone proteins is
without chaperone proteins, which transiently to incubate the bacteria for a few seconds with
a radioactive amino acid and then add a large
excess of the same amino acid in nonradioactive
form. The addition of the nonradioactive amino
acid lowers the specific activity of the radioactive
amino acid sufficiently to halt further incorpora-
tion of radioactivity. This is called “pulse-chase”
measurement. A specific chaperone protein can
be immunoprecipitated from cell extracts at var-
ious times during the “chase,” and the different
radioactive proteins transiently associated with
the chaperone protein, as well as the transit time,
can be visualized by autoradiography after gel
electrophoresis. The flux of proteins through the
chaperone protein can be estimated by measur-
ing the radioactivity in individual protein bands
Fig. 11.20 Model for translational coupling. (A) It separated by gel electrophoresis, as well as the
is postulated that in a polycistronic mRNA, the start total amount of radioactivity that was immuno-
codon for the downstream gene is in a hairpin loop, precipitated during the “chase.”
which thus it is not accessible to the ribosome. (B)
When the ribosome translating the upstream gene
arrives at the stop codon, it disrupts the secondary GroEL/GroES, the GroE system
structure of the mRNA, giving a second ribosome In experiments such as the ones just described,
access to the initiation site in the downstream gene. E. coli and anti-GroEL antibody were used to
monitor the flux of proteins through the GroEL releases them in an unfolded form. Then, the
chaperonin, which cooperates with GroES as actual folding of the completed polypeptide
a complex, called the GroE complex, to pro- takes place when it is transferred to a cavity in
duce proper folding.27 (Chaperone proteins GroEL. GroES is a protein that caps the cavity.
such as GroEL and GroES exist as multisub- (See ref. 31 for a review.)
unit complexes called chaperonins. See note 28 As described next, another chaperone pro-
for a more complete description of GroEL and tein, called trigger factor (encoded by the tig
GroES.) The major conclusions from the pulse- gene), is also ribosome bound and attaches to
chase experiment were as follows. nascent polypeptides as they emerge from the
ribosome, perhaps prior to DnaK. Trigger fac-
1. A variety of newly synthesized proteins (as
tor can replace DnaK in dnaK Δ mutants, and
judged by the number of bands that migrated
DnaK can replace trigger factor in tig Δ mutants.
during gel electrophoresis) transiently asso-
Thus they have overlapping functions.
ciate with GroEL, and this can account for
10 to 15% of the protein during normal
growth temperatures. Trigger factor
2. Upon exposure to heat stress, during which A ribosome-associated protein called trigger
the relative amounts of GroEL and other factor (TF) has been implicated in the folding of
chaperonins increase in the cell, at least certain nascent proteins as they emerge from the
30% of the newly synthesized proteins ribosome, and it may also take part in transfer-
associate with GroEL. (See the discussion ring nascent proteins to the GroEL complex.32
of chaperonins and heat-shock proteins in Trigger factor is believed to be functionally
Section 16.3.6.) redundant with DnaK because mutants of E.
It appears that GroEL binds to the protein coli that are heat sensitive and do not make trig-
after synthesis is complete. This conclusion is ger factors have no defects in growth at interme-
justified by noting that when the polypeptides diate temperatures or in protein folding as long
bound to GroEL were examined, to determine as the bacteria make DnaK.33 Likewise, mutants
their size distribution, they were found to be of that do not make DnaK are fine as long as they
distinct sizes, rather than lying along a contin- make trigger factor. In the absence of trigger
uum, as might be expected if nascent polypep- factor, there is a two- to threefold increase in
tide chains bound to the chaperonin. However, the amount of nascent polypeptide chains that
although GroEL becomes associated with poly- bind to DnaK.
peptides after their release from ribosomes, it In addition to its chaperone activity, trigger
appears that three other chaperone proteins, factor is a peptidyl–prolyl cis/trans isomerase.
DnaK, DnaJ, and GrpE, associate with nascent It is attached to the ribosomal 50S subunit of
polypeptides while on the ribosome. bacterial ribosomes. It binds to nascent proteins
on the ribosomes and promotes proline-limited
folding of nascent proteins, probably cotrans-
A model: First DnaK–DnaJ–GrpE, lationally. Trigger factor catalyzes the cis–trans
then GroE isomerization of the peptide bond that is on the
There is evidence that a chaperone complex N-terminal side of proline residues.
of DnaK, DnaJ, and GrpE can interact with a Normally the hydrogen attached to the sub-
least some nascent polypeptides on ribosomes stituted amino group in peptide bonds is oppo-
and that GroE (GroEL and GroES) binds to site (trans) to the oxygen of the carbonyl group.
the polypeptides after their release from ribo- This is the most stable configuration. However,
somes.29 (See note 30 for a description of the when the nitrogen atom is contributed by pro-
experiments that support this conclusion, and line, the peptide linkage is occasionally cis (in
see Section 16.3.6 for the role of these proteins, about 5% of cases) in native proteins. Probably
as well as the GroE system, in the heat-shock a certain proportion of the prolyl bonds must
response.) Perhaps what happens is that the isomerize from trans to cis for proper folding to
DnaK chaperone complex prevents the poly- be initiated.
peptides from premature folding and aggre- Actually, there are two distinct sites on trig-
gation while they are being synthesized, and ger factor that bind to substrate proteins. One
www.ebook777.com
site, the proline residue site, catalyzes prolyl but does inhibit archaeal and eukaryotic protein
isomerizations. The second site lies elsewhere synthesis. These properties reflect differences in
on trigger factor and binds in a nonspecific the ribosomal and accessory proteins required
fashion to the unfolded portions of newly syn- for translation in these systems.
thesized proteins. In fact, model experiments
have demonstrated that unfolded proteins will
bind to trigger factor regardless of the pres- 11.3 Summary
ence of proline residues; that is, the nonproline Virtually all macromolecules share three
sites are sufficient. The binding to the nonpro- mechanistic features in their synthesis: (1) they
line residue sites is probably important in aid- are built from small molecules that are readily
ing the proline-limited folding of the proteins. synthesized by metabolic pathways; (2) they
One can imagine that binding to the unfolded are energetically unfavorable because they are
protein chains might block folding, and in this dehydrations in an aqueous environment; and
way allow trigger factor to isomerize the proline (3) energy input is required to synthesize them.
residues. Binding to the unfolded protein chains RNA synthesis is initiated at promoter sites,
may also be part of a trigger factor chaperone which are nucleotide sequences upstream of the
function that is important to prevent aggrega- transcription start site. The promoter is rec-
tion or degradation of unfolded nascent protein ognized by a subunit in the polymerase called
chains.34, 35 For a review of prolyl isomerases, sigma factor. The σ70 polymerase recognizes
see ref. 36. consensus sequences at the –10 and –35 regions
where the transcription start site is +1. Other
11.2.13 Inhibitors sigma factors exist but do not recognize the
Some antibiotic inhibitors of protein synthesis same sequence recognized by σ70. The sigma
and their sites of action are listed in Table 11.1. subunit has a key role in both binding the core
Note that chloramphenicol, erythromycin, and RNA polymerase and the promoter site. It
streptomycin inhibit bacterial but not archaeal mediates the formation and loading of an open
or eukaryotic (cytosolic) protein synthesis, and complex into the polymerase before it departs
that diphtheria toxin does not inhibit bacterial during elongation.
Table 11.1 Antibiotics that inhibit protein synthesis
Antibiotic Site of action Bacteria Archaea Eukaryotes
Erythromycin Inhibits translocation + – –

Streptomycina Inhibits initiation and causes mistakes in + – –
reading mRNA at low concentrations
Tetracyclineb Inhibits binding of aminoacyl–tRNA + + +
Chloramphenicol Inhibits peptidyltransferase + – –
Puromycinc Resembles charged tRNA and causes + + +
premature chain termination
Cycloheximide Inhibits peptidyltransferase – – +
Diphtheria toxin Inhibits elongation factor 2 (EF-2), which – + +
is analogous to bacterial EF-G, by ADP
ribosylation
a
Streptomycin binds to the S12 protein in the 30S ribosomal subunit and inhibits the binding of fMet–tRNAfMet to the P site.
Neomycin and kanamycin act in a manner similar to streptomycin.
b
Tetracyclines inhibit bacterial protein synthesis by binding to the 30S ribosomal subunit and preventing attachment
of aminoacylated tRNA. They also inhibit eukaryotic ribosomes, but not at the concentrations used to treat bacterial
infections. This is apparently because eukaryotic membranes are not very permeable to tetracycline.
c
Puromycin resembles the aminoacyl portion of aminoacylated tRNA. It has a free amino group that attacks the ester
linkage in the growing peptide in the P site, with the result that a peptidyl puromycin forms at the A site. However, the
peptidyl puromycin cannot be transferred to the P site (i.e., translocation does not occur), and therefore chain elongation is
terminated as the peptidyl puromycin dissociates from the ribosome.
DNA gyrase is required for transcrip- have been occupied by aminoacylated tRNAs,
tion, to prevent the DNA from being over- the α-amino group of the amino acid in the A
wound downstream from the transcribed site displaces the tRNA at the P site, leaving it
region. Transcription is terminated in either uncharged.
a sequence-dependent manner (Rho indepen- After formation of the aminoacylated pep-
dent) or by accessing a helicase called Rho that tide, translocation takes place. During trans-
disrupts the open complex in the polymerase. location, the ribosome moves one codon with
Transcription is regulated by a variety of posi- respect to the mRNA and the aminoacylated
tive and negative transcription factors that bind peptide is transferred from the A site to the P
to the DNA in the promoter region or upstream site, thus enabling the A site to receive the next
of the promoter (enhancer regions). The tran- aminoacylated tRNA. Translocation itself
scription factors can influence the binding of requires elongation factor EF-G and GTP.
the RNA polymerase to the promoter or the Following each subsequent transfer, the spent
rate at which the open complex forms. In many tRNA that had occupied the P site moves to
operons, RNA synthesis is also regulated by an adjacent exit (E) site. At least two tRNAs
attenuation. within the ribosome maintain contact with
Protein synthesis begins with the aminoacy- the mRNA at all times, helping to preserve the
lation of tRNAfMet in the cytosol. The enzymes reading frame.
that aminoacylate the tRNAs are called amino- The process is repeated until the ribosome
acyl–tRNA synthetases. The tRNA is an adap- reaches the end of the gene. At the end of the
tor molecule, coupling activated amino acids gene there is a termination codon for which
with information present in the anticodon there is no tRNA. When the A site of the ribo-
nucleotides. As the mRNA, which contains the some reaches the termination codon, protein
information to specify the sequence of a pro- termination factors promote hydrolysis of the
tein, moves through the ribosome, its codons linkage between the protein and the tRNA in
are sequentially recognized by the correspond- the P site, thus releasing the protein. The ribo-
ing charged tRNAs, thereby ensuring that the some also dissociates into its subunits by the
amino acids are placed in the correct sequence. action of RRF (ribosome recycling factor) and
The aminoacyl–tRNA synthetases that attach EF-G/GTP. This allows reassembly of the ribo-
the amino acids to the tRNA distinguish among some subunits and initiation of a new round
the different tRNAs and normally attach the of translation at the beginning of a different
amino acid specified by the anticodon with high gene.
specificity. In some cases, the tRNAs are edited Once synthesized, the protein folds into its ter-
by synthetases after initial charging to remove tiary structure. This process is frequently aided
incorrect amino acids. Then, in the presence of by a class of proteins called chaperones. These
initiation factors IF-1, IF-2, and IF-3, as well proteins recognize and sequester misfolded and
as mRNA and the 30S and 50S ribosomal sub- unfolded proteins, then use the energy of ATP
units, the 70S initiation complex is formed. In to transiently unfold them and allow them the
this complex the fMet–tRNAfMet is positioned opportunity to refold correctly.
at the P site in the ribosome. This positioning
is often directed by a sequence of nucleotides Study Questions
called the Shine–Dalgarno sequence, 6 to 10
nucleotides downstream of the transcription 1. What are the three hallmarks of macromo-
start codon. lecular synthesis? Where do the small mol-
The Shine–Dalgarno sequence within the ecules come from to synthesize the large
mRNA hybridizes with the 3′ end of the 16S structures in the cell?
rRNA in the 30S ribosomal subunit and,
2. What is the role of the sigma subunit in RNA
because of this, the start codon is positioned
synthesis? Under what circumstances might
at the P site on the ribosome. The next amino-
a bacterium use different sigma factors?
acylated tRNA enters, delivered as a complex
with EF-Tu complexed with GTP and is posi- 3. Why is the cleft of the core RNA polymerase
tioned at the A site. Once the P site and the A site positively charged?
www.ebook777.com
4. During RNA synthesis, what unwinds REFERENCES AND NOTES

the DNA? Compare this process with the
unwinding during DNA replication. 1. McClure, W. R., 1985. Mechanism and control of
transcription initiation in prokaryotes. Annu. Rev.
5. How might transcription factors increase Biochem. 54:171–204.
or decrease the rates of transcription? 2. Henkin, T. M., 1996. Control of transcription
termination in prokaryotes. Annu. Rev. Genet. 30:
6. How is RNA synthesis terminated? Com- 35–57.
pare and contrast the two available
strategies. 3. Platt, T. 1986. Transcription termination and the
regulation of gene expression. Annu. Rev. Biochem.
7. The addition of tryptophan to the growth 55: p. 339–372.
medium of an E. coli trpR mutant lowers 4. Campbell, E. A., L. F. Westblade, and S. A. Darst.
the expression of the trp operon. What is 2008. Regulation of bacterial RNA polymerase
the explanation for this? sigma factor activity: a structural perspective. Curr.
Opin. Microbiol. 11(2):121–127.
8. What is a termination codon? How is it
5. Ades, S. E. 2004. Control of the alternative
involved in the termination of protein syn- sigma factor sigmaE in Escherichia coli. Curr. Opin.
thesis, and what recognizes this codon? Microbiol. 7(2):157–162.
9. What ensures that the correct amino acid 6. Gruber, T. M., and C. A. Gross, 2003. Multiple
is added to the correct tRNA? If an incor- sigma subunits and the partitioning of bacterial
rect amino acid is added, can the mistake be transcription space. Annu. Rev. Microbiol. 57:441–
466.
corrected?
7. Ishihama, A. 1997. Promoter selectivity control
10. Describe the roles of IF-1, IF-2, IF-3, EF-Tu of RNA polymerase. Mechanisms of Transcription.
and EF-G in protein synthesis. F. Eckstein and D. Lilley (Eds.). Springer-Verlag,
Berlin.
11. What is the direction in which mRNA is
translated (i.e., 5′ to 3′ or 3′ to 5′)? How is 8. Steitz, T. A., et al. 1982. Structural similarity in
the DNA-binding domains of catabolite gene activa-
that characteristic relevant to the coupling tor and Cro repressor proteins. Proc. Natl. Acad. Sci.
between transcription and translation? USA 79(10):3097–3100.
Compare this condition with the polarity
9. The eukaryotic cytosolic ribosome is 80S and has
of transcription. two subunits, 40S and 60S. The 40S subunit contains
12. What are the roles of GTP hydrolysis during an 18S rRNA. The 60S subunit contains a 28S, a 5S,
and a 5.8S rRNA. The rRNA transcript in eukaryotes
translation? How does this differ from the is 45S. It is processed in the nucleus to produce the
role of ATP in biosynthetic reactions, such 18S, 28S, and 5.8S rRNAs, which are present in the
as in the charging of tRNAs with their cog- ribosomes. The 5S rRNA is on a separate transcript.
nate amino acids? 10. Bjork, G. R., et al. 1987. Transfer RNA modifi-
13. Explain the role of chaperone proteins cation. Annu. Rev. Biochem. 56: 263–287.
in protein folding. What is the evidence 11. Certain RNA molecules are known to have
that the binding to chaperone proteins is enzymatic activity. They are called ribozymes. Two
transient? well-studied examples are RNase P and self-splic-
ing group I introns. The self-splicing intron is an
14. What are the similarities and differences endonuclease that cuts itself out of RNA precursor
between the synthesis of RNA and protein molecules and joins the exons together to form the
mature RNA. Another ribozyme is RNase P, which
biopolymers? Specifically, focus on the is actually a nucleoprotein. However, the catalytic
structures of the machines that catalyze activity resides in the RNA. RNase P acts on tRNA
the reaction, the energy input to activate precursors. Transfer RNA is made from longer pre-
substrates for the reaction, the energy cursors that have had nucleotides removed from
input that contributes to conformational the ends. RNase P is an RNA molecule that acts
as an endonuclease and removes nucleotides from
changes and intermediate formation, the 5′ end of tRNA molecules. Self-splicing introns
and the similarities in substrates and and RNase P are widespread in prokaryotes and
products. eukaryotes.
12. Hershey, J. W. B. 1987. Protein synthesis. G or U, respectively, and there is only one codon per
Escherichia coli and Salmonella typhimurium: amino acid and therefore only one tRNA per codon.
Cellular and Molecular Biology, Vol. 1. F. C. However, if the base in the first position in the anti-
Neidhardt et al. (Eds.). ASM Press, Washington, DC. codon is I (inosinate), G, or U, then the third base in
the codon forms a loose pair or “wobble” with its
13. Protein synthesis by cytosolic eukaryotic ribo-
partner in the anticodon. This allows different codons
somes differs in some respects from protein synthesis
for a single amino acid to be read by the same tRNA.
by bacterial ribosomes. (1) The ribosomal subunits
For example, inosinate, which has hypoxanthine as
are 40S and 60S. (2) The first amino acid is methi-
its base, can be found in the first position of the anti-
onine rather than formyl methionine. (But still, a spe-
codon for some amino acids. Inosinate can hydrogen
cial tRNAMet is used for initiation.) (3) There are at
bond weakly to U, C, or A. This means that the third
least nine initiation factors, rather than three. These
base in the codon can be either U, C, or A, and all
are elF3 and el4C, which bind to the 40S subunit and
three codons will bind to the same anticodon (i.e., to
stimulate further steps in initiation; the cap protein
the same tRNA). Likewise, U in the first position in
(or CBPI), which binds to the 5′ end of the mRNA
the anticodon can pair with either A or G in the third
and aids in binding the mRNA to the 40S subunit;
position of the codon, so that in this case two codons
elF2, which aids in binding of Met–tRNAMet to the
can be read by the same tRNA. Similarly, G in the
40S subunit; elF4a, elF4B, and elF4F, which help to
first position of the anticodon can pair with C or U.
locate the start AUG; elF5, which facilitates the dis-
Thus it is possible to have fewer tRNAs than codons.
sociation of the other initiation factors from the 40S
At least 32 different tRNAs are required to read the
subunit so that it can combine with the 60S subunit
61 codons. It has been suggested that the advantages
to form the 80S ribosome; and elF6, which stimulates
of introducing a “wobble” in the third position are
the dissociation of the 80S ribosome not engaged in
that the bonding between the codon and anticodon
protein synthesis to 40S and 60S subunits. The elon-
becomes less tight and that the rate of dissociation of
gation and translocation phases are analogous to
the tRNAs from the ribosome during protein synthe-
what occurs in bacterial ribosomes. The cytosolic
sis is increased.
eukaryotic ribosomes use three elongation factors,
eEF1α, eEF1βγ, and eEF2, which are analogous to 16. Ibba, M., A. W. Curnow, and D. Soll. 1997.
EF-Tu, EF-Ts, and EF-G. Termination is also similar. Aminoacyl–tRNA synthesis: divergent routes to a
Whereas in bacteria there are three release factors that common goal. Trends Biochem. Sci. 22(2): 39–42.
are specific for the three termination codons, protein
17. Becker, H. D., et al. 1997. Existence of two dis-
synthesis by cytosolic eukaryotic ribosomes uses
tinct aspartyl–tRNA synthetases in Thermus ther-
a single release factor, eRF, which recognizes three
mophilus. Structural and biochemical properties of
different termination codons. Eukaryotic mRNA dif-
the two enzymes. Biochemistry. 36(29):8785–8797.
fers from prokaryotic mRNA in not being polycis-
tronic and in not having a special ribosome binding 18. Many organisms possess fewer than 20 different
analogous to the Shine–Dalgarno sequence upstream aminoacyl–tRNA synthetases to activate the 20 stan-
of the initiation codon. Eukaryotes typically use the dard amino acids. That is, in a few instances the same
first AUG at the 5′ end of the mRNA. aminoacyl–tRNA synthetase charges two different
tRNAs with the same amino acid. For example, it has
14. Arnez, J. G., and D. Moras. 1997. Structural
been found that Bacillus subtilis, other gram-positive
and functional considerations of the aminoacylation
bacteria, the gram-negative Rhizobium meliloti,
reaction. Trends Biochem. Sci. 22(6): 211–216.
some archaebacteria, chloroplasts, and mitochondria
15. Although there are 61 codons for the 20 amino use the same enzyme (i.e., GluRS) to charge tRNAGln
acids, there are fewer than 61 tRNAs. This is because (has anticodon for Glu codon) and tRNAGlu (has
some tRNAs can read more than one codon for a anticodon for Gln codon) with glutamate. Then an
particular amino acid. To understand the follow- amidotransferase recognizes the Glu–tRNAGln, and ω
ing discussion, bear in mind that the anticodon and amidates the glutamate to form Gln–tRNAGln, which
the codon have opposite polarity, and thus the third places the Gln at the Gln codon. A similar situation
nucleotide in the codon always pairs with the first occurs in the halophilic archaebacterium Haloferax
nucleotide in the anticodon. Crick realized that the volcanii with Asn–tRNAAsn, where both tRNAAsp and
first two nucleotides in the codon form strong pairs tRNAAsn are charged by the same enzyme with aspar-
with their partners in the anticodon and in fact are tate. The aspartate is then converted to asparagine on
responsible for most of the codon specificity. That tRNAAsn, which places Asn at the Asn codon. Note
is, the third nucleotide in the codon frequently does the similarity to the fate of methionine that is charged
not change the specificity of the codon. Therefore, onto tRNAfMet. A special transformylase recognizes
codons with the same nucleotide in the first two posi- the Met–tRNAfMet and formylates the Met to fMet,
tions but differing in the third nucleotide can often which is used to initiation translation. In all these
code for the same amino acid. This actually depends cases the tRNA is charged with the wrong amino
upon the base at the first position in the anticodon. If acid, which is subsequently modified to the correct
the first base in the anticodon is either C or A, then amino acid. There is also an example of two differ-
the base in the third position of the codon is always ent aminoacyl synthetases activating the same amino
www.ebook777.com
acid and charging the same tRNA. The thermophilic 28. The groE operon contains two genes, groES and
bacterium Thermus thermophilus has two aspartyl– groEL. E. coli requires both GroEL and GroES (some-
tRNA synthetases (AspRS1 and AspRS2) that charge times referred to as the GroE system) for growth at all
tRNAAsp. They are encoded by separate genes. temperatures, although as discussed in Chapter 16,
their synthesis is greatly stimulated by heat shock.
19. The amino acids are attached to the 3′-terminal
Possibly, the excess GroEL at higher temperatures
CCA of tRNAs. Some synthetases attach the amino
protects unstable proteins or aids in refolding par-
acid directly to the 3′-hydroxyl, whereas other syn-
tially denatured proteins when normal temperatures
thetases attach the amino acid to the 2′-hydroxyl, and
are restored. GroEL (also called Hsp60) is a double-
then a transesterification reaction moves the amino
ring structure (“double doughnut”) with 14 identical
acid to the 3′-hydroxyl. A description of the struc-
subunits; there are 7 subunits per ring. GroEL binds
tural and functional features of the aminoacyl–tRNA
to the protein that is to be folded, and the proteins
synthetases is reviewed in Arnez, J. G., and D. Moras.
are believed to be folded in the cavity of the “dou-
1997. Structural and functional considerations of
ble doughnut.” ATP is also necessary. ATP binds to
the aminoacylation reaction. Trends Biochem. Sci.
GroEL, producing a conformational change, and
22(16): 211–216.
increases the affinity of GroEL for the protein sub-
20. Grill, S., et al. 2000. Selective stimulation of strate. The hydrolysis of ATP lowers the affinity and
translation of leaderless mRNA by initiation factor aids in the release of the folded protein. The role of
2: evolutionary implications for translation. EMBO GroES (also called Hsp10), which is a single ring of
J., 19(15): 4101–4110. seven subunits, is to aid in the release of the folded
protein from GroEL One might view the GroEL cav-
21. Ramakrishnan, V. 2009. The Ribosome: Some
ity within which the protein folds as a “safe haven,”
Hard Facts about Its Structure and Hot Air about
where a protein may fold spontaneously without
Its Evolution. Cold Spring Harbor Symposiums in
interacting with another protein to form aggregates.
Quantative Biology, Cold Spring Harbor Press, Cold
Spring Harbon, NY, pp. 1–9. 29. Gaitanaris, G. A., et al. 1994. Successive action of
Escherichia coli chaperones in vivo. Mol. Microbio.
22. Banerjee, R., et al. 2010. tRNAs: cellular bar-
14(5):861–869.
codes for amino acids. FEBS Lett. 584(2):387–395.
30. Following are the methods used by Gaitanaris
23. The end rule refers to the observation that the
et al. in ref. 29 to show that DnaK and DnaJ asso-
identity of the N-terminal amino acid residue in pro-
ciate with nascent polypeptides on ribosomes and
teins specifies its half-life. Amino acids such as leu-
that GroEL transiently associates with completed
cine, phenylalanine, tyrosine, and tryptophan are
polypeptides after their release from ribosomes. Cell
specifically recognized by ClpS and targeted for rapid
lysates of E. coli were centrifuged through a sucrose
enzymatic cleavage by the proteasome. Residues such
gradient. The ribosome fractions were isolated and
as arginine or lysine, which both present a positive
the proteins were separated by gel electrophoresis and
charge, are called secondary destabilizing residues
then transferred (blotted) onto nitrocellulose filters.
because they are substrates for enzymes that attach
The filters were treated with anti-DnaK, anti-DnaJ,
destabilizing amino acids to the N-terminus in a reg-
and anti-GroEL antibodies, and bound antibody was
ulated fashion. Such enzymes utilize charged t-RNAs
detected by means of alkaline phosphatase–coupled
to transfer amino acids directly (reviewed in Tobias,
secondary antibodies. Both DnaK and DnaJ were
J. W., et al. 1991. The N-end rule in bacteria. Science
associated with the ribosomal fractions isolated from
254(5036), 1374–1377.
the sucrose density gradients, but GroEL was not.
24. Frydman, J. 2001. Folding of newly translated When cells were briefly treated with puromycin to
proteins in vivo: the role of molecular chaperones. release nascent polypeptides from ribosomes, DnaK
Annu. Rev. Biochem. 70:603–647. and DnaJ were released from the ribosomes. Newly
synthesized polypeptides were associated with
25. Lund, P. A. 2001. Microbial molecular chaper-
GroEL. This was determined by pulse-labeling cells
ones. Adv. Microb. Physiol. 44:93–140.
with [35S]methionine, chasing with unlabeled methi-
26. The primary structure of proteins refers to the onine for various times, and separating cell fractions
sequence of amino acids. The secondary structure on a sucrose gradient. The proteins were separated
refers to regular arrangements of the polypeptide by gel electrophoresis and blotted onto nitrocel-
chain (e.g., an α-helix in which the polypeptide back- lulose, whereupon GroEL was detected by using
bone is wound around the long axis of the molecule anti-GroEL antibody as just described. Radioactive
in a helix and the R groups of the amino acids point proteins were transiently associated with fractions
out). The tertiary structure refers to the folding of the containing GroEL. However, none of the GroEL-
polypeptide chain. The quaternary structure refers to associated proteins were in the ribosomal fractions,
the spatial arrangement of folded polypeptide sub- indicating that they became associated with GroEL
units in a multimeric protein. after their release from the ribosomes.
27. Ewalt, K. L., et al. 1997. In vivo observation of 31. Fedorov, A. N., and T. O. Baldwin. 1997.
polypeptide flux through the bacterial chaperonin Cotranslational protein folding. J. Biol. Chem.
system. Cell. 90(3):491–500. 272(52): 32715–32718.
32. Stoller, G., et al. 1995. A ribosome-associated independent of proline residues. J. Mol. Biol. 277(3):
peptidyl–prolyl cis/trans isomerase identified as the 723–732.
trigger factor. EMBO J. 14(20):4939–4948.
35. Scholz, C., et al. 1982. Cooperation of enzy-
33. Deuerling, E., et al. 1999. Trigger factor and matic and chaperone functions of trigger factor
DnaK cooperate in folding of newly synthesized pro- in the catalysis of protein folding. EMBO J. 16(1):
teins. Nature. 400(6745):693–696. 54–58.
34. Scholz, C., et al. 1998. Recognition of protein 36. Hunter, T. 1998. Prolyl isomerases and nuclear
substrates by the prolyl isomerase trigger factor is function. Cell. 92(2):141–143.
www.ebook777.com
12
Cell Wall and Capsule Biosynthesis
Studying cell wall and capsule synthesis in antibiotics), the cell would likely swell through
bacteria is instructive in showing how logis- the weak areas and lyse as a result of the internal
tic problems of extracellular biosynthesis are turgor pressures.
solved. For example, the subunits of the cell
wall and capsule polymers are synthesized as The chemical composition of peptidoglycan
water-soluble precursors in the cytosol. How Peptidoglycan is made of glycan strands of alter-
do the subunits traverse the lipid barrier in the nating residues of N-acetylmuramic acid and
cell membrane to the sites of polymer assembly? N-acetylglucosamine linked by β-1,4 glycosidic
A second problem concerns the final stages of bonds between the C1 of N-acetylmuramic acid
peptidoglycan synthesis. During peptidoglycan and the C4 of N-acetylglucosamine (Fig. 12.2).
synthesis, the newly polymerized glycan chains N-Acetylmuramic acid is a modified form of
become cross-linked by peptide bonds on the N-acetylglucosamine in which a lactyl group
outside cell surface. What is the source of energy has been attached to the C3 carbon. Attached
for making the peptide cross-links at a site that to each N-acetylmuramic acid is a tetrapeptide.
lacks ATP? This chapter considers these and The tetrapeptide is L-alanyl-D-glutamyl-γ-L-R3-
other aspects of the biosynthesis of peptidogly- D-alanine. The amino acid in position 3 varies
can and lipopolysaccharide, as well as capsular with the species of bacterium. Gram-negative
and other extracellular polysaccharides. The
biosynthesis of cytosolic polysaccharides is also
considered.
12.1 Peptidoglycan
12.1.1 Structure
Peptidoglycan is a heteropolymer of glycan
chains cross-linked by amino acids (Section
1.2.3). The peptidoglycan is a huge molecule,
since it surrounds the entire cell and appears
Fig. 12.1 The topological relationship of the pep-
to be covalently bonded throughout. A sche- tidoglycan to the cell membrane and rest of the cell
matic drawing of how this might look in gram- wall. In gram-negative bacteria such as E. coli, the
negative bacteria is shown in Fig. 12.1 (see peptidoglycan is a thin layer sandwiched between the
also Fig. 1.8). Peptidoglycan confers strength inner and outer membranes. In gram-positive bacte-
to the cell wall, and if one were to enzymati- ria there is no outer membrane, and the peptidogly-
cally destroy the integrity of the peptidoglycan can is a thick layer usually covalently bonded to other
(with lysozyme) or prevent its synthesis (with molecules (e.g., teichoic acids).
316
cell wall and capsule biosynthesis 317
Fig. 12.2 The disaccharide–peptide subunit in peptidoglycan. The glycan backbone consists of alternating
residues of N-acetylglucosamine (G) and N-acetylmuramic acid (M) linked β-1,4. The carboxyl group of the
lactyl moiety in N-acetylmuramic acid is substituted with a tetrapeptide. The amino acids in the tetrapeptide
are usually L-alanine, D-glutamic, L-R3 (residue 3), which is an amino acid that varies with the species, and
D-alanine. The peptide linkages are all α except for that between D-glutamic and the amino acid in position 3,
which is γ linked. The X can be any one of a large number of side chains, some examples of which are shown.
The α carboxyl of glutamic acid can be free, an amide, or substituted (e.g., by glycine).
bacteria generally have meso-diaminopimelic roseus, and three glycines and two L-serines in
acid. (However, some spirochaetes contain Staphylococcus epidermidis. Sometimes the
ornithine instead of meso-diaminopimelic bridge is from the terminal D-alanine to the
acid.) In contrast, there is much more variability α-carboxyl of D-glutamic acid of another tetra-
in the amino acids in position 3 in gram-positive peptide. Since this is a connection between two
bacteria (Fig. 12.2). carboxyl groups, a bridge of amino acids con-
taining a diamino acid is necessary.
Cross-linking
The tetrapeptide chains are cross-linked by pep-
tide bonds (Fig. 12.3A,B). There is a great deal 12.1.2 Synthesis
of variability in the composition of the cross- Peptidoglycan is made in several stages: (1) the
links among the different groups of bacteria. precursors to the peptidoglycan are UDP deriv-
In fact, amino acid composition and cross-link atives of the amino sugars made in the cytosol;
location have been used for taxonomic pur- (2) the amino sugars are then transferred to a
poses. In most instances the peptide bridge is lipid carrier in the membrane, which carries the
from the carboxyl in the terminal D-alanine in amino sugars across the membrane; (3) the pep-
one tetrapeptide to an amino group in the amino tidoglycan is polymerized on the outer surface
acid in the L-R3 position in another tetrapeptide. of the membrane; and (4) a transpeptidation
In some bacteria the cross-linking is direct— reaction cross-links the peptidoglycan. (For
for example, between D-alanine and diamin- reviews, see refs. 1 and 2.)
opimelic acid in gram-negative bacteria and
many Bacillus species. However, in most gram- Synthesis of the UDP derivatives:
positive bacteria there is a bridge of one or more UDP-N-acetylglucosamine and
amino acids. Some examples are a bridge of five UDP-N-acetylmuramyl pentapeptide
glycine residues in Staphylococcus aureus, three The two amino sugars that are precursors to the
L-alanines and one L-threonine in Micrococcus peptidoglycan are N-acetylglucosamine and
www.ebook777.com
Fig. 12.3 Peptidoglycan cross-linking. (A) Direct cross-link between diaminopimelate and D-alanine as it
occurs in gram-negative bacteria. (B) An amino acid bridge between two tetrapeptides, as in many gram-
positive bacteria. Sometimes the bridge is between the C-terminal D-alanine and the α-carboxyl of D-glutamic
acid. When this occurs, there must be a diamino acid in the bridge. (C) Schematic drawing of cross-linked pep-
tidoglycan. Tetrapeptides are indicated by straight lines. The glycan chains in some bacteria have been found
to contain about 20 to 100 sugar residues, apparently varying in length within the same bacterium. The degree
of cross-linking depends upon the bacterium. In E. coli, which is considered to be a relatively un-cross-linked
peptidoglycan, about 50% of the peptide chains are cross-linked. In some bacteria, about 90% of the chains
are cross-linked. Abbreviations: G, N-acetylglucosamine; M, N-acetylmuramic acid.
N-acetylmuramic acid. Both amino sugars are some is converted to UDP–N-acetylmuramic

made from fructose-6-phosphate (Fig. 12.4). acid (UDP–MurNAc). The UDP–GlcNAc is
In step 1, glutamine donates an amino group converted to UDP–MurNAc by the addition
to fructose-6-phosphate, converting it to of a lactyl group to the sugar in step 5. In this
glucosamine-6-phosphate. Then in step 2, a reaction, the C3 hydroxyl of the sugar displaces
transacylase transfers an acetyl group from the phosphate from the α carbon of phospho-
acetyl–CoA to the amino group on glucosamine- enolpyruvate, forming the enol pyruvate ether
6-phosphate to make N-acetylglucosamine- derivative of the UDP-N-acetylmuramic acid.
6-phosphate, which is isomerized in step 3 Then, in step 6, the enol derivative is reduced to
to N-acetylglucosamine-1-phosphate. The the lactyl moiety by NADPH. The UDPMurNAc
monophosphate attacks UTP, displacing is converted into UDP–MurNAc pentapeptide
pyrophosphate in step 4 to form UDP–N- by the sequential addition of the amino acids
acetylglucosamine. The reaction is driven to L-alanine, D-glutamate, and L-R3 (residue 3),
completion by a pyrophosphatase. the dipeptide D-alanyl-D-alanine, and ATP in
Some of the UDP–N-acetylglucosamine step 7.
(UDP–GlcNAc) is used as the precursor to the Each reaction is catalyzed by a separate
N-acetylglucosamine in peptidoglycan, and enzyme and requires ATP to activate the
Fig. 12.4 Synthesis of N-acetylglucosamine and N-acetylmuramyl pentapeptide. Enzymes: 1, glutamine:

fructose-6-phosphate aminotransferase; 2, glucosamine phosphate transacetylase; 3, N-acetylglucosamine
phosphomutase; 4, UDP–N-acetylglucosamine pyrophosphorylase; 5, enoylpyruvate transferase; 6, UDPN–
acetylglucosamincenol–pyruvate reductase. In 7, the UDP–N-acetylmuramic acid is converted to the pen-
tapeptide derivative by the sequential additions of L-alanine, D-glutamate, L-R3, and D-alanine-D-alanine by
separate enzymes.
carboxyl group of the amino acid. The activated Then an ATP-dependent D-alanyl-D-alanyl
carboxyl is probably an acyl phosphate, which synthetase makes D-alanyl-D-alanine from two
is attacked by the incoming amino group dis- D-alanines. The racemase and synthetase are
placing the phosphate. (See Fig. 8.8.) The prod- inhibited by the antibiotic D-cycloserine. The
ucts are ADP and inorganic phosphate. (The MurNAc pentapeptide is transferred to the lipid
fifth D-alanine is removed during a cross-linking carrier in the membrane, as described next.
reaction described later.) [The synthesis of the
peptide subunit in the archaeal pseudomurein Reactions in the membrane
(Fig. 1.14) takes place via UDP-peptide interme- The lipid carrier is called undecaprenyl phos-
diates, which also appear to form peptide bonds phate or bactoprenol. Undecaprenyl phosphate
via acyl phosphates.3 The D-alanine-D-alanine is is a C55 isoprenoid phosphate whose structure
made by separate enzymes. The first of these is is shown in Fig. 12.5. Undecaprenyl phosphate
a racemase that converts L-alanine to D-alanine. serves not only as a carrier for peptidoglycan
www.ebook777.com
precursors but also as a carrier for the precur- the C1 carbon in UDP–GlcNAc, displac-
sors of other cell wall polymers (e.g., lipopoly- ing the UDP. The product is the disaccharide
saccharide and teichoic acids). Interestingly, precursor to the peptidoglycan, lipid–PP–
eukaryotes also use an isoprenoid phosphate, MurNAc(pentapeptide)–GlcNAc. The disac-
dolichol phosphate, to carry oligosaccharide charide–lipid moves to the other side of the
subunits across the endoplasmic reticulum membrane (Fig. 12.6, step 3). Exactly how the
(ER) membrane to be attached to glycopro- disaccharide-lipid moves through the mem-
teins in the ER lumen. Dolichol phosphate is brane is not understood, but it may be aided
larger than undecaprenyl phosphate but has by unidentified proteins. On the outside sur-
the same structure. The nucleotide sugars dif- face of the membrane, the lipid-disaccharide is
fuse to the membrane, where undecaprenyl transferred to the growing end of the acceptor
phosphate (lipid-P) attacks the UDP–MurNAc glycan chain (Fig. 12.6, step 4). Reaction 4 is
(pentapeptide) displacing UMP (Fig. 12.6, a transglycosylation in which the C4 hydroxyl
step 1). The product is lipid–PP–MurNAc of the incoming GlcNAc attacks the C1 of the
(pentapeptide). The GlcNAc is then trans- MurNAc in the glycan, displacing the lipid–PP
ferred from UDP–GlcNAc to the MurNAc on from the growing glycan chain. This reaction
the lipid carrier (Fig. 12.6, step 2). This occurs is catalyzed by a membrane-bound enzyme
when the C4 hydroxyl in the MurNAc attacks called transglycosylase.
Fig. 12.5 The structure of undecaprenyl phosphate, a C55 isoprenoid phosphate that carries precursors to
peptidoglycan, lipopolysaccharide, and teichoic acids through the cell membrane.
Fig. 12.6 Extension of the glycan chain during peptidoglycan synthesis. 1, The MurNAc pentapeptide (–M)
is transferred to the phospholipid carrier (undecaprenyl phosphate) on the cytoplasmic side of the cell mem-
brane. 2, The GlcNAc (G) is transferred to the MurNAc pentapeptide to form the disaccharide–PP–lipid
precursor. 3, The disaccharide–PP–lipid precursor moves to the external face of the membrane. 4, A transgly-
cosylase transfers the incoming disaccharide to the growing glycan, displacing the lipid–PP from the growing
chain. Thus, the growing chain remains anchored to the membrane by the lipid carrier at the site of the trans-
glycosylase. 5, The lipid–PP released from the growing chain is hydrolyzed to lipid-P by a membrane-bound
pyrophosphatase. Note that the glycan chain grows at the reducing end; that is, displacements occur on the
C1 of muramic acid.
Notice that the growing glycan chain remains transglycosylation and transpeptidation steps;
anchored via the lipid carrier to the membrane that is, they are bifunctional enzymes. (The low
at the site of the transglycosylase. The lipid–PP molecular weight PBPs from E. coli, 4, 5, and 6,
released from the growing glycan chain is appear to be dispensable.) One of the proteins,
hydrolyzed by a membrane-bound pyrophos- PBP3, is specifically required for peptidoglycan
phatase to lipid–P and Pi (Fig. 12.6, step 5). synthesis in the septum during cell division.
This reaction is very important because it helps
drive the transglycosylation reaction to comple-
tion, since the hydrolysis of the phosphodiester 12.2 Lipopolysaccharide
results in the release of substantial energy. This 12.2.1 Structure
hydrolysis also regenerates the lipid–P, which Lipopolysaccharide (LPS) is a complex polymer
is necessary for continued growth of the pep- of polysaccharide and lipid in the outer mem-
tidoglycan as well as other cell wall polymers brane of gram-negative bacteria. The outer
(e.g., lipopolysaccharide and teichoic acids). membrane is described in more detail in Section
The hydrolysis is inhibited by the antibiotic 1.2.3. The best characterized lipopolysaccha-
bacitracin. rides are those of E. coli and Salmonella typh-
imurium (Fig. 12.8).4 The lipolysaccharides of
Making the peptide cross-link other bacteria are very similar.
The problem of providing the energy to make As seen in Fig. 12.8, lipopolysaccharide is
the peptide cross-link outside the cell membrane composed of three parts: (1) a hydrophobic
is solved by using a reaction called transpeptida- region embedded in the outer membrane called
tion. In the transpeptidation reaction, an :NH2 lipid A, which is composed of a backbone of two
group from the diamino acid in the 3 position glucosamine residues linked β-1,6 and esteri-
(e.g., DAP) attacks the carbonyl carbon in the fied via the hydroxyl groups to fatty acids; (2) a
peptide bond holding the two D-alanine residues core polysaccharide region projecting from the
together in the pentapeptide and displaces the outer membrane surface, whose composition is
terminal D-alanine. The result is a new peptide similar in all the Enterobacteriaceae and which
bond (Fig. 12.7). The transpeptidation reaction is connected to lipid A via 3-deoxy-D-manno-
is inhibited by penicillin. octulosonate (KDO: see note 5); and (3) an
outermost polysaccharide region connected to
Penicillin-binding proteins the core. The outermost polysaccharide region,
Bacteria that contain peptidoglycan have in sometimes called the O-antigen or the repeat
their membranes proteins that bind penicillin oligosaccharide, is made up of repeating units
and are called penicillin-binding proteins, or of four to six sugars that vary considerably in
PBPs. For example, E. coli has at least seven composition among different strains of bacte-
such proteins, which are involved in catalyz- ria. The repeating oligosaccharide may have as
ing the late steps of peptidoglycan synthesis. many as 30 units.
The high molecular weight PBPs from E. coli There are 167 different serotypes of
(1a, 1b, 2, 3) have been shown to catalyze the O-antigen in Escherichia coli. The structure of
Fig. 12.7 The transpeptidation reaction. In this nucleophilic displacement, the nucleophilic nitrogen attacks
the carbonyl, displacing the terminal D-alanine. The reaction is inhibited by penicillin. Terminal D-alanine
residues in chains not participating in cross-linking are removed by a D-alanine carboxypeptidase.
www.ebook777.com
abe GlcNac gal hep PP _ etn

man _ rha _ gal glc _ gal _ glc _ hep _ hep _ KDO GlcN _ GlcN
_ antigen n fatty acids
O KDO
KDO lipid A
core
Fig. 12.8 The LPS of Salmonella typhimurium. Abbreviations: abe, abequose; man, mannose; rha, rhamnose;
gal, galactose; GlcNAc, N-acetylglucosamine; glc, glucose; hep, heptose (L-glycero-D-mannoheptose); KDO,
3-deoxy-D-manno-octulosonic acid; etn, ethanolamine; P, phosphate; GlcN, glucosamine.
12.2.2 Synthesis
Lipid A
The lipid A portion of the lipopolysaccharide
is synthesized from UDP–GlcNAc, which as
described in Fig. 12.4 is made from fructose-6
phosphate.6 A model for lipid A synthesis in
E. coli proposed by Raetz is shown in Fig.
12.11.7 Lipid A is synthesized in the cytoplasmic
Fig. 12.9 The structure of KDO. The linkage between membrane, but the early steps are catalyzed by
KDO residues has been reported to be a glycosidic three cytoplasmic enzymes. The first enzyme,
linkage between a C2 hydroxyl in one KDO with
UDP–GlcNAc acyltransferase, transfers
either the C4 or C5 hydroxyl in a second KDO. The
linkage to lipid A has been suggested to be from the
β-hydroxymyristic acid from an ACP derivative
C2 hydroxyl of KDO to the C6 hydroxyl of lipid A. to C3 of UDP–GlcNAc to form the monoacyl
derivative. (Recall from Chapter 10 that fatty
acids are synthesized as ACP derivatives.) We
note parenthetically that the hydroxy acid trans-
KDO is shown in Fig. 12.9. The arrangement fer is actually a branch point in the metabolism
of the LPS in the outer envelope (Fig. 12.10) is of UDP–GlcNAc because some of the UDP–
one in which the LPS is anchored in the outer GlcNAc is incorporated into peptidoglycan. It
envelope by the hydrophobic lipid A region, reacts with PEP to form UDP–MurNAc, and
while the repeating oligosaccharide protrudes it also condenses with UDP–MurNAc to form
into the medium. UDP–MurNAc–GlcNAc. Upon formation of the
monoacyl derivative, a deacetylase removes the
The fatty acids in lipid A acetate from the nitrogen on C2. Interestingly,
Four identical fatty acids are attached directly the gene for the deacetylase is envA, which had
to the glucosamine in lipid A from E. coli and previously been known to be required for cell
Salmonella typhimurium. These are a C14 separation after septum formation.
hydroxy fatty acid, β-hydroxymyristic acid After deacetylation, a second molecule of
(3-hydroxytetradecanoic acid), linked via ester β-hydroxymyristic acid is transferred from
bonds to the 3′-hydroxyls of the glucosamine the ACP derivative by a third enzyme (an
and via amide bonds to the nitrogen of the N-acyltransferase) to the nitrogen to form the
glucosamine. These fatty acids appear to be 2,3-diacyl derivative. Some of the latter loses
found uniquely in lipid A, and their presence in UMP and is converted to 2,3-diacylglucosamine-
a bacterium implies the presence of a lipopoly- 1-P (also called lipid X), which condenses
saccharide-containing lipid A. Esterified to the with UDP-2,3-diacylglucosamine to form
hydroxyls of two of the β-hydroxymyristic the disaccharide linked in β-1,6 linkage. A
acids are long-chain saturated fatty acids. In phosphate is added from ATP to form the
E. coli, these are lauric acid (C12) and myristic 1,4-diphosphate derivative, and this is then
acid (C14). The fatty acids attached to the glu- modified by the addition of KDO from CMP
cosamine anchor the lipopolysaccharide into derivatives and the esterification of fatty acids
the outer membrane. (lauryl and myristoyl) to the hydroxyl of the
O O O
}
O O O
O O O
C C C outer membrane
A A A
P
P P
} peptidoglycan
} cell membrane
Fig. 12.10 The arrangement of the LPS in the outer membrane of some gram-negative bacteria: A, lipid A,
C, core, O, oligosaccharide, P, porin. The outer envelope in the enteric gram-negative bacteria is asymmet-
ric with the lipopolysaccharide confined to the outer leaflet of the lipid bilayer. However, some bacteria
(e.g., penicillin-sensitive strains of Neisseria and Treponema) also have phospholipid in the outer leaflet.
Phospholipids in the outer membrane make the bacterium more sensitive to hydrophobic antibiotics. Refer to
Section 1.2.3 for a more complete discussion of the outer membrane.
β-hydroxymyristic moieties. The fatty acids are subsection entitled How the lipopolysaccharide
donated by ACP derivatives. might be assembled and Fig. 12.13.)
Core O-Antigen
The core in the Enterobacteriaceae contains an The O-antigen region is synthesized by a
inner region consisting of KDO, heptose, etha- mechanism that is different from that of core
nolamine, and phosphate, and an outer region synthesis (Fig. 12.12). Whereas the core is
that consists of hexoses. The biosynthesis of synthesized via the addition of sugars one at
the inner region is not fully understood. The a time to the growing end of the glycan chain,
outer core region grows as hexose units that are the O-antigen is synthesized as a separate poly-
donated one at a time from nucleoside diphos- mer on a lipid carrier and then transferred as a
phate derivatives to the nonreducing end of the unit to the core. The lipid carrier is undecapre-
growing glycan chain attached to the KDO. The nyl phosphate, the same molecule that carries
addition of each sugar is catalyzed by a specific the peptidoglycan precursors across the mem-
membrane-bound glycosyltransferase. The brane. First, the repeat unit of the O-antigen
core–lipid A portion of the LPS is translocated is synthesized on the lipid carrier. This is done
across the cell membrane to the periplasmic sur- by a series of consecutive reactions in which a
face, where the LPS is completed by attachment sugar moiety is transferred from a nucleoside
of the O-antigen.8 (This is discussed later. See diphosphate carrier to the nonreducing end of
www.ebook777.com
Fig. 12.11 Model for synthesis of lipid A in E. coli. See text for details. Source: Adapted from Raetz, C. R.
H. 1987. Structure and biosynthesis of lipid A in Escherichia coli, pp. 498–503. In: Escherichia coli and
Salmonella typhimurium: Cellular and Molecular Biology, Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press,
Washington, DC.
the growing repeat unit (Fig. 12.12, steps 1–4). enters the lipid pyrophosphate pool, where it
Each of these reactions is catalyzed by a differ- is hydrolyzed to lipid–P by a bacitracin-sensi-
ent enzyme. Then the repeat unit is transferred tive enzyme. Finally, the completed oligosac-
as a block to the growing oligosaccharide chain charide chain is transferred to the lipid A–core
(Fig. 12.12, step 5). The lipid–PP of the accep- region (Fig. 12.12, step 6), displacing the lipid
tor oligosaccharide chain is displaced and pyrophosphate.
Fig. 12.12 Synthesis of O-antigen and attachment to core in S. typhimurium. The O-antigen is polymerized
on the same phospholipid carrier that is used for peptidoglycan synthesis. Each sugar is added from a nucle-
otide diphosphate derivative by means of a specific sugar transferase (reactions 1–4). The nucleotide that is
used depends upon the sugar and the specific glycosyltransferase. For example, in S. typhimurium the nucle-
otide precursors are UDP–galactose, TDP–rhamnose, GDP–mannose, and CDP–abequose, in that order. The
first sugar that is added (e.g., galactose in S. typhimurium) is transferred as a phosphorylated derivative to the
lipid carrier. Hence the nucleoside monophosphate (NMP, or UMP in the case of E. coli and S. typhimurium)
is released. When the O-antigen tetrasaccharide subunit is finished, it is transferred to the growing O-antigen
chain by an enzyme called O-antigen polymerase (reaction 5). The completed O-antigen is transferred from its
lipid carrier to the core–lipid A region by an enzyme called O-antigen: lipopolysaccharide ligase (reaction 6).
The lipid–PP that is displaced is hydrolyzed by a bacitracin-sensitive phosphatase (reaction 7). All the reac-
tions take place in the cell membrane. There exist immunoelectron microscopy data to suggest that the ligase
reaction and perhaps the polymerase reaction take place on the periplasmic surface of the cell membrane.
Presumably, the lipid carrier ferries the subunits across the cell membrane. It is not known how the LPS crosses
the periplasm to enter the outer envelope. (See: Mulford, C. A., and M. J. Osborn. 1983. An intermediate step
in translocation of lipopolysaccharide to the outer membrane of Salmonella typhimurium. Proc. Natl. Acad.
Sci. USA 80:1159–1163.)
How the lipopolysaccharide might side, where it is added to the growing O-antigen
be assembled anchored to the membrane by its lipid carrier.
In the model for how the LPS might be assem- The lipid carrier on the growing oligosaccha-
bled depicted in Fig. 12.13 the O-antigen tet- ride is displaced. The core–lipid A region may
rasaccharide subunit is probably synthesized also be assembled on the cytoplasmic surface
on the lipid carrier on the cytoplasmic side of and translocated to the periplasmic surface. The
the membrane, then moves to the periplasmic completion of the lipopolysaccharide would
www.ebook777.com
Fig. 12.13 Model for LPS assembly. The O-antigen subunit is synthesized on the cytoplasmic surface and
then moves on the lipid carrier to the periplasmic surface of the membrane. A polymerase then transfers
the O-antigen subunit to the growing O-antigen chain. The core–lipid A region is also synthesized on the
cytoplasmic surface and is translocated to the periplasmic surface. The O-antigen is transferred to the core
to complete the LPS. The entire LPS is translocated to the outer membrane. Abbreviations: O, O-antigen; C,
core; lip, lipid carrier; A, lipid A.
take place on the periplasmic surface, where the desiccation because they can be hydrated, and
O-antigen is transferred to the core–lipid A, dis- in some cases they protect pathogens from the
placing the lipid–PP. The lipid–P is regenerated host immune system by preventing phagocyto-
via a phosphatase and enters a common pool sis. (See Box 12.1.)
of lipid–P also used in the biosynthesis of pep- The means of attachment of the extracel-
tidoglycan. It is not known how the lipopoly- lular polysaccharides to the bacterial surface
saccharide moves from the outer surface of the varies according to the polysaccharide. Slime
inner membrane into the outer membrane. This polysaccharides are not attached at all and are
might occur at adhesion sites between the inner readily released from the cell surface. Capsular
and outer membranes, or perhaps via a chaper- polysaccharides in gram-negative bacteria are
one carrier that transports the LPS through the generally attached via a hydrophobic mol-
periplasm. For more information, see ref. 9. ecule that anchors them to the lipid portion
of the outer membrane. The hydrophobic
anchor might be lipid A or phosphatidic acid.
12.3 Extracellular Polysaccharide
Lipopolysaccharides (discussed in Section 12.2)
Synthesis and Export in Gram- are attached to the outer membrane via lipid A
Negative Bacteria as shown earlier (Fig. 12.10). Figure 12.14 sum-
12.3.1 Overview marizes the modes of attachments.
For a description of extracellular polysaccha-
rides synthesized by bacteria and the biologi- Escherichia coli capsules
cal roles that they play, refer to Section 1.2.2, E. coli is an example of a gram-negative bac-
which also contains useful reviews.9–16 The terium that often possesses an extracellular
extracellular polysaccharides vary in molec- capsule surrounding its outer membrane.
ular weight from approximately 104 Da to The capsules contribute toward the virulence
about 106 Da. These extracellular polysaccha- of E. coli because they protect the cells from
rides are critical for cell survival in the natu- opsonophagocytosis and complement-medi-
ral habitat, although mutants lacking them ated killing, as indicated in Box 12.1. The cap-
can survive in the laboratory. They either are sule masks the antigenic determinants of the
closely associated with the cell wall (capsular O-polysaccharides in the LPS. An easy way to
polysaccharides) or exist as an amorphous detect the presence of the capsule is to heat the
layer loosely attached to the cell (slime poly- cells. Heating removes the capsule and makes
saccharides). Capsules offer protection from the cells agglutinable with anti-O antigen
slime polysaccharides (EPS)

A
Escherichia coli group-I K-antigens (uncertain)
P ECAPG
some Escherichia coli group-II K-antigens
Neisseria meningitidis CPS
Haemophilus influenzae CPS (uncertain)
B
KDO P
some Escherichia coli group-II K-antigens
LPS O-polysaccharides
C core region lipid A ECALPS
some Escherichia coli group-I K-antigens (KLPS)
Fig. 12.14 Cell surface association of polysaccharides in gram-negative bacteria. (A) No membrane
anchor. (B) Attachment through diacylglycerol phosphate. (C) Attachment through lipid A. Abbreviations:
KDO, 3-deoxy-D-manno-octulosonic acid; P, phosphate. Source: Whitfield, C., and M. A. Valvano. 1993.
Biosynthesis and expression of cell-surface polysaccharides in gram-negative bacteria. Adv. Microb. Physiol.
35:135–246.
BOX 12.1 CAPSULES AS VIRULENCE FACTORS

Capsules can be virulence factors that a plasma protein called activated comple-
protect from phagocytosis. Phagocytosis ment. The coating of the pathogen by
refers to the engulfment and killing of antibody or complement is called opsoniza-
pathogens by white blood cells such as neu- tion, and the antibody or complement is
trophils and macrophages. Neutrophils called an opsonin. Activated complement
kill primarily invading bacteria, whereas can also kill gram-negative bacteria inde-
macrophages kill larger microorganisms pendently of phagocytosis by forming a
as well as bacteria. Phagocytosis requires pore in the bacterial walls, allowing water
that the pathogenic microorganism bind to to rush in and osmotically lyse the cells.
the white blood cell, and the binding pro- E. coli capsules often protect the cells from
cess is greatly stimulated if the pathogenic opsonophagocytosis and complement-me-
microorganism is coated with antibody or diated killing.
antibody. Most capsular antigens are called K acids, and others consist of other constituents in
antigens (from the German Kapselantigene), addition to, or instead of, glucuronic acid, includ-
and there are at least 80 types that differ in their ing N-acetylneuraminic acid (NeuNAc), also
antigenicity and composition. called sialic acid, 3-deoxy-D-manno-octulosonic
The K antigens are acidic polysaccharides. acid (KDO), N-acetylmannosamine, and phos-
Some contain either glucuronic or galacturonic phate. The repeating unit varies from two to six
www.ebook777.com
monosaccharides, depending upon the polysac- intermediates in a pathway similar to the syn-
charide. Many appear to be substituted at their thesis of the oligosaccharide of lipopolysac-
reducing end with lipid A or KDO–phosphatidic charide, and the molecules are translocated
acid, presumably to anchor them in the outer through the membrane in an undecaprenol-de-
membrane (Fig. 12.14B,C). (Another extracel- pendent manner. However, other polysaccha-
lular polysaccharide that is anchored to the outer rides are not synthesized via lipid intermediates
membrane via lipid A is LPS. See Section 12.2.) and are translocated through the membrane in
Interestingly, other bacteria may synthesize an entirely different way, using specific trans-
similar or identical capsular polysaccharides. porters belonging to the family of ATP-binding
For example, Neisseria meningitidis makes a cassette (ABC) transporters. An example of a
capsular polysaccharide that is identical to the polymer secreted by the ABC transporter system
K1 capsule of E. coli. Haemophilus influenzae is the E. coli group 2 capsular polysaccharide
also makes a capsule quite similar to some of (CPS). The polymerization of the polysaccha-
the K polysaccharides of E. coli. However, not ride takes place on the cytoplasmic surface of
all E. coli capsules are made from K-antigens. the cell membrane as glycan residues are added
Some are made from molecules identical to to the growing end of the nascent chain. The
the lipopolysaccharide O-antigen but without polymer moves through the cell membrane
the lipid A or core regions. A classification of via an ABC transporter. Transport across the
E. coli capsules that places them in four groups periplasm requires a PCP protein called KpsE
is based upon criteria such as the organization and an OPX protein called KpsD. This export
of their biosynthetic genes, the regulation of the machinery across the periplasm is analogous
expression of these genes, and the assembly of to another protein system called the Wza–Wzc
the polysaccharides and their translocation to system. The net result of the KpsE/KpsD or
the outer surface. These groups are as follows: Wza–Wzc systems is that the large CPS mol-
group 1 (K30 serotype), group 2 (K1, K5 sero- ecules are exported to the outside surface of the
types), group 3 (K10, K54 serotypes), and group cell. For a more detailed review, the student is
4 (K40, O111 serotypes).14 The O-antigen in referred to ref. 16.
the O111 and other O-antigen capsules is not
linked to LPS. 12.3.2 Polysaccharide synthesis via
undecaprenol diphosphate intermediates
Summary of synthesis of extracellular Several bacteria synthesize certain exopolysac-
polysaccharides charides via undecaprenol intermediates similar
Most extracellular polysaccharides are syn- to the pathway for the biosynthesis of the oli-
thesized from intracellular nucleoside diphos- gosaccharide repeat unit in lipopolysaccharide
phate–sugar precursors and must be transported (Fig. 12.13) and the disaccharide repeat unit in
through the cell membrane to the outside of peptidoglycan (Fig. 12.6). Polysaccharides that
the cell. The nucleoside diphosphate sugars are synthesized via this pathway include the
are synthesized from sugar-1-P and nucleoside group 1 K-antigens of E. coli, the extracellular
triphosphates as described in Section 12.1.2 polysaccharide xanthan made by Xanthomonas
for the synthesis of N-acetylglucosamine. If the campestris, and the capsular polysaccharide of
extracellular polysaccharides are synthesized Klebsiella (Aerobacter) aerogenes. As an exam-
by gram-negative bacteria, they must be trans- ple, we will consider the biosynthesis of the K.
ported through the cell membrane, periplasm, aerogenes capsular polysaccharide.
and outer membrane. As we shall see, not all
polysaccharides are synthesized and exported K. aerogenes capsular polysaccharide
in the same way. The capsule made by K. aerogenes is a poly-
mer of repeating tetrasaccharides composed
Steps in synthesis and assembly of of galactose, mannose, and glucuronic acid
extracellular polysaccharides in in a molar ratio of 2:1:1. The glucuronic acid
gram-negative bacteria is attached as a branch at each mannose resi-
The synthesis of some extracellular poly- due. Its synthesis resembles the synthesis of the
saccharides takes place via undecaprenol oligosaccharide in lipopolysaccharide and is
depicted in Fig. 12.15.17 The tetrasaccharide factor for strains of E. coli that cause septicemia,
repeating unit is synthesized on undecapre- neonatal meningitis, and urinary tract infec-
nol diphosphate from nucleotide diphosphate tions in children. For this and other reasons, the
sugar intermediates. Each repeating unit is then K1 capsule has been of interest for a long time.
added as a block to the growing oligosaccharide Polysialic acid capsules are virulence factors in
attached to undecaprenol diphosphate. part because they resemble polysaccharides on
host tissue and consequently are poorly immu-
12.3.3 Synthesis of E. coli group 2 K1 nogenic. Thus they make the bacteria more
antigen: A pathway that may use an resistant to the immune response. Additionally,
undecaprenol monophosphate capsules containing sialic acid make the cells
intermediate more resistant to complement-mediated killing,
Chemical nature and biological role apparently because they inhibit the activation
of the K1 capsule of complement by the alternative system.
The E. coli K1 capsule (a group 2 antigen),
Neisseria meningitidis group B capsular poly- Biosynthesis of the K1 capsule
saccharide, and capsular polymers of Pasteurella The mechanism of polysialic acid chain elon-
haemolytica and Moraxella nonliquefaciens, gation in E. coli K1 has been well studied
are homopolymers of sialic acid approximately both biochemically and genetically. (See refs.
200 residues long. (For a review, see ref. 18.) The 19 and 20 and references therein.) The poly-
polysialic acid capsule of E. coli is a virulence mer grows by the stepwise addition of single
Fig. 12.15 Synthesis of capsular polysaccharide in Klebsiella aerogenes. Each monosaccharide is donated
to the tetrasaccharide repeat unit from a UDP derivative. This takes place on the inner surface of the cell
membrane. Polymerization takes place while the sugars are attached to undecaprenol diphosphate. The
tetrasaccharide repeat units are transferred from the lipid carrier to the growing chain, presumably on the
outer surface of the cell membrane. See Figs. 12.12 and 12.13 for a model suggesting how this might occur.
Abbreviations: UDP, uridine diphosphate; Gal, galactose; Man, mannose; GlcUA, glucuronic acid, C55,
undecaprenol; P, phosphate.
www.ebook777.com
sialic acid residues, apparently from cytidine GDP–guluronic acid. Another example is cellu-
monophosphate–sialic acid. This can be dem- lose, which is polymerized from UDP–glucose
onstrated by adding CMP [14C] sialic acid by a membrane-bound cellulose synthetase in
to incubation mixtures containing washed Acetobacter xylinum.
membranes some CMP–sialic acid tagged
with 14C. The enzyme that catalyzes the reac- 12.3.5 Export of polysaccharides
tion is called sialytransferase, which is loosely Translocation based on undecaprenol
attached to the cell membrane. The cell phosphate
extracts make undecaprenol monophosphate Because polysaccharides are polymerized from
linked sialic acid residues by transferring nucleotide derivatives of the sugars that are
sialic acid from CMP–sialic acid to undeca- made in the cytoplasm, one can assume that at
prenol phosphate, whereupon the sialic acid least the first steps in polymerization take place
is transferred to endogenous acceptors in the on the cytoplasmic surface of the cell membrane.
membrane (reviewed in ref. 10.) Because of this, there must be a mechanism to
However, doubts have been raised about transport newly synthesized polysaccharide
whether the lipid-linked sialyl residues are across the cytoplasmic membrane. One model
an obligate intermediate in the elongation postulates that the role of undecaprenol phos-
reaction.19 In one model, which does make phate is to serve as part of a transmembrane
this assumption, the sialic acid is transferred assembly process that synthesizes and moves
from CMP to undecaprenol phosphate to form completed polysaccharides to the periplasmic
sialyl monophosphorylundecaprenol, which surface of the cell membrane. This is depicted
transfers the sialic acid to the nonreducing end in Fig. 12.16A.
of the growing polysialic chain. It is not known
whether the growing polymer is attached to
Translocation via ABC transporters
undecaprenol. Moreover, the initial recep-
There is evidence for a specific transport mech-
tor for the first sialic acid is referred to sim-
anism that does not require undecaprenol. This
ply as the endogenous receptor, its identify
evidence, which was derived from examining
not being known. The endogenous receptor,
does not contain sialic acid.21 The completed the genes (kps cluster) for biosynthesis and
export of group 2 capsular polysaccharides
polysaccharide has phospholipid attached to
(including K1 and K5) in E. coli, is depicted
its reducing end, which presumably anchors
in Fig. 12.16B. (See ref. 18 for a review.)
the polysaccharide to the outer membrane. It
Mutations in a particular region of the kps
is not known at what stage of the biosynthesis
gene cluster (region 3) required for biosynthe-
of the polysialyl polymer the phospholipid is
sis of the polysaccharide result in the accumu-
attached.
lation of the polysaccharide in the cytoplasm.
The deduced amino acid sequences of the pro-
12.3.4 Pathways not involving tein products of these genes (KpsM and KpsT)
undecaprenol derivatives indicate that they belong to the family of ABC
The synthesis of some extracellular poly- transporters. Accordingly, they are referred to
saccharides, including some of the group 2 as the KpsMT transporter and probably form
K-antigen capsules such as K5 in E. coli, alg- a protein channel through which the polysac-
inate synthesized by Azotobacter vinelandii charides are translocated either during or after
and Pseudomonas aeruginosa, and cellulose biosynthesis. (See Section 17.3.3 for a discus-
synthesized by Acetobacter xylinum, does not sion of ABC transporters.)
involve undecaprenol intermediates. Where Similar genes have been found in the bio-
such synthesis has been characterized, it is found synthetic gene cluster for extracellular
that the polysaccharides are synthesized from polysaccharide synthesis in other bacteria,
nucleotide diphosphate precursors added to the including Haemophilus, Neisseria, Rhizobium,
growing oligosaccharide chain. For example, Agrobacterium, and may be widespread. Thus,
alginate, which is a linear copolymer of D-man- it can be concluded that many extracellular
nuronic and D-guluronic acid, is synthesized polysaccharides are exported through the cell
by brown algae from GDP–mannuronic and membrane in protein channels by means of
A
outer
membrane
translocation
periplasm and surface
P P assembly
C55 ligation
transfer and reaction
polymerization dephosphorylation cytoplasmic
and lipid recycling membrane
C C55 P P
repeating unit 55
synthesis P P
cytoplasm
precursors
B
translocation
outer
and surface
membrane
assembly
interaction with
periplasm periplasmic
export system
repeating unit ligation

synthesis and reaction
polymerization cytoplasmic
ATP-binding membrane
possibly involving cassette
undecaprenol transporter
precursors
cytoplasm
Fig. 12.16 Models for synthesis and export of polysaccharides in gram-negative bacteria. (A) rfe-Independent
O-polysaccharide biosynthesis in Salmonella enterica. (B) Group 2 capsular polysaccharide biosynthesis in
Escherichia coli. Abbreviations: C55, undecaprenol; P, phosphate. Source: Whitfield, C., and M. A. Valvano.
1993. Biosynthesis and expression of cell-surface polysaccharides in gram-negative bacteria. Adv. Microb.
Physiol. 35:135–246. Copyright Elsevier (1993).
energy derived at least in part from the hydro- dextrans. The synthesis is catalyzed by extracel-
lysis of ATP. (A Dp is also required.) How these lular enzymes called levansucrase or dextransu-
polysaccharides reach the outer membrane sur- crase. The reactions are as follows:
face is not clear. Various possible mechanisms
are reviewed in refs. 10 and 18. y sucrose + (fructose)n
(12.1)
→(fructose)n+y + y glucose
12.4 Levan and Dextran Synthesis
Certain bacteria (e.g., the lactic acid bacteria)
synthesize from sucrose extracellular polymers y sucrose + (glucose)n
(12.2)
of fructose, called levans, or glucose, called → (glucose)n+y + y fructose
www.ebook777.com
In both reactions a monosaccharide is trans- at the expense of ATP. The energy to make the
ferred from the disaccharide to the reducing glycosidic linkages in the cell wall polymers also
end of the growing oligosaccharide chain. The comes from ATP, which drives the synthesis of
polysaccharide chain remains attached to the the sugar–PP–lipid intermediates.
enzyme during its elongation. The dextran is a The formation of the glycosidic linkage is
glucose polymer (glucan) in which the glucose a straightforward displacement of the lipid
residues are attached via α,1-6 linkages. A differ- pyrophosphate from the C1 carbon of the
ent enzyme makes a glucan called mutan, which N-acetylmuramic acid by the hydroxyl on the
has α,1-3 linkages. Streptococcus mutans, the incoming C4 carbon of N-acetylglucosamine.
bacterium that is a principal cause of tooth The reaction is driven to completion by the sub-
decay, forms dextrans, mutans, and fructans. sequent hydrolysis of the lipid pyrophosphate to
The glucans enable the bacteria to adhere to the the lipid phosphate and inorganic phosphate.
surface of the teeth. Lipopolysaccharide synthesis can be thought
of as occurring in four stages: (1) synthesis of the
lipid A portion, (2) synthesis of the core region
12.5 Glycogen Synthesis by adding one sugar at a time to lipid A from
Certain bacteria synthesize glycogen, a poly- nucleoside diphosphate precursors, (3) synthe-
mer of glucose, as an intracellular carbon and sis of the complete repeat oligosaccharide, and
energy reserve. Glycogen is a branched poly- (4) attachment of the repeat oligosaccharide to
saccharide. The linear portion consists of glu- the core. All these events are associated with
cose residues connected by α,1-4 linkages, and the membrane. It appears that the lipid A and
the branches are attached via α,1-6 linkages. core portions are synthesized on the cytoplas-
The oligosaccharides are extended by the addi- mic side of the membrane and then translocated
tion of glucose units donated by ADP-glucose in an unknown manner to the periplasmic sur-
in a reaction catalyzed by glycogen synthetase. face. The oligosaccharide (O-antigen) subunits
A branching enzyme transfers six to eight are assembled on undecaprenyl pyrophosphate
glucose segments from the linear portion of and transferred to a growing oligosaccharide
the oligosaccharide to form an α,1-6-linked chain. When the oligosaccharide is complete, it
branch. The synthesis of nucleoside diphos- is attached to the core. How the lipopolysaccha-
phate derivatives of sugars from sugar-1-P ride enters the outer envelope is not known.
and nucleoside triphosphates was described in Capsular and extracellular polysaccharides
Section 12.1.2. are synthesized and exported in at least two
ways. One pathway is similar to the pathway for
12.6 Summary the synthesis of the oligosaccharide in lipopoly-
saccharide and involves undecaprenol interme-
There are two major problems that must be
diates. A second pathway appears to be more
solved in connection with the synthesis of the
complex, perhaps involving ABC-type trans-
cell wall: how to move the precursors through
porters to move the polysaccharide through the
the cell membrane and how to make peptide
cell membrane.
bonds outside the cell membrane, far from the
cellular ATP pools. Bacteria employ a lipid
carrier, undecaprenyl phosphate, to carry the
Study Questions
cell wall precursors through the membrane.
Interestingly, eukaryotes employ a similar com-
1. O-Antigen synthesis requires a lipid carrier,
pound, dolichol phosphate, to move oligosac-
but core synthesis does not. Offer a plau-
charide precursors through the ER membrane
sible explanation for the difference.
to synthesize glycoproteins.
The peptidoglycan peptide cross-link forms 2. During the synthesis of the pentapeptide
as a result of a transpeptidation reaction. During in the peptidoglycan precursor, an ATP is
the transpeptidation reaction, an amino group expended to make the peptide bond as each
from the diamino acid (e.g., DAP) displaces a amino acid is added to the growing pen-
terminal D-alanine. This is an exchange of a pep- tapeptide. The products are ADP and Pi.
tide bond for one that was made in the cytosol Write a plausible mechanism by which ATP
is used to provide the energy to make a pep- If one carefully examines the pathway for
tide bond between two amino acids. (Note: peptidoglycan synthesis, one can see that
mRNA and ribosomes are not involved. nucleoside triphosphates are indeed used in
The addition of each amino acid is cata- the cytoplasm in reactions that ultimately
lyzed by a separate enzyme, specific for that provide the energy for the transpeptidation
amino acid.) and transglycosidase reactions. Write these
cytoplasmic reactions, and show how they
3. Peptidoglycan and lipid A share a common
drive the transpeptidation and transgly-
pathway early in their syntheses. Outline
cosidase reactions.
the early stages of both pathways up to the
branch point.
4. Carriers of subunit moieties play important
roles in biosynthesis. Usually, the carriers 1. van Heijenoort, J. 1996. Murein synthesis,
are involved in more than one pathway. pp. 1025–1034. In: Escherichia coli and Salmonella:
What carrier is common to both peptido- Cellular and Molecular Biology, Vol. 1. F. C.
glycan and lipopolysaccharide synthesis? Neidhardt et al. (Eds.). ASM Press, Washington, DC.
2. Höltje, J.-V. 1998. Growth of the stress-bearing
5. Important enzymes in cell wall peptidogly- and shape-maintaining murein sacculus of Escherichia
can synthesis are membrane-bound trans- coli. Microbiol. Mol. Biol. Rev. 62:181–203.
glycosidases that transfer carbohydrate
3. Hartmann, E., and H. Konig. 1994. A novel path-
subunits to the growing polymer. What way of peptide biosynthesis found in methanogenic
ensures that the growing end of the polymer archaea. Arch. Microbiol. 162:430–432.
remains at the site of the transglycosidase?
4. Raetz, C. R. 1996. Bacterial lipopolysaccharide:
6. Show how ATP drives the synthesis of a remarkable family of bioactive macroamphiphiles,
pp. 1035–1063. In: Escherichia coli and Salmonella:
UDP–GlcNAc from glucose. Focus on Cellular and Molecular Biology, Vol. 1. F. C.
reactions in which phosphoryl and nucle- Neidhardt et al. (Eds.). ASM Press, Washington, DC.
otide groups are transferred. You must
5. This was formerly called 2-keto-3-deoxyoctonate.
show how ATP drives the synthesis of
UTP. There are two phosphate groups in 6. Rick, P. D. 1987. Lipopolysaccharide biosynthe-
sis, pp. 648–662. In: Escherichia coli and Salmonella
MurNAc(pentapeptide)–PP–lipid. Show
typhimurium: Cellular and Molecular Biology,
how one of them is derived from ATP. What Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press,
eventually happens to this phosphate? Washington, DC.
7. Covalent bond formation in the cytoplasm 7. Raetz, C. R. H. 1987. Structure and biosynthe-
is driven by high-energy molecules such sis of lipid A, pp. 498–503. In: Escherichia coli and
Salmonella typhimurium: Cellular and Molecular
as ATP or other nucleoside triphosphates. Biology, Vol. 1. F. C. Neidhardt et al. (Eds.). ASM
Peptidoglycan synthesis offers some exam- Press, Washington, DC.
ples of how covalent bonds can be formed
8. Mulford, C. A., and M. J. Osborn. 1983. An inter-
outside the cytoplasm without access to a mediate step in translocation of lipopolysaccharide
source of high-energy molecules such as to the outer membrane of Salmonella typhimurium.
nucleoside triphosphates, which are cyto- Proc. Natl. Acad. Sci. USA 80:1159–1163.
plasmic. Two examples are the transpeptida- 9. Bos, M. P., V. Robert, and J. Tommassen. 2007.
tion reaction, which results in the synthesis Biogenesis of the gram-negative bacterial outer mem-
of a peptide bond, and the transglycosidase brane. Ann Rev. Microbiol. 61:191–214.
reaction, which results in the synthesis of 10. Whitfield, C., and M. A. Valvano. 1993.
a glycosidic bond. These reactions are dis- Biosynthesis and expression of cell-surface polysac-
placement reactions in which a diamino acid charides in gram-negative bacteria, pp. 135–246. In:
Advances in Microbial Physics, Vol. 35. A. H. Rose
(e.g., diaminopimelic acid) displaces D-ala- (Ed.). Academic Press, New York.
nine on the pentapeptide (transpeptidation)
and the N-acetylglucosamine moiety of the 11. Rick, P. D., and R. P. Silver. 1996. Enterobacterial
common antigen and capsular polysaccharides, pp.
incoming lipid–disacharide pentapeptide 104–122. In: Escherichia coli and Salmonella: Cellular
displaces the lipid pyrophosphate in the and Molecular Biology, Vol. 1. F. C. Neidhardt et al.
growing glycan chain (transglycosidase). (Eds.). ASM Press, Washington, DC.
www.ebook777.com
12. Roberts, I. S. 1995. Bacterial polysaccharides in 17. Troy, F. A., F. E. Frerman, and E. C. Heath.
sickness and in health. Microbiology 141:2023–2031. 1971. The biosynthesis of capsular polysaccharide in
Aerobacter aerogenes. J. Biol. Chem. 246:118–133.
13. Roberts, I. S. 1996. The biochemistry and genet-
ics of capsular polysaccharide production in bacte- 18. Bliss, J. M., and R. P. Silver. 1996. Coating the
ria. Annu. Rev. Microbiol. 50:285–315. surface: a model for expression of capsular polysialic
acid in Escherichia coli K1. Mol. Microbiol. 21:
14. Whitfield, C., and I. S. Roberts. 1999. Structure,
221–231.
assembly and regulation of expression of capsules in
Escherichia coli. Mol. Microbiol. 31:1307–1319. 19. Steenbergen, S. M., and E. R. Vimr. 1990.
Mechanism of polysialic acid chain elongation in
15. Whitfield, C. 2006. Biosynthesis and assem-
Escherichia coli K1. Mol. Microbiol. 4:603–611.
bly of capsular polysaccharides in Escherichia coli.
Annu. Rev. Biochem. 75:39–68. 20. Cieslewicz, M., and E. Vimr. 1997. Reduced
polysialic acid capsule expression in Escherichia coli
16. Cuthbertson, L., I. L. Mainprize, J. H.,
K1 mutants with chromosomal defects in kpsF. Mol.
Naismith, and C. Whitfield. 2009. Pivotal roles of
the outer membrane polysaccharide export and
polysaccharide copolymerase protein families in 21. Rohr, T. E., and F. A. Troy. 1980. Structure
export of extracellular polysaccharides in gram- and biosynthesis of surface polymers containing
negative bacteria. Microbiol. Mol. Biol. Rev. 73: polysialic acid in Escherichia coli. J. Biol. Chem.
155–177. 255:2332–2342.
13
Inorganic Metabolism
Inorganic molecules such as derivatives of sul- (e.g., lake sediments). However, only a few
fur, nitrogen, and iron are used by prokaryotes of the iron and manganese reducers have
in a variety of metabolic ways related to energy been isolated, and little is known about their
metabolism and biosynthesis. physiology.
3. There are oxidative pathways in which inor-
1. Assimilatory inorganic metabolism refers to
ganic compounds such as H2, NH3, So, H2S,
the reduction of inorganic compounds and
and Fe2+, rather than organic compounds,
their incorporation into organic compounds.
are oxidized as a source of electrons and
These include the reduction of oxidized
energy. Organisms that derive their energy
forms of nitrogen and sulfur and their incor-
and electrons for biosynthesis in this way are
poration into organic materials (e.g. amino
called chemolithotrophs. (If they use CO2 as
acids and nucleotides). Most prokaryotes
the sole or major source of carbon, they are
can do this. In addition, many prokaryotes
called chemolithoautotrophs.)
can utilize nitrogen gas as a source of nitro-
gen, a process called nitrogen fixation.
2. There are dissimilatory pathways in which 13.1 Assimilation of Nitrate and
inorganic compounds are used instead of Sulfate
oxygen as electron acceptors, a process Many bacteria can grow in media in which the
called anaerobic respiration. The reduced only sources of nitrogen and sulfur are inor-
products are excreted into the environment. ganic nitrate salts and sulfate salts. The nitrate
During anaerobic respiration a ∆p is created is reduced to ammonia, and the ammonia is
in the same way as during aerobic respira- incorporated into the amino acids glutamine
tion (see Section 5.6). For example, many and glutamate using the GS/GOGAT system
facultative anaerobes can use nitrate as an (see Section 10.3.1). Glutamate and glutamine
electron acceptor, reducing it to ammonia are the sources of amino groups for the other
or nitrogen gas. Several obligate anaerobes nitrogen-containing organic compounds. The
use sulfate as an electron acceptor, reducing sulfate is reduced to H2S, which is immediately
it to hydrogen sulfide. There also exist bac- incorporated into the amino acid cysteine via
teria that can use Fe3+ or Mn4+ as an electron the O-acetylserine pathway shown earlier (Fig.
acceptor during anaerobic growth.1–3 The 10.19). Cysteine, in turn, is the source of sulfur
latter organisms are responsible for most of for other organic molecules (e.g., such as methi-
the reduction of iron and manganese that onine, coenzyme A, as CoASH; and acyl carrier
takes place in sedimentary organic matter protein, as ACPSH).
335
www.ebook777.com
Fig. 13.1 Assimilatory nitrate reduction. This pathway is present in all bacteria that reduce nitrate to
ammonia, which is then incorporated into cell material. The enzymes are found in the cytosol and are not
coupled to ATP formation. Nitrate is reduced via two-electron steps to nitrite, nitroxyl, hydroxylamine,
and ammonia. The ammonia is incorporated into organic carbon via glutamine synthetase (GS) and the
GOGAT enzyme, or via glutamate dehydrogenase. (See Section 10.3.1.) Enzymes: 1, nitrate reductase;
2, nitrite reductase. Abbreviations: glu, glutamate; α-kg, α-ketoglutarate. (α-Ketoglutarate is also called
2-oxoglutarate, or 2OG.)
13.1.1 Nitrate assimilation 13.1.2 Sulfate assimilation

Nitrate can serve as the source of cellular nitrogen Many bacteria can use sulfate (SO42–) as their
for plants, fungi, and many bacteria. Although principal source of sulfur. The sulfate is first
some bacteria (e.g., Klebsiella pneumoniae) reduced to sulfide (H2S and HS–) and then incor-
assimilate nitrate during aerobic growth, certain porated into cysteine. (See note 6 for an expla-
other bacteria (e.g., E. coli, Salmonella typhimu- nation why H2S and HS– are the major species
rium) assimilate nitrate only during anaerobic of sulfide in the cytoplasm.) Since the oxidation
growth. (See ref. 4 for a review of nitrate trans- level of sulfur in SO42– is +6 and in S2– is –2, a total
port and reduction.) The nitrate is first reduced to of eight electrons is required to reduce sulfate
ammonia. Since the oxidation state for the nitro- to sulfide.
gen in nitrate (NO3–) is +5 and for the nitrogen The first step in the reduction is the formation
in ammonia (NH3) it is –3, eight electrons must of adenosine-5´-phosphosulfate (APS) cata-
be transferred to nitrate to reduce it to ammo- lyzed by the enzyme ATP sulfurylase (Fig. 13.2,
nia. The enzymes involved in the assimilatory reaction 1). Here sulfate acts as a nucleo-
reduction of nitrate to ammonia are cytoplas- phile and displaces pyrophosphate (PPi). (See
mic nitrate reductase, which reduces nitrate to Section 8.1.1 for a discussion of nucleophilic
nitrite, and cytoplasmic nitrite reductase, which displacements.) The pyrophosphate is subse-
reduces nitrite to ammonia. The electron donors quently hydrolyzed to inorganic phosphate,
for the nitrate reductase can be NADH (via a thus driving the synthesis of APS to comple-
separate NADH oxidoreductase), ferredoxin, tion. There is a sound thermodynamic reason
or flavodoxin, depending upon the bacterium, as for making the AMP derivative of sulfate prior
explained in note 5. The electron donor for the to its reduction. Attaching the sulfate to AMP
major nitrite reductase in E. coli is NADH. The raises the reduction potential, making APS a
electron transport pathway is shown in Fig. 13.1 better electron acceptor than free sulfate. The
as proceeding in two-electron transfer steps. reduction potential of sulfate to sulfite is very
The ammonia that is formed is incorporated low (–520 mV), such that its reduction, even by
into glutamine via glutamine synthase (GS). The H2 (= –420 mV) is endergonic.
glutamine then serves as the amino donor for The next reaction, 2, is the phosphoryla-
pyrimidine, purine, and amino sugar biosyn- tion of APS to form adenosine-3′-phosphate-
thesis (Sections 10.2.2, 10.2.3, and 12.1.2), as 5′-phosphosulfate (PAPS), catalyzed by APS
well as for the synthesis of glutamate (glutamate kinase. The reaction is an attack by the 3′
synthetase). Glutamate is the amino donor for hydroxyl of the ribose of APS on the terminal
amino acid biosynthesis via the transamination phosphate of ATP, displacing ADP. The PAPS
reactions described in Section 10.3.1. is then reduced to sulfite with the release of
inorganic metabolism 337
Fig. 13.2 Assimilatory sulfate reduction. Enzymes: 1, ATP sulfurylase; 2, APS phosphokinase; 3, PAPS
reductase; 4, sulfite reductase; 5, O-acetylserine sulfhydrylase. R(SH)2 is reduced thioredoxin.
AMP-3′-phosphate (reaction 3). The reduc- reduction and employ this pathway when oxy-
tant is a sulfhydryl protein called thioredoxin, gen is not available, the use of sulfate as an elec-
which in turn accepts electrons from NADPH. tron acceptor is restricted to obligate anaerobes,
(Thioredoxin is used in other metabolic path- bacteria called sulfate reducers.
ways as a reductant. For example, recall from
Section 10.2.5 that thioredoxin also reduces 13.2.1 Dissimilatory nitrate reduction
the nucleoside diphosphates to the deoxynucle- Dissimilatory nitrate reduction (nitrate respira-
otides. The sulfite is then reduced by NADPH to tion) is usually facultative and occurs as a sub-
hydrogen sulfide (H2S) (reaction 4). Hydrogen stitute for aerobic respiration when the oxygen
sulfide is very toxic and does not accumulate. levels become very low. It takes place in mem-
The sulfide enzymatically displaces acetate from branes, and a ∆p is usually made. The products
O-acetylserine to form cysteine (see Fig. 10.19). of nitrate respiration can be nitrite, ammonia,
The AMP-3′-phosphate is hydrolyzed to AMP or nitrogen gas.
and Pi, thus helping to drive the overall reaction
to completion. Note that three ATPs are used to
reduce sulfate to sulfide, two to make the PAPS Denitrification
derivative, and a third to phosphorylate the When the nitrate (or nitrite) is reduced to nitro-
AMP released from AMP-3′-phosphate to ADP gen gas (or nitric oxide gas, NO, or nitrous
(reaction 3). (The ADP is then converted to ATP oxide gas, N2O), the process is called denitri-
via respiratory phosphorylation or substrate- fication. Denitrification, which can be an eco-
level phosphorylation.) logically important drain of nitrogen from the
soil, occurs when conditions become anaero-
bic (e.g., in water-logged soil) and also during
13.2 Dissimilation of Nitrate and composting and sludge digestion. The denitri-
Sulfate fiers reduce the available nitrate in the soil, but
In the dissimilatory pathways, the nitrate and they are also important in anaerobic niches for
sulfate are used as electron acceptors during the breakdown of undesirable biodegradable
anaerobic respiration. The reduced products materials, plant materials, complex organic
are excreted rather than being incorporated compounds, and so forth.7 The breakdown of
into cell material. Whereas many facultative organic materials by denitrifiers is usually more
anaerobes are capable of dissimilatory nitrate complete than decomposition via a fermentative
www.ebook777.com
process. Many bacteria denitrify, but the most α-ketoglutarate dehydrogenase is replaced by
commonly isolated denitrifiers are Alcaligenes a ferredoxin-dependent enzyme. The path-
and Pseudomonas species, and to a lesser extent way is called the reductive tricarboxylic acid
Paracoccus denitrificans. The electron transport pathway, although it can be used in the oxida-
pathway for denitrification by P. denitrificans is tive direction (Section 14.1.9). Other sulfate
summarized in Section 5.7.2. reducers (e.g., Desulfobacterium autotrophi-
cum, Desulfotomaculum acetooxidans, and
13.2.2 Dissimilatory sulfate reduction the archaeon Archaeoglobus fulgidus) oxidize
General description of the sulfate reducers acetyl–CoA to CO2 by means of the acetyl–CoA
The sulfate reducers are heterotrophic anaerobes pathway described in Section 14.1.3. The use of
that grow in anaerobic muds, mostly in anaero- these pathways in the reductive direction in auto-
bic parts of fresh water, and in seawater.8 They trophic CO2 fixation by facultatively autotrophic
carry out anaerobic respiration during which sulfate reducers is also described in Sections
sulfate is reduced to H2S. Some can be grown 14.1.3 and 14.1.4. Many sulfate reducers are
autotrophically with H2 as the source of electrons known to be able to ferment pyruvate to acetate,
and SO42– as the electron acceptor. Formerly the or to acetate and propionate in the absence of
sulfate reducers were believed to have only a sulfate. This is described in Section 15.8.
limited variety of carbon sources (e.g., formate,
lactate, pyruvate, malate, fumarate, ethanol, The path of electrons to sulfate in
and a few other simple compounds). Recently, Desulfovibrio
however, it has been realized that, depending Desulfovibrio carries out an anaerobic respira-
upon the species, many other carbon sources can tion during which electrons flow in the cell mem-
be used, including straight-chain alkanes and a brane to sulfate as the terminal electron acceptor,
variety of aromatic compounds. Sulfate reducers reducing it to H2S. Electron flow is coupled to
comprise a very diverse group of organisms that the generation of a ∆p, which is used for ATP
include both gram-positive and gram-negative synthesis via respiratory phosphorylation. As
bacteria, as well as archaea. An example of the stated earlier, these electrons may come from the
latter is Archaeoglobus, a hyperthermophile iso- oxidation of organic compounds (e.g., lactate).
lated from sediments near hydrothermal vents. Since dissimilatory sulfate reduction takes place
Gram-positive, spore-forming sulfate reducers in membranes, involves cytochromes, and gener-
belong to the genus Desulfotomaculum, which ates a ∆p, it is very different from the assimila-
is very diverse. The most prominent of the gram- tory pathway, which is a cytoplasmic pathway
negative sulfate reducers belong to the genus and does not generate a ∆p or ATP.
Desulfovibrio, which is also phylogenetically A pathway of electron transport has been pro-
diverse. posed for the genus Desulfovibrio (Fig. 13.3). It
Traditionally, the sulfate reducers are divided is called the hydrogen cycling model. Lactate is
into two physiological groups, I and II. Those in oxidized to pyruvate in the cytoplasm, yielding
group I cannot oxidize acetyl–CoA to CO2 and two electrons (Fig. 13.3, reaction 1). The oxi-
therefore excrete acetate when they are grown on dation of lactate to pyruvate is catalyzed by a
certain carbon sources (e.g., lactate, ethanol). The membrane-bound lactate dehydrogenase that
group I genera include Desulfovibrio and most probably is a flavoprotein. The pyruvate is then
Desulfotomaculum species. Group II organisms oxidized to acetyl–CoA and CO2 by pyruvate–
can oxidize acetyl–CoA to CO2. Group II sulfate ferredoxin oxidoreductase, an enzyme found in
reducers are found in several genera, including other anaerobes (reaction 2). (See Section 8.3.2
Desulfotomaculum and Desulfobacter. for a description of the pyruvate–ferredoxin
There exist two pathways for oxidizing oxidoreductase reaction.) The acetyl–CoA
acetyl–CoA anaerobically to CO2: a modified is used to generate ATP via a substrate-level
citric acid cycle and the acetyl–CoA pathway. phosphorylation, using two enzymes common
Desulfobacter has a modified citric acid cycle in bacteria, phosphotransacetylase and acetate
that resembles that found in aerobes except kinase (reactions 3 and 4). (See Section 8.3.2 for
that (1) instead of a citrate synthase there is an a description of the phosphotransacetylase and
ATP–citrate lyase, and (2) the NAD+-linked acetate kinase reactions.)
Fig. 13.3 Pathway for dissimilatory sulfate reduction in Desulfovibrio. Enzymes: 1, lactate dehydrogenase;
2, pyruvate–ferredoxin oxidoreductase; 3, phosphotransacetylase; 4, acetate kinase; 5, cytoplasmic hydro-
genase; 6, periplasmic hydrogenase; 7, ATP sulfurylase; 8, pyrophosphatase; 9, APS reductase; 10 sulfite
reductase.
Since each lactate that is oxidized to acetyl– ∆p, there would be no ATP left over for growth.
CoA yields four electrons, two lactates must be Some strains of Desulfovibrio can grow on CO2
oxidized to provide the eight electrons to reduce and acetate as the sole sources of carbon, and
one sulfate to sulfide. The model proposes the ∆p produced by the redox reaction between
that the electrons are transferred from lactate H2 and SO42– is the sole source of energy.
dehydrogenase and pyruvate–ferredoxin oxi- There are also facultatively autotrophic
doreductase to a cytoplasmic hydrogenase and strains of sulfate reducers that grow on CO2, H2,
then to H+, producing H2 (reaction 5), and the H2 and SO42– (e.g., Desulfobacterium autotrophi-
diffuses out of the cell into the periplasm. In the cum). All these strains derive their ATP from
periplasm, the H2 is oxidized by a periplasmic the ∆p created during sulfate reduction. Serious
hydrogenase and the electrons are transferred reservations, however, have been expressed
to cytochrome c3 (reaction 6). From cyt c3 the over whether free H2 is actually an electron
electrons travel through a series of membrane- carrier during lactate oxidation by sulfate in
bound electron carriers to APS reductase and Desulfovibrio (i.e., whether the hydrogen-
sulfite reductase in the cytosol (reactions 7, 9, cycling model is generally valid). (See note 9.)
10). A pyrophosphatase pulls the sulfurylation Clearly, there is still much to be learned regard-
of ATP to completion (reaction 8). ing how various species of Desulfovibrio couple
Note that according to the scheme proposed electron transport to the generation of a ∆p.
in Fig. 13.3, the inward flow of electrons across
the membrane leaves the protons from the 13.3 Nitrogen Fixation
hydrogen on the outside, thus generating a ∆p. For reviews, see refs. 10 through 14. From
An examination of the scheme reveals that the an ecological point of view, one of the most
∆p is necessary for growth. The two ATPs made important metabolic processes carried out
via substrate-level phosphorylation from the by prokaryotes is nitrogen fixation (i.e., the
two moles of acetyl–CoA are used up in reduc- reduction of N2 to NH3). As far as is known,
ing the SO42–. This follows because, after reduc- eukaryotes have not evolved this capability.
tion to sulfite, AMP is produced, and the energy Since fixed nitrogen is usually limiting for plant
equivalent of two ATPs is required to make one growth, the ability of prokaryotes to fix nitro-
ATP from one AMP. Thus, without using the gen is necessary to maintain the food chain. The
www.ebook777.com
ammonia that is produced via nitrogen fixation 2. Nonleguminous plants in symbiotic rela-
is incorporated into cell material by means of tionships with nitrogen-fixing bacteria. For
glutamine synthetase and glutamate synthase example, the water fern Azolla makes small
(Section 10.3.1). pores in its fronds within which a nitrogen-
It used to be thought that nitrogen fixa- fixing cyanobacterium, Anabaena azollae,
tion was restricted to a few bacteria such as lives. The Azolla–Anabaena symbiotic sys-
Azotobacter, Rhizobium, Clostridium, and tem is used to enrich rice paddies with fixed
the cyanobacteria. It is now realized that nitrogen. Another example is the alder tree,
nitrogen fixation is a capability widespread which has nitrogen-fixing nodules containing
among many different families of bacte- Frankia, a bacterium resembling the strep-
ria and also occurs in archaea. The enzyme tomycetes. A third example is Azospirillum
responsible for nitrogen fixation, nitroge- lipoferum, which is a nitrogen-fixing rhizo-
nase, is very similar in the different bacte- sphere bacterium that is found around the
ria, and the nitrogen fixation genes have roots of tropical grasses.
homologous regions. This has led to the sug- 3. Many free-living soil and aquatic prokary-
gestion that the nitrogenase gene may have otes. As indicated at the beginning of
been transferred laterally between different the chapter, many different prokaryotes
groups of bacteria. Organisms that fix nitro- fix nitrogen. They include Azotobacter,
gen encompass a wide range of physiological Clostridium, certain species of Desulfovi-
types and include aerobes, anaerobes, facul- brio, the photosynthetic bacteria, and vari-
tative anaerobes, autotrophs, heterotrophs, ous cyanobacteria. Nitrogen fixation is not
and phototrophs. confined to the (eu)bacteria. Some metha-
Nitrogen fixation takes place when N2 is the nogenic archaea have been reported to be
only or the major source of nitrogen because nitrogen fixers.15
the genes for nitrogen fixation are repressed by
Nitrogen fixation is sensitive to oxygen. The
exogenously supplied sources of fixed nitrogen
enzyme that fixes nitrogen, nitrogenase, is
(e.g., ammonia.) It is remarkable that biologi-
inhibited by oxygen. Thus for many prokary-
cal nitrogen reduction takes place at all. The
otes, nitrogen fixation takes place only under
nitrogen molecule is so stable that very high
anaerobic or microaerophilic conditions.
pressures and temperatures in the presence of
However, some prokaryotes can fix nitrogen
inorganic catalysts are necessary to make it
while growing in air. As described later, they
reactive in nonbiological systems. Industrially,
have evolved systems to protect the nitrogenase
nitrogen gas is reduced to ammonia by means
from oxygen.
of the Haber process, which has pressure and
temperature requirements of 200 atm and 800
13.3.2 The nitrogen fixation pathway
°C, respectively. Yet prokaryotes carry out the
Nitrogenase
reduction at atmospheric pressures and ordi-
nary temperatures. As mentioned, the enzyme that reduces nitro-
gen gas is called nitrogenase.16,17 The major
13.3.1 The nitrogen-fixing systems nitrogenase in nitrogen-fixing organisms is
a molybdenum-containing enzyme that con-
Biological nitrogen-fixing systems are found in
sists of two multimeric proteins. One of these,
the following organisms.
usually called the molybdenum–iron protein
1. Rhizobium, Sinorhizobium, Bradyrhizobium, (MoFe protein), is also known as dinitrogenase,
and Azorhizobium in symbiotic relation- or component I. The second protein, called
ships with leguminous plants (soybeans, the iron protein (Fe protein), is also known as
clover, alfalfa, string beans, peas: i.e., plants dinitrogenase reductase, or component II. Both
that bear seeds in pods). The bacteria infect proteins contain FeS centers.
the roots of the plants and stimulate the pro- The MoFe protein is a tetramer (α2β2) of four
duction of root nodules, within which the polypeptides. When it is extracted with certain
bacteria fix nitrogen. The plant responds by solvents, a cofactor called the iron–molybde-
feeding the bacteria organic nutrients made num cluster (FeMoco) is removed. The cofac-
during photosynthesis. tor contains approximately half of the iron and
Fig. 13.4 The nitrogenase reaction. The enzyme system consists of two components. Component I is called
the molybdenum–iron protein (MoFe protein) or dinitrogenase. Component II is called the iron protein (Fe
protein) or dinitrogenase reductase. Both the proteins contain FeS centers. A low potential reductant, either
ferrodoxin or flavodoxin, reduces component II, which transfers the electrons to component I. Component
I reduces N2. There is always some H2 produced. ATP is required even though the overall reduction of N2 by
ferredoxin or flavodoxin is an exothermic reaction.
labile sulfide of the protein. Thus, the MoFe The details of the electron transport pathway
protein contains FeMoco plus additional FeS through the proteins and the role of ATP are not
centers. (See note 18 for more information well understood. (See note 19 for a more com-
about the MoFe protein.) plete discussion of electron flow in the MoFe
The Fe protein is a dimer (γ2) of two identi- protein and the role of ATP.) However, it is clear
cal polypeptides. The dimer contains a single that the hydrolysis of at least two ATP mole-
Fe4S4 cluster, which is responsible for reducing cules is required for every electron transferred
FeMoco during nitrogen fixation. between the two proteins. Therefore, about 16
Twenty-one genes, called the nif genes, have moles of ATP are necessary to convert a mole of
been identified as necessary for the expression nitrogen gas to 2 moles of ammonia. This is a
and regulation of the nitrogenase enzyme sys- great deal of energy (about 800 kJ). Recall that
tem in Klebsiella pneumoniae. The presence only 2 moles of ATP are generated during the
of sufficient NH4+ represses the synthesis of fermentation of one mole of glucose to lactic
nitrogenase. Other nitrogen sources (nitrates, acid, and 38 moles of ATP are produced during
amino acids, urea) also suppress the synthesis of the complete oxidation of one mole of glucose to
nitrogenase, probably by producing ammonia. carbon dioxide and water. Therefore, an organ-
The regulation of expression of the nif genes is ism growing on nitrogen gas must consume a
described in Section 19.4. large fraction of the ATP it produces to reduce
the nitrogen to ammonia. Not surprisingly,
The nitrogenase reaction bacteria do not fix nitrogen gas if an alternative
The nitrogenase reaction is a series of reduc- source of nitrogen is present. As stated previ-
tions during which half a mole of N2 and one ously, this is because the nitrogen fixation genes
mole of H+ are reduced to one mole of NH3 and are repressed when nitrogen sources other than
half a mole of H2 (Fig. 13.4). N2 are available.
During nitrogen reduction, protons are also
Nitrogenase reaction reduced to hydrogen gas. The production of
4e– + 0.5N2 + 4H+ + 8 ATP hydrogen gas appears to be wasteful of elec-
trons and ATP. Indeed, some bacteria (e.g.,
→ NH3 + 0.5H2 + 8 ADP + 8 Pi Azotobacter) are very good at scavenging the
Since the oxidation state of N2 is 0 and the oxi- hydrogen gas with a hydrogenase. The hydro-
dation state of the nitrogen in NH3 is –3, this gen gas is used to generate electrons for the
reaction requires three electrons for every nitro- nitrogenase.
gen atom. A fourth electron is transferred to a
proton to reduce it to hydrogen gas. The elec- Other nitrogenases
trons are transferred one at a time in an ATP- Although the nitrogenase just described is cer-
dependent reaction from the Fe4S4 cluster in the tainly the major one, other nitrogenases have
Fe protein to the MoFe cluster in the MoFe pro- been discovered. For example, Azotobacter
tein and from there to N2. vinelandii can synthesize three nitrogenases,
www.ebook777.com
encoded by three different genes. The three nitro- it has been suggested that photosystem I and an
genases have different metal requirements. The electron donor other than water might reduce
molybdenum-containing nitrogenase (nitroge- nitrogenase in the heterocyst, and similarly
nase I) is made when the organism is grown in that a light-generated reductant might drive the
media containing molybdenum. Nitrogenase reduction of nitrogenase in the green sulfur bac-
II is a vanadium-containing nitrogenase that is teria. Recall that these two photosystems have
synthesized when the cells are grown in media reaction centers that produce a reductant at a
lacking molybdenum but containing vanadium. sufficiently low potential to reduce ferredoxin
Instead of FeMoco, the nitrogenase contains a (Sections 6.3 and 6.4).
vanadium cofactor called FeVaco. When both
molybdenum and vanadium are lacking in the Protecting the nitrogenase from oxygen
media, Azotobacter makes a third nitrogenase, All nitrogenases are rapidly inactivated by oxy-
called nitrogenase III, which requires only iron. gen in vitro. Some nitrogen-fixing microorgan-
isms (e.g., the clostridia or nitrogen-fixing sulfate
The source of electrons for nitrogen reduction reducers) are strict anaerobes. Others (e.g., the
In most known systems the nitrogenases are purple photosynthetic bacteria, Klebsiella spp.)
reduced by ferredoxins (FeS proteins) or fla- are facultative anaerobes; that is, they can grow
vodoxins (flavoproteins). These electron car- aerobically or anaerobically, although they fix
riers have midpoint potentials sufficiently low nitrogen only when they are living anaerobi-
to reduce nitrogenase (–400 to –500 mV). The cally. Some microorganisms are microaero-
source of electrons for the ferredoxins and fla- philic; that is, they can grow only in low levels
vodoxins varies with the metabolism of the of oxygen. Some of these are nitrogen fixers and
organism. For example, during heterotrophic can grow on nitrogen gas under microaerophilic
anaerobic growth, the oxidation of pyruvate conditions. But, what about strict aerobes, or
to acetyl–CoA and carbon dioxide gener- the cyanobacteria that produce oxygen in the
ates reduced ferredoxin (pyruvate:ferredoxin light? Various strategies have evolved to pro-
oxidoreductase, Section 8.3.2) or flavodoxin tect the nitrogenase in microorganisms that fix
(pyruvate:flavodoxin oxidoreductase), which nitrogen in air.20,21
donates electrons to nitrogenase. (Clostridium The Azotobacter species have a very active
pasteurianum uses the ferredoxin enzyme, respiratory system that is suggested to utilize
whereas Klebsiella pneumoniae uses the flavo- oxygen rapidly enough to lower the intracellu-
doxin enzyme.) lar concentrations in the vicinity of the nitroge-
The path of electrons to nitrogenase in aero- nase. These organisms are also able to protect
bic and phototrophic bacteria and in cyanobac- their nitrogenase from inactivation by associat-
teria is not as well understood but is thought ing it with protective proteins. For example, the
to involve ferredoxin or flavodoxin as the nitrogenase of Azotobacter can be isolated as an
immediate electron donor. The ferredoxin or air-tolerant complex with a redox protein called
flavodoxin might be reduced by NAD(P)H (or the Shethna, FeS II, or protective protein.
some other electron carrier) generated during The rhizobia in root nodules exist in plant ves-
metabolism, for example, during the oxidation icles in the inner cortex of the nodule as modified
of carbohydrate. However, a source of energy cells called bacteroids. Oxygen-sensitive micro-
(i.e., the proton motive force) would be neces- electrodes can be used to show that the oxygen
sary to drive the reduction of ferredoxin and concentrations in the inner cortex are much lower
flavodoxin by NAD(P)H (reversed electron than those in the surrounding tissue. The oxygen
transport, Section 5.5) because the midpoint levels in the vicinity of the bacteroid-containing
potential of the NAD(P)+/NAD(P)H couple is cells are controlled by a boundary of densely
–320 mV, versus –400 to –500 mV for the ferre- packed plant cells between the inner and outer
doxins and flavodoxins. cortex. The control of oxygen access to the inner
Alternatively, the cyanobacteria and green cortex is achieved by regulating the intercellular
sulfur bacteria could use light energy to reduce spaces, which either are filled with air or contain
the electron donor for nitrogenase during photo- variable amounts of water, within the boundary
synthetic noncyclic electron flow. For example, layer. However, the bacteroids are dependent
upon oxygen for respiration, and within the nod- has generated much interest. Such cyanobacte-
ule a plant protein called leghemoglobin binds ria include the unicellular Gloethece (formerly
oxygen and delivers it to the bacteroids. called Gloeocapsa), certain strains of the unicel-
It is not understood how the unicellular lular Synechococcus, and the marine filamentous
cyanobacteria or the nonheterocystous filamen- organism Trichodesmum. Nitrogen fixation by
tous cyanobacteria protect their nitrogenase from unicellular and filamentous nonheterocystous
photosynthetically produced or atmospheric cyanobacteria, including protection of the nitro-
oxygen. In fact, most of these strains fix nitrogen genase, have been reviewed by Bergman et al.22
only when grown under anoxic or micro-oxic Several filamentous cyanobacteria protect their
conditions. Anoxic conditions are maintained nitrogenase from oxygen by means of special
in the light only when photosynthetic oxygen cells called heterocysts, in which the organisms
production is experimentally inhibited. Under are able to fix nitrogen under aerobic growth
micro-oxic (also called microaerobic) growth conditions.
conditions, atmospheric oxygen is absent, but
photosynthetically produced oxygen is present. Heterocysts
It is presumed that intracellular oxygen levels Filamentous cyanobacteria such as Anabaena
under these conditions are significantly lower and Nostoc protect their nitrogenase by dif-
than atmospheric levels, hence the term “micro- ferentiating approximately 5 to 10% of their
oxic.” Whether this is indeed the case, especially vegetative cells into special nitrogen-fixing cells
under conditions of high illumination, has not called heterocysts in the absence of combined
been shown. However, some unicellulular and nitrogen23 (Fig. 13.5). Heterocysts differ from
nonheterocystous filamentous cyanobacteria do vegetative cells in the following respects.
fix nitrogen under aerobic conditions (i.e., at oxy-
gen concentrations that are approximately equal 1. Heterocysts express the nitrogenase enzymes.
to atmospheric). How they protect their nitro- 2. They have only photosystem I (PS I) and
genase from oxygen is not known, but the topic therefore do not produce oxygen.
Fig. 13.5 Heterocyst interactions with vegetative cells in Anabaena. The heterocyst reduces dinitrogen to
ammonia, which is incorporated into glutamine via glutamine synthetase. The glutamine then enters the vege-
tative cells, where it serves as a source of fixed nitrogen for growth. The vegetative cells fix carbon dioxide into
carbohydrate by means of the Calvin cycle. Some of the carbohydrate enters the heterocyst, where it serves as
a source of carbon and NADPH. The NADPH reduces the nitrogenase via ferredoxin:NADP oxidoreductase
and ferredoxin. ATP is made via cyclic photophosphorylation in the heterocyst using PS I. Since PS II is lacking
in the heterocyst, oxygen is not produced there.
www.ebook777.com
3. They do not fix CO2. from the oxidation of inorganic compounds;

4. They are surrounded by a thick cell wall con- examples are H2, CO, NH3, NO2–, H2S, S°,
sisting of glycolipid and polysaccharide that S2O32– and Fe2+. This type of metabolism is
is believed to serve as a permeability barrier called lithotrophy, and the organisms are called
to atmospheric oxygen. lithotrophs.24
5. They are not dividing.
The ATP made during cyclic photophosphoryla- 13.4.1 The lithotrophs
tion by photosystem I is used to fix the nitrogen. The lithotrophs are physiologically diverse and
But where do the heterocysts get the reducing exist among several different groups of bacteria
power? What happens is that the heterocysts fix and archaea. Many of the lithotrophs are aer-
N2 and feed reduced nitrogen to the rest of the obes; that is, they carry out electron transport
filament (Fig. 13.5). In turn, the heterocyst is fed with oxygen as the terminal electron accep-
carbohydrate made from CO2 by the vegetative tor. However, some are facultative anaerobes,
cells. The heterocyst oxidizes the carbohydate, using nitrate or nitrite as the electron accep-
reducing ferredoxin, which, in turn, reduces the tor when oxygen is unavailable, and a few are
nitrogenase. Therefore, this is a complex situa- obligate anaerobes that use sulfate or CO2 as
tion in which two different cell types in the fila- the electron acceptor. For most of the lithotro-
ment are feeding each other. phs, the sole or major source of carbon is CO2.
Since the heterocysts do not have photosys- Such organisms are called chemoautotrophs or
tem II, they do not produce oxygen. However, chemolithoautotrophs. The lithotrophs vary with
there is still the problem of protecting the nitro- regard to the autotrophic CO2 fixation pathway
genase from atmospheric oxygen. It has been that they use (Chapter 14). Other lithotrophs
suggested that the crystalline glycolipid and are facultatively heterotrophic. The facultative
polysaccharide cell wall may present a diffusion heterotrophs include all the bacterial hydrogen
barrier to oxygen. Heterocysts also have a high oxidizers, some sulfur-oxidizing thiobacilli, and
rate of respiration, which presumably also con- some thermophilic iron-oxidizing bacteria. Some
tributes to a low internal oxygen environment. representative species are listed in Table 13.1.
Table 13.2 lists midpoint potentials of the
inorganic substrates at pH 7. Reversed elec-
13.4 Lithotrophy tron transport is required to generate NAD(P)
While most organisms derive energy from oxi- H when the electron donor is more electroposi-
dizing organic nutrients (chemo-organotrophs) tive than the NAD+/NADH couple (E´0 = –0.32
or from the absorption of light (phototrophs), V). Reversed electron flow is driven by the ∆p
there exist many prokaryotes that derive energy (Section 5.5). Because of the relatively small ∆Eh
Table 13.1 Chemoautotrophs

Bacterial group Typical species Electron Electron Carbon Product
donor acceptor source
Hydrogen oxidizing Alcaligenes eutrophus H2 O2 CO2 H2O
Carbon monoxide oxidizing Pseudomonas CO O2 CO2 CO2
(carboxydobacteria) carboxydovorans
Ammonium oxidizing Nitrosomonas europaea NH4+ O2 CO2 NO2–
–
Nitrite oxidizing Nitrobacter winogradskyi NO 2
O2 CO2
2–
Sulfur oxidizing Thiobacillus thiooxidans S, S2O 3
O2 CO2 SO42–
2+
Iron oxidizing Thiobacillus ferrooxidans Fe O2 CO2 Fe3+
Methanogenic Methanobacterium H2 CO2 CO2 CH4
thermoautotrophicum
Acetogenic Acetobacterium woodii H2 CO2 CO2 CH3COOH
Source: Schlegel, H. G., and H. W. Jannasch. 1992. Prokaryotes and their habitats, pp. 75–125. In: The Prokaryotes,
Vol. I. A. Balows et al. (Eds.). Springer-Verlag, Berlin.
Table 13.2 Redox potentials of inorganic

compoundsa
Compound E´0 (mV)
CO2/CO –540
SO42–/ HSO3– –516b
H+/H2 –414b
S°/HS– –270b
HSO3–/ HS– –116b
NO3– / NO2– +420
NO3– / NH3 +440
Fe3+/Fe2+ +772b
O2/H2O +818b
a
For comparison, the standard potential at pH 7 Fig. 13.6 Cell yields versus available energy in inor-
for NAD+/NADH is –320 mV.
b
ganic and organic electron sources. Source: Adapted
Data from Thauer, R. K., K. Jungermann,
and K. Decker. 1977. Energy conservation in
from Brock, T. D., and M. T. Madigan. 1991.
chemotrophic anaerobic bacteria. Bacteriol. Rev. Biology of Microorganisms. (Reprinted by permis-
41:100–180. sion of Prentice Hall, Upper Saddle River, NJ.)
between the inorganic electron donor and oxy- Rhizobium. Anaerobic hydrogen oxidizers
gen, and the need to reverse electron transport, include some sulfate-reducing bacteria, as well
the energy yields, and therefore the cell yields, as some archaea, including methanogens grow-
are relatively small in comparison to growth on ing autotrophically on CO2 and certain sulfur-
organic substrates (Fig. 13.6). dependent archaea that use elemental sulfur as
the electron acceptor, reducing it to hydrogen
Aerobic hydrogen-oxidizing bacteria and sulfide.
carboxydobacteria
The hydrogen-oxidizing bacteria are usually Ammonia-oxidizing bacteria
facultative and can live either autotrophically Bacteria that oxidize ammonia as a source
or heterotrophically. However, some always of energy are called nitrifiers.25 There are at
require organic carbon for growth, and these least five genera of nitrifiers: Nitrosomonas,
are called chemolithoheterotrophs. The hydro- Nitrosococcus, Nitrosospira, Nitrosolobus,
gen oxidizers can be found in aerobic or anaer- and Nitrosovibrio. All these nitrifiers are aero-
obic environments where H2 is available. The bic obligate chemolithoautotrophs that assimi-
hydrogen gas itself is produced as a by-product late CO2 via the Calvin cycle. Ammonia that is
of nitrogen fixation (e.g., in the rhizosphere produced in the anaerobic niches by deamina-
of nitrogen-fixing plants and in cyanobacte- tion of amino acids, urea, or uric acid, or via
rial blooms). Hydrogen gas is also produced dissimilatory nitrate reduction diffuses into the
in anaerobic environments via fermentations, aerobic environment, where it is oxidized by the
where some of it escapes into the aerobic atmo- aerobic nitrifiers. The aerobic nitrifiying bacte-
sphere. (However, most of the H2 produced ria are often found at the aerobic–anaerobic
anaerobically is utilized by the sulfate reducers interfaces, where they capture the ammonia as
and methanogens.) Among the hydrogen-oxi- it diffuses from the anaerobic environments,
dizing bacteria are some, called carboxydobac- as well as in the more highly aerobic parts of
teria, that can also grow on carbon monoxide the soil and water. (There are also anaerobic
(CO) as the sole source of energy and carbon, ammonia-oxidizing bacteria that use nitrite
using oxygen, or in some cases nitrates (denitri- as the terminal electron acceptor and produce
fiers), as the electron acceptor. nitrogen gas. This reaction is discussed in note
The hydrogen-oxidizing bacteria and the 26, and reviewed in ref. 27. The bacteria grow
carboxydobacteria are represented by sev- on carbon dioxide, ammonium, and nitrite.)
eral genera, including representatives from Nitrosomonas oxidizes ammonia to nitrite.
Pseudomonas, Arthrobacter, Bacillus, and Along with Nitrobacter, which oxidizes nitrite
www.ebook777.com
to nitrate, it is responsible for a major portion of periplasm membrane cytoplasm
the conversion of ammonia to nitrate, a process

called nitrification. The first oxidation is that 4H+ + HNO2
NH2OH NH2OH + H2O
+ H2O
of ammonia to hydroxylamine, catalyzed by
ammonia monooxygenase (AMO). The reac- HAO
AMO NH3 + O2 + 2H+
tion is:
4e 2e
2H+ + NH3 + 2e– + O2 → NH2OH + H2O c554 UQ
In this reaction, one oxygen atom is incor-

porated into hydroxylamine and the other is
reduced to water. The oxidation of hydroxy- c552
1 +
2 O2 + 2H
lamine to nitrite is catalyzed by the enzyme
cyt. aa3
hydroxylamine oxidoreductase (HAO): oxidase
H2O
2H+ 2H+
NH2OH + H2O → NO2– + 4e– + 5H+
Of the four moles of electrons removed from
one mole of hydroxylamine, two are used
in the ammonia monooxygenase reaction, Fig. 13.7 Model for the electron transport scheme
approximately 1.7 are transferred to oxygen in Nitrosomonas. Ammonia is oxidized to hydroxy-
via cytochrome oxidase, and about 0.3 is used lamine by the enzyme ammonia monooxygenase
to generate NAD(P)H via reversed electron (AMO). Although the oxidation of ammonia is
transport: depicted as occurring in the cytoplasm, the possi-
bility that the reaction takes place in the periplasm
2H+ + 0.5O2 + 2e– → H2O has not been ruled out. Two electrons are required
to reduce one of the oxygen atoms to water. The
One proposed electron transport scheme for the
hydroxylamine diffuses across the membrane to the
topological arrangement of the electron carriers
periplasm, where it is oxidized to nitrite by a complex
(Fig. 13.7) is speculative, being based upon the cytochrome called hydroxylamine oxidoreductase
known location of the enzymes and their redox (HAO). The electrons are passed to a periplasmic
potentials. The scheme proposes that ammonia cytochrome c and from there to ubiquinone in the
oxidation to hydroxylamine takes place in the membrane. Electrons travel from ubiquinone in
cytoplasm, although it may occur either there two branches. One branch passes two electrons
or in the periplasm. (Ammonia monooxyge- to the ammonia monooxygenase, and the second
nase is in the cell membrane, but it is unknown branch leads to oxygen via cytochromes c and aa3.
whether the substrate–binding site is exposed All four electrons end up in water. Source: Adapted
to the cytoplasm or the periplasm.) Assuming from Hooper, A. B. 1989. Biochemistry of the nitri-
fying lithoautotrophic bacteria, pp. 239–265. In:
cytoplasmic oxidation, the hydroxylamine dif-
Autotrophic Bacteria. H. G. Schlegel and B. Bowien
fuses across the cell membrane to the periplasm,
(Eds.). Springer-Verlag, Berlin.
where it is oxidized to nitrite by hydroxylamine
oxidoreductase, a periplasmic enzyme.
During this oxidation four electrons are
transferred to periplasmic cytochrome c554. The Nitrite-oxidizing bacteria
four electrons are then transferred from cyto- The nitrite oxidizers are Nitrobacter,
chrome c554 to ubiquinone in the membrane. Nitrococcus, Nitrospina, and Nitrospira. They
It is suggested that two electrons travel from are aerobic, obligate chemolithoautotrophs,
ubiquinone to the ammonia monooxygenase with the exception of Nitrobacter, which is a
enzyme and two electrons to cytochrome aa3 facultative autotroph (i.e., it can also be grown
via membrane cytochrome c553 and periplasmic heterotrophically). The details of the electron
cytochrome c552. Cytochrome aa3 is presumed transport scheme have not been fully eluci-
to act as a proton pump. The ∆p that is created dated. However, a proposed model is shown in
as a result of the oxidation of ammonia and Fig. 13.8. Electrons travel from nitrite to oxy-
hydroxyamine and the reduction of oxygen is gen via a periplasmic cytochrome c. There is a
the sole source of energy for these bacteria. thermodynamic problem here. Cytochrome c
exists at a midpoint potential (E´0 ) of +270 mV, (Fig. 13.8). In this way, the membrane poten-
whereas the midpoint potential of the nitrate/ tial lowers the potential difference between the
nitrite couple is more electropositive, at +420 nitrate/nitrite couple and the cox/cred couple by
mV. Because electrons do not spontaneously approximately 170 mV. Students may recognize
flow toward the lower redox potential, energy this model as reversed electron transport driven
must be provided to drive the electrons over the by the membrane potential (which is consumed
150 mV difference for nitrite to reduce cyto- in the process).
chrome c. The role of the membrane potential in driv-
Nicholls and Ferguson suggested that the ing electron transport from nitrite to cyto-
membrane potential drives the electrons from chrome c is consistent with the observation
nitrite to cytochrome c.28 They proposed that that experimental procedures that lower the
nitrite oxidation takes place on the cytoplasmic membrane potential (e.g., incubation with
surface of the membrane and that the electrons proton ionophores) decrease electron trans-
flow across the membrane to cytochrome c fer from nitrite to oxygen in inverted mem-
located on the periplasmic side, which is typically brane vesicles from Nitrobacter. The electrons
170 mV more positive than the cytoplasmic side then flow from cytochrome c back across the
periplasm cell membrane cytoplasm
E'm
+
(mV)
–320 — NAD+/NADH
NADH- NAD+ + H+
ubiquinone
oxidoreductase
yH+
NADH
UQ
xH+
bc1
-
cyt. c NO2
+270 — cox/cred - H2O
2e
nitrite oxidase
-
+420 — NO -3/NO -2 NO3 + 2H+
cyt. aa3 1 +
oxidase 2 O2 + 2H
+820 — O2/H2O 2H+ 2H+

H2O
Fig. 13.8 A model for electron transport in Nitrobacter. The electrons travel from nitrite via membrane-
bound nitrite oxidase to cytochrome c, which is at a more negative potential. It is proposed that nitrite oxida-
tion takes place on the cytoplasmic side of the membrane and that the membrane potential drives the electrons
transmembrane to cytochrome c. From cytochrome c, the electrons diverge. Most travel to oxygen at a more
positive potential and a ∆p is created. The coupling site is the cytochrome aa3 oxidase, which is a proton
pump. Other electrons travel to NAD+, which is at a more negative potential. The ∆p drives the electrons in
reversed flow to NAD+. This is accomplished by coupling electron transport with the return of protons down
the proton potential to the cytoplasmic side. The scheme presumes the presence of a bc1 complex as well as a
reversible NADH dehydrogenase complex. Source: Adapted from Nicholls, D. G., and S. J. Ferguson. 1992.
Bioenergetics 2. Academic Press, London.
www.ebook777.com
membrane to oxygen through cytochrome aa3 Sulfur compounds commonly used as sources
oxidase, driven by a favorable midpoint poten- of energy and electrons include hydrogen sul-
tial difference of about +440 mV. A ∆p is cre- fide (H2S), elemental sulfur (So), and thiosulfate
ated by the outward pumping of protons by the (S2O32–), all of which can be oxidized to sulfate.
cytochrome aa3 oxidase. Electron flow is also (Laboratory studies often utilize thiosulfate
reversed from cytochrome c to NAD+ through or tetrathionate because these compounds are
two coupling sites: a ubiquinone–cytochrome c stable in air. Tetrathionate has an additional
oxidoreductase (probably a bc1 complex) and a advantage because unlike thiosulfate, it is stable
reversible NADH–ubiquinone oxidoreductase. at acidic pH values.) Although most sulfur bac-
The model proposes that reversed electron flow teria are aerobes, a few can be grown anaerobi-
to NAD+ is coupled to the influx of protons (i.e., cally by using nitrate as the electron acceptor.
it is driven by the ∆p). The sulfur bacteria can be found in nature grow-
ing near such sources as sulfur deposits, sulfide
Sulfur-oxidizing prokaryotes ores, hot sulfur springs, sulfur mines, and coal
The sulfur-oxidizing prokaryotes include the mines that are sites of iron pyrite (FeS2) depos-
photosynthetic sulfur oxidizers (Chapter 6) and its. Sulfide oxidizers can sometimes be found in
the nonphotosynthetic sulfur oxidizers. (See large accumulations in thin layers between the
refs. 29–31 for reviews.) It is the nonphotosyn- aerobic and anaerobic environments. They are
thetic type that concerns us here, and almost all sometimes confined to the aerobic–anaerobic
the known examples are gram-negative bacte- interface because H2S, which is produced anaer-
ria. Sulfur oxidizers comprise a physiologically obically by sulfate reducers, is rapidly oxidized
heterogeneous group. They may be obligate by oxygen. Thus, to be able to compete with the
autotrophs, which grow only on CO2 as the car- rapid chemical oxidation of sulfide, the sulfide-
bon source, or facultative heterotrophs, which oxidizing bacteria grow where oxygen levels
can also use organic carbon as the source of are relatively low.
carbon. Some are neutrophiles, growing best There has been much confusion in the litera-
around pH 7, whereas others are acidophiles, ture regarding the intermediate stages of sulfur
growing best between pH 1 and 5). oxidation. Probably, more than one pathway
Sulfur-oxidizing acidophiles can be isolated exists. Two pathways that are receiving much
from sulfur mines and coal mines, which pro- research attention are described next.
duce sulfuric acid. An example is Thiobacillus Paracoccus versutus is a neutrophilic facul-
thiooxidans, which can grow at a pH of 1, tative lithoautotroph. It can grow not only on
although the optimum is between 2 and 3. These thiosulfate (but not polythionates), but also on
bacteria can be mesophilic (growth temperature various organic carbon sources. (As reviewed
optimum 25–40 °C) or thermophilic (growth in ref. 29, this organism was formerly called
temperature optimum >55 °C). Some sulfur Thiobacillus versutus.) Oxidation of thiosul-
bacteria (e.g., most Beggiatoa strains) can be fate takes place in the periplasm on a multien-
grown only mixotrophically, that is, using H2S zyme complex (Fig. 13.9). No free intermediates
as the energy source and organic carbon as the are released, and both atoms of thiosulfate are
source of carbon. oxidized to sulfate. Protons are released in the
The nonphotosynthetic sulfur prokaryotes periplasm during the oxidations, and the elec-
include the bacteria Beggiatoa and Thiothrix, trons are transferred electrogenically from the
bacteria belonging to the genera Thiobacillus periplasmic side of the membrane to the cytoplas-
and Paracoccus, and an archaeon belonging to mic side through cytochrome c552 and then cyto-
the genus Sulfolobus, a thermophilic acidophile chrome aa3 oxidase to oxygen. (See Section 4.2.1
that grows in sulfur acid springs, where the tem- for a discussion of electrogenic movement of
perature can be 90 °C and the pH can be between electrons.) A proton current is maintained by the
1 and 5. (See the discussion of Sulfolobus and release of protons in the periplasm during the
other archaea in Section 1.1.1.) The thiobacilli oxidations and their consumption in the cyto-
are the most prominent of the sulfur oxidizers, plasm during the reduction of oxygen.
and their sulfur oxidation pathways are the The expected H+/O ratio (protons produced
most well studied. in the periplasm or translocated from the
periplasm cell membrane cytoplasm
S2O 23-+ 5H2O

8e- c552
enzymes A, B
cytochrome c
2SO 24-+ 10H+
8H+ + 2O2
cyt. oxidase
4H2O
ADP + Pi
+
3H 3H+
ATP synthase
ATP + H2O
Fig. 13.9 Model for thiosulfate oxidation in Parococcus versutus. Thiosulfate is oxidized to sulfate in the
periplasm by a multienzyme complex consisting of enzyme A, enzyme B, and cytochromes c. The electrons
are electrogenically transferred to oxygen via a membrane-bound cytochrome c552 and a cytochrome oxidase
(cytochrome aa3). A ∆p is created by the release of protons in the periplasm via the oxidations, the consump-
tion of protons in the cytoplasm during oxygen reduction, and electrogenic flow of electrons across the mem-
brane to oxygen. A proton-translocating ATP synthase makes ATP. Source: Adapted from Kelly, D. P. 1989.
Physiology and biochemistry of unicellular sulfur bacteria, pp. 193–217. In: Autotrophic Bacteria. H. G.
Schlegel and B. Bowien (Eds.). Springer-Verlag, Berlin.
cytoplasm to the periplasm for every oxygen believed to be transported into the cell, where
atom reduced) can be obtained from Fig. 13.9 it is oxidized to four sulfates via the interme-
(and confirmed experimentally). The ratio is 2.5 diate sulfite (generating 14 electrons). The 14
for thiosulfate oxidation. ATP is synthesized by electrons generated during the tetrathionate
a proton-translocating ATP synthase driven oxidations in the cytoplasm are transferred to a
by the ∆p. If one assumes that H+/ATP for the proton-translocating ubiquinone–cytochrome
ATP synthase is 3 (a consensus value), then the b complex in the cell membrane, which reduces
maximum P/O ratio would be 2.5/3, or 0.83, membrane-bound cytochrome c. Cytochrome
for thiosulfate oxidation. (See Section 5.5.2 for c transfers all the electrons to cytochrome oxi-
a discussion of the relationship between the size dase (probably cytochrome o), which reduces
of the proton current and the upper limit of ATP oxygen. Because the oxidations of tetrathi-
that can be made.) onate and sulfite are cytoplasmic rather than
Thiobacillus tepidarius is a neutrophilic ther- periplasmic, the proton current is due to proton
mophilic lithoautotroph that has been isolated translocation by the ubiquinone–cytochrome
from hot springs. It can be grown on hydrogen b complex (coupled to the cytoplasmic oxida-
sufide, thiosulfate, trithionate, and tetrathion- tions), rather than to periplasmic oxidations, as
ate, as electron donors. The pathway of oxidation is the case for P. versutus.
of thiosulfate to sulfate (called the polythionate Some thiobacilli can incorporate a substrate-
pathway) begins in the periplasm, but the bulk level phosphorylation when oxidizing sulfur.
of the oxidations take place in the cytoplasm. The substrate-level phosphorylation occurs at
Two molecules of thiosulfate are oxidized to the level of sulfite. The sulfite reacts with adenos-
tetrathionate in the periplasm (Fig. 13.10). A ine monophosphate (AMP) to form adenosine
periplasmic cytochrome c accepts the two elec- phosphosulfate (APS), in a reaction catalyzed
trons generated from the thiosulfate oxidation by APS reductase. The APS is oxidized to sul-
and transfers these electrons to a membrane- fate, producing ADP or ATP, using either ADP
bound cytochrome c. The tetrathionate is sulfurylase or ATP sulfurylase.
www.ebook777.com
Fig. 13.10 Model for thiosulfate oxidation in Thiobacillus tepidarius. Thiosulfate is oxidized to tetrathion-
ate in the periplasm by a thiosulfate–cytochrome c oxidoreductase. It is hypothesized that the tetrathionate
is transported into the cell in symport with protons, where it is oxidized to sulfate. The electrons travel to
oxygen via a proton-translocating ubiquinone–cytochrome b system (probably a bc1 complex) and cyto-
chrome oxidase, which appears to be a cytochrome o. Source: Adapted from Kelly, D. P. 1989. Physiology
and biochemistry of unicellular sulfur bacteria, pp. 193–217. In: Autotrophic Bacteria. H. G. Schlegel and
B. Bowien (Eds.). Springer-Verlag, Berlin; and Smith, D. W., and W. R. Strohl, 1991. Sulfur-oxidizing bacte-
ria, pp. 121–146. In: Variations in Autotrophic Life. J. M. Shively and L. L. Barton (Eds.). Academic Press,
New York.
APS reductase Figure 13.11 summarizes the inorganic sulfur

oxidation pathways. Elemental sulfur exists as
SO32– + AMP + APS → 2e– an octet ring of insoluble sulfur (S8°). It is first
activated by reduced glutathione (GSH) to form
ADP sulfurylase a linear polysulfide (G-S-S8-H). Sulfide (S2–) also
reacts with GSH and is oxidized to linear poly-
APS + Pi → ADP + SO42–
sulfide. The sulfur atoms are removed from the
ATP sulfurylase polysulfide one at a time during the oxidation to
sulfite (SO32–).
APS + PPi → ATP + SO42– In another elemental sulfur oxidation path-
way, reported for some thiobacilli, S° is oxygen-
Recall that ATP sulfurylase and APS reductase ated by a sulfur oxygenase:
are used by dissimilatory sulfate reducers to Sulfur oxygenase
reduce sulfate to sulfite in reactions that con-
So + O2 + H2O → H2SO3
sume ATP (Section 13.2.2). By running these
reactions in the oxidative direction, sulfur oxi- However, the oxygenase reaction cannot
dizers can use the energy in pyrophosphate to account for sulfur oxidation under anaerobic
make ATP. conditions when the oxidant provided is nitrate
use either ferrous ion or inorganic sulfur com-

pounds as a source of energy and electrons.
Thiobacillus ferrooxidans can be grown on
ferrous ion (e.g., FeSO4) at optimal external
pH values between 1.8 and 2.4. It maintains an
internal pH of around 6.5 and therefore a ∆pH
of about 4.5. The membrane potential (∆Ψ) at
low pH is reversed, and energy in the ∆p is due
entirely to the ∆pH component (Section 4.3).
(As discussed in Section 16.1.3, the inversion of
the ∆Ψ is necessary for maintenance of the ∆pH
and may be due to electrogenic influx of K+.)
1. Growth on ferrous sulfate

Figure 13.12 illustrates a model for the electron
Fig. 13.11 A summary of sulfur oxidation pathways:
transport pathway in T. ferrooxidans growing
1, oxidation of sulfide to linear polysulfide [S]; 2, con-
version of elemental sulfur to linear polysulfide; 3,
on the aerobic oxidation of ferrous sulfate. In
thiosulfate multienzyme complex; 4, sulfur oxidase; 5, this proposed pathway, extracellular Fe2+ is
sulfite oxidase; 6, APS reductase; 7, ADP–sulfurylase. oxidized to Fe3+ by an Fe3+ complex in the outer
Source: Adapted from Gottschalk, G. 1985. Bacterial membrane. The electrons travel to a periplas-
Metabolism, Springer-Verlag, Berlin. mic cytochrome c. The periplasm also contains
a copper protein called rusticyanin, which is
(T. denitrificans) or Fe3+ (T. ferrooxidans). thought to be part of the electron transport
Furthermore, the oxygenase reaction cannot scheme. From cytochrome c, the electrons flow
conserve energy for the cell, since the electrons across the membrane via cytochrome oxidase
do not enter the respiratory chain. Even though (cytochrome a1) to oxygen, generating a ∆Ψ,
the periplasm might be acidified with H2SO3 and inside negative. The oxidation of ferrous ion
thus produce a ∆pH, this by itself cannot gener- drives the consumption of protons in the cyto-
ate net ATP in a growing cell because protons plasm during oxygen reduction. The ∆pH is
entering via the ATP synthase must not accu- maintained because the rate of respiration and
mulate in the cytoplasm. The extrusion of pro- proton consumption matches the rate at which
tons from the cells or the utilization of protons protons enter the cell through the ATPase or
to form water requires electron transport. through leakage.
From an energetic point of view, one would
Iron-oxidizing bacteria not expect any proton pumping by the cyto-
A few bacteria derive energy from the aerobic chrome oxidase. The reason for this is that the
oxidation of ferrous ion to ferric ion.32 Most of difference in midpoint potentials between the
these are also acidophilic sulfur oxidizers (i.e., Fe3+/Fe2+ couple (pH 2) and the O2/H2O (pre-
they oxidize sulfide to sulfuric acid). The acido- sumed to be at pH 6.5) is very small, perhaps
philic iron oxidizers can be found growing at only 0.08 V or less.33 (See note 34 for a more
the sites of geological deposits of iron sulfide complete discussion.) Under these circum-
minerals [e.g., pyrite (FeS2) and chalcopyrite stances, a two-electron transfer would generate
(CuFeS2)], where water and oxygen are also at the most only 0.16 eV. A simple calculation
present. The iron–sulfide minerals are uncov- reveals that this would not be enough energy to
ered during mining operations, and the presence pump protons against the pH gradient. At an
of acid mine water at these sites is due to the external pH of 2 and an internal pH of 6.5, the
growth of the iron–sulfide oxidizers. Since Fe2+ ∆pH is 4.5. This is equivalent to 0.06 × 4.5 or
is rapidly oxidized chemically by oxygen to Fe3+ 0.27 V at 30 °C (eq. 4.10). Thus, each proton
at neutral pH but only slowly at acid pH, the would have to be energized by approximately
acidic environment is conducive to growth of 0.27 eV to be pumped out of the cell, even in the
the iron oxidizers. An example is Thiobacillus presence of a small ∆Ψ, inside positive, which
ferrooxidans, which is an autotroph able to may be on the order of +0.01 to +0.02 V.
www.ebook777.com
8H2O
2- NADH
16H+ + 2SO4
+ NADH xH+
xH
dehydrogenase
H2S2 NAD+ + H+
2S0
Q
+
2H
yH+
bc1 yH+
2+
2Fe
Fe3+ c
2Fe3+ 1/ O + 2H+
RC 2 2
cytochrome
oxidase
H2O
ADP + Pi
ATP
3H+ synthase 3H+
ATP + H2O
outer periplasm cell membrane cytoplasm

membrane
Fig. 13.12 Electron transfer in Thiobacillus ferrooxidans. Ferrous ion is oxidized to Fe3+ by an Fe3+ complex
in the outer membrane, and the electrons flow through periplasmic cytochrome c and a copper protein, rus-
ticyanin (RC) to a membrane cytochrome oxidase of the a1 type. The ∆p is maintained by the inward flow of
electrons (contributing toward a negative ∆Ψ) and by the consumption of cytoplasmic protons during oxy-
gen reduction, maintaining a ∆pH. The consumption of two protons for every two electrons in the cytoplasm
and the uptake of three protons through the ATP synthase indicate that a maximum of two-thirds of an ATP
can be made for every oxidation of two ferrous ions. When the bacteria are oxidizing iron pyrite (FeS2), the
Fe3+ that is produced outside the cell envelope is chemically reduced to Fe2+ by S2–. The resultant S° is oxidized
to SO24–, and the electrons pass through a proton-translocating bc1 complex to periplasmic cytochrome c and
thence to cytochrome oxidase. The figure also illustrates ∆p-driven reversed electron transport from S° and
from Fe2+ to NAD+ through coupling sites that bring protons into the cell.
ATP is synthesized by a membrane and ferrous ions (Fe2+). The ferrous and disul-
H -translocating ATP synthase driven by a ∆p
+
fide ions are oxidized to ferric ion and sulfuric
of approximately –250 mV. Electrons from Fe2+ acid, respectively, according to the following
must also move toward a lower redox potential overall reaction:
to generate NAD(P)H. Figure 13.12 also illus-
4FeS2 + 2H2O + 15O2 → 4Fe3+ + 8SO42– + 4H+
trates how reversed electron flow from Fe2+ to
NAD+ might occur through a bc1 complex and However, the oxidation is not straightfor-
quinone. Reversed electron flow is probably ward.35 The oxidation of FeS2 is the result of
driven by the ∆p as protons enter the cell down several redox reactions (Fig. 13.12.). When the
the ∆p gradient through the bc1 complex and bacteria are growing on iron pyrite, the Fe2+ is
the NADH dehydrogenase complex. oxidized extracellularly by a complex of Fe3+
in the outer membrane (Fig. 13.12). The outer
2. Growth on pyrite membrane iron complex then transfers the elec-
Pyrite is a crystalline ore of iron that is usually trons to the periplasmic cytochrome c, which
written as ferrous disulfide (FeS2). It is actually a transfers the electrons to cytochrome oxidase.
stable crystal of discrete –S–S– disulfide ions (S22–) The Fe3+ that is formed extracellularly is recycled
to Fe2+ by disulfide from the iron pyrite, accord- However, a comparison of the electrode poten-
ing to the following reaction, which occurs tials points to a problem. The Em″ for NO3–/NO2–
spontaneously: is +0.42 V and the Em″ for NAD+/NADH is –0.32
V. Electrons flow spontaneously only to the
H2S2 + 2Fe3+ → 2So + 2Fe2+ + 2H+
more electropositive acceptor. Thus, to make
The elemental sulfur that is produced is oxi- the electrons flow from NO2– to NAD+, the flow
dized by the bacteria to sulfate, with the elec- must be reversed, and this requires energy. How
trons traveling through a proton-translocating much energy is required? The potential differ-
bc1 complex to the periplasmic cytochrome c ence (Eo″) is –0.32 –0.42, or –0.7 V. Thus each
and from there to cytochrome oxidase. electron would have to be energized by 0.7 eV.
The source of energy is the ∆p. The inward flow
13.4.2 Review of the energetic of protons through the coupling sites down the
considerations for lithotrophic growth ∆p gradient drives electron transport in reverse
(see Section 4.7.1).
ATP synthesis
As discussed in Section 4.7.1, the energy from
respiratory oxidation–reduction reactions is 13.5 Summary
first converted into a ∆p, which is then used to When it comes to inorganic metabolism,
drive ATP synthesis via the ATP synthase. For prokaryotes are versatile creatures. They can
illustrative purposes, we will consider growth reduce oxidized forms of sulfur and nitrogen, as
on nitrite in an aerated culture as a source of well as nitrogen gas, for incorporation into cell
energy and electrons. The Em″ for NO3– / NO2– is material, and they can oxidize a variety of inor-
+0.42 V. The Em″ for O2/H2O is +0.82 V. The ganic substrates, trapping the energy released as
difference in potential is therefore +0.82 –0.42 a ∆p. They can also carry out anaerobic respira-
or +0.40 V. Recall that in the respiratory chain, tion, using oxidized forms of iron, manganese,
each coupling site is associated with pairs of nitrogen, and sulfur as electron acceptors. One
electrons traveling over a midpoint potential can add CO2, which is used by the methanogens,
difference of approximately 0.2 V or more. The to this list of electron acceptors.
conclusion is that there is sufficient energy for a There is both an assimilatory and a dissimi-
coupling site between nitrite and oxygen. latory route for nitrate reduction. The assimi-
Another way of examining this question is latory route is catalyzed by cytosolic enzymes
to consider the amount of energy required to that reduce nitrate to ammonia. The ammonia
synthesize an ATP. The energy required to syn- is then incorporated into amino acids via the
thesize ATP will vary with the concentrations of GOGAT enzyme system. All bacteria that grow
ATP, ADP, and Pi. However, we will assume a on nitrate as a source of cell nitrogen must have
value of 0.4 to 0.5 eV, which is a reasonable esti- the assimilatory pathway.
mate. Therefore, two electrons traveling over a The dissimilatory route differs in being mem-
redox gradient of 0.40 V should give 0.80 eV of branous and in producing a ∆p that can be used
energy, which is sufficient for the synthesis of an for ATP synthesis. A common dissimilatory
ATP. (See note 36.) route reduces nitrate to N2. Nitrate dissimilation
to N2 is also called denitrification. Organisms
NADH reduction requires reversed electron that carry out denitrification are widespread.
transport However, the enzymes for denitrification are
To grow, all cells must be able to make NAD(P)H oxygen sensitive and furthermore are formed
because this is a major source of electrons for only under anaerobic conditions or when oxy-
reductions that occur during biosynthesis. For gen tensions are very low. These bacteria are
example, the Calvin cycle, which reduces CO2 generally facultative anaerobes and will carry
to the level of carbohydrate, requires two moles out an aerobic respiration when oxygen is avail-
of NADH for every CO2. What can the cells use able. Several bacteria that are not denitrifiers
as a source of electrons to reduce NAD+ when can carry out an anaerobic respiration in which
they are growing on NO2– and CO2? Clearly nitrate is reduced to nitrite or to ammonia by
not CO2, since that is already the highest oxi- a membrane-bound nitrate reductase, creating
dized form of carbon. So we are left with NO2–. a ∆p.
www.ebook777.com
Most bacteria can use sulfate as a source of Many lithotrophs are aerobic autotrophs.
cell sulfur. They employ a cytosolic assimila- Electron transport takes place in the membrane,
tory pathway in which the sulfate is activated and a ∆p is created. The ∆p is used to drive ATP
by ATP to form adenosine-3′-phosphate-5′- synthesis and reversed electron flow so that
phosphosulfate (PAPS). The formation of NADH can be generated for biosynthesis.
PAPS raises the Eh of sulfate so that it is a bet- Iron-oxidizing bacteria such as Thiobacillus
ter electron acceptor. The reduction of PAPS ferrooxidans live in acid environments (around
yields sulfite, which is reduced to H2S via a pH 2) where iron–sulfide minerals and oxygen
soluble sulfite reductase. The H2S does not are available. By taking up cytoplasmic protons
accumulate but is immediately incorporated during the reduction of oxygen to water, the
into O-acetylserine to form L-cysteine. The bacteria maintain a ∆pH of around 4.5 units.
L-cysteine donates the sulfur to the other sulfur- The ∆pH drives the synthesis of ATP via a mem-
containing compounds. brane ATP synthase.
Certain strict anaerobes can use sulfate as an Lithotrophic activities are of immense eco-
electron acceptor for anaerobic respiration and logical significance because they are necessary
reduce it to H2S. These are the sulfate reducers. for the recycling of inorganic nutrients through
Organic carbon is oxidized, and the electrons the biosphere. For example, consider the nitro-
are transferred to protons via an hydrogenase gen cycle. All organisms use ammonia and
to form H2. It is suggested that the H2 diffuses nitrate, or amino acids, as the source of nitro-
into the periplasm, where it is oxidized by a gen. These are called “fixed” forms of nitrogen.
cytochrome c that transfers the electrons to Approximately 97% of all the nitrogen incor-
membrane carriers. Electrons travel across the porated in living tissue comes from fixed nitro-
membrane to the inner surface, creating a mem- gen. However, the majority of the nitrogen on
brane potential. The protons resulting from this planet is in the form of N2 and unavailable
the periplasmic oxidation of H2 remain on the to most organisms. Furthermore, the denitrifi-
outer surface or in the periplasm, contributing cation activities of bacteria living anaerobically
toward the proton gradient. Dissimilatory sul- cause a constant drain of fixed nitrogen from
fate reduction differs from assimilatory reduc- the biosphere. That is, living systems eventually
tion in that APS is reduced rather than PAPS, oxidize reduced forms of nitrogen to nitrate,
the electron carriers are membranous, and a ∆p which is then reduced to nitrogen gas by the
is created. denitrifying bacteria. Therefore, we all depend
Nitrogen fixation is carried out by a diversity upon the prokaryotes that reduce N2 for our
of prokaryotes, including both bacteria and supply of fixed nitrogen.
archeae. These include cyanobacteria, photo- The sulfate-reducing bacteria, which are
synthetic bacteria, strict heterotrophic anaer- responsible for much of the H2S produced,
obes (Clostridium pasteuranium, Desulfovibrio feed sulfide to the photosynthetic sulfur bacte-
vulgaris), and several obligate and facultative ria and aerobic sulfur-oxidizing bacteria. The
aerobes (e.g., rhizobia, Azotobacter, Klebsiella, combined activities of the sulfate reducers and
methanogens). The reduction of N2 is an ATP- the sulfide oxidizers account for much of the
dependent process catalyzed by the enzyme elemental sulfur found in deposits worldwide.
nitrogenase. Hydrogen gas is always a by- Hydrogen sulfide can also have deleterious
product of nitrogen fixation. The nitrogenase effects. Since H2S is toxic, it can occasionally
is oxygen sensitive and therefore must be pro- produce harmful effects on fish, waterfowl,
tected from oxygen, either in specialized cells and even plants when the soil becomes anaero-
(heterocysts) in the case of cyanobacteria, or in bic, as may occur in rice paddies. The H2S can
leguminous nodules, or perhaps by an unusu- also cause corrosion of metal pipes in anaero-
ally high respiratory rate (Azotobacter), by bic soils and waters.
binding to protective proteins, or by growth in
an anaerobic environment (e.g., as with some Study Questions
photosynthetic bacteria).
The oxidation of inorganic substances by 1. What are the major features distinguish-
oxygen is the source of energy for lithotrophs. ing assimilatory and dissimilatory nitrate
reduction and assimilatory and dissimila- 176:2551–2559; Gangeswaran, R., D. J. Lowe, and
tory sulfate reduction? R. R. Eady. 1993. Purification and characterization
of the assimilatory nitrate reductase of Azotobacter
2. During assimilatory sulfate reduction, vinelandii. Biochem. J. 289:335–342.
O-acetylserine reacts with H2S to form 6. Most of the reduced intracellular sulfur consists of
cysteine. The O-acetylserine is formed from H2S and HS–, rather than S2–. This is because the pK1 of
serine and acetyl–CoA. Write a reaction H2S is 7.04, the pK2 is 11.96, and the cytoplasmic pH
showing the chemical structures to suggest is usually close to 7. For example, neutrophilic bacte-
ria have a cytoplasmic pH of approximately 7.5.
how O-acetylserine might be formed from
serine and acetyl–CoA. 7. Casella, S., and W. J. Payne. 1996. Potential of
denitrifiers for soil environment protection. FEMS
3. Clostridia can reduce nitrogen gas to Microbiol. Lett. 140:1–8.
ammonia by means of electrons derived 8. Hansen, T. 1994. Metabolism of sulfate-reducing
from pyruvate (generated from glucose via prokaryotes. Antonie van Leeuwenhoek 66:165–185.
the Embden–Meyerhof–Parnas pathway). 9. Thauer has argued that free H2 may not be the
The immediate reductant for the nitroge- electron carrier for lactate oxidation by sulfate.
nase is reduced ferredoxin. How is reduced Even at very low H2 partial pressures, the reduction
ferredoxin generated from pyruvate? of H+ by lactate is thermodyamically unfavorable.
Furthermore, it has been demonstrated that H2 does
4. What is the minimum number of Fe2+ ions not affect lactate oxidation by sulfate by Desulfovibrio
that must be oxidized by oxygen to Fe3+ to vulgaris, and Desulfovibrio sapovorans growing on
lactate plus sulfate does not contain hydrogenase.
reduce one NAD+? Solve this problem by
Reviewed in: Thauer, R. K. 1989. Energy metabo-
focusing on electron volts. lism of sulfate-reducing bacteria, pp. 397–413. In:
5. The ∆pH drives the synthesis of ATP in the
(Eds.). Science Tech Publishers, Madison, WI.
acidophilic iron-oxidizing bacteria. How
do they maintain the ∆pH? 10. Reviewed by: Postgate, J. 1987. Nitrogen
Fixation. Edward Arnold, London.
6. What are some of the mechanisms used by 11. Reviewed in: Stacey, G., R. H. Burris, and H.
nitrogen fixers to protect the nitrogenase J. Evans (Eds.). 1992 Biological Nitrogen Fixation.
from oxygen? Chapman & Hall, New York.
12. Reviewed in: Dilworth, J. J., and A. R. Glenn
(Eds.). 1991. Biology and Biochemistry of Nitrogen
Fixation. Elsevier, Amsterdam.
13. Peuppke, S. G. 1996. The genetic and biochemi-
1. Lovley, D. R. 1991. Dissimilatory Fe(III) and cal basis for nodulation of legumes by rhizobia. Crit.
Mn(IV) reduction. Microbiol. Rev. 55:259–287. Rev. Biotechnol. 16:1–51.
2. Lovley, D. R. 1993. Dissimilatory metal reduc- 14. Dixon, R., and D. Kahn. 2004. Genetic regu-
tion. Annu. Rev. Microbiol. 47:263–290. lation of biological nitrogen fixation. Nat. Rev.
3. Nealson, K. H., and D. Saffarini. 1994. Iron and
manganese in anaerobic respiration: environmental 15. Lobo, A. L., and S. H. Zinder. 1992. Nitrogen
significance, physiology, and regulation. Annu. Rev. fixation by methanogenic bacteria, pp. 191–211.
Microbiol. 48:311–343. In: Biological Nitrogen Fixation. G. Stacey, H. R.
Burris, and H. J. Evans (Eds.). Chapman & Hall,
4. Lin, J. T., and V. Stewart. 1998. Nitrate assimila- New York.
tion by bacteria. Adv. Microb. Physiol. 39:1–30.
16. Dean, Dennis R., J. T. Bolin, and L. Zheng. 1993.
5. In Klebsiella pneumoniae the electron donor for Nitrogenase metalloclusters: structures, organiza-
nitrate reductase appears to be an NADH-dependent tion, and synthesis. J. Bacteriol. 175:6737–6744.
oxidoreductase. Thus, the electrons travel from
17. Peters, J. W., K. Fisher, and D. R. Dean. 1995.
NADH to the oxidoreductase to the nitrate reductase.
Nitrogenase structure and function: a biochemical–
Other bacterial nitrate reductases use reduced ferre-
genetic perspective. Annu. Rev. Microbiol. 49:335–
doxin or flavodoxin. The major cytoplasmic nitrite
366.
reductase in E. coli uses NADH as the electron
donor. For example, see the following references: 18. The MoFe protein has pairs of metalloclusters: a
Lin, J. T., B. S. Goldman, and V. Stewart. 1994. The pair of iron–sulfur clusters (Fe8S7–8), known as P clus-
nas FEDCBA operon for nitrate and nitrite assimi- ters, and a pair of iron–sulfur–molybdate clusters
lation in Klebsiella pneumoniae M5a1. J. Bacteriol. (Fe7S9Mo-homocitrate), known as FeMo cofactors.
www.ebook777.com
19. There are two types of metallocluster in the 24. Kelly, D. P. 1990. Energetics of chemolithotro-
MoFe protein, a FeMo cofactor and a P cluster. phs, pp. 479–503. In: The Bacteria, Vol. XII. T. A.
The P cluster consists of two Fe4S4 clusters. Each αβ Krulwich (Ed.). Academic Press, New York.
dimer has a FeMo cofactor and a P cluster, making
25. Hooper, A. B. 1989. Biochemistry of the nitri-
two copies of each type of cluster for every tetramer.
fiying lithoautotrophic bacteria, pp. 239–265. In:
One model of electron flow proposes that the elec-
tron travels in the Fe protein in an ATP-dependent
reaction from the Fe4S4 cluster to a P cluster in the
MoFe protein. From the P cluster the electron moves 26. Bacteria that oxidize ammonia anaerobically
to the FeMo cofactor, and from there to N2. The pre- can be found in the order Planctomycetales. They
cise role of ATP in electron transfer is not known, carry out the following reaction:
although it has been shown to bind as the Mg2+ che-
late to the Fe protein. It is presumed that binding NH4+ + NO2– → N2 + 2H2O
and/or hydrolysis of MgATP causes conformational
changes in the nitrogenase that facilitate electron They have been detected in anaerobic parts of marine
transfer. For example, conformational changes in waters and sediments, and they make a significant
the proteins might alter the environment and redox contribution to the production of N2 in these environ-
potential of the metal clusters, or bring the Fe4S4 clus- ments. Probably, the nitrite is produced from nitrate
ter in the Fe protein and the P cluster in the FeMo in the anaerobic zone by heterotrophic denitrifiers.
protein into closer proximity. Specifically, it has been The source of the nitrate is ammonia oxidized by the
proposed that binding of ATP to the Fe protein low- aerobic nitrifiers living in the oxic water zones above
ers the redox potential of the Fe4S4 cluster and also the sediments.
leads to association (complex formation) between
27. Strous, M., and M. S. M. Jetten. 2004. Anaerobic
the Fe protein and the MoFe protein. Association of
oxidation of methane and ammonium. Annu. Rev.
the two proteins causes ATP hydrolysis and electron
transfer. One model proposes that complex forma-
tion and hydrolysis of ATP is accompanied by a con- 28. Nicholls, D. G., and S. J. Ferguson. 1992.
figurational change in the MoFe protein that moves Bioenergetics 2. Academic Press, London.
a P cluster close to a Fe4S4 cluster in the Fe protein,
29. Kelly, D. P. 1989. Physiology and biochemis-
thus facilitating electron transfer. The two proteins
try of unicellular sulfur bacteria, pp. 193–217. In:
dissociate after electron transfer. Proposals about
how the MoFe protein might reduce N2 are derived
from model systems. The chemical reactivity of N2
is increased when it binds to a transition metal, espe- 30. Kelly, D. P., W. P. Lu, and R. K. Poole. 1993.
cially in a complex with other ligands, such as sul- Cytochromes in Thiobacillus tepidarius and the
fur. This has been studied in model systems, where respiratory chain involved in the oxidation of
chemically synthesized metal complexes containing thiosulphate and tetrathionate. Arch. Microbiol.
molybdenum bound to N2 were reduced with artifi- 160:87–95.
cial reductants. It is thought that the N2 becomes acti-
31. Friedrich, C. G. 1998. Physiology and genetics of
vated when it binds to the molybdenum in the FeMo
sulfur-oxidizing bacteria, pp. 235–289. In: Advances
cofactor in nitrogenase and, while bound, is reduced
in Microbial Physiology. R. K. Poole (Ed.). Academic
to ammonia.
Press, New York.
20. Eady, R. R. 1992. The dinitrogen-fixing bac-
32.Ingledew, W. J. 1990. Acidophiles, pp. 33–54. In:
teria, pp. 535–553. In: The Prokaryotes, Vol. I. A.
Microbiology of Extreme Environments. C. Edwards
Balows, H. G. Truper, M. Dworkin, W. Harder, and
(Ed.). McGraw-Hill, New York.
K.-H. Schleifer (Eds.). Springer-Verlag, Berlin.
33. Ingledew, W. J. 1982. Thiobacillus ferrooxi-
21. Yates, M. G., and F. O. Campbell. 1989. The
dans: the bioenergetics of an acidophilic chemolitho-
role of oxygen and hydrogen in nitrogen fixation, pp.
troph. Biochim. Biophys. Acta 683:89–117.
383–416. In: SGM Symposium, Vol. 42, The Nitrogen
and Sulphur Cycles. J. A. Cole and S. Ferguson (Eds.). 34. The amount of energy available from the oxida-
Cambridge University Press, Cambridge. tion of Fe2+ by O2 depends upon whether the oxygen
is reduced on the periplasmic side of the membrane
22. Bergman, B., J. R. Gallon, A. N. Rai, and L.
or on the cytoplasmic side. At pH 2, the Em of Fe3+/
J. Stal. 1997. N2 fixation by non-heterocystous
Fe2+ is usually stated to be about 0.77 to 0.78 V. At
cyanobacteria. FEMS Microbiol. Rev. 19:139–185.
pH 7, the Em of O2/H2O is 0.82 V; but at pH 2, it is
23. Haselkorn, R., and W. J. Bulikema. 1992. 1.12 V. If the oxygen were reduced in the cytoplasm,
Nitrogen fixation in cyanobacteria, pp. 166–190. then 0.82 – 0.77, or 0.05 eV, would be available per
In: Biological Nitrogen Fixation. G. Stacey, R. H. electron. If the oxygen were reduced in the periplasm,
Burris, and H. J. Evans (Eds.). Chapman & Hall, then 1.12 – 0.77, or 0.35 eV would be available
New York. per electron. However, as pointed out by Nicholls
and Ferguson in Bioenergetics 2 (Nicholls, S., and 35. Ehrlich, H. L., J. W., Ingledew, and J. C. Salerno.
S. J. Ferguson. 1992. Academic Press, London) 1991. Iron- and manganese-oxidizing bacteria, pp.
periplasmic reduction of oxygen with consumption 147–170. In: Variations in Autotrophic Life. J. M.
of periplasmic protons would not be helpful because Shively and L. L. Barton (Eds.). Academic Press, New
a ∆p would not develop. (There would be no elec- York.
trogenic electron flow and no differential consump-
tion of protons between periplasm and cytoplasm.) 36. According to the chemiosmotic theory, the
Attempting to reduce oxygen by bringing protons actual number of ATPs made will depend upon the
to the periplasm from the cytoplasm would require ratio of protons translocated per electron to the num-
energy to overcome the proton concentration gradi- ber translocated by the ATP synthase, that is, (H+/e–)/
ent between periplasm and cytoplasm. H+/ATP). (See Section 5.5.2.)
www.ebook777.com
14
C1 Metabolism
Many prokaryotes can grow on C1 compounds There are three major autotrophic CO2 fixa-
as their sole source of carbon. The following tion pathways in prokaryotes. These are the
are some common C1 compounds that support Calvin cycle (also known as the Calvin–Benson–
growth: Bassham, or CBB, cycle), the acetyl–CoA
pathway, and the reductive tricarboxylic acid
1. Carbon dioxide (CO2)
pathway. The Calvin cycle is the most promi-
2. Methane (CH4)
nent of the autotrophic CO2 fixation systems
3. Methanol (CH3OH)
and is found in photosynthetic eukaryotes,
4. Methylamine (CH3NH2)
most photosynthetic bacteria, cyanobacteria,
A few strictly anaerobic prokaryotes that use and chemoautotrophs. It does not occur in the
carbon dioxide as a source of cell carbon also archaea, nor is it present in certain obligately
use it as an electron acceptor, reducing it to anaerobic or microaerophilic bacteria. Instead,
methane and deriving ATP from the process. the acetyl–CoA pathway and the reductive car-
These are the methanogens, which are placed boxylic acid pathway are found in these organ-
in the Archaea, one of the two prokaryotic isms. The reductive carboxylic acid pathway
domains (Section 1.1.1). occurs in the green photosynthetic bacteria
Chlorobium (anaerobes), in Hydrogenobacter
(microaerophilic), in Desulfobacter (anaer-
14.1 Carbon Dioxide Fixation obes), and in the archaeons Sulfolobus and
Systems Thermoproteus. The acetyl–CoA pathway is
Organisms that use CO2 as the sole or major more widespread and is found in methanogenic
carbon source are called autotrophs, and the archaea, some sulfate-reducing bacteria, and in
pathway of CO2 assimilation is called auto- facultative heterotrophs that synthesize acetic
trophic CO2 fixation. The autotrophs include acid from CO2 during fermentation. The latter
the photoautotrophs and the chemoautotrophs. bacteria are called acetogens.
The prefix indicates the source of energy and the
term concludes with the name of the source of
carbon. The photoautotrophs, which use light as 14.1.1 The Calvin cycle
their source of energy, include the plants, algae, The Calvin cycle is a pathway for making phos-
and photosynthetic bacteria. The chemoauto- phoglyceraldehyde (PGALD) completely from
trophs use inorganic chemicals as their source CO2.1 The pathway operates in the chloroplasts
of energy (e.g., hydrogen gas, ammonia, nitrite, of plants and algae. However, in bacteria, where
ferrous ion, inorganic sulfur). Thus far, the only there are no similar organelles, the Calvin cycle
known chemoautotrophs are bacteria. takes place in the cytosol.
358
C1 metabolism 359
Summing the reactions of the Calvin cycle think of the Calvin cycle as occurring in two
During the Calvin cycle, three CO2 molecules stages:
are reduced to PGALD, which is at the oxida- 1. Stage 1 is a reductive carboxylation of ribu-
tion level of glyceraldehyde (C3H6O3). Twelve lose-1,5-bisphosphate (RuBP) to form phos-
electrons are required, and these are provided phoglyceraldehyde (PGALD).
by six moles of NAD(P)H. [Recall that each 2. Stage 2 consists of sugar rearrangements
NAD(P)H carries a hydride ion (one proton regenerating RuBP by using some of the
and two electrons) (Section 9.3)]. However, PGALD produced in stage 1.
NAD(P)H is not a sufficiently strong reductant
to reduce CO2 without an additional source of Stage 1 The first reaction is the carboxylation
energy. Each reduction thus requires an ATP. of RuBP, forming two moles of 3-phospho-
Therefore, six ATPs are required for the six glycerate (PGA). The enzyme is called RuBP
reductions. However, the overall reaction, carboxylase or ribulose-1,5-bisphosphate
which follows, indicates that nine ATPs are carboxylase (“rubisco”). The reaction is
required: shown in Fig. 14.1, along with its formal
mechanism. An enolate anion forms that is
3CO2 + 9 ATP + 6 NAD(P)H + 6H+ + 5H2O carboxylated on the C2. Hydrolysis of the
intermediate yields two moles of 3-phospho-
→ PGALD + 9 ADP + 8 Pi + 6 NAD(P)+ glycerate (PGA).
The PGA is then reduced to PGALD via 1,3-bis-
The three extra ATPs are used to recycle an inter- phosphoglycerate (BPGA). These reactions also
mediate, ribulose-1,5-bisphosphate (RuBP), take place in glycolysis. The reactions of stage
which is catalytically required for the pathway. 1 (see 1 and 2 in Fig. 14.2) can be summarized
as follows:
The Calvin cycle can be divided into two 3CO2 + 3 RuBP + 3H2O → 6 PGA
stages
6 PGA + 6 ATP → 6 BPGA + 6 ADP
The Calvin cycle is a complicated pathway,
but it will be familiar because it resembles the 6 BPGA + 6 NAD(P)H + 6H+
pentose phosphate pathway in certain key → 6 PGALD + 6 NAD(P)+ + 6 Pi
reactions (Section 9.5). It is convenient to
Fig. 14.1 Carboxylation of RuBP: the “rubisco” reaction. The carbonyl group in RuBP attracts electrons,
resulting in the dissociation of a proton from the C3 carbon upon enzyme binding. Electrons then shift to form
an enolate anion that is coordinated by an enzyme-bound magnesium (not shown) that directs carboxylation
at the C2 position. Hydrolysis of the carboxylated intermediate yields two 3-PGA molecules.
www.ebook777.com
Fig. 14.2 The Calvin cycle. Enzymes: 1, ribulose-1,5-bisphosphate carboxylase; 2, 3-phosphoglycerate kinase
and triosephosphate dehydrogenase; 3, triosephosphate isomerase; 4, fructose-1,6-bisphosphate aldolase; 5,
fructose-1,6-bisphosphate phosphatase; 6 and 9, transketolase; 7, sedoheptulose-1,7-bisphosphate aldolase;
8, sedoheptulose-1,7-bisphosphatase; 10, phosphopentose epimerase; 11, ribose phosphate isomerase; 12,
phosphoribulokinase.
C1 metabolism 361
3CO2 + 3 RuBP + 3H2O + 6 ATP the pentose phosphates that are formed from
+ 6NAD(P)H + 6H+ sedoheptulose-7-phosphate in the next step
from being converted back to phosphoglyc-
→ 6 PGALD + 6 ADP + 6 Pi + 6 NAD(P)+ eraldehyde. The sedoheptulose-7-phosphate
Stage 2 The whole point of the sugar rearrange- is a two-carbon donor in a second transke-
ments of stage 2 is to regenerate three RuBPs tolase reaction in which the last of the phos-
from five of the six PGALDs made in stage phoglyceraldehyde is used as an acceptor
1. For reference, see Fig. 14.2. (The Calvin (reaction 9). The products are the pentose
cycle is summarized without chemical struc- phosphates ribose-5-phosphate and xylulose-
tures shortly. See the subsection entitled. 5-phosphate, both of which are isomerized
The relationship of the Calvin cycle to gly- to ribulose-5-phosphate (reactions 10 and
colysis and the pentose phosphate pathway.) 11). The ribulose-5-phosphate is then phos-
Since RuBP has five carbons and PGALD has phorylated to ribulose-1,5-bisphosphate by a
three carbons, a two-carbon fragment must ribulose-5-phosphate kinase that is unique to
be transferred to PGALD. That is, a transke- the Calvin cycle (reaction 12). Thus, the pen-
tolase reaction is required. The C2 donor is tose isomerase reactions and the transketolase
fructose-6-phosphate, which is made from reactions are the same in both the Calvin cycle
PGALD (reactions 3, 4, and 5, Fig. 14.2). and the pentose phosphate pathway. The reac-
The synthesis of fructose-6-phosphate from tions of stage 2 are summarized as follows:
PGALD uses glycolytic enzymes and fruc-
tose-1,6-bisphosphate phosphatase. One 2 PGALD → 2 DHAP
molecule of PGALD first isomerizes to dihy- PGALD + DHAP → FBP
droxyacetone phosphate (DHAP), which
condenses with a second molecule of PGALD FBP + H2O → F6P + Pi
to form fructose-1,6-bisphosphate. The F6P + PGALD → erythrose-4-P + xylulose-5-P
fructose-1,6-bisphosphate is then dephos-
phorylated to fructose-6-phosphate by the Erythrose-4-P + DHAP
phosphatase (reaction 5). The fructose-6- → sedoheptulose-1,7-bisphosphate
phosphate donates a two-carbon fragment
in a transketolase reaction to PGALD, form- Sedoheptulose-1,7-bisphosphate + H2O
ing erythrose-4-phosphate and xylulose-5- → sedoheptulose-7-P + Pi
phosphate (reaction 6).
Sedoheptulose-7-P + PGALD
The xylulose-5-phosphate is isomerized to rib- → ribose-5-P + xylulose-5-P
ulose-5-phosphate (reaction 10). The eryth- 2 Xylulose-5-P → 2 ribulose-5-P
rose-4-phosphate is condensed with the second
dihydroxyacetone phosphate to form a seven- Ribose-5-P → ribulose-5-P
carbon ketose diphosphate, sedoheptulose-1,7- 3 Ribulose-5-P + 3 ATP
bisphosphate (reaction 7). The formation of
sedoheptulose-1,7-bisphosphate is analogous → 3 ribulose-1,5-bisphosphate + 3 ADP
to the condensation of dihydroxyacetone
phosphate with PGALD to form fructose-1, 5 PGALD + 2H2O + 3 ATP
6-bisphosphate, but it is catalyzed by a differ- → 3 ribulose-1,5-bisphosphate + 2 Pi + 3 ADP
ent aldolase, sedoheptulose-1,7-bisphosphate
aldolase. (However, as discussed later, the Summing stages 1 and 2 yields
bacterial sedoheptulose-1,7-bisphosphate 3CO2 + 9 ATP + 6 NAD(P)H + 6H+ + 5H2O
aldolase and the fructose-1,6-bisphosphate
→ PGALD + 9 ADP + 8 Pi + 6 NAD(P)+
aldolase may actually be a single bifunc-
tional enzyme.) A phosphatase hydrolyti-
cally removes the phosphate from the C1 to The carbon balance
form sedoheptulose-7-phosphate (reaction 8). Complex pathways can be seen in simpler
This is an irreversible reaction and prevents perspective by examining the carbon balance.
www.ebook777.com
The carbon balance for the Calvin cycle is as The relationship of the Calvin cycle to
follows: glycolysis and the pentose phosphate
3 C1 + 3 C5 → 6 C3 pathway
Most of the reactions of the Calvin cycle also
2 C3 → C6 take place in glycolysis and the pentose phos-
C6 + C3 → C4 + C5 phate pathway. The 12 reactions in Fig. 14.3
comprise the Calvin cycle. Reaction 1 is the
C4 + C3 → C7 carboxylation of ribulose-1,5-bisphosphate
C7 + C3 → 2 C5 to form 3-phosphoglycerate. This reaction is
unique to the Calvin cycle. The 3-phosphoglyc-
3 C1 → C3 erate enters the glycolytic pathway. Reactions
2 through 6 take place during the reversal of
The carbon balance for the Calvin cycle illus- glycolysis, whereby 3-phosphoglycerate is
trates that the pathway produces one C3 from transformed into fructose-6-phosphate (see
three C1 molecules. Section 9.1). The fructose-6-phosphate enters
Fig. 14.3 Relationships between the Calvin cycle, glycolysis, and the pentose phosphate pathway. Reactions
1 through 12 are the Calvin cycle. Note that the only reactions unique to the Calvin cycle are reactions 1,
8, 9, and 12. Reaction 1 is catalyzed by ribulose-1,5-bisphosphate carboxylase. Reactions 2 through 5 are
glycolytic reactions. Reaction 6 is catalyzed by fructose-1,6-bisphosphatase. Reaction 7 is the transketolase
reaction also found in the pentose phosphate pathway. Reactions 8 and 9 are the sedoheptulose–bisphosphate
aldolase and the sedoheptulose-1,7-bisphosphatase reactions. Reaction 10 is the transketolase reaction, also
found in the pentose phosphate pathway. Reactions 11 are the pentose epimerase and isomerase reactions
also present in the pentose phosphate pathway. Reaction 12 is the RuMP kinase reaction. The pentose phos-
phate pathway differs from the Calvin cycle only in that the pentose phosphate pathway synthesizes sedohep-
tulose-7-phosphate via a reversible transaldolase reaction (dashed lines), whereas the Calvin cycle synthesizes
sedoheptulose-7-phosphate via aldolase and phosphatase reactions (reactions 8 and 9). Since the phosphatase
is irreversible, the Calvin cycle irreversibly converts phosphoglyceraldehyde to pentose phosphates.
C1 metabolism 363
the pentose phosphate pathway at the level of is used by methanogens for methanogenesis.
the sugar rearrangement reactions (reaction 7) Sections 14.1.3 and 14.1.4 describe the indi-
(Section 9.5). Reaction 7 is the transketolase vidual reactions of the acetyl–CoA pathway in
reaction that forms erythrose-4-phosphate and acetogens and in methanogens.
xylulose-5-phosphate.
Now the Calvin cycle diverges from the Making acetyl–CoA from CO2 via the
pentose phosphate pathway. Whereas the acetyl–CoA pathway
pentose phosphate pathway synthesizes The acetyl–CoA pathway can be viewed as
sedoheptulose-7-phosphate from erythrose- occurring in four steps. The first step is a series
4-phosphate via the reversible transaldo- of reactions that result in the reduction of CO2
lase reaction (TA), the Calvin cycle uses the to a methyl group [CH3]. The methyl group is
aldolase (reaction 8) and the irreversible bound to tetrahydrofolic acid (THF) in bacteria
phosphatase (reaction 9) to synthesize sedo- and to tetrahydromethanopterin (THMP) in
heptulose-7-phosphate. This has important archaea. The methyl group is then transferred to
consequences regarding the directionality of the enzyme carbon monoxide dehydrogenase.
the Calvin cycle. Because the phosphatase Carbon monoxide dehydrogenase also catalyzes
(reaction 9) is an irreversible reaction, the the reduction of a second molecule of carbon
Calvin cycle proceeds only in the direction dioxide to a bound carbonyl group [CO], which
of pentose phosphates, whereas the pen- will become the carboxyl group of acetate.
tose phosphate pathway is reversible from During autotrophic growth, the electrons are
sedoheptulose-7-phosphate. (An irrevers- provided by hydrogen gas via a hydrogenase.
ible Calvin cycle makes physiological sense,
since the sole purpose of the Calvin cycle is CO2 + 3H2 + H+ → [CH3] + 2H2O (14.1)
to regenerate ribulose-1,5-bisphosphate for CO2 + H2 → [CO] + H2O (14.2)
the initial carboxylation reaction.)
Once the sedoheptulose-7-phosphate has The bound [CH3] and [CO] are then condensed
formed, the synthesis of ribulose-5-phosphate by carbon monoxide dehydrogenase to form
(reactions 10 and 11) takes place via the same bound acetyl, [CH3CO].
reactions as in the pentose phosphate pathway. [CO] + [CH3] → [CH3CO] (14.3)
Reaction 12 of the Calvin cycle is the phos-
phorylation of ribulose-5-phosphate to form The bound acetyl reacts with bound CoASH,
ribulose-1,5-bisphosphate catalyzed by RuMP [CoAS], to form acetyl–CoA, which is released
kinase, a second reaction that is unique to the from the enzyme:
Calvin cycle. The sedoheptulose-1,7-bisphos- [CH3CO] + [CoAS] → CH3COSCoA (14.4)
phate aldolase and the sedoheptulose-1,7-
bisphosphate phosphatase also occur in the The acetyl–CoA either is converted to acetate via
ribulose monophosphate pathway, which is a phosphotransacetylase and acetate kinase (gen-
formaldehyde-fixing pathway that uses Calvin erating an ATP) and excreted or is assimilated
cycle reactions (Section 14.2.1).2 into cell material. As mentioned, many anaero-
bic bacteria can grow autotrophically with the
acetyl–CoA pathway, using H2 as the source of
14.1.2 The acetyl–CoA pathway electrons. These bacteria must generate ATP
Bacteria that use the acetyl–CoA pathway from the Δp generated during the reduction of
include methanogens, acetogenic bacteria, and the CO2 by H2. The Δp is required for net ATP
most autotrophic sulfate-reducing bacteria.3 synthesis because the ATP generated during the
(See note 4 for a definition of the acetogenic bac- conversion of acetyl–CoA to acetate is balanced
teria.) This section presents a general summary by the ATP used for the formation of formyl–
of the acetyl–CoA pathway without describing THF (see Section 14.1.3, Fig. 14.4).
the individual reactions. The first part explains
how acetyl–CoA is made from CO2 and H2, Incorporating acetyl–CoA into cell material
the second part explains how the acetyl–CoA For acetogens growing autotrophically, part of
is incorporated into cell material, and the third the acetyl–CoA that is produced must be incor-
part summarizes how the acetyl–CoA pathway porated into cell material. The glyoxylate cycle,
www.ebook777.com
which is generally associated with aerobic The pyruvate can be phosphorylated to phos-
metabolism, is not present in these organisms, phoenolpyruvate by using PEP synthetase. The
and therefore a different pathway must oper- phosphoenolpyruvate is used for biosynthesis
ate. As shown in eqs. 14.5 and 14.6, the acetyl– in the usual way. A more detailed discussion
CoA is reductively carboxylated to pyruvate via of acetyl–CoA assimilation in methanogens is
pyruvate synthase, a ferredoxin-linked enzyme. given in Section 14.1.4.
Fig. 14.4 The acetyl–CoA pathway in Clostridium thermoaceticum. During heterotrophic growth, the car-
boxyl group in some pyruvate molecules is directly transferred to CODH without being released as free CO2.
The evidence for this is that even in the presence of CO2, pyruvate is required in cell-free extracts for the
synthesis of acetic acid from CH3–THF. Furthermore, radioisotope experiments confirm that the carboxyl
in acetate is derived from the carboxyl in pyruvate without going through free CO2. However, the cells can
be grown on CO2 and H2. Under these circumstances, the CODH reduces CO2 to CO–CODH via hydroge-
nase and electrons from H2 (reactions 15 and 16). Enzymes: 1, glycolytic enzymes; 2, pyruvate:ferredoxin
oxidoreductase; 3, phosphotransacetylase; 4, acetate kinase; 5, formate dehydrogenase; 6, formyltetrahy-
drofolate (HCO–THF) synthetase; 7, methenyltetrahydrofolate (CH–THF) cyclohydrolase; 8, methylenetet-
rahydrofolate (CH2–THF) dehydrogenase; 9, methylenetetrahydrofolate reductase; 10, methyltransferase;
11, corrinoid enzyme; 12–15, carbon monoxide dehydrogenase (CODH); 16, hydrogenase. Abbreviations:
CODH, carbon monoxide dehydrogenase; Co, corrinoid (See note 6.)
C1 metabolism 365
Pyruvate synthase corrinoid enzyme, [Co]-E, which transfers the

methyl group to the CO–CODH (reactions 10
Acetyl–CoA Fdred + CO2 and 11).6 The CODH makes the acetyl moiety,
→ pyruvate Fdox + CoASH (14.5) [CH3CO] (reaction 12). The CODH also has a
binding site for CoASH and synthesizes acetyl–
PEP synthetase CoA from the bound intermediates (reactions
13 and 14). The CODH can also bind free CO2
Pyruvate + ATP + H2O and reduce it to the level of carbon monoxide
→ PEP + AMP + Pi (14.6) (reaction 15).
14.1.3 The acetyl–CoA pathway in Autotrophic growth

Clostridium thermoaceticum Following the entry of CO2 at reactions 5
The acetyl–CoA pathway was first investigated and 15 (Fig. 14.4), the acetyl–CoA pathway
in C. thermoaceticum, an acetogenic bacte- allows autotrophic growth on CO2 and H2.
rium that converts one mole of glucose to three Carbon dioxide is reduced to [CH3] and [CO]
moles of acetate.5 Under these circumstances, (reactions 5–10 and 15). The [CH3] and [CO]
the production of acetate is used as an electron combine with CoASH to form acetyl–CoA
sink during fermentation rather than for auto- (reactions 11–14). During acetogenesis, an
trophic growth. The pathway is illustrated in ATP can be made via substrate-level phospho-
Fig. 14.4. Glucose is converted to pyruvate via rylation from acetyl–CoA (reactions 3 and 4).
the Embden–Meyerhof–Parnas pathway (reac- However, because an ATP is required to incor-
tions 1). Two of the acetates are synthesized porate formate (reaction 6), there can be no net
from the decarboxylation of pyruvate by means ATP synthesis via substrate-level phosphory-
of pyruvate–ferredoxin oxidoreductase, phos- lation alone.
photransacetylase, and acetate kinase (reac- How do organisms using the acetyl–CoA
tions 2–4). The third acetate is synthesized from pathway during autotrophic growth make net
CO2 via the acetyl–CoA pathway. ATP? It is necessary to postulate that ATP is
In the acetyl–CoA pathway, one of the carbon made by a chemiosmotic mechanism coupled to
dioxides that is removed from pyruvate is not set the reduction of CO2 to acetate. For example,
free but instead becomes bound to the enzyme there may be electrogenic electron flow follow-
carbon monoxide dehydrogenase (CODH), ing periplasmic oxidation of H2; or perhaps a
where it will be used for acetate synthesis quinone loop or a proton pump may be oper-
(reaction 2a). This bound CO2 will eventually ating. (Mechanisms for generating a Δp were
become the carbonyl group in acetyl–CoA. The reviewed in Sections 4.7.2 and 5.6.)
second pyruvate is decarboxylated to release It appears that homoacetogenic clostridia do
free CO2 (reaction 2b). In reactions 5 through indeed create a Δp during the reduction of CO2
9, the free CO2 is reduced to bound methyl. to acetate and that the Δp drives the synthesis of
The free CO2 first becomes reduced to formate ATP via a H+-ATP synthase.7
(HCOOH) by formate dehydrogenase (reaction Not all acetogenic bacteria use a proton
5). The formate is then attached to tetrahydro- circuit, however. Acetobacterium woodii
folic acid (THF) in an ATP-dependent reaction couples the reduction of CO2 to acetyl–CoA
to make formyl–THF (HCO–THF) (reaction to a primary sodium ion pump.8 The sodium
6). (See Section 10.2.4 for a discussion of tetra- ion translocating step has not been identified
hydrofolate reactions.) The formyl–THF is then but is thought to be either the reduction of
dehydrated to form methenyl–THF (CH–THF) methylene–THF to methyl–THF (reaction 9)
(reaction 7). The methenyl–THF is reduced or some later step in the synthesis of acetyl–
by NADPH to methylene–THF (CH2–THF) CoA because these are highly exergonic reac-
(reaction 8). The methylene–THF is reduced tions. A. woodii also has an ATP synthase that
by ferredoxin to methyl–THF (CH3–THF) is sodium ion dependent. Thus when growing
(reaction 9), a reaction thought to be coupled on CO2 and H2, these bacteria rely on a sodium
to ATP formation by a mechanism involving ion current to make ATP. A similar situation
Δp. The methyl group is then transferred to a exists in Propionigenium modestum, where
www.ebook777.com
a Na+-translocating methylmalonyl–CoA 14.1.4 The acetyl–CoA pathway in

decarboxylase generates a Na+ electrochemi- methanogens
cal potential that drives the synthesis of ATP Acetyl–CoA synthesis from CO2 and H2
(Section 4.8.1). Among the archaea are a group called methano-
gens that can grow autotrophically on CO2 and
Carbon monoxide dehydrogenase H2 (Section 1.1.1). During autotrophic growth,
the CO2 is first incorporated into acetyl–CoA
It is clear that carbon monoxide dehydrogenase
in a pathway very similar to the acetyl–CoA
(CODH), also called carbon monoxide dehy-
pathway that exists in the bacteria (e.g., in
drogenase/acetyl–CoA synthase (CODH/ACS),
C. thermoaceticum).13–16 In other words, one
is a crucial enzyme for acetogenesis, for auto-
molecule of CO2 is reduced to [CO] bound to
trophic growth in anaerobes, and for methano-
CODH. A second molecule of CO2 is reduced
genesis from acetate. Because of the importance
to bound methyl, [CH3], which is transferred
of this enzyme, some of the evidence for its par-
to the CODH. Then the CODH makes acetyl–
ticipation in synthesizing and degrading acetyl–
CoA from the bound CO, CH3, and CoASH.
CoA is described next.
There are, however, differences between the
The CODH catalyzes the breakage and for-
acetyl–CoA pathway found in the methano-
mation of the carbon-carbon bond between the
genic archaea and that in the bacteria. The dif-
methyl group [CH3] and the carbonyl group
ferences have to do with the cofactors and the
[CO] in the acetyl moiety. Evidence for this is
utilization of formate. For reference, the struc-
the following exchange reaction carried out by
tures of the cofactors are drawn in Fig. 14.5.
CODH9:
The differences between the two pathways are
14
CH3–CODH + CH3COSCoA as follows.
U CH3–CODH + 14CH3COSCoA 1. The carrier for the formyl and more reduced
C1 moieties in the methanogens is not tet-
The CODH catalyzes the breakage and for- rahydrofolic acid (THF) but rather tetra-
mation of the thioester bond between CoASH hydromethanopterin (THMP), a molecule
and the carbonyl group of the acetyl moiety. resembling THF. (For a comparison of the
Evidence for this is the following exchange structures, see Fig. 14.5, where the THMP is
reaction carried out by CODH in conjunc- labeled “methanopterin.”)
tion with another enzyme called disulfide 2. The methanogens do not reduce free CO2 to
reductase,10 in which an asterisk refers to radio- formate. Nor do they incorporate formate
active CoASH: into THMP to form formyl–THMP. What
CH3COSCoA + *CoASH the methanogens do instead is fix CO2 onto
a C1 carrier called methanofuran (MFR) and
U CH3COS*CoA + CoASH reduce it to formyl–MFR. The formyl group
is then transferred to THMP. (Recall that
The following exchange has also been demon-
the acetogenic bacteria reduce free CO2 to
strated to be catalyzed by CO dehydrogenase11:
formate and then incorporate the formate
[1–14C]Acetyl–CoA + 12CO into formyl–THF.)
U 14
CO + [1–12C]acetyl–CoA Figure 14.6 is a diagram showing how the
acetyl–CoA pathway is thought to operate in
The three exchange reactions just described methanogens. (The same pathway is shown
show that the CODH is capable of disassem- in Fig. 14.7, but with the chemical structures
bling acetyl–CoA into its components, [CH3], drawn.) The CO2 is first condensed with metha-
[CO], and [CoASH], and resynthesizing the nofuran (MFR) and reduced to formyl–MFR
molecule. Therefore the presence of CODH (reaction 1). The formyl group is then trans-
is sufficient for making acetyl–CoA from the ferred to tetrahydromethanopterin (THMP)
bound components. The crystal structure of the to form formyl–THMP (reaction 2). Then
enzyme and a Ni-Fe-Cu center at the active site a dehydration produces methenyl–THMP
have been described. (reaction 3). The methenyl–THMP is reduced
C1 metabolism 367
Fig. 14.5 Structures of coenzymes in the acetyl–CoA pathway and in methanogenesis. (A) Methanofuran
(4-[N-(4,5,7-tricarboxyheptanoyl-n-l-glutamyl-n-l-glutamyl)-p-(β-aminoethyl) phenoxymethyl]-2-
(aminomethyl)furan). The CO2 becomes attached to the amino group on the furan and is reduced to the
oxidation level of formic acid (formyl–methanofuran). (B) Tetrahydrofolic acid and methanopterin; both are
derivatives of pterin, shown in (C). In formyl–methanopterin the formyl group is bound to N5, whereas in
tetrahydrofolic acid it is bound to N10. (C) Pterin. (D) Oxidized coenzyme F420: this 5-deazaflavin carries two
electrons but only one hydrogen (on N1); compare this structure to flavins that have a nitrogen at position 5
and therefore carry two hydrogens (Section 5.2.1). (E) Factor B or HS–HTP (7-mercaptoheptanoylthreonine
phosphate). (F) Coenzyme M (2-mercaptoethanesulfonic acid. (G) Methylcoenzyme M. (H) Coenzyme F430:
this prosthetic group is a nickel–tetrapyrrole.
to methylene–THMP (reaction 4), which is For the synthesis of acetyl–CoA, the methyl
reduced to methyl–THMP (reaction 5). The group is transferred to a corrinoid enzyme
pathway diverges at the methyl level. One (Figs. 14.6 and 14.7, reaction 9). A second
branch synthesizes acetyl–CoA in reactions molecule of CO2 is reduced by carbon monox-
quite similar to the synthesis of acetyl–CoA in ide dehydrogenase (CODH) to bound carbon
the bacteria (Fig. 14.4), and the other branch monoxide [CO] (reaction 14). The CODH
synthesizes methane (Section 14.1.5). then catalyzes the synthesis of acetyl–CoA
www.ebook777.com
Fig. 14.6 The acetyl–CoA pathway in methanogens. There are two points of entry of CO2, and these are
reactions 1 and 14. In reaction 14 CO2 is bound to carbon monoxide dehydrogenase and is reduced to the
oxidation level of carbon monoxide [CO]. The electron flow is from H2 via a hydrogenase to ferredoxin to
CODH. In reaction 1 CO2 is attached to methanofuran (MF) and becomes reduced to the level of formic acid
(formyl–MF). The formyl group is then transferred to a second carrier, tetrahydromethanopterin (THMP) to
form formyl–THMP (reaction 2). The formyl–THMP is reduced to methyl–THMP (CH3–THMP) (reactions
3–5). The CH3–THMP donates the methyl group to a corrinoid enzyme (Co-E) for the synthesis of acetyl–CoA
(reactions 9–13), or to CoMSH for the synthesis of methane (reactions 6–8). Enzymes: 1, formyl–MFR dehy-
drogenase (an iron–sulfur protein); 2, formyl–MFR:H4MPT formyltransferase; 3, N5,N10-methenyl–H4MPT
cyclohydrolase; 4, N5,N10-methylene–H4MPT dehydrogenase; 5, N5,N10-methylene–H4MPT reductase; 6,
N5-methyl–H4MPT: CoMSH methyltransferase; 7, methyl–S–CoM reductase; 8, heterodisulfide reductase;
9, methyltransferase; 10–13, carbon monoxide dehydrogenase. THMP is also written as H4MPT.
from [CH3], [CO], and CoASH (reactions form CH3–SCoM (Figs. 14.6 and 14.7, reac-
11–13). tion 6). The terminal reduction is catalyzed
by the methylreductase system (reactions 7
14.1.5 Methanogenesis from CO2 and H2 and 8).17 The methylreductase system has two
In addition to incorporating the [CH3] into components. One component is a methylre-
acetyl–CoA as just described, the methano- ductase that reduces CH3–SCoM to CH4 and
gens can reduce it to methane (CH4) in a series CoM–S–S–HTP. The electron donor for the
of reactions not found among the bacteria. methylreductase is HTP–SH.18 A nickel-con-
For methane synthesis, the methyl group is taining tetrapyrrole, F430, is an electron carrier
transferred from CH3–THMP to CoMSH to that is part of the methylreductase. The second
C1 metabolism 369
Fig. 14.7 The acetyl–CoA pathway in methanogens. See the legend to Fig. 14.6.
component is an FAD-containing heterodisul- CoM–S–S–HTP (Fig. 14.7, reaction 8) occurs

fide reductase that reduces CoM–S–S–HTP to in the membranes and is accompanied by the
CoMSH and HTP–SH. The crystal structure generation of a Δp that drives ATP synthesis via
of methylreductase has been published, and a a H+-translocating ATP synthase. (See note 20
model for how it works has been proposed.19 for experimental evidence.) The mechanism of
the generation of the Δp is not known but could
Energy conservation during methanogenesis include H2 oxidation on the outer membrane
The production of methane yields ATP and is surface depositing protons on the outside, fol-
the only means of ATP formation for the metha- lowed by electrogenic movement of electrons
nogens. Probably, energy is conserved for ATP across the membrane to CoM–S–S–HTP on
synthesis at two sites in the pathway of metha- the inside surface. This is a tactic often used by
nogenesis. It appears that electron transfer to bacteria to generate a Δp coupled to periplasmic
www.ebook777.com
oxidations. For example, recall the periplasmic acetate in which the carbonyl group is oxidized
oxidation of H2 by Wolinella succinogenes to CO2, and the electrons are used to reduce the
coupled to the cytoplasmic reduction of fumar- methyl group to methane according to the fol-
ate described in Section 5.7.4, or the reduc- lowing overall reaction:
tion of sulfate by Desulfovibrio described in
Section 13.2.2. CH3 CH2– + H+ → CH4 + CO2
There may also be a second energy-coupling The acetate is first converted to acetyl–CoA.
site. The transfer of the methyl group from Methanosarcina uses acetate kinase and phos-
CH3–THMP to CoMSH (Fig. 14.7, reaction photransacetylase to make acetyl–CoA from
6) is an exergonic reaction accompanied by the acetate:
uptake of Na+ into inverted vesicles, creating Acetate kinase
a primary sodium motive force (inside posi-
tive). How might the sodium motive force be Acetate ATP → acetyl phosphate + ADP
used? The sodium motive force probably drives Phosphotransacetylase
ATP synthesis in whole cells, since it is known
to energize the synthesis of ATP in washed Acetyl phosphate CoA → acetyl–CoA
inverted vesicles made from Methanosarcina + inorganic phosphate
mazei Gö1. There appear to be two different
ATP synthases present, one coupled to protons Methanothrix uses acetyl–CoA synthetase to
and the other coupled to sodium ions.21 It has make acetyl–CoA:
been speculated that the sodium motive force
Acetyl–CoA synthetase
created during methyl transfer to CoMSH may
also serve to drive the formation of formyl– Acetate ATP → acetyl–AMP
MFR from free CO2, using H2 as the reductant
+ pyrophosphate
(Fig. 14.7, reaction 1). The latter reaction has a
standard free energy change of approximately Acetyl–AMP + CoA → acetyl–CoA + AMP
+8 kJ/mol.
Reactions similar to those of the acetyl–CoA
pathway in methanogens convert acetyl–CoA
Unique coenzymes in the Archaea to methane and CO2, deriving ATP from the
As discussed in Section 1.1.1, there are sev- process. Examine Fig. 14.6 again, beginning
eral unique aspects to the biochemistry of the with acetyl–CoA: in reactions 13 and 12, acetyl–
Archaea. For example, there are several coen- CoA combines with carbon monoxide dehydro-
zymes in archaea that are not found in bacte- genase (CODH), which catalyzes the breakage
ria or eucarya. These are represented by five of the C–S bond and the release of CoA. In reac-
coenzymes used in methanogenesis: methano- tion 11, the CODH catalyzes the breakage of
furan (formerly called CDR, for carbon dioxide the C–C bond forming the carbonyl [CO] and
reduction enzyme), methanopterin, coenzyme methyl [CH3] groups. Then the carbonyl group
M (CoSM), F430, and HS–HTP, also known as is oxidized to CO2 (reaction 14) and the methyl
factor B or coenzyme B (Fig. 14.5). Coenzyme group is transferred to a corrinoid enzyme (reac-
M, F430, and HS–HTP are found only in metha- tion 10). (Note that the electrons removed from
nogens. The other coenzymes occur in other the carbonyl group are not released in hydro-
archaea. The electron carrier F430 also exists in gen gas as implied in reaction 14 but are used
bacteria and eucarya. to reduce the methyl group to methane.) The
methyl group is transferred to tetrahydrometh-
14.1.6 Methanogenesis from acetate anopterin (THMP) (reaction 9).24 (Acetate-
Species of methanogens within the genera grown cells of Methanosarcina species also
Methanosarcina and Methanothrix can use ace- contain large amounts of a derivative of THMP,
tate as a source of carbon and energy.22 In fact, tetrahydrosarcinapterin (H4SPT), which has
acetate accounts for approximately two-thirds also been reported to serve as a methyl group
of the biologically produced methane. (See note carrier during methanogenesis from acetate.25)
23.) The methanogens that convert acetate to The methyl group is then transferred to coen-
methane and CO2 carry out a dismutation of zyme M (CoMSH) (reaction 6) and reduced to
C1 metabolism 371
methane by means of the electrons derived from It was reported that ATP can also be made dur-
the oxidation of the carbonyl group (reaction ing the oxidation of [CO] to CO2 and H2.26
7). The electron donors for the reduction of the
methyl group are HTP–SH and CH3–S–CoM.
14.1.7 Incorporation of acetyl–CoA into
Each of these contributes one electron from its
respective sulfur atom, and the oxidized prod-
cell carbon by methanogens
uct is the heterodisulfide HTP–S–S–CoM. The The glyoxylate cycle, one of the means of incor-
reduction of HTP–S–S–CoM to CoMSH and porating net acetyl–CoA into cell material, is not
HTP–SH (reaction 8) is coupled to the oxida- present in prokaryotes that use the acetyl–CoA
tion of the carbonyl group (reaction 14). The pathway. How do they grow on the acetyl–CoA?
electron transport pathway that links the oxi- The acetyl–CoA must be converted to phospho-
dation of the carbonyl group with the reduction enolpyruvate, which feeds into an incomplete
of HTP–S–S–CoM is thought to generate ATP, citric acid pathway and into gluconeogenesis.
presumably via a Δp.19 The enzymes to make phosphoenolpyruvate
Methanogenesis from acetate is summarized from acetyl–CoA are widespread in anaerobes.
as follows: A suggested pathway for acetyl–CoA incorpo-
ration by methanogens is discussed next.
CH3CO–CoA → [CO] + CoA (14.7) Figure 14.8 shows a proposed pathway for
[CO] + H2O → CO2 + 2[H] (14.8) acetyl–CoA incorporation by methanogens.27
The first reaction is the carboxylation of acetyl–
[CO3] + 2[H] + ADP + Pi → CH4 + ATP (14.9) CoA to form pyruvate, a reaction catalyzed by
Fig. 14.8 Assimilation of acetyl–CoA into cell material in methanogens. The acetyl–CoA is carboxylated to
form pyruvate, which is then phosphorylated to make phosphoenolpyruvate (PEP). In M. thermoautotrophi-
cum, the PEP is carboxylated to oxaloacetate (OAA), which is reduced to α-ketoglutarate or to phosphoglyc-
eraldehyde. M. barkeri synthesizes α-ketoglutarate by a different pathway: the oxaloacetate condenses with
another acetyl–CoA to form citrate, and the citrate is then oxidized to α-ketoglutarate via aconitate, isocitrate,
and oxalosuccinate (dashed lines). Enzymes: 1, pyruvate synthase; 2, PEP synthetase; 3, PEP carboxylase; 4,
malate dehydrogenase; 5, fumarase; 6, fumarate reductase; 7, succinyl–CoA synthetase; 8, α-ketoglutarate
synthase; 9, citrate synthase; 10, aconitase; 11, isocitrate dehydrogenase; 12, enolase, mutase, phosphoglyc-
erate kinase, and in triosephosphate dehydrogenase.
www.ebook777.com
pyruvate synthase. The pyruvate is then phos- Desulfobacter and Chlorobium, and the aer-
phorylated via PEP synthase to form phospho- obic Hydrogenobacter.30 (See note 31.) The
enolpyruvate. The phosphoenolpyruvate has pathway is also present in the Archaea and
two fates. Some of it enters the gluconeogenic is thought to have evolved earlier than the
pathway, and some is carboxylated to oxaloace- Calvin cycle as an autotrophic carbon diox-
tate via PEP carboxylase. In Methanobacterium ide fixation pathway.32 The overall reaction
thermoautotrophicum, the oxaloacetate is is the synthesis of one mole of oxaloacetate
reduced to α-ketoglutarate via an incomplete from four moles of carbon dioxide. In the
reductive citric acid pathway. That is, the reductive tricarboxylic acid pathway, phos-
reactions between citrate and α-ketoglutarate phoenolpyruvate (PEP) is carboxylated to
do not take place. M. thermoautotrophicum form oxaloacetate (OAA) via the enzyme
lacks citrate synthase and so must make its PEP carboxylase (Fig. 14.9, reaction 1). The
α-ketoglutarate this way. However, another oxaloacetate is reduced to succinate, which is
methanogen, M. barkeri, does have citrate derivatized to succinyl–CoA. The succinyl–
synthase and synthesizes α-ketoglutarate in an CoA is carboxylated to α-ketoglutarate (the
incomplete oxidative pathway via citrate and precursor to glutamate). This is a reductive
isocitrate. Thus, no methanogen seems to have carboxylation requiring reduced ferredoxin
a complete citric acid pathway.28 and is carried out by α-ketoglutarate syn-
thase. The α-ketoglutarate is carboxylated
14.1.8 Using the acetyl–CoA pathway to to isocitrate, which is converted to citrate.
oxidize acetate to CO2 anaerobically Therefore, this is really a reversal of the cit-
Several anaerobes can oxidize acetate to CO2 ric acid pathway, substituting fumarate
without the involvement of the citric acid cycle. reductase for succinate dehydrogenase and
They do this by operating the acetyl–CoA path- α-ketoglutarate synthase for α-ketoglutarate
way in reverse. These include some of the sulfate dehydrogenase. Thus far, one mole of phos-
reducers that grow heterotrophically on acetate phoenolpyruvate (C3) has been converted to
and oxidize it to CO2.29 The anaerobes begin one mole of citrate (C6).
the oxidation with reactions 4 and 3 to make The citrate is split by a special enzyme,
acetyl–CoA (Fig. 14.4) and then reverse reac- called ATP-dependent citrate lyase, to acetyl–
tions 14 through 11. This produces CH3–[Co-E] CoA (C2) and oxaloacetate (C4). [The ATP-
and CO–[CODH]. In reaction 15 of Fig. 14.4, dependent citrate lyase is found in eukaryotic
CO–[CODH] is oxidized to CO2. The CH3 is cells but is rare in prokaryotes. (See note 33.)
oxidized to CO2 via reactions 10 through 5. In Most prokaryotes use the citrate synthase
this way, both carbons in acetate are converted reaction, which proceeds only in the direction
to CO2. A Δp is made during electron transport. of citrate.] The oxaloacetate can be used for
(However, it may be that the enzymes involved growth.
in the oxidative acetyl–CoA pathway are not The acetyl–CoA is used to regenerate the
identical to those of the reductive acetyl–CoA phosphoenolpyruvate as follows. The acetyl–
pathway that operates in autotrophs.) This CoA (C2) is reductively carboxylated to pyru-
pathway is used by Desulfotomaculum acetoxi- vate (C3) by using reduced ferredoxin and the
dans and some other group II sulfate reducers enzyme pyruvate synthase. There is a ther-
(complete oxidizers). Still other group II sulfate modynamic reason to explain why reduced
reducers use the reductive tricarboxylic acid ferredoxin and not NADH is the electron
pathway to oxidize acetate (Section 14.1.9). donor. NADH is not a sufficiently strong
Archaean sulfate reducers (e.g., Archaeoglobus reductant to reduce acetyl–CoA and CO2 to
fulgidus) may also use the acetyl–CoA pathway pyruvate. Reduced ferredoxin has a potential
to oxidize acetate to CO2 anaerobically. more electronegative than that of NADH and
is therefore a stronger reductant. The pyru-
14.1.9 The reductive tricarboxylic acid vate is then converted to phosphoenolpyru-
pathway (reductive citric acid cycle) vate via PEP synthetase, thus regenerating
Bacteria that use the reductive TCA pathway the phosphoenolpyruvate. This discussion
are strict anaerobes belonging to the genera and Section 14.1.2 emphasize the importance
C1 metabolism 373
Fig. 14.9 The reductive tricarboxylic acid pathway. Enzymes: 1, PEP carboxylase; 2, malate dehydrogenase;
3, fumarase; 4, fumarate reductase; 5, succinyl–CoA synthetase (succinate thiokinase); 6, α-ketoglutarate
synthase; 7 and 8, isocitrate dehydrogenase; 9 and 10, aconitase; 11, ATP–citrate lyase; 12, pyruvate syn-
thase; 13, PEP synthetase.
of pyruvate synthase and PEP synthetase for 2. α-Ketoglutarate synthase replaces the NAD+-
anaerobic growth on acetate. linked α-ketoglutarate dehydrogenase.
Note that to reverse the citric acid cycle, three 3. ATP-dependent citrate lyase replaces citrate
new enzymes are required: synthase.
1. Fumarate reductase replaces succinate dehy- In addition, pyruvate synthase replaces pyruvate
drogenase. This commonly occurs in other dehydrogenase, and PEP synthetase replaces
prokaryotes that synthesize succinyl–CoA pyruvate kinase.
from oxaloacetate under anaerobic condi- Summary of the reductive tricarboxylic acid
tions. pathway:
CO2 + PEP → OAA + Pi

OAA + NADH + H+ + FADH2 + Fdred + ATP + CO2 → α-ketoglutarate + NAD+ + FAD + Fdox + ADP + Pi
α-Ketoglutarate + NADPH + H+ + CO2 → citrate
Citrate + ATP + CoASH → OAA + acetyl–CoA + ADP + Pi
Acetyl–CoA + Fdred + CO2 → pyruvate + Fdox
Pyruvate + ATP → PEP + AMP + Pi
4CO2 + 2 NAD(P)H + 2H+ + 2 Fdred + FADH2 + 3 ATP

→ OAA + 2 NAD(P)+ + FAD + 2 Fdox + 2 ADP + AMP + 4 Pi
www.ebook777.com
Anaerobic acetate oxidation by reversal of methanol but not on methane, are believed to be
the reductive tricarboxylic acid pathway involved in methane oxidation.
As mentioned previously, some group II sulfate The methyltrophs that grow on either C1
reducers reverse the reductive tricarboxylic compounds (methanol or methylamine) or
acid pathway to oxidize acetate to CO2.28 An multicarbon compounds are called facultative
example is Desulfobacter postgatei. As shown methylotrophs. The facultative methylotrophs
in Fig. 14.9, by reversing reactions 11 through are found in many genera and consist of both
6, acetyl–CoA is oxidized to CO2. The oxaloac- gram-positive and gram-negative bacteria. They
etate used in reaction 11 is regenerated in reac- include species belonging to the genera Bacillus,
tions 5 through 2. Acetyl–CoA is made from Acetobacter, Mycobacterium, Arthrobacter,
acetate by transferring the CoA from succinyl– Hyphomicrobium, Methylobacterium, and
CoA. Hence, reaction 5 is replaced by a CoA Nocardia. Some species of Mycobacterium can
transferase. grow on methane as well as on methanol or
multicarbon compounds.
14.2 Growth on C1 Compounds Other Methylotrophs assimilate the C1 carbon
source via either the ribulose–monophosphate
than CO2: The Methylotrophs (RuMP) pathway or the serine pathway. The
Many aerobic bacteria can grow on com- RuMP pathway assimilates formaldehyde into
pounds other than CO2 that do not have cell material, whereas the serine pathway assim-
carbon–carbon bonds. These bacteria are ilates carbon dioxide and formaldehyde. (See
called methylotrophs. For reviews, see refs. note 40.) There also exist bacteria that grow on
34–36. Compounds used for methylotrophic methanol and oxidize it to CO2, which is assim-
growth include single-carbon compounds ilated via the ribulose bisphosphate (RuBP)
such as methane (CH4), methanol (CH3OH), pathway (Calvin cycle). These microorganisms
formaldehyde (HCHO), formate (HCOOH), have been called “pseudomethylotrophs” or
and methylamine (CH3NH2), as well as mul- autotrophic methylotrophs.
ticarbon compounds without C–C bonds
such as trimethylamine [(CH3)3N], dimethyl 14.2.1 Growth on methane
ether [(CH3)2O], and dimethyl carbonate As mentioned previously, methanotrophic
(CH3OCOOCH3). These compounds appear bacteria utilize methane as their sole source of
in the natural habitat as a result of fermen- carbon and energy. For a review of the biochem-
tations and breakdown of plant and animal istry of methane oxidation, see ref. 41. Methane
products, and pesticides. (Note 37 points out is oxidized aerobically to CO2 in a series of four
where these compounds are found.) reactions. (Methane can also be oxidized anaer-
The methylotrophs are divided according to obically. See note 42 for an explanation and a
whether they can also grow on multicarbon com- reference to a review article.) The first oxida-
pounds. Those that cannot are called obligate tion is to methanol, and it uses a mixed-function
methylotrophs. The obligate methylotrophs that oxidase called methane monooxygenase. There
grow on methanol or methylamine but not on are two different methane monooxygenases:
methane are aerobic gram-negative bacteria of one in the membrane and one that is soluble.
the genera Methylophilus and Methylobacillus. All but one genus of methanotrophs have the
(Some strains of Methylophilus can use glucose as membrane-bound enzyme, and a small subset
a sole carbon and energy source.38) The obligate of methanotrophs produce both the membrane-
methylotrophs that grow on methane or metha- bound and membrane-soluble forms.
nol are called methanotrophs. (See note 39.) In the first oxidation reaction, one atom of
They are gram-negative and fall into five genera, oxygen is incorporated into methanol and the
Methylomonas, Methylococcus, Methylobacter, other atom is reduced to water, using NADH as
Methylosinus, and Methylocystis. All the the reductant:
methanotrophs form extensive intracellular
membranes and resting cells, either cysts or
CH4 + NADH + H+ + O2
exospores. The intracellular membranes, which
are not present in methylotrophs that grow on →CH3OH + NAD+ + H2O
C1 metabolism 375
than being oxidized to CO2. Depending upon

the particular bacterium, one of two formalde-
hyde fixation pathways is used: the serine path-
way or the ribulose–monophosphate cycle.
The serine pathway

Fig. 14.10 Structure of pyrroloquinoline–quinone
The serine pathway produces acetyl–CoA from
(PQQ).
formaldehyde and carbon dioxide (Fig. 14.11).
The second reaction is the oxidation of metha- The formaldehyde is incorporated into glycine
nol to formaldehyde, catalyzed by methanol to form serine in a reaction catalyzed by serine
dehydrogenase, which has as its prosthetic hydroxymethyltranseferase (reaction 1). In this
group a quinone called pyrroloquinoline qui- reaction, the formaldehyde is first attached to
none (PQQ) (Fig. 14.10): tetrahydrofolic acid to form methylene–THF.
(Tetrahydrofolic acid reactions are described in
CH3OH + PQQ →HCHO + PQQH2 Section 10.2.4). Methylene–THF then donates
the C1 unit to glycine to form serine, regenerating
In the third reaction the formaldehyde is oxidi-
THF. The serine is then converted to hydroxy-
zed to formate by formaldehyde dehydrogenase:
pyruvate via a transaminase, which aminates
HCHO + NAD+ + H2O glyoxylate, regenerating glycine (reaction 2).
Hydroxypyruvate is reduced to glycerate (reac-
→ HCOOH + NADH + H+ tion 3), which is phosphorylated to 3-phospho-
And the formate is oxidized to CO2 by formate glycerate (reaction 4). The 3-phosphoglycerate is
dehydrogenase: then converted to 2-phosphoglycerate (reaction
5), which is dehydrated to phosphoenolpyru-
HCOOH + NAD+→CO2 + NADH + H+ vate (reaction 6). The phosphoenolpyruvate is
Because the formaldehyde and formate dehy- carboxylated to oxaloacetate (reaction 7). The
drogenases are soluble enzymes that use NAD+, oxaloacetate is reduced to malate (reaction 8).
they are probably located in the cytoplasm. Of Malyl–CoA synthetase then converts malate
the two NADHs produced, one is reutilized in to malyl–CoA (reaction 9), which is split by
the monooxygenase reaction and the second malyl–CoA lyase to acetyl–CoA and glyoxylate
is fed into the respiratory chain in the usual (reaction 10). Thus, the glyoxylate is regener-
way. However, methanol dehydrogenase is a ated and the product is acetyl–CoA. So far, the
periplasmic enzyme in gram-negative bacteria. following has happened:
It is thought that the methanol diffuses into the HCHO + CO2 + 2 NADH + 2H+ + 2 ATP
periplasm, where it is oxidized by the dehydro-
genase. The electrons are transferred to periplas- + CoASH → acetyl–CoA + 2 NAD+
mic cytochromes c, which in turn transfer the + 2 ADP + 2 Pi
electrons to a membrane cytochrome oxidase,
which probably pumps protons out of the cell But how is acetyl–CoA incorporated into cell
during inward flow of electrons to oxygen. material? In some methylotrophs, the serine
A Δp is established by the inward flow of pathway goes around a second time to generate
electrons, the outward pumping of protons by a second oxaloacetate. Then the second oxalo-
the cytochrome aa3 oxidase, and the release of acetate condenses with the acetyl–CoA to form
protons in the periplasm and consumption in citrate (reaction 11). The citrate isomerizes to
the cytoplasm. (See Section 5.7.2 for a descrip- isocitrate (reaction 12). The isocitrate is cleaved
tion of electron transport and the generation of by isocitrate lyase to form succinate and glyoxy-
a Δp during methanol oxidation in Paracoccus late (reaction 13). The succinate can be assimi-
denitrificans.) A small number of gram-positive lated into cell material via oxaloacetate and
bacteria are also methylotrophic. Some of the PEP (Section 9.13). The second glyoxylate can
formaldehyde produced during methane oxi- be used for the second round of the serine path-
dation is incorporated into cell material rather way, which produces the second oxaloacetate.
www.ebook777.com
Fig. 14.11 The serine–isocitrate lyase pathway. Enzymes: 1, serine hydroxymethylase; 2, transaminase; 3,
hydroxypyruvate reductase; 4, glycerate kinase; 5, phosphoglycerate mutase; 6, enolase; 7, PEP carboxylase;
8, malate dehydrogenase; 9, malyl–CoA synthetase; 10, malyl–CoA lyase; 11, citrate synthase; 12, cis-aco-
nitase; 13, isocitrate lyase.
Therefore, the serine pathway and the isocitrate converted via the serine pathway to 3-phos-
lyase pathway (called the serine–isocitrate lyase phoglycerate, which is assimilated into cell
pathway) can be described by the following material.
overall reaction: The reactions of Fig. 14.11 are summarized
as follows:
2HCHO + 2CO2 + 3 NADH + 3H+ + 3 ATP
→succinate + 3 NAD+ + 3 ADP + 3 Pi + 2H2O 1. 2CH2O + 2 glycine → 2 serine
2. 2 Serine + 2 glyoxylate
Alternatively, the succinate can be converted → 2 glycine + 2 hydroxypyruvate
to the second oxaloacetate via fumarate and 3. 2 Hydroxypyruvate + 2 NADH + 2H+
malate, and the second glyoxylate can be → 2 glycerate + 2 NAD+
C1 metabolism 377
4. 2 Glycerate + 2 ATP (a one-carbon compound), stage 1 must be

→ 2 3-PGA + 2 ADP + 2 Pi repeated three times.
5. 2 3-PGA → 2 2-PGA In stage 2, one of the three fructose-6-phos-
6. 2 2-PGA → 2 PEP + 2H2O phates is split into two C3 compounds. There
7. 2 PEP + 2CO2 → 2 OAA + 2 Pi are two pathways for the cleavage, depend-
8. OAA + NADH + H+ → malate + NAD+ ing upon the organism. In one pathway,
9. Malate + ATP + CoASH fructose-6-phosphate is phosphorylated to
→ malyl–CoA + ADP + Pi fructose-1,6-bisphosphate, which is cleaved
10. Malyl–CoA → glyoxylate + acetyl–CoA via the fructose-1,6-bisphosphate aldolase
11. Acetyl–CoA + OAA → citrate to phosphoglyceraldehyde and dihydroxy-
12. Citrate → isocitrate acetone phosphate. This is called the FBPA
13. Isocitrate → glyoxylate + succinate pathway, for fructose bisphosphate aldolase.
If this route is taken, the net product of the
2CH2O + 2CO2 + 3 NADH + 3H+ + 3 ATP → pathway is dihydroxyacetone phosphate. In
succinate + 3 NAD+ + 3 ADP + 3 Pi + 2H2O the second pathway, fructose-6-phosphate is
isomerized to glucose-6-phosphate, which is
Note: The serine–isocitrate lyase pathway oxidatively cleaved to phosphoglyceraldehyde
depicted in Fig. 14.11 for formaldehyde (HCHO) and pyruvate by means of the enzymes of the
assimilation requires the continuous regenera- Entner–Doudoroff pathway (Section 9.6). This
tion of glyoxylate (reactions 11–13), and this is called the KDPGA pathway, for the 2-keto-
requires the enzyme isocitrate lyase (reaction 3-deoxy-6-phosphogluconate aldolase. If this
13). However, only a few methylotrophs have route is taken, the net product of the path-
isocitrate lyase, so an alternative pathway for way is pyruvate. Both pathways also produce
the regeneration of glyoxylate must exist. And, phosphoglyceraldehyde.
indeed this is the case. It is called the ethylmal- Stage 3 is a sugar rearrangement stage dur-
onyl–CoA pathway and is described in detail in ing which the phosphoglyceraldehyde pro-
refs. 34 and 43. The overall reaction converts duced in stage 2 and two fructose-6-phosphates
two acetyl–CoA molecules into glyoxylate and produced in stage 1 are used to regenerate the
succinyl–CoA: three ribulose-5-phosphates. In some bacteria
the rearrangements take place via the pentose
phosphate cycle enzymes (called the TA path-
2 Acetyl–CoA + 2CO2 + 2H
way, for transaldolase), whereas in other bac-
→ 1 glyoxylate + H+ + 1 succinyl–CoA teria the rearrangements use the enzymes of the
closely related Calvin cycle (called the SBPase
pathway, for sedoheptulose-1,7-bisphosphate
The ribulose–monophosphate cycle aldolase).
There are several methanotrophs that use the Taking into consideration the alternative
ribulose–monophosphate (RuMP) pathway pathways in stages 2 and 3, there are four
instead of alternative pathways for formal- variations of the RuMP cycle that might occur
dehyde assimilation.44 The pathway is shown in the methylotrophs. The obligate metha-
in Fig. 14.12. It is convenient to divide the notrophs (Methylococcus and Methylomonas)
RuMP pathway into three stages. Stage 1 and obligate methylotrophs (Methylophilus,
begins with condensation of formaldehyde Methylobacillus) that have been examined all
with ribulose-5-phosphate to form hexulose- use the KDPGA mode of cleavage and the TA
6-phosphate (reaction 1). The reaction is pathway of sugar arrangement. The facultative
catalyzed by hexulose phosphate synthase. methylotrophs thus far examined use the FBPA
The hexulose phosphate is then isomerized to mode of cleavage. Some use the SBPase sugar
fructose-6-phosphate by hexulose phosphate rearrangement pathway, and some use the TA
isomerase (reaction 2). To synthesize a three- pathway. Use of the KDPG aldolase and the
carbon compound (e.g., dihydroxyacetone SBPase pathways in combination has not yet
phosphate or pyruvate) from formaldehyde been found.
www.ebook777.com
Fig. 14.12 The ribulose–monophosphate cycle. Enzymes: 1, hexulose-6-phosphate synthetase; 2, hexu-

lose-6-phosphate isomerase; 3, phosphofructokinase; 4, fructose-1,6-bisphosphate (FBP) aldolase; 5, glu-
cose phosphate isomerase; 6, glucose-6-phosphate dehydrogenase; 7, 6-phosphogluconate dehydratase; 8,
2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase; 9, 12, 15, and 19, transketolase; 10, 14, 16, and 21,
ribulose-5-phosphate epimerase; 11, transaldolase; 13 and 20, ribose-5-phosphate isomerase; 17, sedohep-
tulose-1,7-bisphosphate aldolase; 18, sedoheptulose-1,7-bisphosphatase. Abbreviations: RuMP, ribulose-5-
phosphate; F6P, fructose-6-phosphate; FBP, fructose-1,6-bisphosphate; PGALD, phosphoglyceraldehyde;
DHAP, dihydroxyacetone phosphate; KDPG, 2-keto-3-deoxy-6-phosphogluconate; xyl-5-P, xylulose-5-
phosphate; rib-5-P, ribose-5-phosphate; sed-BP, sedoheptulose-1,7-bisphosphate.
14.3 Summary phosphate pathway and several glycolytic reac-

There are three characterized CO2 fixation path- tions to reduce CO2 to phosphoglyceraldehyde.
ways used for autotrophic growth: the Calvin It bypasses the transaldolase reaction and syn-
cycle, the acetyl–CoA pathway, and the reduc- thesizes sedoheptulose-7-P via an aldolase and
tive tricarboxylic acid pathway. The Calvin an irreversible phosphatase and proceeds only
cycle is the only one in the aerobic biosphere. It in the direction of pentose phosphate. The path-
is present in green plants, algae, cyanobacteria, way has two unique enzymes, RuMP kinase and
chemoautotrophs, and most photosynthetic RuBP carboxylase.
bacteria. The Calvin cycle uses the transketo- The acetyl–CoA pathway is widespread
lase and isomerization reactions of the pentose among anaerobes and occurs in both the archaea
C1 metabolism 379
and the bacteria. However, the archaea use dif- The phosphoenolpyruvate is regenerated from
ferent coenzymes to carry the C1 units. Not only acetyl–CoA via its carboxylation to pyruvate
is the pathway used reductively for autotrophic by means of pyruvate synthase and phospho-
growth, it also can be used for anaerobic oxi- rylation of the latter to phosphoenolpyruvate
dation of acetate to CO2 (e.g., by certain group using PEP synthetase. Therefore, the bacteria
II sulfate reducers). An analogous pathway is are synthesizing oxaloacetate by carboxylat-
used by methanogens to form CH4 and CO2 ing phosphoenolpyruvate and regenerating the
from acetate. A key enzyme in the acetyl–CoA phosphoenolpyruvate via a reductive citric acid
pathway is carbon monoxide dehydrogenase pathway. In some group II sulfate reducers, the
(CODH), which is capable of reducing CO2 to pathway can operate in the oxidative direction
the level of carbon monoxide and catalyzing its and oxidize acetate to CO2.
condensation with bound methyl and CoA to There are many bacteria that grow aerobi-
form acetyl–CoA. The bound methyl is made by cally on C1 compounds such as methane or
reducing a second molecule of CO2 via a sepa- methanol. They oxidize the C1 compounds
rate series of enzymatic reactions and transfer- to CO2, deriving ATP from the respiratory
ring the methyl group to the CODH. pathway in the usual way, whereupon they
The reduction of CO2 to bound methyl also incorporate the rest of the C1 at the level of
takes place during methanogenesis. In metha- formaldehyde, into cell carbon. Bacteria that
nogenesis, the bound methyl is further reduced do this are called methylotrophs. A subclass
to methane, an energy-yielding reaction, rather of methylotrophs are the methanotrophs, bac-
than being transferred to CODH. Growth on teria that are able to grow on methane. Three
acetyl–CoA requires the synthesis of phospho- pathways for formaldehyde incorporation
enolpyruvate because the glyoxylate pathway have been found in these bacteria. They are the
is not present in these organisms. It involves a serine–isocitrate lyase pathway, which incor-
ferredoxin-linked carboxylation of acetyl–CoA porates both formaldehyde and carbon diox-
to pyruvate catalyzed by pyruvate synthase, fol- ide, the ribulose–monophosphate cycle, which
lowed by the phosphorylation of pyruvate to incorporates only formaldehyde, and the eth-
phosphoenolpyruvate via the PEP synthetase. ylmalonyl–CoA pathway for formaldehyde
The incorporation of phosphoenolpyruvate incorporation, which was referenced but not
into cell material takes place by means of reac- described in this chapter.
tions common to heterotrophs. The use of C1 compounds is an important
The third CO2 fixation pathway has been part of the carbon cycle. The methane produced
found among the photosynthetic green sulfur by methanogens escapes into the aerobic atmo-
bacteria, Hydrogenobacter, and Desulfobacter sphere and is transformed back into carbon
species. The pathway is a reductive citric acid dioxide by aerobic methane oxidizers. The car-
pathway and synthesizes oxaloacetate from bon dioxide is reduced to organic carbon in the
four moles of CO2. There are two carboxylation aerobic environments by both photosynthetic
reactions. One carboxylation is common, even eukaryotes and chemoautotrophic prokaryotes.
among heterotrophs. It is the carboxylation of It is also reduced to organic carbon anaerobi-
phosphoenolpyruvate to form oxaloacetate cat- cally by photosynthetic prokaryotes, acetogens,
alyzed by PEP carboxylase. The oxaloacetate is and methanogens.
reduced to succinyl–CoA. The second carboxy- Methanogenesis occurs in a variety of anaer-
lation is found only in some strict anaerobes. It obic environments, including swamps and
is the ferredoxin-linked carboxylation of succi- marshes, the mammalian rumen, the intestine
nyl–CoA to form α-ketoglutarate catalyzed by of termites, and anaerobic microenvironments
α-ketoglutarate synthase. The α-ketoglutarate in soil, lake muds, and rice paddies. Indeed,
is converted to citrate via a reversal of the reac- close to 70% of the atmospheric methane is pro-
tions of the citric acid cycle, and the citrate is duced by methanogens. This is approximately
cleaved to oxaloacetate and acetyl–CoA via an 108 tons of methane per year. (The other 30%
ATP-dependent citrate lyase. The bacteria can or so originates from abiogenic sources such as
incorporate the oxaloacetate into cell carbon. biomass burning and coal mines.) Vast amounts
www.ebook777.com
of methane are produced by the methanogens, 6. Corrinoid enzymes are proteins containing vita-
which are responsible for cycling much of the min B12 derivatives as the prosthetic group. Vitamin
B12 is a cobalt-containing coenzyme.
earth’s carbon into a gaseous form for reutiliza-
tion by the methane oxidizers and eventually 7. Reviewed in ref. 3.
the CO2-fixing organisms. Clearly these micro- 8. Heise, R., V. Muller, and G. Gottschalk. 1993.
organisms play a critical life-supporting role in Acetogenesis and ATP synthesis in Acetobacterium
the biosphere. woodii are coupled via a transmembrane pri-
mary sodium ion gradient. FEMS Microbiol. Lett.
112:261–268.
Study Questions 9. Lu, W.-P., S. R. Harder, and S. W. Ragsdale.
1990. Controlled potential enzymology of methyl
1. Determine the number of ATPs required to transfer reactions involved in acetyl–CoA synthesis
by CO dehydrogenase and the corrinoid/iron–sulfur
fix three moles of CO2 into phosphoglycer-
protein from Clostridium thermoaceticum. J. Biol.
aldehyde by means of the Calvin cycle, the Chem. 265:3124–3133.
acetyl–CoA pathway, and the reductive car-
10. Pezacka, E., and H. G. Wood. 1986. The auto-
boxylic acid pathway. trophic pathway of acetogenic bacteria: role of CO
2. Describe the role of carbon monoxide dehy- dehydrogenase disulfide reductase. J. Biol. Chem.
261:1609–1615.
drogenase in methane and carbon dioxide
production from acetate, and in acetyl–CoA 11. Ragsdale, S. W., and H. G. Wood. 1985. Acetate
synthesis from CO2. biosynthesis by acetogenic bacteria: evidence that
carbon monoxide dehydrogenase is the condensing
3. Why can anaerobes but not aerobes carbox- enzyme that catalyzes the final steps of the synthesis.
ylate acetyl–CoA and succinyl–CoA? J. Biol. Chem. 260:3970–3977.
12. Doukov, T. I., T. M. Iverson, J. Seravalli, S. W.
Ragsdale, and C. L. Drennan. 2002. A Ni-Fe-Cu cen-
ter in a bifunctional carbon monoxide dehyhdroge-
REFERENCES AND NOTES nase/acetyl–CoA synthase. Science 298:567–572.
1. The properties of the Calvin cycle enzymes and 13. Fuchs, G. 1990. Alternatives to the Calvin cycle
their regulation are reviewed in: Bowien, B. 1989. and the Krebs cycle in anaerobic bacteria: pathways
Molecular biology of carbon dioxide assimila- with carbonylation chemistry, pp. 13–20. In: The
tion in aerobic chemolithotrophs, pp. 437–460. In: Molecular Basis of Bacterial Metabolism. G. Hauska
Autotrophic Bacteria. H. G. Schlegel and B. Bowien and R. Thauer (Eds.). Springer-Verlag, Berlin.
(Eds.). Science Tech Publishers, Madison, WI, and 14. Reviewed in: Jetten, M. S. M., A. J. M. Stams,
Springer-Verlag, Berlin. and A. J. B. Zehnder. 1992. Methanogenesis from
2. It appears that the sedoheptulose-1,7-bisphos- acetate: a comparison of the acetate metabolism in
phatase and the fructose-1,6-bisphosphatase may Methanothrix soehngenii and Methanosarcina spp.
be a single bifunctional enzyme, capable of using FEMS Microbiol. Rev. 88:181–198.
either fructose-1,6-bisphosphate or sedoheptulose- 15. Wolfe, R. S. 1990. Novel coenzymes of archae-
1,7-bisphosphate as the substrate. In addition, the bacteria, pp. 1–12. In: The Molecular Basis of
fructose-1,6-bisphosphate aldolase and the sedo- Bacterial Metabolism. G. Hauska and R. Thauer
heptulose-1,7-bisphosphate aldolase may be a single (Eds.). Springer-Verlag, Berlin.
bifunctional enzyme.
16. Weiss, D. S., and R. K. Thauer. 1993. Methano-
3. Reviewed in: Wood, H. G., and L. G. Ljungdahl. genesis and the unity of biochemistry. Cell 72:819–822.
1991. Autotrophic character of the acetogenic bacte-
ria, pp. 201–250. In: Variations in Autotrophic Life. 17. Olson, K. D., L. Chmurkowska-Cichowlas, C.
J. M. Shively and L. L. Barton (Eds.). Academic Press, W. McMahon, and R. S. Wolfe. 1992. Structural
New York. modifications and kinetic studies of the substrates
involved in the final step of methane formation
4. The “acetogenic bacteria” are defined as anaero-
in Methanobacterium thermoautotrophicum. J.
bic bacteria that synthesize acetic acid solely from
Bacteriol. 174:1007–1012.
CO2 and secrete the acetic acid into the media. Many
of the acetogenic bacteria are facultative heterotrophs 18. HTP–SH is also known as factor B or coenzyme
that use the acetyl–CoA pathway to reduce CO2 to B. It is 7-mercaptoheptanolythreonine phosphate.
acetic acid as an electron sink, but these can also be
19. Ermler, U., W. Grabarse, S. Shima, M. Goubeaud,
grown autotrophically on CO2 and H2.
and R. K. Thauer. 1997. Crystal structure of methyl–
5. The early history is reviewed in: Wood, H. G. 1985. coenzyme M reductase: the key enzyme of biological
Then and now. Annu. Rev. Biochem. 54:1–41. methane formation. Science 278:1457–1462.
C1 metabolism 381
20. There are several reasons for suggesting that a 32. Reviewed in: Schonheit, P., and T. Schafer.
Δp is generated during electron transfer to the mixed 1995. Metabolism of hyperthermophiles. World J.
disulfide: (1) M. barkeri creates a Δp during catabo- Microbiol. Biotechnol. 11:26–57.
lism of acetate; (2) the membranes of M. thermophila
33. In eukaryotic cells, the ATP-dependent citrate
and M. barkeri contain electron carriers, including
lyase is used to generate acetyl–CoA from citrate.
cytochrome b and FeS proteins; and (3) over 50% of
The acetyl–CoA is then used as a precursor to fatty
the heterodisulfide reductase is in the membranes.
acids and steroids in the cytosol.
21. Becher, B., and V. Müller. 1994. ΔμNa+ drives
34. Chistoserdova, L., M. G., Kalyuzhnaya, and
the synthesis of ATP via an ΔμNa+-translocating F1F0–
M. E. Lidstrom. 2009. The expanding world of
ATP synthase in membrane vesicles of the archaeon
methylotrophic metabolism. Annu. Rev. Microbiol.
Methanosarcina mazei Göl. J. Bacteriol. 176:2543–
63:477–499.
2550.
35. Reviewed in: Colin Murrell, J., and H. Dalton
22. Ferry, J. G. 1992. Methane from acetate. J.
(Eds.). 1992. Methane and Methanol Utilizers.
Bacteriol. 174:5489–5495.
Plenum Press, New York and London.
23. The methanogens are paramount in recycling
36. Reviewed in: Lidstrom, M. E. 1992. The aero-
organic carbon into gaseous forms of carbon in
bic methylotrophic bacteria, pp. 431–445. In: The
anaerobic habitats (Section 14.4.) Most of the meth-
Prokaryotes, Vol. 1, 2nd ed. A. Balows, H. G. Truper,
ane that enters the aerobic atmosphere is oxidized to
M. Dworkin, W. Harder, and K.-H. Schleifer (Eds.).
CO2 by the aerobic methanotrophs (Section 13.2).
Springer-Verlag, Berlin.
24. Fischer, R., and R. K. Thauer. 1989.
37. Methanol appears in the environment as a result
Methyltetrahydromethanopterin as an intermediate
of the microbial breakdown of plant products with
in methanogenesis from acetate in Methanosarcina
methoxy groups (e.g., pectins and lignins). Formate
barkeri. Arch. Microbiol. 151:459–465.
is a fermentation end product excreted by ferment-
25. Grahame, D. A. 1991. Catalysis of acetyl–CoA ing bacteria. Methylamines can result from the
cleavage and tetrahydrosarcinapterin methylation breakdown products of some pesticides and certain
by a carbon monoxide dehydrogenase–corrinoid other compounds, including carnitine and lecithin
enzyme complex. J. Biol. Chem. 266:2227–2233. derivatives. Carnitine is a trimethylamine derivative
that is present in many organisms and in all animal
26. Bott, M., B. Eikmanns, and R. K. Thauer. 1986.
tissues, especially muscle. It functions as a carrier
Coupling of carbon monoxide oxidation to CO2
for fatty acids across the mitochondrial membrane
and H2 with the phosphorylation of ADP in acetate-
into the mitochondria, where the fatty acids are oxi-
grown Methanosarcina barkeri. Eur. J. Biochem.
dized. Lecithin is phosphatidylcholine, which is a
159:393–398.
phospholipid containing a trimethylamine group.
27. Stupperich, E., and G. Fuchs. 1984. Autotrophic (Section 10.1.2.)
synthesis of activated acetic acid from two CO2
38. Green, P. N. 1992. Taxonomy of methylotrophic
in Methanobacterium thermoautotrophicum. I.
bacteria, pp. 23–84. In: Methane and Methanol
Properties of the in vitro system. Arch. Microbiol.
Utilizers. J. Colin Murrell and H. Dalton (Eds.).
139:8–13.
Plenum Press, New York and London.
28. Reviewed in: Danson, M. J. 1988. Archaebacteria:
39. The methanotrophs are responsible for oxidizing
the comparative enzymology of their central meta-
about half the methane produced by methanogens.
bolic pathways, pp. 165–231. In: Advances in
The rest of the methane escapes to the atmosphere.
Microbial Physiology, Vol. 29. A. H. Rose and D. W.
Tempest (Eds.). Academic Press, New York. 40. Most of the nonmethanotrophs that are obligate
methylotrophs use the RuMP pathway. Within the
29. Hansen, T. A. 1993. Carbon metabolism of methanotrophs, some use the RuMP pathway and
sulfate-reducing bacteria, pp. 21–40. In: The Sulfate- some the serine pathway.
Reducing Bacteria: Contemporary Perspectives.
J. M. Odom and R. Singleton Jr. (Eds.). Springer- 41. Hakemian, A. S., and A. C. Rosenzweig. 2007.
Verlag, New York. The biochemistry of methane oxidation. Annu. Rev.
Biochem. 76:223–241.
30. Reviewed in: Amesz, J. 1991. Green photosyn-
thetic bacteria and heliobacteria, pp. 99–119. In: 42. For many years it was thought that methane
Variations in Autotrophic Life. J. M. Shively and L. was oxidized only aerobically using a monooxyge-
L. Barton (Eds.). Academic Press, New York. nase via the pathway described in Section 14.2.1.
However, recent observations indicate that anaero-
31. Desulfobacter is a group II sulfate-reducing bac- bic oxidation of methane exists in anaerobic sedi-
terium that grows on fatty acids, especially acetate; ments where methane production (methanogenesis)
it uses sulfate as an electron acceptor and produces is taking place.
sulfide (Section 12.2.2). Chlorobium is a green sul-
fur photoautotroph. Hydrogenobacter is an obligate
aerobic, chemolithotrophic hydrogen oxidizer. CH4 + SO42– + H+ → CO2 +HS– +2H2O
www.ebook777.com
This has been reported in sediments at the bottom of of methane, as well as ammonium, read: Strous, M.,
the Black Sea and in deep-sea sediments that over- and M. S. M. Jetten. 2004. Anaerobic oxidation of
lie natural gas deposits where methane is produced. methane and ammonium. Annu. Rev. Microbiol.
At this time, the anaerobic oxidation of methane is 58:99–117.
believed to require two different organisms that live
43. Peyraud, R., P. Kiefer, C. P. Massou, J. C.
in mutual dependence (syntrophism) but have not yet
Portais, and J. A. Vorholt. 2009. Demonstration of
been grown in pure culture. The methane oxidizer is
the ethylmalonyl–CoA pathway using 13C metabo-
a methanogenic archaeon that is thought to oxidize
lomics. Proc. Natl. Acad. Sci. USA. 106:4846–
methane via a pathway very similar to the reverse of
4851.
the methanogenic pathway shown in Fig. 14.6. The
electrons removed from the methane are transferred 44. Dijkhuizen, L., P. R. Levering, and G. E. De
via an unknown carrier to sulfate-reducing bacte- Vries. 1990. The physiology and biochemistry of
ria living in the consortium and are used to reduce aerobic methanol-utilizing gram-negative and gram-
sulfate to sulfide. The “electron shuttles” suggested positive bacteria, pp. 149–181. In: Methane and
include formate and acetate. How the methanogen Methanol Utilizers. Biotechnology Handbooks, Vol.
derives energy from this reversal of methanogenesis 5. J. Colin Murrell and H. Dalton (Eds.). Plenum
is not known. For a review of anaerobic oxidation Press, New York and London.
15
Fermentations
In the numerous anaerobic niches in the carbohydrate fermentations can be grouped

biosphere (muds, sewage, swamps, etc.) are into six main classes: lactic, ethanol, butyric,
prokaryotes that can grow indefinitely in the mixed acid, propionic, and homoacetic. A
complete absence of oxygen, a capability that homoacetic fermentation of glucose is described
is rare among eukaryotes. (Eukaryotes require in Section 14.1.3 and Fig. 14.4.
oxygen to synthesize unsaturated fatty acids and
sterols.) Anaerobically growing prokaryotes
reoxidize NADH and other reduced electron 15.1 Oxygen Toxicity
carriers either by anaerobic respiration (e.g., For a review of protection against oxygen tox-
using nitrate, sulfate, or fumarate as the electron icity, see ref.1 Many anaerobic prokaryotes
acceptor) or by carrying out a fermentation. (strict anaerobes) are killed by even small traces
A fermentation is defined as a pathway in of oxygen. Strict anaerobes are killed by oxy-
which NADH (or some other reduced electron gen because toxic products of oxygen reduc-
acceptor that is generated by oxidation reaction accumulate in the cell. The toxic products
tions in the pathway) is reoxidized by metab- of oxygen are produced when single electrons
olites produced by the pathway. The redox are added to oxygen sequentially.2 These toxic
.
reactions occur in the cytosol rather than in the products are hydroxyl radical (OH ), superox-
membranes, and ATP is produced via substrate- ide radical (O2–), and hydrogen peroxide (H2O2).
level phosphorylation. A more complete discussion of oxygen toxicity
It should not be concluded that fermentation can be found in Chapter 16.
occurs only among prokaryotes. Eukaryotic The superoxide radical forms because oxy-
microorganisms (e.g., yeast) can live fermen- gen is reduced by single-electron steps:
tatively, although as mentioned, oxygen is O2 + e– → O2–
usually necessary unless the medium is sup-
plemented with sterols and unsaturated fatty A small amount of superoxide radical is always
acids. Furthermore, certain animal cells (e.g., released from the enzyme when oxygen is
muscle cells, human red blood cells) are capable reduced by electron carriers such as flavopro-
of fermentation. teins or cytochromes. This is because the elec-
Fermentations are named after the major trons are transferred to oxygen one at a time.
end products they generate. For example, yeast The hydroxyl radical and hydrogen peroxide
carry out an ethanol fermentation, and muscle are derived from the superoxide radical.
cells and red blood cells carry out a lactic acid Because aerobic and aerotolerant organ-
fermentation. Microorganisms perform many isms have an enzyme called superoxide dis-
different types of fermentation. However, the mutase, missing in strict anaerobes, they do not
383
www.ebook777.com
accumulate superoxide radicals. The superox- bacteria reveals a variety of methods, which
ide dismutase catalyzes the following reaction: were described in earlier chapters. Some of these
are reviewed in Fig. 15.1. Most fermenting bac-
O2– + O2– + 2H2 → H2O2 + O2
teria make most or all of their ATP via substrate-
Notice that one superoxide radical transfers its level phosphorylations, and they create a ∆p
extra electron to the second radical, which is (needed for solute transport, motility, etc.) by
reduced to hydrogen peroxide. Strict anaerobes reversing the membrane-bound ATPase. Many
also lack catalase, the enzyme that converts anaerobic bacteria also generate a ∆p by reduc-
hydrogen peroxide to water and oxygen: ing fumarate. This is an anaerobic respiration
that can occur during fermentative metabolism
H2O2 + H2O2 → 2H2O + O2
when fumarate is produced. During fumar-
Catalase catalyzes the transfer of two electrons ate reduction, NADH dehydrogenases donate
from one hydrogen peroxide molecule to the sec- electrons to fumarate via menaquinone, and a
ond, oxidizing the first to oxygen and reducing ∆p is created, perhaps via a quinone loop. (See
the second to two molecules of water. Table 15.1 Section 5.6.1 for a discussion of quinone loops.)
shows the distribution of catalase and superox- The production of a ∆p by fumarate respiration
ide dismutase in aerobes and anaerobes. in fermenting bacteria probably spares ATP
If the H2O2 is not disposed of, it can oxidize that would normally be hydrolyzed to maintain
transition metals such as free iron (II) in the the ∆p.
Fenton reaction and form the free hydroxyl Some anaerobes (e.g., Wolinella succino-
.
radical, OH : genes) carry out a periplasmic oxidation of
. electron donors, such as H2 or formate in the
Fe2+ + H+ + H2O2 → Fe3+ + OH + H2O
case of W. succinogenes, to create a ∆p. (See
(See note 3 for a definition of transition metals.) Section 5.7.4.) There are several other means
available to anaerobic bacteria for generating a
15.2 Energy Conservation by proton motive force or a sodium motive force.
Anaerobic Bacteria Some anaerobes and facultative anaerobes are
capable of creating an electrochemical ion gra-
An examination of mechanisms of energy con-
dient by symport of organic acids out of the
servation and ATP production in anaerobic
cell with protons or sodium ions. The organic
acids are produced during fermentation, and
Table 15.1 The distribution of catalase and
the energy to create the electrochemical gradi-
superoxide dismutase ent is due to the concentration gradient of the
excreted organic acid (high inside). This has
Bacterium Superoxide Catalase been demonstrated for lactate excretion (pro-
dismutase ton symport) by the lactate bacteria and for
succinate excretion (sodium ion symport) by a
Aerobes or facultative
anaerobes rumen bacterium, Selenomonas ruminantium.
Escherichia coli + + (See Section 4.8.3 for a discussion of lactate and
Pseudomonas species + + succinate efflux in symport with protons and
Deinococcus + + sodium ions.)
radiodurans
Klebsiella pneumoniae is capable of generat-
Aerotolerant bacteria
Butyribacterium + – ing an electrochemical sodium ion gradient by
rettgeri using the energy released from the decarboxyla-
Streptococcus faecalis + – tion of oxaloacetate to pyruvate (Section 4.8.1).
Streptococcus lactis + – The decarboxylase pumps Na+ out of the cell.
Strict anaerobes
The sodium potential that is created is used to
Clostridium – –
pasteurianum drive the uptake of oxaloacetate, which is used
Clostridium – – as a carbon and energy source. Oxalobacter
acetobutylicum formigenes creates a proton potential by cata-
Source: Stanier, R. Y., J. L. Ingraham, M. L. Wheelis, and lyzing an electrogenic anion exchange coupled
P. R. Painter. 1986. The Microbial World. Reprinted by per- to the decarboxylation of oxalate to formate
mission of Prentice-Hall, Upper Saddle River, NJ. (Section 4.8.2).
fermentations 385
Fig. 15.1 Energy conservation in anaerobic bacteria. (A) Substrate-level phosphorylation: 1, the PGA kinase
reaction; 2, the pyruvate kinase reaction; 3, the acetate kinase reaction. (B) Fumarate respiration. When the
electron donor is NADH, a Q loop probably operates to translocate protons out of the cell. When the electron
donor is periplasmic, proton translocation is not necessary. (C) Efflux of an organic acid coupled to protons
or sodium ions (e.g., the coupled efflux of protons and lactate by the lactate bacteria). (D) Decarboxylation
of an organic acid coupled to Na+ efflux (e.g., Klebsiella). (E) Electrogenic oxalate:formate exchange in
Oxalobacter.
15.3 Electron Sinks electron acceptor (e.g., oxygen or nitrate). Since

What to do with the electrons removed in the fermentations usually occur in the absence of an
course of oxidations poses a major problem exogenously supplied electron acceptor, the fer-
that must be addressed during fermentation. mentation pathways themselves must produce
For example, how is the NADH reoxidized? the electron acceptors for the electrons pro-
Respiring organisms do not have this prob- duced during oxidations. The electron acceptors
lem, since the electrons travel to an exogenous are called “electron sinks” because they dispose
www.ebook777.com
of the electrons removed during the oxidations,

and the reduced products are excreted into the
medium. Consequently, fermentations are char-
acterized by the excretion of large quantities of
reduced organic compounds such as alcohols,
organic acids, and solvents. Frequently, hydro-
gen gas is also produced, since hydrogenases use
protons as electron acceptors. Generally, one
can view fermentations in the following way,
where AH2 is the compound being fermented,
and B is the electron sink generated from AH2:
AH2 + NAD+ + Pi + ADP → B + NADH + H+ + ATP
Fig. 15.2 The anaerobic food chain.
B + NADH + H+ → BH2 (excreted) + NAD+
and
acids, alcohols, hydrogen gas, and carbon diox-
Hydrogenase ide. These fermentation end products are oxi-
NADH + H+ → H2 + NAD+ dized to CO2, H2, and acetate by organisms that
have been only partially identified. Finally, the
Thus there is the lactate fermentation: methanogens grow on the acetate, H2, and CO2,
Glucose + 2 ADP + 2 Pi → 2 lactate + 2 ATP converting these to CH4 and CO2.
or, to choose another example, an ethanol

fermentation:
15.4.1 Interspecies hydrogen transfer
Some anaerobic bacteria use protons as the
Glucose + 2 NAD+ + 2 ADP + 2 Pi → major or sole electron sink. They include the
2 pyruvate + 2 NADH + 2 H+ + 2 ATP obligate proton–reducing acetogens that oxi-
2 Pyruvate → 2 acetaldehyde + 2CO2 dize butyrate, propionate, ethanol, and other
compounds to acetate, H2, and CO2. The physi-
2 Acetaldehyde + 2 NADH + 2H+ → ology of these organisms is not well understood.
2 ethanol + 2 NAD+ Probably the electrons travel from the organic
Glucose + 2 ADP + 2 Pi → 2 ethanol + substrate to an intermediate electron carrier
2CO2 + 2 ATP (e.g., from NAD+ to hydrogenase to H+). Some
of these oxidations are as follows:
15.4 The Anaerobic Food Chain CH3CH2CH2COO– + 2H2O →
The fermentation of amino acids, carbohy- 2CH3COO– + H+ + 2H2 ∆G90= +48.1 kJ
drates, purines, and pyrimidines to organic acids CH3CH2COO– + 3H2O → CH3COO–
and alcohols (acetate, ethanol, butanol, propi- + HCO3– + H+ + 3H2 ∆G'0 = +76.1 kJ
onate, succinate, butyrate, etc.) by prokaryotes,
CH3CH2OH + H2O → CH3COO–
and the conversion of these fermentation end
+ H+ + 2H2 ∆G90 = +9.6 kJ
products to CO2, CH4, and H2 by the combined
action of bacteria of several different types, Notice that these reactions are thermodynami-
is called the anaerobic food chain (Fig. 15.2). cally unfavorable under standard conditions at
Anaerobic environments include muds, lake pH 7. In fact, the proton reducers live symbi-
bottoms, and sewage treatment plants. otically with H2 utilizers that keep the H2 levels
The anaerobic food chain is an important part low and pull the reaction toward H2 produc-
of the carbon cycle and serves to regenerate gas- tion. The H2 utilizers are methanogens and sul-
eous carbon (i.e., carbon dioxide and methane), fate reducers. (Under conditions of sufficient
which is reutilized by other microorganisms and sulfate, the sulfate reducers predominate over
plants throughout the biosphere. As illustrated methanogens. The sulfate reducers also oxi-
in Fig. 15.2, the process can be viewed as occur- dize ethanol, lactate, and other organic acids to
ring in three stages. Fermenters produce organic acetate.)
fermentations 387
The symbiotic relationship between the obli- that the C/O ratio is 1. This changes C2H6O
gate proton reducers and the hydrogen utiliz- to C2H8O2. Acetic acid is C2H4O2. Since the
ers is called a syntrophic association.4 It is also C/O ratio is already 1, nothing further need
called interspecies hydrogen transfer. In fact, be done. The formula for carbon dioxide is
many other fermenting bacteria besides the CO2. To make the C/O ratio 1, it is necessary
obligate proton reducers have dehydrogenases to subtract H2O, that is: [CO2 – H2O]. The
that transfer electrons from NADH to protons result is C(–2H)O.
and can use protons as an electron sink when 2. Now compare the number of hydrogens in
grown in the presence of hydrogen gas utilizers. the modified formula with (CH2O)n, which
An example is Ruminococcus albus, which is has the same number of carbons as in the
discussed in Section 15.12. modified formula. For ethanol, C2H8O2 is
Hydrogenases are also coupled to the ther- compared with (CH2O)2.There are 8 hydro-
modynamically favored oxidation of reduced gens in C2H8O2 but only 4 in (CH2O)2. Thus,
ferredoxin, as in pyruvate:ferredoxin oxi- ethanol has an additional 4 hydrogens.
doreductase found in the clostridia, sulfate Carbon dioxide, C(–2H)O, however, has
reducers (Section 13.2.2), and Ruminococcus 4 fewer hydrogens than CH2O. Acetic acid
albus (see later: Fig. 15.12). Ferredoxins often (C2H4O2) and (CH2O)2 have the same num-
have two iron–sulfur clusters, each one capa- ber of hydrogens.
ble of carrying an electron (also indicated in 3. Add –1 for each additional 2H and +1 for a
Fig. 15.12). decrease in 2H. Thus, the O/R for ethanol is
–2, for CO2 it is +2, and for acetic acid it is 0.
15.5 How to Balance a Fermentation Since both the oxidized and reduced fermenta-
A written fermentation is said to be balanced tion end products originate from the substrate,
when the hydrogens produced during the oxi- in a balanced fermentation, the sum of the O/R
dations equal the hydrogens transferred to the of the products equals the O/R of the substrate.
fermentation end products. Only under these For example, the O/R for glucose is 0. When
conditions can all of the NADH and reduced one mole of glucose is fermented to two moles
ferredoxin be recycled to the oxidized forms. It of ethanol and two moles of carbon dioxide, the
is important to know whether a fermentation O/R of the products is (–2 × 2) + (+2 × 2) = 0.
is balanced because if it is not, the overall writ- Often one simply takes the ratio (+/–) of the
ten reaction is incorrect. There are two methods O/R of the products when a carbohydrate is
used to balance fermentations, the oxidation/ fermented. For a balanced fermentation, +/–
reduction (O/R) method and the available should be 1.
hydrogen method; both are bookkeeping meth-
ods to keep track of the hydrogens.
15.5.2 The available hydrogen method
Like the O/R method, the available hydrogen
15.5.1 The O/R method
procedure is merely one of bookkeeping. It has
The O/R method entailing the computation of nothing to do with the chemistry of the reac-
an oxidation/reduction balance is computed as tions. According to this method, one “oxidizes”
described here. One arbitrarily designates as the molecule to CO2 by using water to obtain
zero the O/R value for formaldehyde (CH2O) the “available hydrogen.” For example:
and multiples thereof [(CH2O)n] and uses that
formula as a standard against which to com- C6H12O6 + 6H2O → 24H + 6CO2
pare the reduction level of other molecules.
Thus glucose has 24 available hydrogens.
Following are the steps in determining the O/R
The available H in all the products must add
value of any molecule.
up to the available H in the starting material.
1. Add or subtract water to the molecule in Table 15.2 gives the concentration of products
question to make the C/O ratio 1. This will per 100 moles of glucose used. The available H
allow a comparison to (CH2O)n. For examin the glucose is 24 × 100 = 2,400. The available
ple, the formula for ethanol is C2H6O. The H in the products adds up to 2,242. Thus, the
C/O ratio is 2. One water must be added so balance is 2,400/2,242 = 1.07.
www.ebook777.com
Table 15.2 Balancing an acetone–butanol fermentation

Substrate and Moles per Carbon O/R O/R value Available Available H
products 100 moles of (mol) value (mol/100 mol) H (mol/100 mol)
substrate
Substrate
Glucose 100 600 0 — 24 2,400
Products
Butyrate 4 16 –2 8 20 80
Acetate 14 28 0 — 8 112
CO2 221 221 +2 +442 0 —
H2 135 — –1 –135 2 270
Ethanol 7 14 –2 –14 12 84
Butanol 56 224 –4 –224 24 1,344
Acetone 22 66 –2 –44 16 352
Total 569 –425, +442 2,242
Source: Gottschalk, G. 1986. Bacterial Metabolism. Springer-Verlag, Berlin.
15.6 Propionate Fermentation via yielding 2[H] again (reaction 2). The acetyl–
the Acrylate Pathway CoA is converted to acetate and ATP via
acetyl–P (reactions 3 and 4). During the oxida-
The genus Clostridium comprises a heteroge-
tions, 4[H] are produced, which must be reuti-
neous group of bacteria consisting of gram-
lized. The electron acceptor is created from a
positive, anaerobic, spore-forming bacteria
second and third molecule of lactate (actually
that cannot use sulfate as a terminal electron
lactyl–CoA). The lactate acquires a CoA from
acceptor. The clostridia can be isolated from
propionyl–CoA (reaction 5). The lactyl–CoA is
anaerobic regions (or areas of low oxygen lev-
dehydrated to yield the unsaturated molecule,
els) in soil. They ferment organic nutrients to
acrylyl–CoA (reaction 6). Each acrylyl–CoA
products that can include alcohols, organic
is then reduced to propionyl–CoA, using up
acids, hydrogen gas, and carbon dioxide. C.
the 4[H] (reaction 7). The fermentation is thus
propionicum oxidizes three moles of lactate to
balanced. The propionate, which is produced
two moles of propionate, one mole of acetate,
during the CoA transfer step (reaction 5), is
and one mole of carbon dioxide, to produce one
catalyzed by CoA transferase, an enzyme that
mole of ATP. The pathway is called the acry-
occurs in many anaerobes.
late pathway because one of the intermediates
What can we learn from this pathway? The
is acrylyl–CoA. The bacteria derive ATP via
bacteria use a standard method for making ATP
a substrate-level phosphorylation during the
under anaerobic conditions. They decarboxy-
conversion of acetyl–P to acetate catalyzed by
late pyruvate to acetyl–CoA and then, using a
acetate kinase. Since only one acetate is made
phosphotransacetylase (to make acetyl–P) and
for every three lactates used, the pathway yields
an acetate kinase, they make ATP and acetate.
one-third of an ATP per lactate. Growth yields
These reactions are widespread among fer-
are proportional to the amount of ATP pro-
menting bacteria. The production of acetate is
duced, and it is to be expected that the growth
presumed to be associated always with the syn-
yields for these organisms are very low. (The
thesis of two moles of ATP per mole of acetate
molar growth yield for ATP is 10.5 g of cells per
if the bacteria are growing on glucose and using
mole of ATP synthesized, Section 2.3.)
the Embden–Meyerhof–Parnas pathway. One
ATP is produced from acetyl–CoA, and the sec-
15.6.1 The fermentation pathway of ond ATP is produced during the production of
Clostridium propionicum the pyruvate in the EMP pathway.
A molecule of lactate is oxidized to pyruvate, Another common reaction among ferment-
yielding 2[H] (Fig. 15.3, reaction 1). The pyru- ing bacteria is the transfer of coenzyme A from
vate is then oxidized to acetyl–CoA and CO2, one organic molecule to another, a reaction
fermentations 389
Fig. 15.3 Propionate fermentation via the acrylate pathway. Enzymes: 1, lactate dehydrogenase; 2, pyruvate–
ferredoxin oxidoreductase; 3, phosphotransacetylase; 4, acetate kinase; 5, CoA transferase; 6, lactyl CoA
dehydratase 7, a dehydrogenase.
catalyzed by CoA transferase. The other way of characteristic flavor of this cheese is due to the
attaching a coenzyme A molecule to a carboxyl propionate, and the holes in the cheese are due
group is to transfer an AMP or a phosphate to to the carbon dioxide produced.
the carboxyl group from ATP, making an acyl The pathway illustrated in Fig. 15.4 shows
phosphate or an acyl–AMP, and then displac- that three molecules of lactate are oxidized to
ing the AMP or phosphate with CoASH. (Recall pyruvate (reaction 1). This yields six electrons.
the activation of fatty acids prior to their deg- Then, one pyruvate is oxidized to acetate, CO2,
radation, Section 10.1.1.) However, fermenting and ATP, yielding two more electrons (reac-
organisms must conserve ATP. The CoA trans- tion 2). We now have eight electrons to use up.
ferase reaction is one way this can be done. The other two pyruvates are carboxylated to
yield two molecules of oxaloacetate (reaction 5).
The reaction is catalyzed by methylmalonyl–
15.7 Propionate Fermentation via CoA transcarboxylase. The two oxaloacetates
the Succinate–Propionate Pathway are reduced to two malates, consuming a total
Many bacteria that yield propionic acid as a of four electrons (reaction 6). The two malates
product of fermentation use a pathway dif- are dehydrated to two fumarates (reaction 7).
ferent from the acrylate pathway. The other The two fumarates are reduced via fumarate
pathway, called the succinate–propionate reductase to two succinates, consuming four
pathway, yields more ATP than the acrylate electrons (reaction 8). The latter reduction is
pathway per mole of propionate formed. One coupled to the generation of a ∆p. The fermen-
of the organisms that utilizes this pathway, tation is now balanced.
Propionibacterium, ferments lactate as well as The two succinates are converted to two mol-
hexoses to a mixture of propionate, acetate, ecules of succinyl–CoA via a CoA transferase
and CO2. Propionibacterium is a gram-positive (reaction 9). The two molecules of succinyl–
anaerobic, nonsporulating, nonmotile, pleo- CoA are isomerized to two molecules of meth-
morphic rod that is part of the normal flora ylmalonyl–CoA in an unusual reaction in which
in the rumen of herbivores; it is also found on COSCoA moves from the α carbon to the β car-
human skin and in dairy products (e.g., cheese). bon in succinyl–CoA to form methylmalonyl–
Propionibacterium is used in the fermenta- CoA (reaction 10). The reaction can be viewed
tion process that produces Swiss cheese. The as an exchange between adjacent carbons of a
www.ebook777.com
Fig. 15.4 Propionate fermentation by the succinate–propionate pathway. Enzymes: 1, lactate dehydroge-
nase (a flavoprotein); 2, pyruvate dehydrogenase (an NAD+ enzyme); 3, phosphotransacetylase; 4, acetate
kinase; 5, methylmalonyl–CoA–pyruvate transcarboxylase; 6, malate dehydrogenase; 7, fumarase; 8, fumar-
ate reductase; 9, CoA transferase; 10, methylmalonyl–CoA racemase.
hydrogen for a COSCoA. The enzyme that car- Section 15.10.) The electron donors besides
ries out this reaction is methylmalonyl–CoA race- lactate include NADH, H2, formate, and
mase, and it requires vitamin B12 as a cofactor. The glycerol-3-phosphate. The ∆p that is estab-
two molecules of methylmalonyl–CoA donate the lished can be used for ATP synthesis, for sol-
carboxyl groups to pyruvate via the transcarbox- ute uptake, or to spare ATP that might be
ylase and in turn become propionyl–CoA (reac- hydrolyzed to maintain the ∆p.
tion 5). The propionyl–CoA donates the CoA to 2. The transcarboxylase reaction spares an
succinate via the CoA transferase and becomes ATP. One way to carboxylate pyruvate
propionate (reaction 9). Notice the important role is to use pyruvate carboxylase and CO2
of transcarboxylases and CoA transferases. These (Section 9.9). However, this requires an
enzymes allow the attachment of CO2 and CoA to ATP. Propionibacterium has a transcarbox-
molecules without the need for ATP. ylase called methylmalonyl–CoA–pyruvate
When comparing Figs. 15.3 and 15.4, we transcarboxylase that transfers a carboxyl
learn that in metabolism there is sometimes more group from methylmalonyl–CoA to pyru-
than one route to take from point A to point B. vate, hence an ATP is not used (Fig. 15.4,
This is demonstrated in the following remarks reaction 5). By using the transcarboxylase,
about two important enzyme reactions. the bacteria save energy by substituting one
covalent bond for another. However, not all
1. The fumarate reductase serves as a coupling fermenting bacteria that produce propionate
site. Propionibacterium and other bacteria from pyruvate use the methylmalonyl–CoA–
that use the succinate–propionate pathway pyruvate transcarboxylase. For example,
use a circuitous route, but one that sends Veillonella alcalescens and Propionigenum
electrons through an energy-coupling site via modestum use a sodium-dependent decar-
the membrane-bound fumarate reductase. boxylase to remove the carboxyl group
Electron flow to fumarate requires a qui- from methylmalonyl–CoA while generat-
none, and presumably the ∆p is generated ing an electrochemical potential (Section
via a redox loop involving the quinone 4.8.1). The sodium-dependent decarboxy-
(Section 5.6). The use of fumarate as an lase pumps sodium ions out of the cell, gen-
electron acceptor during anaerobic growth erating a sodium ion potential, which can be
is widespread among bacteria. (See the later used as a source of electrochemical energy
discussion of the mixed-acid fermentation in (e.g., for solute uptake or ATP synthesis).
fermentations 391
15.7.1 The PEP carboxytransphosphorylase discussed in the context of the reductive citric
of propionibacteria and its physiological acid pathway (Section 9.8).
significance The pathway from PEP or pyruvate to suc-
The reaction cinate is extremely important for anaerobes
Propionibacteria growing on carbon sources and serves three purposes: (1) the fumarate is an
such as glucose that enter the glycolytic path- electron sink, enabling NADH to be reoxidized;
way can produce succinate as well as propionate (2) the fumarate reductase is a coupling site (i.e.,
as an end product of fermentation. This means a ∆p is generated); and (3) the succinate can be
that they must have an enzyme to carboxylate converted to succinyl–CoA, which is required
a C3 intermediate to form the C4 product. The for the biosynthesis of tetrapyrroles, lysine,
C3 intermediate that is carboxylated is phospho- diaminopimellic acid, and methionine. With
enolpyruvate (an intermediate in glycolysis). respect to fumarate acting as an electron sink,
The phosphoenolpyruvate is carboxylated to Gest has suggested that the carboxylation of
oxaloacetate, which is then reduced to succinate the C3 glycolytic intermediate and the reductive
via reactions 6, 7, and 8 shown in Fig. 15.4. The pathway from oxaloacetate to succinate may
enzyme that catalyzes the carboxylation of phos- have evolved when the earth’s atmosphere was
phoenolpyruvate is called PEP carboxytrans- still anaerobic, serving the purpose of balancing
phosphorylase, and it catalyzes the following fermentations and thus sparing one of the two
reaction: (See note 5 for a description of other pyruvates derived from glucose for biosynthe-
C3 carboxylases.) sis.6 The reactions between oxaloacetate and
succinate may have become part of the oxida-
PEP + CO2 + Pi → oxaloacetate + PPi tive citric acid cycle later, during the evolution
During the carboxylation, a phosphoryl group of aerobic metabolism.
is transferred from PEP to inorganic phosphate
to form pyrophosphate. Pyrophosphate is a
high-energy compound, and propionibacteria
15.8 Acetate Fermentation
have enzymes that phosphorylate fructose-6- (Acetogenesis)
phosphate to fructose-1,6-bisphosphate and As discussed in Section 14.1.3 in connection
serine to phosphoserine, using pyrophosphate with the acetogenic bacterium Clostridium
as the phosphoryl donor. thermoaceticum, some bacteria use CO2 as an
electron sink and reduce it to acetate as a fer-
Physiological significance mentation end product via the acetyl–CoA
The carboxylation of phosphoenolpyruvate or pathway. This is called acetogenesis. Another
pyruvate to oxaloacetate and the reduction of acetogenic bacterium is the sulfate reducer
the oxaloacetate to succinate is a widespread Desulfotomaculum thermobenzoicum, when it
pathway among fermenting bacteria (see, e.g., is growing on pyruvate in the absence of sulfate.7
the later discussion of mixed-acid fermentation (Another sulfate reducer, Desulfobulbus propi-
in Section 15.10). These reactions were also onicus, uses the succinate–propionate pathway
Fig. 15.5 Acetogenesis from pyruvate by Desulfotomaculum thermobenzoicum. Enzymes: 1, pyruvate dehy-
drogenase; 2 and 7, phosphotransacetylase; 3 and 8, acetate kinase; 4, enzymes of the acetyl–CoA pathway; 5
and 6, carbon monoxide dehydrogenase.
www.ebook777.com
to ferment pyruvate to a mixture of acetate and bifidum pathway, is found in Bifidobacterium

propionate.) The acetogenic pathway is shown bifidum.
in Fig. 15.5. Four pyruvate molecules are oxida-
tively decarboxylated to four acetyl–CoA mol- 15.9.1 Homofermentative lactate
ecules, producing eight electrons (reaction 1). fermentation
The acetyl–CoA is converted to acetyl phos- The homofermentative pathway produces pri-
phate via phosphotransacetylase (reaction 2). marily lactate. The bacteria use the glycolytic
The acetyl phosphate is converted to acetate pathway to oxidize glucose to pyruvate. This
and ATP via the acetate kinase (reaction 3). nets them two ATPs per mole of glucose via
Six of the electrons are used to reduce CO2 to the oxidation of phosphoglyceraldehyde. The
bound methyl, [CH3] (reaction 4). This requires NADHs produced during this oxidative step
an ATP in the acetyl–CoA pathway to attach are used to reduce the pyruvate, forming lac-
the formic acid to the THF (see Section 14.1.3). tate. The overall reaction is as follows:
The remaining two electrons are used to reduce
a second molecule of CO2 to bound carbon Glucose + 2 ADP + 2 Pi → 2 lactate + 2 ATP
monoxide, [CO] (reaction 5). The [CH3], [CO], Whenever lactate is produced during fermenta-
and CoASH combine to yield acetyl–CoA (reactions, it is always the result of the reduction of
tion 6). The fourth acetyl–CoA is converted to pyruvate. Since the ATP yield per pyruvate is
acetate with the formation of an ATP (reactions one, it can be assumed that one ATP is made per
7 and 8). mole of lactate produced.
15.9 Lactate Fermentation 15.9.2 Heterofermentative lactate

The lactic acid bacteria are a heterogeneous fermentation
group of aerotolerant anaerobes that fer- The heterofermentative lactate fermentation
ment glucose to lactate as the sole or major produces lactate via the decarboxylation and
product of fermentation. They include the isomerase reactions of the pentose phosphate
genera Lactobacillus, Sporolactobacillus, pathway (Fig. 15.6). The glucose is oxidized to rib-
Streptococcus, Leuconostoc, Pediococcus, and ulose-5-phosphate by means of glucose-6-phos-
Bifidobacterium. Lactic acid bacteria are found phate dehydrogenase and 6-phosphogluconate
living on the skin of animals, in the gastrointes- dehydrogenase (reactions 1–3). Four electrons
tinal tract, and in other places (e.g., mouth and are produced in the form of two NADHs. The
throat). ribulose-5-phosphate is isomerized to xylulose-
Some genera live in vegetation and in dairy 5-phosphate by using the epimerase (reaction 4).
products. Several lactic acid bacteria are medi- Then an interesting reaction occurs during which
cally and commercially important organisms. the xylulose-5-phosphate is cleaved with the aid
These include the genus Streptococcus, several of inorganic phosphate to form phosphoglyc-
species of which are pathogenic. eraldehyde and acetyl phosphate (reaction 5).
The lactic acid bacteria are also important in The enzyme that catalyzes this reaction is called
various food fermentations (e.g., the manufac- phosphoketolase, and it requires thiamine pyro-
ture of butter, cheese, yogurt, pickles, and sau- phosphate (TPP) as a cofactor. The phosphoglyc-
erkraut). Although they can live in the presence eraldehyde is oxidized to pyruvate via reactions
of air, they metabolize glucose only fermenta- also found in the glycolytic pathway yielding
tively and derive most or all of their ATP from an ATP and a third NADH, and the pyruvate is
substrate-level phosphorylation. Under certain reduced to lactate by one of the three NADHs
growth conditions they may transport lactate (reactions 9–14). The acetyl phosphate produced
out of the cell in electrogenic symport with H+, in the phosphoketolase reaction is reduced to
creating a ∆Ψ (Section 4.8.3). There are two ethanol via the second and third NADH (reac-
major types of lactate fermentation: homofer- tions 6–8). Thus, the fermentation is balanced.
mentative and heterofermentative. The former The overall reaction is as follows:
uses the Embden–Meyerhof–Parnas pathway
(glycolysis), and the latter uses the pentose Glucose + ADP + Pi → ethanol + lactate
phosphate pathway. A third pathway, called the + CO2 + ATP
fermentations 393
Fig. 15.6 Heterofermentative lactate fermentation. Enzymes: 1, hexokinase; 2, glucose-6-phosphate dehy-

drogenase; 3, 6-phosphogluconate dehydrogenase; 4, ribulose-5-phosphate epimerase; 5, phosphoketolase;
6, phosphotransacetylase; 7, acetaldehyde dehydrogenase; 8, alcohol dehydrogenase; 9, PGALD dehydroge-
nase; 10, PGA kinase; 11, phosphoglycerate mutase; 12, enolase; 13, pyruvate kinase; 14, lactate dehydro-
genase.
Note that the heterofermentative pathway pro- 2 Glucose + 5 ADP + Pi → 3 acetate

duces only one ATP per glucose in contrast to + 2 lactate + 5 ATP
the homofermentative pathway, which pro- The ATP yields are therefore greater than for
duces two ATPs for every glucose. the homo- or heterofermentative pathways.
The pathway uses reactions of the pentose
15.9.3 Bifidum pathway phosphate pathway (Section 9.5) and the
The bifidum pathway ferments two glucoses to homofermentative pathway. Two glucose
two lactates, and three acetates with the pro- molecules are converted to two molecules
duction of 2.5 ATPs per glucose. The overall of fructose-6-P, requiring two ATPs. One
reaction is: of the fructose-6-P molecules is cleaved by
www.ebook777.com
fructose-6-phosphate phosphoketolase to 15.10 Mixed-Acid and Butanediol

erythrose-4-P and acetyl–P. The acetyl–P is Fermentation
converted to acetate via acetate kinase, with
The enteric bacteria are facultative anaerobes.
the formation of an ATP. The erythrose-4-P
In the absence of oxygen, as part of the adapta-
reacts with the second fructose-6-P in a transal-
tion to anaerobic growth, the following physi-
dolase reaction to form glyceraldehyde-3-P
ological changes take place.
and sedoheptulose-7-P. These then react in
a transketolase reaction to form xyulose-5-P 1. Terminal reductases replace the oxidases in
and ribose-5-P. The latter isomerizes to form the electron transport chain.
a second xylulose-5-P. The two xylulose- 2. The citric acid cycle is modified to become
5-P molecules are cleaved by xylulose-5- a reductive pathway. α-Ketoglutarate dehy-
phosphate phosphoketolase to two molecules drogenase and succinate dehydrogenase
of glyceraldehyde-3-P and two of acetyl–P. are missing or occur at low levels, the latter
The two glyceraldehyde-3-P molecules are being replaced by fumarate reductase.
converted to two lactates with the production 3. Pyruvate–formate lyase is substituted for
of four ATPs, using reactions of the homo- pyruvate dehydrogenase. This means that
fermentative pathway, and the two acetyl–P the cells oxidize pyruvate to acetyl–CoA and
molecules are converted to two acetates with formate, rather than to acetyl–CoA, CO2,
the production of two more ATPs. Thus, seven and NADH.
ATPs are produced for every two glucose mol- 4. The bacteria then carry out a mixed-acid or
ecules fermented, but since two ATPs were butanediol fermentation.
used to make the two fructose-6-P molecules,
The mixed-acid and butanediol fermentations
the net gain in ATP per glucose is 5/2, or 2.5.
are similar in that both produce a mixture
Note that since glucose-6-P is not oxidized to
of organic acids, CO2, H2, and ethanol. The
6-phosphogluconate, the acetyl–P can serve as
butanediol fermentation is distinguished by
a phosphoryl donor for ATP synthesis rather
producing large amounts of 2,3-butanediol,
than being reduced to ethanol.
acetoin, more CO2 and ethanol, and less acid.
The mixed-acid fermenters belong to the gen-
15.9.4 Synthesis of acetyl–CoA era Escherichia, Salmonella, and Shigella. All
or acetyl–P from pyruvate by lactic three can be pathogenic, causing intestinal
acid bacteria infections such as dysentery, typhoid fever
Not all the pyruvate produced by lactic (Salmonella typhimurium), or food poisoning.
acid bacteria needs be converted to lactate. Butanediol fermenters are Serratia, Erwinia,
Depending upon the species, the lactic acid bac- and Enterobacter.
teria may have one of three enzymes or enzyme
complexes for decarboxylating pyruvate
15.10.1 Mixed-acid fermentation
(Section 8.3.2). Streptococci have pyruvate
The products of the mixed-acid fermentation are
dehydrogenase, usually found in aerobically
succinate, lactate, acetate, ethanol, formate, car-
respiring bacteria; several lactic acid bacteria
bon dioxide, and hydrogen gas (Fig. 15.7). Each
are known to have pyruvate formate lyase, an
of these is made from one phosphoenolpyruvate
enzyme also found in the Enterobacteriaceae;
and CO2 or one pyruvate. For example, the for-
Lactobacillus plantarum and L. delbruckii use
mation of succinate is due to a carboxylation of
the flavoproteins pyruvate oxidase and lactate
phosphoenolpyruvate to oxaloacetate followed
oxidase in coupled reactions that convert two
by two reductions to form succinate (reactions
pyruvates to one acetyl–P and one lactate, as
10–13). All the other products are formed from
follows:
pyruvate. The formation of lactate from pyru-
Pyruvate oxidase vate is simply a reduction (reaction 4). Pyruvate
Pyruvate + Pi + FAD → acetyl–P + CO2 + is also decarboxylated to acetyl–CoA and for-
FADH2 mate by means of the pyruvate–formate lyase
(reaction 3). The acetyl–CoA can be reduced
Lactate oxidase to ethanol (reactions 6 and 7) or converted to
Pyruvate + FADH2 → lactate + FAD acetate and ATP via acetyl–P (reactions 8 and
fermentations 395
Fig. 15.7 Mixed-acid fermentation. Enzymes: 1, glycolytic enzymes; 2, pyruvate kinase; 3, pyruvate–formate
lyase; 4, lactate dehydrogenase; 5, formate–hydrogen lyase; 6, acetaldehyde dehydrogenase; 7, alcohol dehy-
drogenase; 8, phosphotransacetylase; 9, acetate kinase; 10, PEP carboxylase; 11, malate dehydrogenase; 12,
fumarase; 13, fumarate reductase. Note the ATP yields: per succinate, approximately 1; per ethanol, 1; per
acetate, 2; per formate, 1; per CO2 and H2, 1; per lactate, 1. Energy equivalent to approximately 1 ATP is
conserved per succinate formed because the fumarate reductase reaction takes place in the cell membrane and
generates a ∆p. Note also the reducing equivalents used in the production of the end products: per succinate,
4; per ethanol, 4; per acetate, 0; per lactate, 2; per formate, 0. The number of reducing equivalents used must
equal the number produced during glycolysis. Therefore, only certain ratios of end products are compatible
with a balanced fermentation.
9). The formate can be oxidized to CO2 and H2 fermentation, two electrons must be used for
by the enzyme system formate–hydrogen lyase each phosphoenolpyruvate or pyruvate formed.
(reaction 5). This system actually consists of The pathways that utilize four electrons per
two enzymes; formate dehydrogenase and an phosphoenolpyruvate or pyruvate formed spare
associated hydrogenase. The formate dehydro- the second phosphoenolpyruvate or pyruvate
genase oxidizes the formate to CO2 and reduces for biosynthesis. However, they may do this at
the hydrogenase, which transfers the electrons the expense of an ATP. For example, the reduc-
to two protons to form hydrogen gas. Shigella tion of acetyl–CoA to ethanol uses 4H, but this
and Erwinia do not contain formate–hydrogen is at the expense of an ATP that can be formed
lyase, and therefore do not produce gas. (See when acetyl–CoA is converted to acetate. In this
note 8.) context, the production of succinate is particu-
Each of the pathways following phosphoe- larly valuable. The pathway utilizes 4H and also
nolpyruvate or pyruvate can be viewed as a met- includes a coupling site (fumarate reductase).
abolic branch that accepts different amounts of
reducing equivalent, that is, 0, 2[H], or 4[H], 15.10.2 Butanediol fermentation
depending upon the pathway. The reducing The butanediol fermentation is character-
equivalents in the different branches are succi- ized by the production of 2,3-butanediol and
nate (4), ethanol (4), lactate (2), acetate (0), and acetoin (Fig. 15.8). Glucose is oxidized via the
formate (0). During glycolysis, two electrons glycolytic pathway to pyruvate (reactions 1).
are produced for each phosphoenolpyruvate There are three fates for the pyruvate. Some of
or pyruvate formed. Therefore, to balance the it is reduced to lactate (reaction 10), some is
www.ebook777.com
Fig. 15.8 Butanediol formation. Enzymes: 1, glycolytic enzymes; 2, pyruvate–formate lyase; 3, formate–hy-
drogen lyase; 4, acetaldehyde dehydrogenase; 5, alcohol dehydrogenase; 6 and 7, α-acetolactate synthase; 8,
α-acetolactate decarboxylase; 9, 2,3-butanediol dehydrogenase; 10, lactate dehydrogenase.
converted to acetyl–CoA and formate (reac- How thiamine pyrophosphate catalyzes the
tion 2), and some is used for the synthesis of decarboxylation of α-ketocarboxylic acids
2,3-butanediol (reactions 6–9). The formate The decarboxylation of α-ketocarboxylic
is converted to CO2 and H2 (reaction 3), and acids presents a problem because there is no
the acetyl–CoA is reduced to ethanol (reac- electron-attracting carbonyl group β to the
tions 4 and 5). The first free intermediate in the C–C bond to withdraw electrons, as there is
butanediol pathway is α-acetolactate, formed in β-ketocarboxylic acids. (See Section 9.11.2
by the enzyme α-acetolactate synthase, which for a description of the decarboxylation of
decarboxylates pyruvate to enzyme-bound β-ketocarboxylic acids.) The problem is solved
“active acetaldehyde” (reaction 6); this is a by using thiamine pyrophosphate (TPP). A pro-
reaction that depends upon thiamine pyro- posed mechanism is illustrated in Fig. 15.9. A
phosphate (TPP). The active acetaldehyde is proton dissociates from the thiamine pyrophos-
transferred by the α-acetolactate synthase to phate to form a dipolar ion, which is stabilized
pyruvate to form α-acetolactate (reaction 7). by the positive charge on the nitrogen atom.
The α-acetolactate, a β-ketocarboxylic acid, The negative center of the dipolar ion adds to
is decarboxylated to acetoin (reaction 8). The the keto group of the α-ketocarboxylic acid.
acetoin is reduced by NADH to 2,3-butanediol Then, the electron-attracting =N+– group pulls
(reaction 9). The production of butanediol is electrons away from the C–C bond, facilitat-
favored under slightly acidic conditions and is ing the decarboxylation. The product is an
a way for the bacteria to limit the decrease in α-ketol called “active aldehyde.” The flow
external pH caused by the synthesis of organic of electrons is reversed, and the “active alde-
acids from pyruvate. hyde” then attacks a positive center on another
fermentations 397
Fig. 15.9 Proposed mechanism for reactions catalyzed by thiamine pyrophosphate (TPP). Step 1. The TPP enzyme
loses a proton to form a dipolar ion. The anion is stabilized by the positive charge on the nitrogen. The anionic cen-
ter is nucleophilic and can attack positive centers such as carbonyl carbons. Step 2. The dipolar ion condenses with
pyruvate to form a TPP adduct. Step 3. The electron-attracting N in the TPP facilitates the decarboxylation to form
“active acetaldehyde.” The “active acetaldehyde” can form acetaldehyde (step a) or α-acetolactate (step b).
molecule. The product varies, depending upon 15.11 Butyrate Fermentation

the enzyme. Butyrate fermentations are carried out by the
As shown in Fig. 15.9, α-acetolactate synthase butyric acid clostridia. The clostridia are a hetero-
catalyzes the condensation of “active acetalde- geneous group of anaerobic spore-forming bac-
hyde” with pyruvate to form α-acetolactate, teria that can be isolated from anaerobic muds,
and pyruvate decarboxylase catalyzes the addi- sewage, feces, or other anaerobic environments.
tion of a proton to form acetaldehyde. Pyruvate They all are classified in the genus Clostridium.
dehydrogenase and α-ketoglutarate dehydroge- Some are saccharolytic (i.e., they ferment car-
nase catalyze the condensation with lipoic acid bohydrates) and/or proteolytic. The prote-
to form the acyl–lipoate derivative. olytic clostridia are important in the anaerobic
www.ebook777.com
decomposition of proteins, called “putrefaction.”

Other clostridia are more specialized and will fer-
ment only a few substrates (e.g., ethanol, acetate,
certain purines, certain amino acids).
The butyric acid clostridia ferment carbohy-
drates to butyric acid. The fermentation prod-
ucts also include hydrogen gas, carbon dioxide,
and small amounts of acetate. The bacteria first
oxidize the glucose to two moles of pyruvate
via the Embden–Meyerhof–Parnas pathway
(Fig. 15.10). This produces two molecules of
NADH, plus two of ATP. The pyruvate is then
decarboxylated to acetyl–CoA, CO2, and H2 by
means of pyruvate–ferredoxin oxidoreductase
and hydrogenase (reaction 2). The acetyl–CoA
is condensed to form acetoacetyl–CoA (reac-
tion 3), which is reduced to β-hydroxybutyryl–
CoA by means of one of the two NADHs
(reaction 4). The β-hydroxybutyryl–CoA is
reduced to butyryl–CoA by using the second
NADH (reactions 5 and 6). The CoASH is dis-
placed by inorganic phosphate (reaction 7), and
the butyryl–phosphate donates the phosphoryl
group to ADP to form ATP and butyrate (reac-
tion 8). This pathway therefore utilizes three
substrate-level phosphorylations, the phos-
phoryl donors being 1,3-bisphosphoglycerate,
phosphoenolpyruvate, and butyryl–P. Note
the role of the hydrogenase (reaction 9) in the
hydrogen sink. (For a discussion of acetyl–CoA
condensations, see Section 9.11.1.)
15.11.1 Butyrate and butanol–acetone

fermentation in Clostridium
acetobutylicum
Some clostridia initially make butyrate during fer-
mentation and, when the butyrate accumulates
and the pH drops (owing to the butyric acid), the
Fig. 15.10 The butyrate fermentation. The glucose is
fermentation switches to a butanol–acetone fer- degraded via glycolysis to pyruvate, which is then oxi-
mentation. As we shall see, the butyrate is actu- datively decarboxylated to acetyl–CoA. Two molecules
ally taken up by the cells and converted to butanol of acetyl–CoA condense to form acetoacetyl–CoA,
and acetone. The accumulation of butyric acid in which is reduced to butyryl–CoA. A phosphotrans-
media of low pH can be toxic because the undisso- acetylase makes butyryl–P, and a kinase produces
ciated form of the acid is lipophilic and can enter butyrate and ATP. Two ATPs are produced in the gly-
the cell, acting as an uncoupler and also result- colytic pathway for every glucose and one ATP from
ing in a decrease in the ∆pH. (The pK of butyric butyryl–P. Note the production of hydrogen gas as an
acid is 4.82.) Butanol and acetone production by electron sink. This actually allows the production of
the third ATP from butyryl–P, rather than reducing the
the clostridia was at one time the second largest
butyryl–SCoA to butanol to balance the fermentation.
industrial fermentation process, after ethanol.
Enzymes: 1, glycolytic enzymes; 2, pyruvate–ferredoxin
The solvents are now synthesized chemically. oxidoreductase; 3, acetyl–CoA acetyltransferase (thio-
As an example of the butyrate fermentation and lase); 4, β-hydroxybutyryl–CoA dehydrogenase; 5,
the shift to the butanol–acetone fermentation, we crotonase; 6, butyryl–CoA dehydrogenase; 7, phospho-
will consider the fermentation of carbohydrates transbutyrylase; 8, butyrate kinase; 9, hydrogenase.
fermentations 399
carried out by C. acetobutylicum.9 During expo- and are converted to butanol, acetone, and etha-
nential growth, in what is called the acidogenic nol. This is called the solventogenic phase.
phase, the bacteria produce butyrate, acetate, Pentoses are also fermented, and these are
H2, and CO2. When the culture enters stationary converted to fructose-6-phosphate and phos-
phase, the acids are taken up by the cells, concom- phoglyceraldehyde via the pentose phosphate
itant with the fermentation of the carbohydrate, pathway (Fig. 15.11). The pyruvate formed
Fig. 15.11 Butyrate and butanol–acetone fermentation in C. acetobutylicum. Carbohydrates are oxidized to
acetyl–SCoA. Pentose phosphates are converted to fructose-6-phosphate and phosphoglyceraldehyde via the
pentose phosphate reactions. Glucose is oxidized to pyruvate via the Embden–Meyerhof–Parnas pathway.
The pyruvate is oxidized to acetyl–SCoA. In the butyric acid fermentation, the acetyl–SCoA is converted to
acetate and butyrate (solid lines). When the acetate and butyrate levels rise, they are taken up by the cells and
converted to butanol, and ethanol, while carbohydrates continue to be fermented (dashed lines). During the
butanol–acetone fermentation, the acetoacetyl–SCoA donates the CoASH to butyrate and acetate and becomes
acetoacetate, which is decarboxylated to acetone. Enzymes: 1, pyruvate–ferredoxin oxidoreductase; 2, acetyl–
CoA acetyltransferase (thiolase); 3, hydroxybutyryl–CoA dehydrogenase; 4, crotonase; 5, butyryl–CoA dehy-
drogenase; 6, phosphotransbutyrylase; 7, butyrate kinase; 8, phosphotransacetylase; 9, acetate kinase; 10,
acetoacetyl–SCoA:acetate/butyrate:CoA transferase; 11, acetoacetate decarboxylase; 12, acetaldehyde dehy-
drogenase; 13, ethanol dehydrogenase; 14, butyraldehyde dehydrogenase; 15, butanol dehydrogenase; 16,
NADH–ferredoxin oxidoreductase and hydrogenase; 17, hydrogenase. Source: Adapted from Jones, D. T.,
and D. R. Woods. 1986. Acetone–butanol fermentation revisited. Microbiol. Rev. 50:484–524.
www.ebook777.com
from the sugars is oxidatively decarboxylated to mixed populations can influence fermentation
acetyl–CoA and CO2 via the pyruvate–ferredoxin end products. The fermentation is easy to under-
oxidoreductase (reaction 1). The acetyl–CoA stand. The glucose is first oxidized to two moles
has two fates. Some of it is converted to butyrate of pyruvate, yielding two moles of NADH. Then
and ATP as described in Fig. 15.10 (reactions each mole of pyruvate is oxidized to acetyl–CoA,
2–7). Some acetyl–CoA is also converted to CO2, and H2 by means of the pyruvate–ferre-
acetate via acetyl–P in a reaction that yields an doxin oxidoreductase and hydrogenase. One of
additional ATP (Fig. 15.11, reactions 8 and 9). the acetyl–CoA molecules is converted to acetate,
The ability to send electrons to hydrogen via the allowing an ATP to be made. The second acetyl–
NADH:ferredoxin oxidoreductase (reaction 16) CoA is reduced to ethanol via the two NADHs.
allows the bacteria to produce more acetate, Thus the fermentation is balanced. The overall
hence more ATP, rather than reducing acetyl– reaction is the conversion of one mole of glucose
CoA to butyrate. Notice that the reaction gener- to one mole of ethanol, one mole of acetate, two
ates twice as much ATP as butyrate per acetate. moles of hydrogen, and two moles of carbon
However, reduction of protons by NADH is not dioxide yielding, three ATPs.
favored energetically and is limited by increasing Something different happens when the bacte-
concentrations of hydrogen gas. rium is grown in a coculture with a methanogen.
The NADH oxidoreductase can be pulled in the Growth with a methanogen shifts the fermenta-
direction of hydrogen gas by other bacteria that tion in the direction of acetate with the concomi-
utilize hydrogen in a process called interspecies tant production of more ATP. This is explained
hydrogen transfer, explained in Section 15.4.1. In in the following way. R. albus has a hydrogenase
the solventogenic phase, butyrate and acetate are that transfers electrons from NADH to H+ to
taken up by the cells and reduced to butanol and produce hydrogen gas. When hydrogen accu-
ethanol (dashed lines in Fig. 15.11). The acids mulates in the medium, the hydrogenase does
are converted to their CoA derivatives by accept- not oxidize NADH because the equilibrium
ing a CoA from acetoacetyl–CoA (reaction 10). favors NAD+ reduction. The NADH therefore
The reaction is catalyzed by CoA transferase. reduces acetyl–CoA to ethanol. However, the
(Recall that CoA transferase is also used in the methanogen utilizes the hydrogen gas for meth-
acrylate pathway for propionate fermentation, ane production and keeps the hydrogen levels
Section 15.6.) The acetoacetate that is formed is very low. In the presence of the methanogen, the
decarboxylated to acetone and CO2 (reaction 11). NADH in R. albus reduces protons to hydrogen
The acetyl–CoA is reduced to ethanol (reactions gas instead of reducing acetyl–CoA. The result is
12 and 13), and the butyryl–CoA is reduced to that R. albus converts the acetyl–CoA to acetate.
butanol (reactions 14 and 15). This is also an advantage to the methanogen,
The molar ratios of the fermentation end prod- since methanogens can also use acetate as a car-
ucts in clostridial fermentations will vary accord- bon and energy source. The transfer of hydrogen
ing to the strain.10 For example, C. acetobutylicum gas from one species to another is called inter-
is an important solvent-producing strain, and species hydrogen transfer and is an example of
when grown at pH0 below 5 it produces butanol nutritional synergism found among mixed pop-
and acetone in the molar ratio of 2:1 with small ulations of bacteria. (See the discussion of inter-
amounts of isopropanol, whereas C. sporogenes species hydrogen transfer in Section 15.4.1.)
and C. pasteurianum produce very little solvent.
Clostridium butyricum forms butyrate and ace- 15.13 Summary
tate in a ratio of about 2:1, whereas C. perfrin-
Fermentations are cytosolic oxidation–reduc-
gens produces these acids in a ratio of 1:2, with
tion pathways in which the electron acceptor is
significant amounts of ethanol and lactate.
an organic compound, usually generated in the
pathway.
15.12 Ruminococcus albus The source of ATP in fermentative path-
Ruminococcus albus is a rumen bacterium ways is substrate-level phosphorylation. For
that can use the glycolytic pathway to ferment sugar fermentations, these are the phospho-
glucose to ethanol, acetate, H2, and CO2 (Fig. glycerate kinase, pyruvate kinase, acetate
15.12). It is a good fermentation to examine kinase, and butyrate kinase reactions. In other
because it illustrates how growth of bacteria in words, ATP is made from bisphosphoglycerate,
fermentations 401
Fig. 15.12 Fermentation of glucose by Ruminococcus albus. R. albus ferments glucose to a mixture of etha-
nol, acetate, CO2, and H2. Methanogens draw off the H2, thus stimulating electron flow to H2. The result is a
shift in the fermentation end products toward acetate, accompanied by a greater ATP yield. The production
of H2 by one species and its utilization by another is called interspecies hydrogen transfer. The methanogens
can also utilize the acetate produced by R. albus.
phosphoenolpyruvate, acetyl–P, and butyryl–P. pyruvate plus CO2. Protons can also be used
Butyryl–P is a phosphoryl donor during butyrate as electron sinks, and many fermenting bacte-
fermentations. ria have hydrogenases that reduce protons to
However, other means of conserving energy hydrogen gas.
are available for fermenting bacteria. For exam- The excreted end products of fermentations,
ple, a ∆p can be created during electron flow to including hydrogen gas, are used by other
fumarate, the fumarate being generated during anaerobic bacteria so that an anaerobic food
fermentation. Other means of creating a proton chain develops. At the bottom of the food chain
potential or sodium potential exist in certain are the methanogens, which convert hydrogen
groups of bacteria. These include efflux of car- gas, carbon dioxide, and acetate to CO2 and
boxylic acids in symport with protons or sodium CH4. These are recycled to organic carbon in
ions, decarboxylases that function as sodium the biosphere for use by autotrophic and metha-
ion pumps, and oxalate:formate exchange. notrophic organisms as a source of carbon.
Besides making ATP, fermenting bacte-
ria must have a place to unload the electrons Study Questions
removed during oxidation reactions. Of course,
the reason for this is that they must regenerate 1. Write a fermentation balance using both the
the NAD+, oxidized ferredoxin, and FAD to O/R and the available hydrogen method for
continue the fermentation. The electron accep- the following:
tors are sometimes referred to as “electron
C6H12O6 (glucose) + H2O → C2H4O2 (acetate)
sinks.” During a fermentation the electron sinks
+ C2H6O (ethanol)
are created from the carbon source. In fact, all
+ 2H2 + 2CO2
the electron sinks for the major carbohydrate
fermentations either are pyruvate itself or are If the EMP pathway is used, what is the
synthesized from pyruvate or phosphoenol yield of ATP?
www.ebook777.com
2. C. propionicum and Propionibacterium Offer an explanation to account for the

both carry out the following reaction: shift from lactate to acetate in the coculture
3 Lactate → acetate + 2 propionate + CO2 compared to the pure culture. How might
you expect this to affect the growth yields
C. propionicum nets one ATP, but of Selenomonas?
Propionibacterium derives more. How
might Propionibacterium make more ATP
from the same overall reaction? REFERENCES AND NOTES
3. Use the Entner–Doudoroff pathway to 1. Imlay, J. A. 2008. Cellular defenses against super-
write an ethanol fermentation. Contrast oxide and hydrogen peroxide. Annu. Rev. Biochem.
77:755–776.
the ATP yields with an ethanol fermenta-
tion by means of the EMP pathway. 2. Henle, E. S., and S. Linn. 1997. Formation, pre-
vention, and repair of DNA damage by iron/hydro-
4. Consider the following fermentation data gen peroxide. J. Biol. Chem. 272:19095–19098.
for Selenomonas ruminantium products 3. Transition elements are defined as those elements
formed per millimole of glucose: in which electrons are added to an outer electronic
shell before one of the inner shells is complete. There
Product Amount formed (nmol) are nine of them. They are all metals, and they include
iron, which is the one that occurs in highest amounts
Lactate 0.31 in living systems.
Acetate 0.70 4. Shink, B. 1997. Energetics of syntrophic coopera-
Propionate 0.36 tion in methanogenic degradation. Microbiol. Mol.
Succinate 0.61 Biol. Rev. 61:262–280.
Source: Michel, T. A., and J. M. Macy. 1990. 5. Other enzymes besides PEP carboxytransphos-
Generation of a membrane potential by sodium- phorylase that carboxylate C3 glycolytic interme-
dependent succinate efflux in Selenomonas diates to oxaloacetate are PEP carboxylase and
ruminantium. J. Bacteriol. 172:1430–1435. pyruvate carboxylase (Section 9.9), and PEP car-
Find the fermentation balance according boxykinase (Section 9.13). The latter enzyme usually
operates in the direction of PEP synthesis; however,
to both the O/R method and the available in some anaerobes (e.g., Bacteroides) it functions to
hydrogen method. What percentage of the synthesize oxaloacetate.
glucose carbon is recovered in end products?
6. Gest, H. 1983. Evolutionary roots of anoxygenic
5. Consider the following data for fermenta- photosynthetic energy conversion, pp. 215–234.
tion products made by Selenomonas rumi- In: The Phototrophic Bacteria: Anaerobic Life in
the Light. Studies in Microbiology, Vol. 4. J. G.
nantium in pure culture and in coculture Ormerod (Ed.). Blackwell Scientific Publications,
with Methanobacterium ruminantium (in Oxford.
moles per 100 moles of glucose). 7. Tasaki, M., Y. Kamagata, K. Nakamura, K.
Okamura, and K. Minami. 1993. Acetogenesis from
Proportion (mol/100 mol glucose) pyruvate by Desulfotomaculum thermobenzoicum
Product Selenomonas Selenomonas + and differences in pyruvate metabolism among three
sulfate-reducing bacteria in the absence of sulfate.
Methanobacterium
FEMS Microbiol. Lett. 106:259–264.
8. Gas production is generally observed as a bubble
Lactate 156 68
in an inverted vial placed in the fermentation tube.
Acetate 46 99
The bubble is due to H2, since CO2 is very soluble in
Propionate 27 20
water.
Formate 4 0
Methane 0 51 9. Jones, D. T., and D. R. Woods. 1986. Acetone–
CO2 42 48 butanol fermentation revisited. Microbiol. Rev.
50:484–524.
Source: Chen, M., and M. J. Wolin. 1977. Influence
of CH4 production by Methanobacterium ruminan- 10. Hamilton, W. A. 1988. Energy transduction in
tium on the fermentation of glucose and lactate by anaerobic bacteria, pp. 83–149. In: Bacterial Energy
Selenomonas ruminantium. Appl. Environ. Microbiol. Transduction. C. Anthony (Ed.). Academic Press,
34:756–759. New York.
16
Responses to Environmental Stress
Bacteria often find themselves to be in an have an internal pH of 8.4 to 9. Therefore, alka-

external environment that imposes stress to liphiles maintain a negative ∆pH of about 1.5
which the bacteria must adapt in order to sur- to 2 units, inside acid. (This lowers the ∆p. For a
vive. The topics covered in this chapter include discussion of this point, see Section 4.10.)
some of the mechanisms by which they do this.
The adaptive mechanisms result in maintaining 16.1.2 Demonstrating pH homeostasis
an intracellular pH close to neutrality, maintain- One method to demonstrate pH homeostasis is
ing an appropriate internal osmotic pressure, to perturb the cytoplasmic pH by changing the
protection from shifts to higher temperatures, external pH and then study the recovery process.
repair of damaged DNA, and protection from When E. coli is exposed to a rapid change in the
toxic forms of oxygen such as hydrogen perox- external pH (e.g., of 1.5–2.5 units), the inter-
ide and oxygen radicals. nal pH initially changes in the direction of the
external pH. However, a recovery soon occurs
16.1 Maintaining a ∆pH as the internal pH bounces back to its initial
16.1.1 Neutrophiles, acidophiles, and value (Fig. 16.1). In other bacteria, pH homeo-
alkaliphiles stasis may be so rapid that a temporary change
in the internal pH is not even measurable.
Bacteria can be found growing in habitats that
vary in pH from approximately pH 1 to 2 in acid
springs to as high as pH 11 in soda lakes and 16.1.3 The mechanism of
alkaline soils. However, regardless of the exter- pH homeostasis
nal pH, the internal pH is usually maintained Proton pumping
within 1 to 2 units of neutrality, which is neces- One would expect to find that many factors
sary to maintain viability (Table 16.1).1–6 Thus, influence the intracellular pH. These include the
bacteria maintain a pH gradient (∆pH) across buffering capacity of the cytoplasm and meta-
the cell membrane. For example, bacteria that bolic reactions that produce acids and bases.
grow optimally between pH 1 and 4 (i.e., aci- However, it is generally believed that the regu-
dophiles) have an internal pH of about 6.5 to 7. lation of intracellular pH is, to a large extent, a
That is, they maintain a ∆pH of greater than 2.5 consequence of controlling the flow of protons
units, inside alkaline. Those that grow optimally across the cell membrane.
between pH 6 and 8 (i.e., neutrophiles) have an The reason for believing this is that the ∆pH
internal pH of around 7.5 to 8.0 and can main- is dissipated when the proton pumps are inhib-
tain a ∆pH of 0.5 to 1.5 units, inside alkaline. ited or when proton ionophores are added to
Those that grow best at pH 9 or above (often the medium. (Recall that proton ionophores
in the range of pH 10–12, i.e., alkaliphiles), equilibrate protons across the cell membrane.)
403
www.ebook777.com
The idea is that when the cytoplasm becomes

8.3
too acid, protons are pumped out. For bulk
proton flow to take place (e.g., by bringing K+
8.2
into the cell), the pumping must be done elec-
troneutrally. When the cytoplasm becomes too
8.1
basic, protons are brought in via exchange with
outgoing K+ or Na+ (Fig. 16.2). This implies the
8.0
existence of feedback mechanisms by which
the intracellular pH can signal proton pumps pHin
7.9
and antiporters. Regulation at this level is not
understood. Moreover, even the influx of pro-
7.8
tons depends upon the outgoing proton pumps.
This is because the antiporters that bring in pro-
tons in exchange for sodium or potassium ions
0 5 10 20
are driven by the ∆p: by the ∆pH, outside acid, minutes
and/or by the membrane potential, ∆Ψ. When
the inflow of protons is electrogenic (i.e., when Fig. 16.1 pH homeostasis in E. coli. Growing E. coli
the ratio of H+ to Na+, or K+, on the antiporter cells were shifted from pH 7.2 to pH 8.3 (arrow). The
exceeds 1), ∆Ψ drives the antiporter. Therefore, ∆pH (pHin–pHout) was measured by using a weak acid or
inhibition of the proton pumps should lead to a a weak base as described in Section 4.5.2, and the pHin
was calculated. Immediately after the cells had shifted
to pH 8.3, the cytoplasmic pH rose to 8.3. However,
Table 16.1 pH homeostasis in bacteria within a few minutes the cytoplasmic pH was restored
to approximately its original value. The mechanism by
Bacteria pHout pHin ∆pH (pHin—pHout)
which E. coli acidifies the cytoplasm to maintain pH
homeostasis is uncertain. Data from Zilberstein, D., V.
Neutrophile 6–8 7.5–8 +
Agmon, S. Schuldiner, and E. Padan. 1984. Escherichia
Acidophile 1–4 6.5–7 +
Alkaliphile 9–12 8.4–9 –
coli intracellular pH, membrane potential, and cell
growth. J. Bacteriol. 158:246–252.
1 and 2 together raise

intracellular pH
1
K+
H+
2
Na+
5 completes 5
Na+ circuit R Na+
3
K+
H+
4
3 and 4 lower
intracellular pH
H+
Fig. 16.2 Mechanisms of pH homeostasis. (1) Primary proton pumps create a membrane potential. (2) The
uptake of K+ dissipates the membrane potential, allowing extrusion of protons via the pumps and an alka-
linization of the cytoplasm. The bulk solution is kept electrically neutral because of the counterions that had
neutralized the K+. Therefore, a ∆Ψ is changed into a ∆pH, inside alkaline. (3) and (4) Cation/proton antiport-
ers pump Na+ and K+ out and bring in H+. These are suggested to be the major mechanism for acidifying the
cytoplasm. (5) Sodium ion uptake systems complete the sodium circuit.
responses to environmental stress 405
situation where pHin = pHout (i.e., a collapse of

the ∆pH), and this can be tested.
The two major classes of proton pumps are
those coupled to respiration and the proton-
translocating ATPase. The former can be inhib-
ited by respiratory poisons such as cyanide,
and the latter by inhibitors of the ATPase [e.g.,
N,N´ -dicyclohexylcarbodiimide (DCCD)] or
by mutation. One can also short-circuit the pro-
ton flow by using proton ionophores. The iono-
phores will dissipate the proton potential, thus
neutralizing the pumps.7 These experiments
show that in the absence of proton pumping,
the ∆pH falls as the protons tend to equilibrate
across the cell membrane.
The role of K+ in maintaining a ∆pH

A problem in using the proton pumps to move
protons out of the cell is that the pumps are elec-
trogenic, and the number of protons that can be
pumped out of the cell is limited by the mem-
brane potential that develops. (See the discus-
sion of membrane capacitance, Section 4.2.2.)
Thus, for proton pumping to raise the intracel-
lular pH, the excess membrane potential must
be dissipated either by the influx of cations or
by the efflux of anions. In neutrophilic bacteria,
K+ influx dissipates the membrane potential,
allowing more protons to be pumped out of
the cell. This means that K+ influx is required to
raise the intracellular pH. Fig. 16.3 Uptake of K+ in E. coli is associated with
pH homeostasis. Washed E. coli cells were sus-
This is seen in Fig. 16.3, where the addition
pended in buffer without K+ at pH 5.3 (open circles),
of K+ to E. coli cells suspended in media of low
6.8 (solid circles), 7.15 (open squares), or 7.6 (solid
pH caused an immediate increase in intracellu- squares). The cytoplasmic pH (pHin) was determined
lar pH, which stabilized at approximately pH by using the distribution of a weak acid. Glucose was
7.6. There is much more uncertainty about how added shortly after 0 min, and K+ was added at 10
neutrophiles might decrease their intracellular min. Potassium uptake was complete within 10 min
pH. Potassium ions and sodium ions may play of being added. Immediately upon the addition of
a role in lowering the intracellular pH by bring- K+, the intracellular pH, which had been lowered by
ing protons back into the cell via the H+/K+ and the extracellular pH, rose and stabilized at approxi-
H+/Na+ antiporters (Fig. 16.2). However, the mately pH 7.6. The rise in the cytoplasmic pH was
evidence for this suggestion is not as strong in due to pumping protons out of the cell in response
to the depolarization of the membrane by the influx
neutrophiles as it is in alkaliphiles.
of potassium ions. The mechanism for acidification
of the cytoplasm (i.e., recovery from overshoot of
In alkaliphiles the Na+/H+ antiporter the pHin) is unknown. Source: Kroll, R. G., and I.
acidifies the cytoplasm R. Booth. 1983. The relationship between intracel-
Since alkaliphiles live in a very basic medium, lular pH, the pH gradient and potassium transport in
their main problem is keeping a cytoplasmic Escherichia coli. Biochem. J. 216:709–716.
pH more acid than the external pH, perhaps
by as much as 2 units. In other words, they
must always be bringing protons into the
cell. (See ref. 8 for a review of bioenergetics in
www.ebook777.com
alkaliphiles, including pH homeostasis.) In con- maintaining a ∆pH in acidophiles and neutro-

trast to research with neutrophiles, a strong case philes may be similar in relying on K+ influx to
(reviewed in ref. 5) can be made for the acidifi- depolarize the membrane. It has also been sug-
cation of the cytoplasm of alkaliphiles by Na+/ gested that the efflux of K+ in acidophiles may be
H+ antiporters (Fig. 16.2). When alkaliphiles slowed when pHin falls, thus limiting the entry of
are placed in a medium without Na+ at pH 10 protons against a positive ∆Ψ.
or 10.5, the internal pH quickly rises to the Although the protons are not being pumped
value of the external pH. However, when Na+ against a membrane potential when the exter-
is present in the external medium, the internal nal pH is acid, they are being pumped against
pH does not rise upon shifting to the more basic a proton gradient. The ∆pH in acidophiles can
medium. (See note 9 for more detail.) be 4 to 5 units, which is equivalent to 240 to
Furthermore, mutants of alkaliphiles that 300 mV (i.e., 60 ∆pH). This is a large concentra-
cannot grow at pH values above 9 are defective tion gradient against which to pump protons.
in Na+/H+ antiporter activity. The antiporter is The energy to pump the protons is derived from
electrogenic (H+ > Na+) and in this case is driven aerobic respiration. However, iron-oxidizing
by the ∆Ψ, the membrane potential, which is acidophilic bacteria do not generate sufficient
generated by the primary proton pumps of energy during electron transport to pump pro-
the respiratory chain. The sodium ion circuit tons out of the cell because the ∆Eh between
is completed when sodium ion enters the cell Fe3+/Fe2+ and O2/H2O is very small (<100 mV).
via Na+/solute symporters that are also driven These bacteria appear to regulate their ∆pH by
by the ∆Ψ. The use of Na+/solute symporters consuming cytoplasmic protons during respi-
is advantageous because solute transport is ration, rather than by proton pumping. This is
driven by the sodium potential rather than the discussed in Section 13.4.1.
proton potential, the latter being low because
of the inverted ∆pH. 16.2 Osmotic Pressure and Osmotic
Potential
pH Homeostasis in acidophiles 16.2.1 Osmotic pressure
Acidophiles differ from other bacteria in that When two solutions are separated by a mem-
the external pH is several units lower than the brane that allows the passage of water but not of
cytoplasmic pH.6 The maintenance of the large solute, the water will diffuse from the less con-
∆pH requires an inverted ∆Ψ at low pHout; oth- centrated to the more concentrated side, thereby
erwise proton efflux would be limited by a posi- equalizing the water concentration. What is
tive ∆Ψ as well as a low pHout, and proton influx happening is that the concentration of water
would be promoted by a negative ∆Ψ as well (actually the thermodynamic activity of water)
as a high pHin. Accordingly, acidophilic bacte- in the less concentrated solution is higher than
ria have small membrane potentials, which can in the concentrated solution. Thus the water is
be inside positive at acidic pHout. For example, simply following its concentration gradient.
the membrane potential of Thiobacillus fer- The diffusion of water into the more concen-
rooxidans is +10 mV at an outside pH of 2, trated solution is called osmosis. When water is
and the membrane potential of Bacillus acido- diffusing from a side with pure water, the pres-
caldarius is +20 to +30 mV at an outside pH of sure that must be applied to stop this osmotic
2.5. However, acidophiles pump protons out of flow is called the osmotic pressure and is given
the cell during electron transport, generating a the symbol Π. If a solution is sufficiently dilute
membrane potential that is outside positive as that one can discount the interactions between
in other bacteria. the solute molecules, then Π = RTCs, where R is
How is the membrane potential inverted? It the gas constant, T is temperature in kelvins, and
has been suggested that the maintenance of the Cs is the molar concentration of solute particles.
inverted membrane potential in acidophiles is It is important to point out that Cs represents the
due to an inward flux of K+ greater than an out- concentration of independent particles that con-
ward flux of protons. This might be due to the tribute to the osmotic pressure. For example,
electrogenic influx of K+ catalyzed by a known Cs is equal to the sum of the concentration of
ATP-dependent K+ pump. Thus, the method of ions produced when a salt completely ionizes in
solution. Thus, Cs is equal to the osmolarity of the Equation 16.3 points out that solutions with
solution (osM), and the units can be expressed as higher concentrations of solutes have more neg-
moles per liter or osmoles per liter. For example, ative osmotic potentials.12 Thus, water flows
a one-molar (1 M) solution of NaCl is 2 osmoles into these solutions.
per liter because there is one mole each of Na+
and Cl– per liter. The gas constant, R, is equal
16.2.3 Turgor pressure and its
to 0.0281 liter-atm K–1 mol–1, and the units of
importance for growth
osmotic pressure are usually expressed as mil-
Because the cytoplasm of most bacterial cells
limeters of mercury (mmHg). At higher solute
is much more concentrated in particles than is
concentrations one must take into account the
the medium in which the cells are suspended,
interactions between solute molecules, and Π/
the cytoplasm has a more negative osmotic
RT is not equal to Cs but, rather, to the sum of
potential (a more positive osmotic pressure)
the effective molar concentrations of the solutes.
than the medium, and water flows into the cell.
Often concentrations are expressed as molality
The incoming water expands the cell mem-
(moles of solute per kilogram of solvent) rather
brane, which exerts a pressure directed out-
than molarity, and the units of osmotic pressure
ward against the cell wall. The pressure exerted
are given as osmolality.
against the cell wall is called the turgor pressure.
The turgor pressure is equal to the difference in
16.2.2 Osmotic potential
osmotic pressure between the medium and the
For a review of osmotic potential and its regula-
cytoplasm
tion, see ref. 10. Csonka has pointed out that the
term “osmotic potential” is more useful than P = ∆Π = Πin—Πout
“osmotic pressure” in that it emphasizes that = RT(osmin—osmout) (16.4)
water flows from solutions of a high osmotic
where P is the turgor pressure and osm is the
potential to solutions of a low osmotic poten-
total concentration (in units of molarity) of
tial.11 [The osmotic potential is numerically
osmotically active solutes in the cells and in the
equal to the osmotic pressure but, as shown
medium.
shortly, has a negative sign (eq. 16.3).] The
The turgor pressures in gram-positive bacte-
osmotic potential, π, is a function of the activity
ria are about 15 to 20 atm and in gram-negative
(a) of the solvent. For water, this would be
bacteria between 0.8 and 5 atm.13–15 This, of
π = (RT/Vw)ln aw (16.1) course, is the reason that bacterial cell walls must
be so strong. In bacteria, the tensile strength
where R is the universal gas constant, T is the
of the cell wall is due to the peptidoglycan. It
absolute temperature, Vw is the partial molal
is important to realize that the turgor pressure
volume of water, and aw is the activity of water.
provides the force that expands the cell wall and
The activity of pure water is defined as one,
is necessary for the growth of the wall and cell
making the osmotic potential zero. Solutes tend
division.16 In fact, sudden decreases in turgor
to lower the activity of water, therefore mak-
pressure brought on by increasing the osmolar-
ing the osmotic potential of solutions negative.
ity of the suspending medium result in a cessa-
This is because in an ideal solution (i.e., one in
tion of growth accompanied by the inhibition of
which the interactions between solute and sol-
a variety of physiological activities (e.g., nutri-
vent molecules are independent of concentra-
ent uptake and DNA synthesis). Therefore, the
tion), the activity of the solvent is equal to its
physiological significance of osmotic homeo-
mole fraction. For example, for water,
stasis is that it maintains an internal turgor pres-
aw = nw/(nw + ns) (16.2) sure necessary for growth.
Bacteria have the capability of adjusting
where nw is the number of water molecules and
their internal osmolarity to changes in exter-
ns is the number of solute molecules. For dilute
nal osmolarity for the sake of maintaining
solutions, the osmotic potential is related to the
cell turgor. How bacteria detect differences in
molar concentration (moles of solute per liter),
external osmolarity and transfer the appropri-
Cs, as follows:
ate signals to the adaptive cellular machinery
π = –RTCs (16.3) is largely unknown. The problems associated
www.ebook777.com
with analyzing the signaling are discussed in the biosynthesis of organic compatible solutes
Section 16.2.7. Before we address these prob- have been reviewed.15
lems, we will examine the evidence for osmotic
regulation and identify the molecules primarily Osmotic homeostasis in halobacteria
responsible for maintaining the osmotic differ- As described in Section 1.1.1, the halobacteria
ential between the cytoplasm and the external (archaeons) live in waters having NaCl concen-
medium. trations from 3 to 5 M. To prevent water from
exiting the cell via osmosis and also to maintain
a positive internal turgor pressure, the cyto-
16.2.4 Adaptation to high-osmolarity plasm is kept quite salty. However, the cation is
media K+ (KCl), rather than Na+ (NaCl). Potassium ion
When cells are placed in media of high osmo- uptake systems maintain internal K+ concentra-
larity, they increase the intracellular concentrations on the order of 3 M. The Na+ is exported
tions of certain solutes called osmolytes, thus from the cytoplasm.18 Of course for the proteins
ensuring that the internal osmolarity is always to cope with such a high ionic strength, there
higher than the external and that cell turgor is must be adaptations in the cytoplasm. It is well
therefore maintained. The osmolytes used by known that ionic interactions are weakened in
bacteria are sometimes called compatible sol- the presence of high concentrations of salt, and
utes to reflect their relative nontoxicity. Some this can have a profound effect on the tertiary
compatible solutes are not synthesized by the and quaternary structure of proteins, leading to
cells but are accumulated intracellularly from unfolding and dissociation of subunits. In fact,
the medium via transport. Organic compatible however, the proteins and structures in extreme
solutes that are accumulated via transport but halophiles require high concentrations of salt
are not synthesized are called osmoprotectants. (at least 1 M) for stability and activity. Even the
One of the most important compatible sol- cell envelope of halobacteria disintegrates when
utes in bacteria is K+, which has an intracellular the salt concentration is lowered. (See ref. 19 for
concentration high enough to make this ion a a review of the effect of salt on halophilic mac-
major contributor to the internal osmolarity romolecules and ref. 20 for a general review of
and hence turgor. (See note 17.) Bacteria that the biology of the halophilic bacteria.)
live in high-osmolarity media have proportion-
ally higher intracellular concentrations of K+ Osmotic homeostasis in E. coli
because of uptake. Therefore, K+ is important It is possible to observe osmotic homeostasis
for two major aspects of homeostasis: mainte- by rapidly increasing the external osmolarity
nance of cell turgor and maintenance of a ∆pH of the medium and identifying the intracellular
(Section 16.1.3). Bacteria employ other com- compatible solutes that maintain turgor. To
patible solutes in addition to K+. The situation illustrate osmotic homeostasis, some experi-
with regard to compatible solutes can be sum- ments performed with E. coli will be consid-
marized as follows. ered.21 When E. coli is shifted to a medium
Solutes that increase intracellularly in many of high osmolarity, there is a series of adjust-
different bacteria in response to high external ments (Fig. 16.4). First, there is an influx of K+
osmolarity include K+, the amino acids gluta- via the potassium uptake systems discussed in
mate, glutamine, and proline, the quaternary Chapter 17. These are thought to respond to a
amine betaine (also called trimethylglycine or decrease in turgor pressure.22
glycinebetaine), and certain sugars (e.g., treha- For there to be an influx of K+, however, two
lose, which is a disaccharide of glucose). Very problems must be solved by the cell. Steps must
few chemoheterotrophic bacteria can synthe- be taken to (1) preserve electrical neutrality in
size betaine de novo; probably the others find the cytoplasm and (2) prevent depolarization of
it in the environment. Betaine is, however, syn- the cell membrane, which would lead to a drop
thesized by cyanobacteria and phototrophic in the ∆p. In fact, E. coli handles these problems
bacteria and is probably abundant in saline in two ways. A transient alkalinization of the
environments, where these phototrophs grow. cytoplasm as protons are pumped out both pre-
Betaine is also synthesized by several halophilic vents depolarization of the membrane and aids
and halotolerant archaebacteria. Pathways for in maintaining cytoplasmic neutrality during
Fig. 16.4 Changes in intracellular osmolyte concentrations when E. coli is shifted to a medium of higher
osmolarity. Growing cells were shifted to a medium containing 0.5 M NaCl at time zero. The intracellular
osmolyte concentrations are the differences between the concentrations in cells subjected to osmotic upshock
and in control cells that were not. Upon a shift to a medium of higher osmolarity, the cells accumulated potas-
sium ion and synthesized glutamate. The potassium glutamate was replaced by newly synthesized trehalose.
The addition of proline to the medium (arrow) caused the displacement of trehalose by proline. Proline was
also capable of inducing an early efflux of K+ if added earlier. Source: Data adapted from Dinnbier, U., et al.
1988. Transient accumulation of potassium glutamate and its replacement by trehalose during adaptation
of growing cells of Escherichia coli K-12 to elevated sodium chloride concentrations. Arch. Microbiol. 150:
348–357.
the early rapid uptake of K+. However, the cyto- readily accumulated from media of high osmo-
plasm reacidifies, and the main reason for cyto- larity and can even suppress the uptake of other
plasmic neutrality is the rapid accumulation osmoprotectants, as well as causing the excre-
during reacidification of glutamic acid, which tion of K+ and the catabolism of trehalose, thus
ionizes, providing counterions to the increased replacing these osmolytes.
levels of K+.16,23 The accumulation of glu-
tamic acid is due to either increased synthesis, Effect of osmolarity on transcription and on
decreased utilization, or a combination of the activities of enzymes
two. Thus, the initial major compatible solute As part of adaptation to a change in media
is potassium glutamate. However, after several osmolarity, bacteria synthesize new enzymes
minutes the potassium glutamate is replaced by or transporters, which are responsible for
the newly synthesized sugar, trehalose, which the biosynthesis of compatible solutes or the
then becomes the major compatible solute. This transport of these into the cells. Increased tran-
is due to the excretion of the K+ and the catabo- scription of some of the genes activated by an
lism or excretion of glutamate, as well as the osmotic upshift is due to increased levels of the
synthesis of trehalose. transcription factor σs (sigma S). The transcrip-
If proline or betaine is present in the medium, tion factor σs is a subunit of RNA polymerase
then E. coli transports these into the cell and holoenzyme that is responsible for recognizing
replaces the trehalose or the excess K+, indicat- σs-dependent promoters that are activated dur-
ing that E. coli preferentially uses proline and ing osmotic upshifts as well as during starvation.
betaine as osmotic stabilizers rather than tre- (See Section 2.2.2 and ref. 25 for a review.) The
halose or K+. (See note 24.) (The K+ is excreted increase in σs is regulated by medium osmolar-
and the trehalose is catabolized.) The preferen- ity at the level of translation and protein turn-
tial use of some osmoprotectants over others over. Following are a few examples of the effect
is common among bacteria. In many bacteria of osmolarity on transcription and enzyme
betaine is a preferred osmoprotectant that is activities.
www.ebook777.com
1. E. coli activates enzymes for the synthe- body, also repress the transcription of ompF.
sis of periplasmic oligosaccharides when Regardless of the physiological significance of
shifted to low-osmolarity media. This will the switch in porins, the system is of interest to
be explained in Section 16.2.5. us here because it is regulated in some way by
2. Staphylococcus aureus activates a preexist- osmotic pressure, and the signaling pathway
ing proline uptake system when shifted to from the membrane to the genome is being dis-
high-osmolarity media.26 Proline is an osmo- sected experimentally.28
protectant.
3. E. coli and S. typhimurium increase the tran- 16.2.5 Adaptation to low-osmolarity
scription of proU, an operon that codes for a media
proline (and betaine) transport system when Thus far we have been considering the adap-
shifted to high-osmolarity media. tation of bacteria to high-osmolarity media,
4. Another set of genes whose transcription is which tend to suck water out of the cytoplasm
increased in high-osmolarity media is the and lower the turgor pressure. As just discussed,
kdp operon in Escherichia, which codes for bacteria respond to high osmolarity by raising
a high-affinity K+ transport system. As men- the intracellular concentrations of compatible
tioned, K+ is a major compatible osmolyte in solutes. Many bacteria have means for adjust-
most bacteria. ing to low osmolarity media, thus limiting cell
One can ask how bacteria sense changes in the turgor pressure. Few details are known.
external osmolarity and transmit the appropri- One response of bacteria to low-osmolarity
ate signals to the genome or to certain enzymes. media is to decrease the concentration of cyto-
The answers are largely unknown. Perhaps the plasmic osmolytes. This might occur via specific
best-studied signaling system that responds excretion of osmolytes or as a result of their
to changes in external osmolarity includes catabolism. For example, E. coli excretes K+ via
the genes for the OmpF and OmpC porins in special transporters that may respond to cell
Escherichia and Salmonella (see Section 1.2.3). turgor.29 An important means of adjusting to a
The total amounts of OmpF and OmpC are drop in external osmolarity involves mechano-
fairly constant, but the ratio changes with the sensitive channels, as discussed next.
osmolarity and temperature of the medium. In
high-osmolarity media, the transcription of the Mechanosensitive channels
ompF gene, which codes for the larger OmpF Mechanosensitive (MS) channels, which have
channel, is repressed and the transcription of the been found in cell membranes of eukaryotes,
ompC gene, which codes for the smaller OmpC archaea, bacteria, have been most studied
channel, is increased. The result is a switch to in Escherichia coli.30,31 These channels offer
a smaller porin channel in high-osmolarity protection against osmotic stress when cells
media. are placed in a sufficiently dilute (hypo-os-
Why bacteria should switch from one porin to motic) medium, that is, when they are hypo-
the other is not clear, but it probably has noth- osmotically shocked. The channels are gated,
ing to do with osmotic homeostasis. The smaller which means that when water rushes into the
OmpC channel is probably an advantage in cell from an hypo-osmotic (dilute) environ-
the intestinal tract, where these bacteria live, ment, thus increasing the turgor pressure, the
because of the presence of toxic molecules (e.g., increased tension in the cell membrane causes a
bile salts).27 The argument is that in the intesti- conformational change in the MS channel pro-
nal tract, the osmotic pressures are higher, thus teins such that the channels transiently open.
favoring the smaller OmpC channels, whereas As a consequence, numerous solutes (e.g., K+,
in ponds and streams, where the bacteria are ATP, glutamate, and compatible solutes) exit.
also found, the lower osmotic pressures favor This lowers the internal osmotic pressure, and
the larger OmpF channels, which may allow thus less water rushes in and the cells are not
more efficient uptake of nutrients. Consistent osmotically lysed. The channels actually allow
with this hypothesis is the observation that transport in both directions. At the same time
higher temperatures, expected in the intestines that there is an efflux of solutes from the cyto-
of animals as opposed to habitats outside the plasm, Na+ and H+ enter the cell.
Osmotic homeostasis in the periplasm oligosaccharides is to raise the osmolarity in

The periplasm is reportedly filled with a gel the periplasm. The glucans generally have
whose volume is still a matter of controversy anionic groups because they are substituted
(Section 1.2.4). (See note 32.) It has been sug- with phosphorylated compounds (phospho-
gested that gram-negative bacteria adapt to glycerol, phosphoethanolamine, and phos-
low-osmolarity media by raising the osmolar- phocholine, and sometimes succinic acid). The
ity of the periplasm so that the cytoplasm never anionic oligosaccharides would be expected to
actually “sees” the low-osmolarity external be very effective in raising the osmolarity of the
medium. In this way, swelling of the cell mem- periplasm because cations would accumulate
brane and concomitant compression of the in the periplasm in response to the negatively
periplasm are minimized, as well as the turgor charged nonpermeable oligosaccharides. (See
pressure across the cell membrane. (The option note 37.) However, E. coli mutants unable to
of lowering the osmolarity of the cytoplasm by synthesize the MDOs show no growth defects
excreting solutes is of limited usefulness. The in low-osmolarity media.38 If it is necessary to
cytoplasm must maintain a minimum concen- maintain a high periplasmic osmolarity when
tration of salts and other solutes, approximately the external osmolarity is decreased, then there
300 mosM, to support growth.) In fact, it has must exist alternative mechanisms besides the
been reported that the periplasm remains as a synthesis of MDOs.
separate compartment under all conditions of
external osmolarity, with an osmolarity appar- 16.2.6 Conceptual problems
ently iso-osmotic with that of the cytoplasm.33 It has been suggested that the periplasm is
However, the nonspecific diffusion channels maintained iso-osmotic with respect to the
formed by the porins render the outer membrane cytoplasm. If indeed this is the case, there are
of Escherichia and Salmonella and, by infer- important consequences for our understand-
ence, other gram-negative bacteria, permeable ing of how the cell walls of gram-negative bac-
to small molecules of molecular weight less than teria are able to resist high turgor pressures. It
600. This being the case, one can ask how bacteria is generally believed that the turgor pressure is
maintain an osmolarity in the periplasm higher exerted across the cell membrane, not the outer
than that of the external medium. One possi- membrane, and that the overlying peptidogly-
bility is that gram-negative bacteria synthesize can acts as a strong retainer against which the
and/or accumulate periplasmic osmolytes when cell membrane is pressed. If the periplasm were
grown in low-osmolarity media. Indeed, many iso-osmotic with the cytoplasm, the turgor pres-
gram-negative bacteria besides E. coli, includ- sure would not be across the cell membrane and
ing those belonging to the genera Salmonella, against the peptidoglycan but, rather, against
Pseudomonas, Agrobacterium, Acetobacter, the outer membrane. However, the outer mem-
Klebsiella, Enterobacter, Bradyrhizobium, brane is not built to withstand high turgor
Brucella, Xanthomonas, Alcaligenes, and pressures.39
Rhizobium, synthesize periplasmic β-glucan If indeed the turgor pressure were exerted
oligosaccharides (called membrane-derived oli- against the outer membrane, one would have to
gosaccharides, or MDO, in E. coli) when grown suppose that the peptidoglycan reinforced the
in media of low osmolarity.34–36 It has been outer membrane by being tightly bonded to it
shown that E. coli, Agrobacterium, Rhizobium, at numerous sites. Perhaps the lipoprotein mol-
and Bradyrhizobium increase the synthesis ecules that are covalently bonded to the pepti-
of periplasmic β-glucans when grown in low- doglycan are anchored sufficiently to the outer
osmolarity media. The enzymes that synthesize membrane via hydrophobic bonding to provide
the oligosaccharides are constitutive; therefore the needed stability (Section 1.2.3). Another
their activities are increased when the cells are conceptual difficulty regarding a periplasm iso-
grown in low-osmolarity media.29 osmotic with the cytoplasm entails the sensing
The increase in the synthesis of the periplas- of turgor pressure by cell membrane proteins:
mic oligosaccharides when certain bacte- without differential pressure across the cell
ria are grown in media of low osmolarity membrane, it is uncertain how iso-osmoticity
has led to the suggestion that the role of the could be maintained. Clearly, much more needs
www.ebook777.com
to be learned about turgor pressure and osmotic 16.3 Heat-Shock Response (HSR)
regulation in the periplasm. 16.3.1 Heat-shock proteins
When E. coli is shifted to a higher temperature,
16.2.7 What is the nature of the signal it responds by transiently increasing the rate of
sensed by the osmosensors? synthesis of a group of proteins called the heat-
The signals to which the putative osmosensors shock proteins (Hsps) relative to other proteins.
respond are not understood.5 Indeed, the dif- For a review, see refs. 40 and 41. (Cold-shock
ferent osmosensors may not even respond to proteins, which may function during adapta-
the same type of signal. Several possibilities for tion to low temperature also are induced in
osmosensor responses have been discussed, bacteria.42–44) Upon a downshift from the higher
including membrane proteins that are sensi- temperature, the synthesis of heat-shock pro-
tive to (1) pressure against the peptidoglycan teins is decreased. For example, when E. coli is
sacculus, (2) membrane stretch, (3) changes shifted from 30 °C to 42 °C, the cells increase
in the concentrations of intracellular solutes the rate of synthesis of more than 30 Hsps for
(e.g., K+), and (4) water activity. It is important 5 to 10 min. The increased rate is about 10 to
to know whether the periplasm is truly iso- 20 times greater than the rate of synthesis of the
osmotic with the cytoplasm as just discussed. majority of proteins, and during this period over
For under these conditions, there should be no 20% of all the proteins synthesized are Hsps.
pressure differential across the cell membrane, As will be explained later, the cell has a need
and mechanisms (1) and (2) just listed would for several of these proteins at all temperatures,
not apply. not simply high temperatures. However, the
It should also be understood that after the need is greater at the elevated temperatures,
cells have adapted to an osmotic upshift and hence the increased rate of synthesis. This
begun to grow again, the osmotic differential type of response to a temperature shift is not
between the cytoplasm and medium is pre- unique to bacteria and occurs in cells of animals
sumably restored, and therefore systems still (including humans), plants, and eukaryotic
activated under these circumstances cannot microorganisms. In fact, it was first discovered
be responding to a change in turgor pressure in Drosophila, where it is manifested in the syn-
or related events such as membrane stretch- thesis of new chromosomal puffs associated
ing. They could, however, be responding to with the production of Hsps in salivary gland
increased concentrations of specific solutes or chromosomes.
some other parameters not dependent upon a The heat-shock proteins are classified into
changed turgor pressure. families according to their molecular weights.
For example, when E. coli is shifted to a high- The heat-shock proteins with a molecular
osmolarity medium, it continues to repress weight of about 70 kDa are in the Hsp70 family
the ompF gene and stimulate the ompC gene, (e.g., DnaK), those with a molecular weight of
even though adaptation to the high-osmolarity around 60 kDa are in the Hsp60 family (e.g.,
medium has taken place. Similarly, the proU GroEL), those with a molecular weight of about
genes that specify proline and glycinebetaine 40 kDa are in the Hsp40 family (e.g., DnaJ), and
uptake systems in E. coli and S. typhimurium those with a molecular weight around 10 kDa
remain induced in high-osmolarity media, also are in the Hsp10 family (e.g., GroES). The pro-
implying that the sensor is not responding to teins made during the heat-shock response are
a changed osmotic pressure differential on remarkably similar across phylogenetic lines.
both sides of the cell membrane. On the other For example, the DnaK protein from E. coli is a
hand, the kdp genes for potassium ion uptake member of the Hsp70 family and is about 50%
are quickly but only transiently induced by an homologous with heat-shock proteins in the
upshift in osmolarity. This fact, plus the find- Hsp70 family found in humans.
ing that only nonpermeable solutes induce tran- What do the Hsps do? Some of the Hsps
scription of the kdp operon, suggests that the aid in the proper folding of other proteins, the
osmosensor for these genes may indeed detect export of proteins, and some are proteases
turgor pressure or something closely related, that degrade unfolded proteins. It is clear that
such as membrane stretch. in E. coli at least some of them are absolutely
required for growth to occur above 20 °C. The and, as a consequence, its amounts increase. The
evidence for this is that null mutations (loss-of- proteases involved in the turnover of σ32 have
function mutants) in a gene (rpoH) encoding been identified.47 Apparently, the free pools of
a sigma factor (σ32) required to transcribe the the cytoplasmic proteins DnaK and DnaJ are
hsp genes prevent growth above 20 °C.45 (Most being sensed during heat stress. At nonstress
of the E. coli genes are transcribed by an RNA temperatures (30 °C) the available DnaK and
polymerase that has σ70 as the sigma factor.) In DnaJ bind to σ32, and the sigma factor is subject
addition to the absolute requirement for Hsp to proteolysis by several proteases including
proteins above 20 °C, there is an enhanced effect FtsH, Hs1VU, ClpAP, and Lon. At high tem-
on growth by the Hsps at all temperatures, as peratures (42 °C) the DnaK and DnaJ preferen-
judged by the slower growth of null mutants of tially bind to denatured proteins, and the sigma
rpoH in comparison to wild-type cells.44 Thus, factor binds to RNAP. Binding of σ32 to RNAP
as will be made clear in Section 16.3.6, the Hsps protects the sigma factor from proteolysis and
not only function to repair or eliminate pro- results in a holoenzyme that transcribes the
teins damaged by heat stress, but they also play Hsp σ32 regulon. Thus, it is not the temperature
important roles in growth at all temperatures. per se that increases the activity and amount of
This is because, as mentioned earlier, several σ32, but rather the amount of denatured pro-
of the Hsps (e.g., DnaK, DnaJ, GrpE, GroEL, tein. This is in agreement with the finding that
GroES) are important for proper protein fold- increased synthesis of the Hsps is induced not
ing and protein export at all temperatures, and only by high temperatures, but also by ethanol,
the increased amount at higher temperatures starvation, and oxidative stress.
reflects an increased demand for these proteins. Additionally, there is an increase in the rate
The following discussion concerns how of translation of the mRNA for σ32 during the
E. coli uses alternative sigma factors to respond period of increased synthesis of the heat-shock
to environmental stresses, one of these being proteins. Interestingly, after a temperature
heat shock. Although the heat-shock response is downshift, the activity of σ32 decreases, and
universal, the mechanisms that regulate the syn- this, rather than a decrease in concentration
thesis of the heat-shock proteins differ in vari- of σ32, results in a lowered rate of synthesis of
ous prokaryotes. For example, the expression of the heat-shock proteins. Thus, the regulation of
heat-shock genes in several gram-positive bacte- σ32 expression is complex and occurs at several
ria thus far investigated indicates that an alter- levels, including stability of the protein, activ-
native sigma factor is not involved. On the other ity of the protein, and translation of the mRNA.
hand, Bacillus subtilis does induce an alternative The student is referred to the review by Gross,
sigma factor during the heat-shock response.46 which discusses possible mechanisms of regula-
tion.48 Also, see the discussion in Section 2.2.2
16.3.2 The σ32 (RpoH) regulon of sigma factor σs, which directs the transcrip-
tion of genes induced by starvation and osmotic
Following a temperature upshift (e.g., from
upshift, and is also regulated at several levels, as
30 °C to 42 °C), there is a transient increase in the
well as ref. 49 and note 50.
amount of sigma factor σ32, also called RpoH,
which is responsible for the synthesis of at least
30 Hsps that act in the cytoplasm. Thus, σ32 rec- 16.3.3 The σE (σ24) regulon
ognizes the promoters of genes in a major heat- Another regulon that in E. coli is activated by
shock regulon, the σ32 regulon. Mutants that do stress is the σE regulon, which is activated at
not make σ32 cannot grow at temperatures above very high temperatures (45–50 °C) and protects
20 °C. A major portion of the research discussed against damage to extracytoplasmic proteins.51,52
next has been done with E. coli. Stress conditions such as high heat (45–50 °C)
Several factors contribute toward the accu- or ethanol can result in the denaturation (mis-
mulation of σ32 at higher temperatures. During folding) of proteins in the outer membrane or
steady state growth at any temperature, σ32 is periplasm of E. coli. As a consequence of mis-
an unstable protein with a half-life of only one folding, a signaling pathway results in the acti-
minute. After a shift from 30 °C to 42 °C, how- vation of sigma factor σE (σ24) in the cytoplasm.
ever, the protein is stabilized for a few minutes As a consequence of the activation of σE, the σE
www.ebook777.com
regulon, which consists of at least 11 genes, is (1) the folding of newly synthesized proteins at
transcribed. The proteins synthesized include all temperatures, (2) the export of proteins at
a periplasmic peptidylprolyl isomerase (FkpA) all temperatures, (3) the refolding of misfolded
and a periplasmic protease (DegP) that are pos- polypeptides, and (4) the proteolysis of improp-
tulated to be involved in the folding, refolding, erly folded or otherwise abnormal proteins.48,56
or degradation of misfolded envelope proteins.
Sigma factor σE is required for growth at all tem- Several of the Hsps are chaperone proteins
peratures. It is suggested that under nonstress If an Hsp aids in the folding of newly synthe-
conditions σE is kept at low activity by being sized proteins, the refolding of improperly
bound to an anti-sigma factor that is localized in folded proteins, or the export of proteins, it
the inner membrane (cell membrane). Envelope belongs to the class of proteins called chaper-
stress results in release of the sigma factor so that one proteins. These are proteins that take part
it can bind to the RNAP. in the folding (or prevent folding), assembly, or
export of other proteins but are not part of the
16.3.4 The σs (RpoS) regulon final protein or protein complex. An example
One of the sigma factors, σs (or RpoS), has been of a chaperone protein is SecB, described in
referred to as a “master regulator” of the gen- Section 18.1 as part of the Sec transport system
eral stress response in E. coli. It increases the that exports proteins.
expression of many genes in response to stress
conditions, including starvation, entry into sta- The Hsps that aid in folding newly
tionary phase, hyperosmolarity (0.3 M NaCl), synthesized proteins
and low pH (pH 5).53 RpoS is kept at low levels Several of the Hsps are chaperone proteins that
during logarithmic growth via a protease called play important roles not only at elevated tem-
ClipXP and an adaptor protein called RssB peratures, but (as discussed in Section 11.2.12)
that targets RpoS for degradation by ClipXP.54 at ordinary growth temperatures as well,
This is a complex regulatory process, and vari- because they aid in the folding of newly synthe-
ous antiadaptor proteins are now known that sized proteins. Newly synthesized proteins usu-
protect RpoS from degradation during stress ally do not spontaneously fold into their correct
responses that require RpoS. For an example of structures without the help of chaperone pro-
this, see ref. 55. RpoS and some of the genes that teins that transiently bind to the proteins during
it regulates are discussed in Section 2.2.2. the folding process. Chaperone proteins also
prevent newly synthesized proteins from aggre-
16.3.5 The Cpx system gating with each other before proper folding has
Another system that responds to misfolded occurred. This is especially important at higher
proteins is a two-component signaling system temperatures, where improper folding or pro-
called the Cpx system. It is induced by several tein aggregation may occur more frequently.
conditions, including elevated pH and the pro- Three important Hsp chaperone proteins
duction of misfolded envelope proteins. The that assist folding are DnaK (Hsp70 family),
membrane histidine kinase (HK) that is part of DnaJ (Hsp40 family), and GrpE, which act
this two-component system is CpxA, and the together and with two other Hsp chaperone
response regulator (RR) is CpxR. The Cpx sys- proteins, GroEL (Hsp60 family) and GroES
tem is also involved in controlling part of the (Hsp10 family), which also cooperate to pro-
σE regulon (DegP), as well as the expression of duce correct folding. (This is discussed in more
genes in the Cpx regulon. The genes in the Cpx detail in Section 11.2.12 and the referenced
regulon encode envelope folding factors. notes therein.) It appears that DnaK, DnaJ, and
GrpE can be ribosome associated and can bind
16.3.6 Functions of the E. coli to the protein as a team while the protein is still
heat-shock proteins attached to the ribosome and then, after the pro-
Four roles for the Hsps tein has left the ribosome, pass the protein into a
Let us now ask what the Hsps do in the cell. cavity of the multimeric GroEL. The final stages
Depending upon the particular protein, an of folding take place in the GroEL cavity, which
Hsp may function in the following way: is capped by GroES. The chaperone function of
GroEl–GroES is ATP dependent. (For a review, part, to unfold the protein prior to proteolysis.
see ref. 48. See note 57 for more information on One of these, Lon, is of major importance in
folding.) Chaperone proteins, such as GroEL, degrading damaged proteins.
exist as multisubunit complexes of stacked rings Lon is an ATP-dependent protease that is
called chaperonins. Apparently mutants that encoded by the capR (lon) gene.61 Like the other
make no σ32 grow at temperatures below 20 °C Hsps, Lon functions not only during the heat-
(albeit much more slowly than the wild type) shock response, but also during ordinary growth
because the recognition of weak promoters in temperatures, and it is of interest to examine the
the σ32 regulon by other sigma factors results in different roles that this protease plays in cells
some transcription of the gro and dna genes.41,49 regardless of the temperature. Lon is important
for degrading abnormal proteins: for example,
The Hsps that aid in protein export proteins that may result from nonsense or mis-
Some chaperone proteins seem to play a sense mutations, or denatured proteins in gen-
role in protein export similar to that of SecB, eral. In fact, it appears that Lon is responsible
a non-Hsp chaperone protein discussed in for the degradation of the majority of abnormal
Section 18.1.1. SecB is the major chaperone proteins in E. coli. In addition, several normal
protein in E. coli. It prevents folding and brings cellular proteins are also degraded by Lon, two
unfolded presecretory proteins to SecA and examples of which follow.
the translocation machinery at the membrane. Mutations in capR (lon) give rise to strains that
Some of the Hsps have been reported to function synthesize excess capsular polysaccharide and
in the export of some of the SecB-independent form mucoid colonies. This is because Lon rec-
proteins. The latter are proteins that use the Sec ognizes and degrades the RcsA protein, which is a
machinery but not SecB. For example, DnaK positive regulator of capsular polysaccharide syn-
and DnaJ are used for the export of some (not thesis. Hence, the lack of functional Lon results in
all) SecB-independent proteins. and DnaK and increased levels of RcsA and consequently large
DnaJ can substitute for SecB in strains lacking amounts of capsular polysaccharide accumulate.
SecB.58,59 Other unidentified Hsps can substi- Lon mutants have an additional phenotype. In the
tute for SecB in mutants that do not make SecB absence of Lon, cells are very sensitive to ultravio-
or have low amounts of available SecB. let radiation as a consequence of the UV-induced
SOS response discussed in Section 16.5. When
The Hsps that refold denatured proteins the SOS response is induced in lon mutants, cell
In Escherichia coli, an Hsp called ClpB is an division is inhibited, resulting in cell filamenta-
ATPase that is thought to disentangle thermally tion and death. Cell division is defective because
aggregated proteins and transfer them to the an inhibitor of cell septation called SulA accu-
DnaK–DnaJ–GrpE chaperone system. DnaK– mulates and is stable in UV-irradiated cells in the
DnaJ–GrpE chaperones and the GroEL-GroES absence of Lon. (SulA inhibits the polymerization
chaperonins cooperate to refold proteins that of FtsZ, a critical protein for septum formation
have misfolded under stress conditions. They and cell division. See Section 16.5.1 for the role
are aided by another chaperone, Hsp IbpB, of SulA in the SOS response.) SulA is another sub-
which stabilizes partially folded proteins until strate for Lon, accounting for why cells that do
they can be refolded. If the proteins are too dam- not make Lon accumulate SulA. In UV-irradiated
aged to be refolded, or if the folding chaperones wild-type cells, the inhibition of cell division is
are saturated, heat-shock proteases, described transient because Lon degrades the accumulated
next, will destroy the damaged proteins. SulA. Thus, the Lon protease can be viewed as
an enzyme that not only degrades abnormal pro-
The Hsps that are ATP-dependent proteases
teins but also helps to regulate the levels of certain
At least six of the Hsps are ATP-dependent
regulatory proteins.
proteases.48,60 Some of these proteases may func-
tion, in part, to degrade improperly folded or
denatured proteins at all growth temperatures. 16.4 Repairing Damaged DNA
Some also play a role in the degradation of spe- There are several ways in which bacteria repair
cific native proteins during ordinary growth damaged DNA. A more complete discussion of
temperatures. The ATP may be required, in this subject can be found in refs. 62 through 65.
www.ebook777.com
Fig. 16.5 Formation of thymine dimers by ultraviolet light. Ultraviolet light causes the formation of cyclobu-
tane pyrimidine dimers. Usually this occurs between adjacent thymine dimers in the same strand of DNA and
involves C5 and C6 of both molecules.
We will begin with the kinds of damage that can

occur.
16.4.1 Kinds of DNA damage

A particularly severe type of damage called pho-
todimerization of pyrimidines occurs when DNA
bases are excited by ultraviolet radiation (Fig.
16.5). As a consequence of the dimerization of
adjacent pyrimidines such as thymine, DNA
replication is blocked. Pyrimidine dimers are
removed by specific excision processes or photo-
reactivation, as described next. Other malformita-
tions of DNA can include mismatched base pairs
in the duplex that arise during DNA replication,
breaks or gaps in the DNA, and bases modified by
chemicals, many of which are carcinogens.
Fig. 16.6 Photoreactivation of thymine dimers.
16.4.2 Repairing UV-damaged DNA by Thymine dimers form in the presence of ultraviolet
photoreactivation light. Most organisms other than placental mammals
Exposure to ultraviolet radiation can cause the possess a flavoprotein enzyme called photolyase,
formation of thymine dimers (Fig. 16.6). (For which will reverse the dimerization upon absorp-
a review of this subject as well as other DNA tion of blue light (350–500 nm). DNA photolyase in
repair systems and mutagenesis, see ref. 64.) E. coli is encoded by the phr gene.
During photoreactivation, a special enzyme
called DNA photolyase, encoded by the phr primary chromophore is MTHF, which absorbs
gene in E. coli, reverses the dimerization reac- 60 to 80% of the light absorbed by the enzyme at
tion upon absorbing blue light (300–500 nm). 385 nm. The energy is transferred from MTHF
The enzyme contains noncovalently bound to the flavin. The excited flavin molecule cleaves
FADH2 (reduced flavin adenine dinucleotide) the pyrimidine dimer. (For more details of the
and, in E. coli, also a folic acid derivative model, see note 66.)
(5,10-methenyltetrahydrofolyl)polyglutamate DNA photolyase is widespread, occurring in
(MTHF), both of which absorb light. The both prokaryotic and eukaryotic microorgan-
enzyme uses the absorbed light energy to cleave isms, plants, and animals. However, it is not
the dimer into two monomers, thus restoring found in placental mammals. Purified DNA pho-
the original coding properties of the DNA. The tolyases from other organisms have FADH2 but
differ with respect to the second chromophore. segment of DNA, aided by UvrD, which is a heli-
Some, like E. coli, have MTHF, whereas oth- case (helicase II), leaves a single-stranded gap
ers have 8-hydroxy-5-deazaflavin, instead. The that is filled by DNA polymerase I and sealed by
folate-containing enzymes have an absorption DNA ligase (Fig. 16.7). The endonuclease has
maximum of about 380 nm, whereas the deaza- been shown in vitro to recognize other kinds
flavin enzymes have a slightly higher absorption of damage such as chemically modified bases
maximum (i.e., about 440 nm). (e.g., alkylated bases if they are bulky enough)
that cause distortion and similarly remove the
16.4.3 Repairing damaged DNA via damaged DNA. The role of the uvr genes was
nucleotide excision repair originally discovered when it was found that
In E. coli the products of three genes, uvrA, mutants in the uvr genes are very sensitive to
uvrB, and uvrC, encode UvrABC endonuclease, killing not only by ultraviolet light but also by
an enzyme that recognizes the distortion in the different chemicals that damage DNA.
double helix most commonly caused by the
UV-induced pyrimidine dimer. This endonu- 16.4.4 Repairing damaged DNA by
clease also cuts the DNA eight nucleotides away recombination
on the 59 side of the dimer and four nucleotides The excision mechanisms thus far discussed
away on the 39 side of the dimer. Removal of the require that one of the two so-called sister
Fig. 16.7 Nucleotide excision repair (NER). Certain kinds of DNA damage (e.g., thymine dimers) can be
repaired by excision repair. The distortion in the double helix is recognized by a complex of proteins called
UvrABC, an endonuclease that cuts the damaged DNA eight nucleotides away on the 59 side of a thymine
dimer and four nucleotides away on its 39 side. Removal of the segment of DNA, aided by the helicase UvrD,
leaves a single-stranded gap, which is filled with DNA polymerase I and sealed with DNA ligase.
www.ebook777.com
strands have no errors and be able to serve as the nuclease
template. When both strands are damaged, for

example, by gaps (discontinuities) in the daugh-
ter strand made during the copying of a damaged
template, RecA is used in the process referred to
as recombinational repair, or daughter strand
gap (DSG) repair. For example, the damage can A nuclease, RecA
occur when pyrimidine dimers form as a conse-
quence of UV irradiation and the region of DNA
containing the pyrimidine dimer is not replicated.
Replication begins again about 1,000 base pairs
past the damaged DNA, resulting in a single-
stranded gap. The gap is filled, but not by rep-
crossover intermediate forms
lication; rather a segment of the complementary B DNA polymerase 1,
strand with the correct sequence from the sister ligase
duplex moves into the gap strand exchange. (The

term “sister duplex” designates a pair of DNA
cut cut
strands comprising a single unit and resulting
from replication at the replication fork.)
In the model for the repair process outlined in
Fig. 16.8, when a gap is present in the daughter
C cutting and resolution
strand opposite a lesion in the template strand,
a RecA-dependent recombinational event takes
place, resulting in the transfer of the comple-
mentary strand from the sister duplex to the gap
in the daughter strand. Then a crossover inter-
mediate forms, which after cutting and sealing
results in filling the gap with the complementary
DNA from the sister strand. The resultant gap Fig. 16.8 Postreplication daughter strand gap
in the sister strand is filled by DNA polymerase (DSG) repair; the daughter strand is the strand that
I (Pol I). The gaps are sealed by DNA ligase. The is being copied from the template strand. The pres-
faulty nucleotide can now be removed by using ence of a structural deformation in the duplex (e.g.,
one imposed by the presence of a thymine dimer in
nucleotide excision repair (UvrABC) or sim-
the template strand) causes a gap in the daughter
ply diluted out during growth. Postreplicative
strand opposite the lesion. The gap results from
repair is generally error free, as opposed to SOS a collapse of the replication fork and a restart
mutagenesis described in Section 16.5.2. of DNA replication at an upstream site, and it is
repaired by RecA and other enzymes. (A) The good
template in the undamaged sister duplex is nicked
16.4.5 Repair of deaminated bases
and moves into the gap in the daughter strand of
(base-excision repair) the damaged sister duplex. This is referred to as
A common type of DNA damage is the deamina- sister strand exchange and requires nuclease activ-
tion of bases. As a consequence of deamination, ity and RecA. Some DNA replication may occur in
an amino group is replaced by a keto group. the daughter strand of the receiving duplex to fill
For example, cytosine might be deaminated to any gap that may be present between the 39 end of
uracil, adenine to hypoxanthine, and guanine the incoming strand and the 59 end of the strand
to xanthine. (See Fig. 10.8 for the structures with the gap. (B) A crossover intermediate forms.
of the bases.) The deaminated bases pair with (C) The crossover intermediate is cut and the DNA
the wrong base during replication, resulting in is resolved into two separate duplexes. During
the process, all breaks are sealed, resulting in the
mutations. For example, hypoxanthine pairs
incorporation of the sister strand into the gap in
with cytosine so that an AT pair is replaced by the daughter strand, and leaving a gap in the sis-
a GC pair. Also, uracil will pair with adenine, ter template strand. As the figure indicates, DNA
resulting in a GC pair being changed to an AT replication in the donor template strand probably
pair. Deamination can occur spontaneously, or begins before the crossover intermediate has been
it can be caused by various chemical reagents resolved.
(e.g., nitrous acid). Deaminated bases (as well these repair pathways are sometimes called spe-
as other damaged bases such as alkylated bases) cific repair pathways. After the base has been
are removed by specific enzymes called DNA removed, AP endonucleases cleave the phos-
glycosylases. (See note 67 for a discussion of phodiester linkage next to the AP site, usually
alkylation of bases.) The enzyme catalyzes the on the 59 side. This generates a free 39-hydroxyl
breakage of the N-glycosyl bond between the end, which is extended by DNA polymerase I (in
base and the sugar, leaving an apyrimidinic or E. coli) as the 59-exonuclease activity of DNA
apurinic site (Fig. 16.9). Such sites are referred polymerase I removes a portion of the damaged
to as AP sites. Each glycosylase is specific for strand ahead. After the damaged portion has
each type of damaged base that is removed. Thus been replaced, DNA ligase seals the nick.
16.4.6 The GO system and

protection from 8-oxoguanine
(7,8-dihydro-8-oxoguanine)
Reactive forms of oxygen such as superox-
ide radical, hydrogen peroxide, and hydroxyl
radicals can oxidatively damage guanine to
form 8-oxoguanine (GO). Such damage is
called a lesion. (See ref. 68 for a review and
Section 15.1 of this book, Oxygen Toxicity, for
a description of how reactive forms of oxygen
are made in the cell.) The structures of guanine
and 8-oxoguanine are shown in Fig. 16.10A.
Mutations will occur if 8-oxoguanine then
Fig. 16.10 Structures of guanine and 8-oxoguanine

Fig. 16.9 Base excision repair. (A) A damaged base (8-oxoG) and 8-oxoG-A base pairs (dR, deoxyriboside).
is removed by DNA glycosylase, leaving an AP site Oxygen free radicals can hydroxylate deoxyguanosine
(an apurinic or apyrimidinic site). (B) An AP endonu- at the C8 position. (A) One of the tautomeric forms is
clease cleaves the phosphodiester bond on the 59 side 8-oxodeoxyguanosine, which is the major one under
of the AP site. (C) DNA polymerase I extends the 39 physiological conditions. (B) Sometimes the DNA poly-
end while removing a portion of the damaged strand merase will insert dAMP rather than CMP opposite
with its 59-39-exonuclease activity. (D) The nick is 8-oxoG. Unless this error is repaired, when the DNA
sealed by DNA ligase. Newly synthesized DNA is replicates there will be a change of a GC base pair to an
shown as a rectangular box. AT base pair.
www.ebook777.com
mispairs with adenine (Fig. 16.10B), causing 8-oxoguanine can be removed. (See Fig.
GC pairs to be converted to AT pairs.69 The GO 16.10B.) When this happens, MutY, which is
system, however, prevents this from happen- an N-glycosylase, removes the adenine to pro-
ing. The GO system consists of three proteins: duce an AP site opposite the 8-oxoguanine (Fig.
MutM, MutT, and MutY. 16.11). The DNA is cut by an AP endonuclease
at the AP site (apurinic site) and the DNA is
MutM repaired as diagrammed in Fig. 16.11, yielding
MutM is an N-glycosylase that removes a C–8-oxoG–C base pair, which can be repaired
8-oxoguanine (Fig. 16.11). (See Section 16.4.5 by MutM. This avoids the formation of an AT
for a discussion of DNA glycosylases and AP base pair from a GC base pair.
site endonucleases.) After removal of the 8-ox-
oguanine and cutting of the strand, a portion of MutT
the damaged strand is removed by exonuclease Reactive forms of oxygen also convert the
activity and repaired by DNA polymerase I and guanine in dGTP to 8-oxoguanine to form
DNA ligase. 8-oxo-dGTP, which is a substrate for DNA
polymerase. MutT is a phosphatase that con-
MutY verts 8-oxo-dGTP to 8-oxo-dGMP, thus pre-
Occasionally, as a result of replication, venting the incorporation of 8-oxoguanylic
AMP is paired with 8-oxoguanine before the acid into DNA.
Fig. 16.11 Repair of 8-oxoguanine (GO) lesions. Oxidative damage results in the oxidation of guanine resi-
dues to 8-oxoguanine. The MutM protein removes the 8-oxoguanine and cuts the damaged DNA with its AP
endonuclease activity. The damaged DNA is removed, and the gap is repaired by DNA polymerase and DNA
ligase. Sometimes an AMP is inserted opposite the GO. When this happens, the MutY protein can remove
the adenine, leaving an apurinic (AP) site. An AP endonuclease activity, which appears to be associated with
MutY, cuts the damaged strand near the AP site. A portion of the damaged strand is removed, and the gap
is filled by DNA polymerase and sealed with DNA ligase. This generates a GO–C base pair, affording the
opportunity for MutM to remove the GO. The MutY protein can correct A/G mismatches as well as A/GO
mismatches.
Other Mut proteins repair the damaged DNA. We now turn to a dis-
Several other Mut proteins have been described. cussion of these genes and their products.
They are MutD, which is commonly known as What is the “SOS signal” that the bacteria
the ε subunit of DNA polymerase III holoen- use? The signal is generated when RecA binds
zyme, which is responsible for its 39-exonuclease to single-stranded DNA (ssDNA). This occurs
editing function, and MutS, MutL, and MutH, when ssDNA is produced at a stalled replica-
which function in mismatch repair. Mutations tion fork as the cell attempts to replicate dam-
in any of the mut genes increase the spontaneous aged DNA, or during infection with mutant
rate of mutation because they prevent the repair ssDNA phage that are defective in synthesis
of damaged DNA or the removal of incorrectly of the complementary (minus) strand.70 (RecA
paired nucleotides. They are called mut genes is also important for recombinational repair,
because of the mutator phenotype that results as discussed in Section 16.4.4.) When the sig-
from the mutations. nal is received, the SOS regulon is induced by
inactivating a repressor (LexA) of the SOS reg-
16.5 The SOS Response ulon. Precisely how the nucleoprotein com-
plex of ssDNA and RecA inactivates LexA is
The term “SOS,” used to describe the bacterial described next.
response to stress, alludes to the Morse code
signal, introduced at the beginning of the twen-
tieth century as a call for help to be transmitted 16.5.1 Regulation of the SOS response
via telegraph. (For more about the interest- Role of RecA and LexA
ing history of the Morse code and “SOS,” see We will begin with the transcription repressor
Box 16.1.) In the context of genetics, the use protein LexA, which interferes with the bind-
of “SOS” indicates that a cell is responding to ing of RNA polymerase by binding to LexA-
a distress signal generated by DNA damage. binding sites in the SOS box, located near the
When DNA is damaged or DNA replication operator of each SOS gene or operon in the SOS
is inhibited, a signal is produced that results in regulon (Fig. 16.12). In an undamaged cell LexA
the induction of over 20 unlinked genes, some represses the transcription of over 20 genes in
of which were discussed earlier in this chapter. the SOS regulon, including recA and lexA.
The products of the induced genes perform vari- When damage occurs to the DNA, RecA
ous functions to cope with the DNA damage, becomes activated (RecA*), and this results
including repairing the damaged DNA, allow- in the inactivation of LexA (Fig. 16.12). The
ing DNA replication to proceed past the dam- sequence proceeds in the following way. RecA
aged site (a process called translesion synthesis), becomes activated to RecA* when it forms a
and stalling cell division to give the cell time to nucleoprotein filament by polymerizing on
BOX 16.1 HISTORICAL PERSPECTIVE: THE MORSE CODE

The Morse code is named after Samuel numbers. The code phrase SOS (in Morse
Finley Breese Morse (1791–1872), an code: . . . _ _ _ . . . ) was first used in 1910.
American painter and inventor. In 1838 he The letters were arbitrarily chosen as an
refined one of his inventions, the telegraph, easy way to transmit via the Morse code
which he patented in 1854. He also devel- a call for help. They are not, as commonly
oped the telegraphic code that is named thought, an acronym for “save our ship” or
after him. “save our souls.”
In the Morse code messages are trans-
mitted by means of sequences of dots and Source: Barnhart, R. K. 1995. The Barnhart
dashes, or short and long sounds or flashes Concise Dictionary of Etymology. Harper-
that represent letters of the alphabet and Collins, New York.
www.ebook777.com
Fig. 16.12 Regulation of the SOS response. The model proposes that RecA becomes activated when it binds
to single-stranded DNA resulting from replicative bypass of damaged DNA. Activated RecA (i.e., RecA*)
leads to the inactivation of LexA by autoproteolytic cleavage. LexA is a repressor for all the genes induced
during the SOS response. There are over 20 genes regulated by LexA. Some of these genes are shown: recA,
for example, encodes the RecA protein; sulA inhibits FtsZ and thus cell division. UvrA, UvrB, and UvrD func-
tion during excision repair. PolB is DNA polymerase II. The gene umuDC encodes proteins that comprise the
subunits of Pol V, which copies damaged DNA, namely, UmuD9 (two subunits) and UmuC (one subunit).
single-stranded DNA, which can result as a con- Table 16.2 Some SOS responses and induced genes
sequence of an imperfect attempt to replicate a of E. coli
damaged DNA template. LexA, a transcription Induced gene Response
repressor, then binds to the RecA*–DNA com-
plex, and RecA* promotes the autoproteolytic umuDC, recA UV mutagenesis of
cleavage of LexA at the Ala84–Gly85 bond, thus bacterial chromosome
inactivating the repressor. The genes in the SOS sulA Inhibition of cell division
(filamentation)
regulon include genes for excision repair and
uvrA, uvrB, uvrD uvr-Dependent excision
recombinational repair, as well as genes that repair
allow damaged DNA to be copied (Table 16.2). recA, ruvAB Daughter strand gap
Some of these genes are described next. repair
recA, recN, ruvAB Double-strand break
Genes whose transcription is stimulated repair
during SOS response Source: Adapted from Walker, G. C. 1996. The SOS res-
Some of the genes activated during the SOS ponse of Escherichia coli, pp. 1400–1416. In: Escherichia
coli and Salmonella: Cellular and Molecular Biology, Vol. 1.
response as well as the physiological roles for the
F. C. Neidhardt et al. (Eds.). ASM Press, Washington, DC.
gene products are listed in Table 16.2; several
were discussed earlier. The list does not include
uvrC or phr (the gene that encodes the photore- has been accomplished. The levels of SulA
activating enzyme) because they are not under increase during the SOS response, but they are
LexA control. The product of the uvrD gene quickly reduced by the action of Lon protease
(helicase II) is not actually required for endonu- when the DNA is repaired, at which time cell
clease activity but is required for the release of division resumes (Section 16.3.6). Presumably,
the oligonucleotide fragment and of UvrC from it is an advantage to the cell not to divide while
the complex after the cuts in the DNA have been it is busy repairing its DNA.
made. (These genes are involved in nucleotide The SOS response also includes the induction
excision repair. See Section 16.4.3.) The list also of certain prophages such as lambda. This is
includes sulA, which, as discussed earlier, codes because RecA* promotes the cleavage of repres-
for a protein that inhibits septation and causes sors of these bacteriophages in a fashion similar
the cells to grow as filaments until DNA repair to the induced proteolysis of LexA. Prophage
induction cannot be a benefit to the damaged sites, thymine dimers, and photoproducts. In
host, but it benefits the virus in that after induc- view of its catalytic activity, Umu D92 C is now
tion, the virus can find a host whose DNA is not called Pol V. Compared with the normal repli-
damaged. Other genes induced include recA, cative DNA polymerase, Pol III, Pol V is error
umuC, and umuD. As we shall see, the products prone. This is because Pol V does not have a
of these genes allow DNA replication past sites stringent proofreading capability, which is
of DNA lesions, but at the same time increase why it can insert nucleotides nonspecifically
the frequency of mutations. opposite DNA lesions; moreover, it does have
a lower base-pairing fidelity, which is why it
accepts damaged bases in its active site. Thus,
16.5.2 SOS-induced translesion synthesis
mutations can occur opposite the lesion as well
and mutagenesis
as downstream of the lesion as Pol V continues
For a review, see ref. 64. Replication of the replication. In addition to Pol V, translesion
DNA past a lesion in the template strand, such synthesis requires RecA and single-stranded
as an abasic site or a UV-induced thymine binding protein. It may be that Pol III also inter-
dimer, is called translesion synthesis (TLS). acts with Pol V and plays a role at the replication
TLS cannot be performed by the replicative fork during translesion synthesis, as reviewed in
DNA polymerase III (DNA Pol III), which stalls ref. 64.
at such sites. A new polymerase, called DNA E. coli does not rely solely on Pol V to help it
Pol V, is called upon for translesion synthesis. replicate DNA that has been damaged. In fact,
Translesion synthesis increases the survival of there are two other SOS-induced DNA poly-
the damaged cell, but as we shall soon learn, merases (Pol II and Pol IV) that can be called
also increases the rate of mutation, since DNA upon to replicate DNA opposite lesions at a
Pol V makes errors as nucleotides are inserted damaged fork. These are discussed in note 71,
nonspecifically opposite lesions in the template and reviewed in refs. 63 and 72.
strand. The process of making mutations dur-
ing translesion synthesis is referred to as SOS
mutagenesis or error-prone repair. As we shall 16.6 Oxidative Stress
see, the catalytic unit of DNA Pol V is encoded 16.6.1 Toxic forms of oxygen
by the umuC gene, one of the genes in the SOS For a review, see ref. 73. When cells are exposed
regulon. to toxic forms of oxygen, they are said to be
Three genes are required for translesion subject to oxidative stress.74 As discussed ear-
synthesis. They are recA, umuD, and umuC. lier, aerobic organisms form several deriva-
The latter two are in the umuDC operon. For tives of oxygen that can be toxic to cells. (See
translesion synthesis to to occur, RecA must be Section 15.1 for an explanation of how toxic
activated. As mentioned previously, the activa- forms of oxygen originate inside cells.) The
.
tion of RecA occurs when the protein binds to toxic forms include hydroxyl radical (HO ),
–
regions of single-stranded DNA generated when superoxide radical (O 2), and hydrogen perox-
Pol III attempts to replicate the damaged DNA ide (H2O2), which can damage molecules such
template. The RecA/ssDNA nucleoprotein is as DNA, proteins, and lipids. Since electrons
the signal that the DNA is damaged. UmuD are passed to oxygen one at a time via the cyto-
interacts with the RecA nucleoprotein, which chromes, the superoxide radical is an interme-
activates the autocleavage of UmuD to an active diate in the reduction of oxygen to water. The
form, called UmuD9. (Recall that RecA/ssDNA hydrogen peroxide is derived from superoxide
also activates the autocleavage of LexA.) and can react with transition metal ions to form
The finding that UmuC can catalyze transle- hydroxyl radicals. (See the discussion of the
sion synthesis in vitro in the absence of DNA Fenton reaction in Section 15.1.)
Pol III led to the following model to account There does not exist a mechanism to detox-
for translesion synthesis and SOS mutagen- ify hydroxyl radicals, and thus protection from
esis (reviewed in ref. 64). UmuC, which when toxic forms of oxygen must be obtained by
complexed with an activator, consisting of two eliminating superoxide and hydrogen perox-
UmuD9, subunits, replicates DNA past DNA ide. The enzyme superoxide dismutase (Sod)
lesions in the template strand, such as a basic converts superoxide to hydrogen peroxide
www.ebook777.com
and oxygen, and the enzyme catalase con- 16.6.3 Transcriptional regulation of
verts hydrogen peroxide to water and oxygen. genes induced by oxidative stress
(See Section 15.1 for the reactions that these Two transcriptional regulatory systems in
enzymes catalyze.) It is of interest that the E. coli control the production of proteins in
reactive forms of oxygen as well as nitric oxide response to oxidative stress: the SoxRS system
(which also induces the superoxide-induced and the OxyR system, which control the soxRS
stress response, discussed next) are also pro- and oxyR regulons, respectively.75 These two
duced in animals by phagocytic cells (e.g., regulatory pathways control the expression of
activated macrophages and neutrophils) as genes that encode antioxidant enzymes. For
cytotoxic agents to kill bacteria that have been a review of the OxyR and SoxRS systems, see
ingested by phagocytosis. refs. 76 and 77. We will begin with SoxRS,
which includes SoxR, a transcriptional activa-
16.6.2 Proteins made in response to tor of soxS.
oxidative stress
To study the response to oxidative stress, one can SoxR and SoxS
add to cultures of bacteria H2O2, or redox-cycling SoxR is an iron–sulfide protein that is acti-
reagents such as methyl viologen (paraquat) or vated (probably by undergoing a conforma-
phenazine methosulfate. The redox-cycling tional change) in response to superoxide,
reagents accept from NADH or NADPH a sin- which causes a redox reaction in the iron–
gle electron, which they transfer directly to O2 to sulfide centers, or to nitric oxide (NO), which
.
form the superoxide radical, O2 . This results in causes nitrosylation. For a more complete
an increase in the synthesis of many proteins that explanation, see note 78. Activated SoxR acti-
can be detected by two-dimensional gel electro- vates transcription of the soxS gene. The SoxS
phoresis. The functions of many of the induced protein then activates the transcription of tar-
proteins are not known. However, some of the get genes whose products confer resistance
proteins have an obvious role in protecting the to a diverse array of toxic elements including
cell from oxidative damage. For example, in E. superoxide, nitric oxide produced by activated
coli the induced enzymes include Mn2+– super- macrophages, multiple antibiotics, organic
oxide dismutase, catalase (HPI catalase), and solvents, and heavy metals (Fig. 16.13). The
endonuclease IV. The latter functions in exci- activated genes include sodA (codes for Mn2+–
sion repair of damaged DNA, including damage superoxide dismutase) and nfo (codes for
caused by superoxide. (Endonuclease IV cuts the the DNA repair enzyme, endonuclease IV).79
DNA at apurinic or apyrimidine sites produced (SoxS autoregulates its own synthesis by being
by DNA glycosylases, which remove damaged a negative transcription regulator of the gene
bases from DNA.) that encodes it, soxS.)
Fig. 16.13 The SoxRS system. SoxR is activated by oxidative stress and possibly other environmental stress
factors. Activated SoxR stimulates the transcription of the soxS gene. The SoxS protein stimulates transcrip-
tion of a set of genes that aid the bacterium in coping with a toxic environment. Source: Hidalgo, E., H. Ding,
and B. Demple. 1997. Redox signal transduction via iron–sulfur clusters in the SoxR transcription activator.
Trends Biochem. Sci. 22:207–210. Copyright 1997, with permission from Elsevier.
OxyR and phosphoadenosine phosphosulfate (PAPS)

When cells are exposed to H2O2, OxyR becomes reductase, which is important for assimilatory
activated and as a consequence binds to target sulfate reduction (Section 13.1).
promoters. OxyR is also activated in vitro by The thioredoxins and glutaredoxins have
O2. OxyR is the transcriptional activator of pairs of cysteine residues separated by two
several genes including katG, which codes for a amino acids, Cys-X1-X2-Cys. The cysteines
catalase. (Catalases catalyze the transfer of two donate electrons to disulfide residues in other
electrons from one molecule of hydrogen per- proteins, producing sulfhydryl groups in the
oxide to another, resulting in the formation of target proteins; in turn, their own cysteine resi-
one molecule of oxygen formed by the electron dues are oxidized to cystine disulfides. They
donor, and two molecules of water, formed by are sometimes called thiol–disulfide exchange
the electron acceptor. See Section 15.1.) OxyR proteins. Reduced thioredoxin is regenerated
activates transcription, apparently through by NADPH via a reductase, and reduced glu-
contacts with one of the subunits of RNA poly- taredoxin is regenerated by reduced glutathi-
merase. There is no increase in the level of OxyR one (GSH). (Actinomycetes do not contain
protein, simply in its activity. OxyR is a homo- glutathione. See note 81.) Glutathione is a trip-
tetramer, and the activated form has an intra- eptide containing cysteine (γ-glutamyl-cysteine-
molecular disulfide bond between two cysteine glycine). The oxidized form of glutathione is
residues, forming a cystine disulfide that results glutathione disulfide (GSSG), which consists
in a conformational change in the protein. The of two glutathione molecules linked by a cys-
disulfide bond forms when the two cysteine tine disulfide. Glutathione is regenerated from
residues are oxidized by H2O2 or another oxi- GSSG via an NADPH reduction catalyzed by
dant. A repressor of photosynthetic genes in the enzyme glutathione reductase.
bacteria, CrtJ, is activated by the formation of In summary then, NADPH regenerates both
an intramolecular disulfide bond in the pres- reduced thioredoxin and reduced glutaredoxin,
ence of oxygen. In addition, the transcription of and the reduced thioredoxin and reduced glu-
the oxyS gene is activated by OxyR. The RNA taredoxin in turn regenerate reduced forms of
product of the oxyS gene is not translated, and other disulfide proteins (e.g., OxyR). See the
its function is not understood at present. It has discussion in Section 18.6.2 of thiol–redox
been called an antimutator because it protects enzymes in the periplasm.
against mutagenesis.
16.7 Summary
The role of thioredoxins and glutaredoxins in Bacteria must maintain a fairly constant inter-
reducing disulfide bonds in OxyR and other nal pH and adjust their osmolarity in response
proteins in the cytoplasm to the external osmolarity. Very little is known
The student is urged to consult the review in ref. about the details of how this is done. In broad
80 for a description of the very interesting area outline, however, the internal pH is understood
of research into thiol–redox enzymes in bacteria to be adjusted by using proton pumps to extrude
and other organisms. OxyR is only transiently protons and antiporters that bring protons
activated by the formation of disulfide bonds into the cell in exchange for sodium or potas-
during oxidative stress. There exist small thiol– sium ions. What is not clear is precisely how the
redox enzymes in bacteria, called thioredoxins activities of the pumps and antiporters are regu-
and glutaredoxins, whose function is to reduce lated by the external pH, although it is reason-
disulfide bonds in other proteins (e.g., OxyR) to able to suggest that they respond to the internal
sulfhydryl groups. When this occurs to OxyR, pH, by some sort of feedback mechanism. The
the form of the protein that senses oxidative acidophilic iron-oxidizing bacteria represent a
stress is regenerated. Other proteins whose di- special problem in that there is very little energy
sulfide bonds are reduced by thioredoxin and obtainable by Fe2+ oxidation to pump protons
glutaredoxin are various reductases such as ribo- out of the cell. These bacteria maintain a ∆pH
nucleotide reductase, which generates deoxy- by consuming cytoplasmic protons during oxy-
nucleotides for DNA synthesis (Section 10.2.5), gen respiration.
www.ebook777.com
Bacteria must maintain an osmotic differ- rates of synthesis of proteins called heat-shock
ential across the cell membrane because the proteins. In E. coli, the increased rates of synthe-
resulting turgor pressure is essential for growth sis are due to increased activity of a sigma fac-
of the cell wall and for cell division. When the tor, σ32. Although the biological roles for most
turgor pressure is suddenly decreased, a variety of these proteins are not known, it is clear that
of physiological activities come to a halt (e.g., some of them aid bacteria in coping with the
nutrient uptake, DNA synthesis). When the increase in temperature. The heat-shock pro-
external osmotic pressure is increased (causing teins include chaperone proteins that aid in the
decreased turgor), bacteria respond by increas- folding of newly synthesized proteins and also
ing the concentration of internal osmolytes to function in the export of certain proteins, and
raise the internal osmotic pressure and turgor. ATP-dependent proteases that degrade specific
An important molecule in this regard is potas- proteins and abnormal proteins. Several of the
sium ion, which increases, in some cases tran- heat-shock proteins have been shown to play a
siently, in response to an increase in external role during growth at ordinary temperatures.
osmolarity. A counterion must also increase, In addition to σ32, σE is important in respond-
and in E. coli this is glutamate. In E. coli, the K+ ing to stress, including heat stress. The σE regu-
is subsequently replaced by trehalose. If proline lon encodes proteins responsible for the folding,
or betaine is in the medium, it replaces the K+ refolding, or degradation of misfolded envelope
and trehalose. proteins.
It is clear that transcriptional changes occur DNA can be damaged in several differ-
when bacteria are shifted into media of different ways, and various systems exist to repair
ent osmolarities. For example, when E. coli is the damage. There are specific repair systems
grown in high-osmolarity media, the transcrip- called base excision repair (BER) systems that
tion of several genes is increased. These include use N-glycosylases to remove damaged bases
proU, the operon that codes for the uptake of and AP endonucleases, DNA polymerase I,
two osmoprotectants, proline and betaine, and DNA ligase to repair the DNA. Bases can
and kdp, the operon that codes for the proteins be damaged in several ways, including deami-
required for the uptake of K+. Also, when E. coli nation and alkylation. Guanine can also be
and S. typhimurium are grown under condi- oxidatively damaged. The cell protects itself
tions of high external osmolarity, the gene for from oxidative guanine damage by using the
the OmpC porin is activated, whereas the gene MutM, MutY, and MutT proteins. MutM and
for the OmpF porin is repressed. This leads to MutY are N-glycosylases and function in base
more OmpC and less OmpF. A protein in the excision systems. However, MutT is not an
cell membrane called EnvZ has a periplasmic N-glycosylase; rather, it works by preventing
domain and is thought to be an osmosensor the incorporation of oxidized guanylic acid into
(Chapter 19). DNA.
There is also osmoregulation of the periplasm Pyrimidine dimers can be produced by
in gram-negative bacteria. Gram-negative bac- UV irradiation, and the dimerization can be
teria adapt to media of low osmolarity by syn- reversed by photoreactivation. Pyrimidine dim-
thesizing membrane-derived oligosaccharides ers can also be removed by a nucleotide excision
in the periplasm. The idea is that the multiple- repair (NER) system.
charged anionic MDO molecules accumulate There are also general repair systems. These
cations in the periplasm, thus raising its osmo- include nucleotide excision repair, postrepli-
larity. However, mutants of E. coli defective cation recombination repair, and mismatch
in MDO synthesis show no growth defects in repair.
low-osmolarity media. It must be concluded Bacteria respond to damaged DNA by induc-
that other mechanisms of osmoregulation of ing the genes of the SOS system. The protein
the periplasm are present and that osmoregula- RecA becomes activated and leads to the inacti-
tion of the periplasm is not well understood at vation of a repressor, LexA. As a consequence,
this time. the transcription of the genes of the SOS reg-
Upon a temperature upshift, bacteria, as well ulon is stimulated. The protein products of
as other organisms, transiently increase the some of these genes effect repairs by nucleotide
excision, postreplication recombination repair, 8. Discuss the role of Lon in UV resistance

and translesion synthesis (TLS), which refers to and in polysaccharide synthesis.
replication of the DNA past a lesion in the tem-
9. Suppose a gap occurs in a daughter DNA
plate strand.
strand during replication. How might it
Oxidative stress is due to the production of
be repaired?
toxic forms of oxygen that damage cell mole-
cules. The transcription of several genes whose 10. Describe two different pathways that can
protein products protect bacteria from oxida- remove or correct thymine dimer damage
tive stress is stimulated by oxidative stress. In in the template strand.
E. coli, these proteins include superoxide dis-
11. Mutations in mutM, mutY, and mutT
mutase, catalase, and endonuclease IV. There
increase the spontaneous rate of mutation.
are two transcriptional regulatory systems that
Explain why this is the case. The increase
control the expression of genes induced by oxi-
in the mutation rates are additive; that is,
dative stress. These are the SoxRS and the OxyR
the rates are higher when two or three of
systems, which consist of transcriptional acti-
the mut genes are mutated than when only
vators that are activated by superoxide, nitric
one of the genes is mutated. Explain why
oxide, and hydrogen peroxide.
this is the case.
Study Questions 12. What is meant by the SOS response? How

are the genes regulated?
1. Na+/H+ and K+/H+ antiporters are an impor- 13. What is the explanation for SOS
tant way to lower the intracellular pH. What mutagenesis?
is the evidence for this in alkaliphiles?
14. What is meant by oxidative stress, and
2. No one knows for sure how acidophiles how does E. coli respond to it?
establish an inverted membrane potential.
How might an inverted membrane poten- REFERENCES AND NOTES
tial affect pH homeostasis? What experi-
ments would support the hypothesis that 1. Padan, E., D. Zilberstein, and S. Schuldiner. 1981.
an electrogenic K+ pump was responsible pH homeostasis in bacteria. Biochim. Biophys. Acta
for the inverted membrane potential in 650:151–166.
acidophiles? 2. Bakker, E. P. 1990. The role of alkali–cation
transport in energy coupling of neutrophilic and aci-
3. Upon shifting E. coli to a higher osmolar- dophilic bacteria: an assessment of methods and con-
ity, there is a transient uptake of K+. What is cepts. FEMS Microbiol. Rev. 75:319–334.
the role of K+ in this regard? What keeps the 3. Krulwich, T. A., A. A. Guffanti, and D. Seto-
cytoplasm neutral? How is the membrane Young. 1990. pH homeostasis and bioenergetic work
potential maintained? Is there a complica- in alkaliphiles. FEMS Microbiol. Rev. 75:271–278.
tion with the ∆pH? Describe the role of K+ 4. Matin, A. 1990. Keeping a neutral cytoplasm;
in pH homeostasis. the bioenergetics of obligate acidophiles. FEMS
4. What is meant by turgor pressure? What is 5. Krulwich, T. A., and D. M. Ivey. 1990.
the evidence that turgor pressure is neces- Bioenergetics in extreme environments, pp. 417–
sary for cell growth? 447. In: The Bacteria, Vol. XII. T. A. Krulwich (Ed.).
Academic Press, San Diego, CA.
5. What is meant by periplasmic osmotic
6. Booth, I. R. 1985. Regulation of cytoplasmic pH
homeostasis? What might be the role of
in bacteria. Microbiol. Rev. 49:359–378.
MDOs?
7. Harold, F. M., E. Pavlasova, and J. R. Baarda.
6. What is meant by heat-shock proteins, and 1970. A transmembrane pH gradient in Streptococcus
what is the evidence that they are required faecalis: origin, and dissipation by proton conduc-
for growth? tors and N, N´-dicyclohexylcarbodiimide. Biochim.
Biophys. Acta 196:235–244.
7. Which heat-shock proteins are required for 8. Ivey, D. M., M. Ito, R. Gilmour, J. Zemsky, A.
growth at all temperatures? Why is that? A. Guffanti, M. G. Sturr, D. B. Hicks, and T. A.
www.ebook777.com
Krulwich. 1998. Alkaliphile bioenergetics, pp. 181– 21. Dinnbier, U., E. Limpinsel, R. Schmid, and E. P.
210. In: Extremophiles: Microbial Life in Extreme Bakker. 1988. Transient accumulation of potassium
Environments. K. Horikoshi and W. D. Grant (Eds.). glutamate and its replacement by trehalose during
John Wiley & Sons, New York. adaptation of growing cells of Escherichia coli K-12
to elevated sodium chloride concentrations. Arch.
9. Careful measurements have been made of the ∆pH
Microbiol. 150:348–357.
of the facultative alkaliphile Bacillus firmus to show
that the bacteria are capable of maintaining a ∆pH of 22. Epstein, W. 1986. Osmoregulation by potas-
up to 2 units as the external pH is raised. When pHout sium transport in Escherichia coli. FEMS Microbiol.
was 7.5, pHin was also 7.5. When the external pH Rev. 39:73–78.
was increased from 7.5 to 10.5, the cytoplasmic pH
23. McLaggan, D., T. M. Logan, D. G. Lynn, and W.
increased only to 8.2. At an external pH of 11.4, the
Epstein. 1990. Involvement of γ-glutamyl peptides
cytoplasmic pH was only 9.5. Growth rates did not
in osmoadaptation of Escherichia coli. J. Bacteriol.
begin to slow until the external pH was greater than
172:3631–3636.
10.6. For a review, see ref. 8.
24. Betaine is synthesized by oxidizing the hydroxyl
10. Csonka, L. N., and A. D. Hanson. 1991.
group in choline to a carboxyl group, that is:
Prokaryotic osmoregulation: genetics and physiol-
ogy. Annu. Rev. Microbiol. 45:569–606. (CH3)3N+–CH2CH2OH (choline) →
11. Csonka, L. N. 1989. Physiological and genetic (CH3)3N+-CH2COOH (betaine)
responses of bacteria to osmotic stress. Microbiol.
Betaine and glycinebetaine (N,N,N-trimethylglycine)
Rev. 53:121–147.
are different names for the same molecule.
12. A derivation of eq. 16.3 can be found in the
25. Hengge-Aronis, R. 1996. Back to log phase: σs
review article by Csonka (ref. 11).
as a global regulator in the osmotic control of gene
13. Ingraham, J. L. 1986. Effect of temperature, expression in Escherichia coli. Mol. Microbiol.
pH, water activity and pressure on growth, pp. 21:887–893.
1543–1554. In: Escherichia coli and Salmonella
26. Townsend, D. E., and B. J. Wilkinson. 1992.
typhimurium: Cellular and Molecular Biology,
Proline transport in Staphylococcus aureus: a high-
Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press,
affinity system and a low-affinity system involved in
Washington, DC.
osmoregulation. J. Bacteriol. 174:2702–2710.
14. Koch, A. L., and M. F. S. Pinette. 1987.
27. Nikaido, H., and M. Vaara. 1985. Molecular
Nephelometric determination of turgor pressure
basis of bacterial outer membrane permeability.
in growing gram-negative bacteria. J. Bacteriol.
169:3654–3668.
28. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989.
15. Walsby, A. E. 1986. The pressure relationships
Protein phosphorylation and regulation of adap-
of halophilic and non-halophilic prokaryotic cells
tive responses in bacteria. Microbiol. Rev. 53:450–
determined by using gas vesicles as pressure probes.
490.
FEMS Microbiol. Rev. 39:45–49.
29. Bakker, E. P., I. R. Booth, U. Dinnbier, W.
16. Koch, A. L. 1991. Effective growth by the simplest
Epstein, and A. Gajewska. 1987. Evidence for mul-
means: the bacterial way. ASM News 57:633–637.
tiple potassium export systems in Escherichia coli. J.
17. Glutamate and K+ are present in high concentra- Bacteriol. 169:3743–3749.
tions in most (eu)bacteria. The former is the major
30. Stokes, N. R., H. D. Murray, C. Subramaniam,
organic anion, and the latter is the major inorganic
R. L. Gourse, P. Louis, W. Bartlett, S. Miller, and I. R.
cation. Glutamate is generally synthesized rather
Booth. 2003. A role for mechanosensitive channels
than accumulated from the medium, using either
in survival of stationary phase: regulation of chan-
glutamate dehydrogenase or glutamate synthase
nel expression by RpoS. Proc. Natl. Acad. Sci. USA
discussed in Section 10.3.1. In E. coli, potassium
100:15959–15964.
ion is taken up by four different transport systems,
including the Kdp and Trk systems discussed later 31. Booth, I. R., and P. Louis. 1999. Managing
(Section 17.3.3). Potassium ion efflux is catalyzed by hypoosmotic stress: aquaporins and mechano-
two systems, the KefB and KefC transporters. sensitive channels in Escherichia coli. Curr. Opin.
18. Galinski, E. A. 1995. Osmoadaption in bacteria.
Adv. Microb. Physiol. 37:273–328. 32. The volume of the periplasm in Escherichia and
Salmonella has been reported by some investigators
19. Lanyi, J. K. 1974. Salt-dependent properties
to be as high as 20 to 40% of the total cell volume and
of proteins from extremely halophilic bacteria.
by others as low as 5% of the total cell volume. One
Bacteriol. Rev. 38:272–290.
must conclude that there is some uncertainty regard-
20. Vreeland, R. H., and L. I. Hochstein (Eds.). 1992. ing the size of the periplasm and that, furthermore,
The Biology of Halophilic Bacteria. CRC Press, Boca the periplasm of different types of bacteria may vary
Raton, FL. significantly in physical dimensions.
2
33. Stock, J. B., B. Rauch, and S. Roseman. 1977. [K+]1([K+]1 – [R–]1) = K+ 2 (e)
Periplasmic space in Salmonella typhimurium and
Escherichia coli. J. Biol. Chem. 252:7850–7861. and rearranging, we have
2 2
34. Miller, K. J., E. P. Kennedy, and V. N. Reinhold. [K+ ]2 – [K+]1 [R–]1 = [K+ ]2
1986. Osmotic adaptation by gram-negative bacte-
ria: possible role for periplasmic oligosaccharides. Thus, the concentration of K+ on the side without R–
Science 231:48–51. (side 2) is less than the concentration on the side with
R– (side 1) by [K+]1[R–]1. Note that if R– is multiva-
35. Weissborn, A. C., M. K. Rumley, and E. P. lent, then even more K+ will accumulate because each
Kennedy. 1992. Isolation and characterization of negative charge on R– is balanced by a K+.
Escherichia coli mutants blocked in production of
38. Kennedy, E. P. 1982. Osmotic regulation and
membrane-derived oligosaccharides. J. Bacteriol.
the biosynthesis of membrane-derived oligosaccha-
174:4856–4859.
rides in Escherichia coli. Proc. Natl. Acad. Sci. USA
36. Kennedy, E. P. 1996. Membrane-derived oligo- 79:1092–1095.
saccharides (periplasmic β-D-glucans) of Escherichia
39. The difficulty in reconciling a periplasm iso-
coli. pp. 1064–1071. In: Escherichia coli and osmotic with the cytoplasm with the fragility of the
Salmonella: Cellular and Molecular Biology. F. C. outer membrane was suggested by Arthur Koch.
Neidhardt et al. (Eds.). ASM Press, Washington,
DC. 40. Guisbert, E., T., Yura, V. A. Rhodius, and C. A.
Gross. 2008. Convergence of molecular, modeling,
37. Whenever a solution of (1) a nonpermeant ion and systems approaches for and understanding of
is separated from another solution (2) (e.g., by a the Escherichia coli heat shock response. Microbiol.
membrane permeable to other ions), there will be Mol. Biol. Rev. 72:545–554.
an unequal distribution of ions at equilibrium. The
compartment with the nonpermeant ion will contain 41. Yura, T., M. Kanemori, and M. T. Morita.
a higher concentration of permeant ions and hence a 2000. The heat shock response: regulation and
higher osmotic pressure (or lower osmotic potential). function, pp. 3–18. In: Bacterial Stress Responses.
For the case of the periplasm and the outer membrane, G. Storz and R. Hengge-Aronis (Eds.). ASM Press,
suppose the nondiffusible oligosaccharide is R– and Washington, DC.
the diffusible ions are K+ and Cl–. Then K+ will pas- 42. Phadtare, S., K. Yamanaka, and M. Inouye.
sively accumulate in the periplasm as the counterion 2000. The cold shock response, pp. 33–45. In:
to R–. At equilibrium, the concentration of K+ will be Bacterial Stress Responses. G. Storz and R. Hengge-
greater on the same side of the membrane as R– (the Aronis (Eds.). ASM Press, Washington, DC.
periplasm), whereas the concentration of Cl– will
be greater on the other side (the medium). Because 43. Graumann, P., and M. A. Marahiel. 1996. Some
the K+ can diffuse across the outer membrane and R– like it cold: response of microorganisms to cold
cannot, a voltage potential, outside positive, develops shock. Arch. Microbiol. 166:293–300.
across the membrane. This is the Donnan potential, 44. Graumann, P., T. M. Wendrich, M. H. W.
which is really a K+ diffusion potential. The following Weber, K. Schröder, and M. A. Marahiel. 1997. A
equations explain the relationships between the dif- family of cold shock proteins in Bacillus subtilis is
fusible and nondiffusible ions. At equilibrium, essential for cellular growth and for efficient protein
[K+]1 = [R–]1 + [Cl–]1 (a) synthesis at optimal and low temperatures. Mol.
and
45. Zhou, Y.-N., K. Noriko, J. W. Erickson, C. A.
[K+]2 = [Cl–]2 (b) Gross, and T. Yura. 1988. Isolation and charac-
Equations a and b describe electrical neutrality in the terization of Escherichia coli mutants that lack the
two solutions. heat-shock sigma factor σ32. J. Bacteriol. 170:3640–
Because almost all the K+ that diffuses across 3649.
the membrane must be coupled with Cl– [otherwise 46. Hecker, M., W. Schumann, and U. Völker. 1996.
the diffusion potentials would prevent further cat- Heat-shock and general stress response in Bacillus
ion (or anion) flow], the diffusion kinetics can be subtilis. Mol. Microbiol. 19:417–428.
described by a second-order rate equation
47. Kanemori, M., K. Nishihara, H. Yanagi, and T.
k[K+]1[Cl–]1 = k[K+]2[Cl–]2 Yura. 1997. Synergistic roles of Hs1VU and other
ATP-dependent proteases in controlling in vivo turn-
where k is the rate constant for diffusion through the
over of σ32 and abnormal proteins in Escherichia coli.
membrane. Since, however, k cancels out, we have
J. Bacteriol. 179:7219–7225.
[K+]1[Cl–]1 = [K+]2[Cl–]2 (c)
48. Gross, C. A. 1996. Function and regulation of the
+ –
But, eq. b states that [K ]2 = [Cl ]2, therefore, heat-shock proteins, pp. 1382–1399. In: Escherichia
2 coli and Salmonella: Cellular and Molecular Biology.
[K+]1[Cl–]1 = [K+ ]2 (d)
But, eq. a states that [Cl–]1 = [K+]1 – [R–]1, therefore, Washington, DC.
www.ebook777.com
49. Lange, R., and R. Hengge-Aronis. 1994. The 58. Wild, J., E. Altman, T. Yura, and C. A. Gross.
cellular concentration of the σs subunit of RNA poly- 1992. DnaK and DnaJ heat-shock proteins partici-
merase in Escherichia coli is controlled at the levels pate in protein export in Escherichia coli. Genes Dev.
of transcription, translation, and protein stability. 6:1165–1172.
Genes Dev. 8:1600–1612.
59. Kusukawa, N., T. Yura, C. Ueguchi, Y.
50. The expression of the rpoS gene, which codes Akiyama, and K. Ito. 1989. Effects of mutations in
for σs, is regulated at the transcriptional and transla- heat-shock genes groES and groEL on protein export
tional levels, as well as by altering the stability of the in Escherichia coli. EMBO J. 8:3517–3521.
protein (see ref. 49). Transcription is stimulated when 60. Suzuki, C. K., M. Rep, J. M. van Dijl, K. Suda, L.
the cells enter stationary phase in rich media and by A. Grivell, and G. Schatz. 1997. ATP-dependent pro-
an increase in medium osmolarity. Translation of teases that also chaperone protein biogenesis. Trends
the rpoS mRNA is stimulated when cells enter sta- Biochem. Sci. 22:118–123.
tionary phase and also during an osmotic upshift.
The sigma factor itself is very unstable during expo- 61. Chung, C. H., and A. L. Goldberg. 1981. The
nential growth and becomes stabilized when cells product of the lon (capR) gene in Escherichia coli
enter stationary phase. The molecular mechanisms is the ATP-dependent protease, protease La. Proc.
responsible for the changes in the rate of transla- Natl. Acad. Sci. USA 78:4931–4935.
tion and for the alterations in protein stability are 62. Rupp, W. D. 1996. DNA repair mechanisms,
unknown. However, it is clear that these processes, pp. 2277–2294. In: Escherichia coli and Salmonella:
as well as transcriptional control, are responsible for Cellular and Molecular Biology, Vol. 2, F. C.
the increase in σs in response to starvation and osmo- Neidhardt et al. (Eds.). ASM Press, Washington,
larity signals. DC.
51. Raivio, T. L., and T. J. Silhavy. 2000. Sensing 63. Friedberg, E. C., G. C. Walker, and W. Siede
and responding to envelope stress, pp. 19–32. In: (Eds.). 1995. DNA Repair and Mutagenesis. ASM
Bacterial Stress Responses. G. Storz and R. Hengge- Press, Washington, DC.
64. Walker, G. C., B. T. Smith, and M. D. Sutton.
52. Raivio, T., and T. J. Silhavy. 2001. Periplasmic 2000. The SOS response to DNA damage, pp. 131–
stress and ECF sigma factors. Annu. Rev. Microbiol. 144. In: Bacterial Stress Responses. G. Storz and
55:591–624. R. Hennge-Aronis (Eds.). ASM Press, Washington,
53. Weber, H., T. Polen, J. Heuveling, V. F. Wendisch, DC.
and R. Hengge. 2005. Genome-wide analysis of the 65. Sutton, M. D., B. T. Smith, V. G. Godoy, and G.
general stress response network in Escherichia coli: C. Walker. 2000. The SOS response: recent insights
σs -dependent genes, promoters, sigma factor selec- into umuDC-dependent mutagenesis and DNA dam-
tivity. J. Bacteriol. 187:1591–1603. age tolerance. Annu. Rev. Gene 34:479–497.
54. Bougdour, A., C. Cunning, P. J. Baptiste, T. 66. The model for the E. coli photolyase proposes
Elliot, and S. Gottesman. 2008. Multiple pathways that light energy is absorbed by MTHF, raising an
for regulation of σs (RPoS) stability in Escherichia electron to an excited state (*MTHF). When the
coli via the action of multiple anti-adaptors. Mol. electron returns to the ground state, energy is trans-
Microbiol. 68:298–313. ferred to deprotonated reduced flavin, FADH–, excit-
55. Merrikh, H., A. E. Ferrazzoli, A. Bougdour, ing an electron to an excited state, *FADH–. Then
A. Olivier-Mason, and S. T. Lovett. 2009. A DNA *FADH– is thought to transfer the excited electron
damage response in Escherichia coli involving the to the pyrimidine dimer, resulting in the cleavage
alternative sigma factor, RpoS. Proc. Natl. Acad. Sci. of the C–C bonds holding the dimer together. The
USA. 106:611–616. reaction is completed when the electron is returned
to FADH0, regenerating FADH–. For details of the
56. Mujacic, M., M. W. Bader, and F. Baneyx. 2004. model, see the discussion of photoreactivation of
Escherichia coli Hsp31 functions as a holding chap- DNA in E. C. Friedberg, G. C. Walker, and W. Siede
erone that cooperates with the DnaK–DnaJ–GrpE (Eds.). 1995. DNA Repair and Mutagenesis. ASM
system in the management of protein misfolding Press, Washington, DC.
under severe stress conditions. Mol. Microbiol.
51:849–859. 67. Alkyl groups include methyl (CH3–) and ethyl
(CH3CH2–). These can be added to the bases by alky-
57. The crystal structures of GroEL and GroES have lating agents such as ethylmethanesulfonate (EMS)
been published. This work plus electron microscopic and nitrosoguanidine (NTG). The alkyl groups are
studies have suggested that a cavity exists in GroEL often added to the N3 of adenine and the N7 of
within which the target protein is folded. For references, guanine to form N-methyl or N-ethyl derivatives of
see the following review: Rassow, J., O. von Ahsen, U. the bases. Nitrosoguanidine will also methylate the
Bömer, and N. Pfanner. 1997. Molecular chaperones: O6 of guanine and the O4 of thymine to form the
towards a characterization of the heat-shock protein O-methyl derivatives. The alkylated bases mispair,
70 family. Trends Cell Biol. 7:129–132. causing mutations.
68. Michaels, M. L., and J. H. Miller. 1992. The GO Physiology, Vol. 46. R. K. Poole (Ed.). Academic
system protects organisms from the mutagenic effect Press, San Diego, CA.
of the spontaneous lesion 8-hydroxyguanine (7,8- 77. Hidalgo, E., H. Ding, and B. Demple. 1997.
dihydro-8-oxoguanine). J. Bacteriol. 174:6321– Redox signal transduction via iron–sulfur clusters in
6325. the SoxR transcription activator. Trends Biochem.
69. Kouchakdjian, M., V. Bodepudi, S. Shibutani, Sci. 22:207–210.
M. Eisenberg, F. Johnson, A. P. Grollman, and D. 78. SoxR is a homodimer and has a [2Fe–2S] center
J. Patel. 1991. NMR structural studies of the ioniz- in each subunit of the dimer. The activity of SoxR is
ing radiation adduct 7-hydro-8-oxodeoxyguanosine regulated by redox reactions. One way that this was
(8-oxo-7H-dG) opposite deoxyadenosine in a DNA demonstrated is as follows. The midpoint potential
duplex. 8-Oxo-7H-dG(syn)–dA(anti) alignment at for the [2Fe–2S] clusters at pH 7.6 is about –285 mV.
lesion site. Biochemistry 30:1403–1412. When the protein was titrated in vitro, its transcrip-
70. Higashitani, N., A. Higashitani, A. Roth, and tional activity was inhibited at –380 mV, that is, when
K. Horiuchi. 1992. SOS induction in Escherichia the FeS centers are more than 95% reduced. This was
coli by infection with mutant filamentous phage that reversible; that is, when the FeS centers were reoxi-
are defective in initiation of complementary-strand dized by titration, activity of the protein was restored.
DNA synthesis. J. Bacteriol. 174:1612–1618. Candidates for the in vivo reductants and oxidants
for the iron–sulfide centers in SoxR include NADPH,
71. Two DNA polymerases besides Pol V are NADH (reductants), O2, (oxidants). According to
induced by the SOS response as a result of DNA the model, SoxR works by binding to the soxS pro-
damage. One of these is DinB (Pol IV), encoded moter regardless of the oxidation state of SoxR; the
by the dinB gene in the SOS regulon. DinB repli- oxidized form of SoxR, however, undergoes an allos-
cates templates that have a bulged-out nucleotide. teric modification, enabling it to activate the soxS
The other SOS-induced DNA polymerase is Pol II, promoter, apparently by stimulating the formation
encoded by the polB gene. Pol II catalyzes “replica- of the “open” complex required to initiate transcrip-
tion restart,” a fast reinitiation of DNA replication tion. Note the similarities and differences from the
after UV damage, after which Pol III takes over. In mechanism proposed for oxygen regulation of FNR
mutants lacking both Pol II and Pol V, replication activity discussed in Section 19.2.2. It is proposed that
resumes in UV-irradiated cells after a long (90 min) oxygen can inactivate FNR by oxidizing the [4Fe–4S]
delay, indicating that there is yet another mechanism clusters and that prolonged exposure to oxygen actu-
for restarting replication. ally results in the loss of the iron–sulfide clusters. Both
72. Rangarajan, S., R. Woodgate, and M. F. the protein with the oxidized iron–sulfide clusters and
Goodman. 1999. A phenotype for enigmatic DNA the apoprotein (missing the clusters) bind very poorly
polymerase II: a pivotal role for Pol II in replication to DNA and do not influence transcription.
restart in UV-irradiated Escherichia coli. Proc. Natl. Nitric oxide also activates SoxR, but the mecha-
Acad. Sci. USA 96:9224–9229. nism is different from that of oxidizing the iron–sul-
fide centers. Nitric oxide displaces the sulfide from
73. Imlay, J. A. 2008. Cellular defenses against the iron–sulfide centers, forming a nitrosylated SoxR
superoxide and hydrogen peroxide. Annu. Rev. (dinitrosyl–iron–cysteine complex). The nitrosylated
Biochem. 77:755–776. SoxR activates transcription to the same extent as
74. Storz, G., and M. Zheng. 2000. Oxidative SoxR with oxidized iron–sulfide centers.
stress, pp. 47–59. In: Bacterial Stress Responses. 79. Li, Z., and B. Demple. 1994. SoxS, an activator
G. Storz and R. Hengge-Aronis (Eds.). ASM Press, of superoxide stress genes in Escherichia coli. J. Biol.
Washington, DC. Chem. 269:18371–18377.
75. Lynch, A. S. and E. C. S. Lin. 1996. Responses to 80. Ritz. D., and J. Beckwith. 2001. Roles of thiol–
molecular oxygen, pp. 1526–1538. In: Escherichia redox pathways in bacteria. Annu. Rev. Microbiol.
coli and Salmonella: Cellular and Molecular Biology, 55:21–48.
Vol. I. F. C. Neidhardt et al. (Eds.). ASM Press,
81. Most bacteria contain glutathione as a compo-
Washington, DC.
nent of their redox buffer. The actinomycetes, includ-
76. Pomposiello, P. J., and B. Demple. 2002. Global ing Mycobacteria, which have a cysteine-containing
adjustment of microbial physiology during free radi- sugar-based compound called mycothiol, are an
cal stress, pp. 319–341. In: Advances in Microbial exception.
www.ebook777.com
17
Solute Transport
Bacterial cell membranes consist in large part might move along amino acid side chains that
of a phospholipid matrix that acts as a perme- face into the channel. For example, the lactose/
ability barrier blocking the diffusion of water- proton symporter from E. coli probably forms
soluble molecules into and out of cells (see 12 transmembrane loops.1 Another model that
Section 1.2.5). This has the advantage of allow- might apply to other carriers is that there is a
ing the bacterium to maintain an internal envi- single binding site for the solute and that the pro-
ronment different from the external, and one tein carrier can have two alternating conforma-
conducive to growth. For example, metabolites tions in which the binding site is inward facing
can be maintained at an intracellular concen- or outward facing. Thus, transport of the sol-
tration that is orders of magnitude higher than ute would occur by interconversion of the two
the extracellular concentration. This has two conformations. The importance of transport-
important consequences: (1) the promotion ers is easy to demonstrate. Mutants that lack
of more rapid enzymatic reactions and (2) the a transporter for a required nutrient will grow
retention of metabolic intermediates within the very poorly in low concentrations of the nutri-
cell. The lipid barrier also minimizes the passive ent. In the natural environment, such mutants
diffusion of ions, including protons, and thus would not be expected to survive. But they can
functions to maintain the electrochemical probe isolated during a screen for slow growers and
ton and sodium ion gradients that are impor- maintained as laboratory cultures. Transport
tant for driving ATP synthesis, solute transport, through the outer membrane of gram-negative
and other membrane activities. bacteria is also an important area of research.
Since the phospholipid presents a perme- The role of porins and of TonB in outer mem-
ability barrier, virtually everything that is not brane transport is discussed in Sections 1.2.3
lipid soluble enters and leaves the cell through and 1.2.4, respectively.
integral membrane proteins that have various
names, including transporters, carriers, por-
ters, or permeases. The protein carriers them- 17.1 The Use of Proteoliposomes to
selves do not diffuse across the membrane but Study Solute Transport
facilitate the transport of the solutes in other Proteoliposomes are artificial membrane ves-
ways. For example, the amino acid sequences icles of protein and phospholipid that are of
of a few transporters, deduced from nucleotide enormous value in studying solute transport,
sequences, suggest that these proteins form and it is instructive to describe how they are
multiple transmembrane loops that fold in the made and some of their properties. There are
membrane, perhaps making an internal channel several methods of preparing proteoliposomes,
through which the solute is passed. The solute but they are all similar and rely upon the ability
432
solute transport 433
Fig. 17.1 Preparation of proteoliposomes. Phospholipids are dispersed in water and sonicated. This is fol-
lowed by the formation of small vesicles, each surrounded by a lipid bilayer. The vesicles are mixed with
detergent-solubilized protein and diluted into buffer. Proteoliposomes form, with the protein incorporated
into the bilayer. The proteoliposomes can be loaded with substrate or ATP by including these in the dilution
buffer.
of membrane proteins solubilized in detergent Salmonella typhimurium.3–6 Proteoliposomes

to integrate into phospholipid bilayers when the are also used to study electron transport. For
detergent is removed by dilution or dialysis.2 example, they have been important in experi-
One way to prepare phospholipid bilayers is to ments that establish that certain cytochrome
first disperse the phospholipids in water, where oxidases are proton pumps.
they spontaneously aggregate to form spheri-
cal vesicles called liposomes, which consist of
concentric layers of phospholipid (Fig. 17.1). 17.2 Kinetics of Solute Uptake
The liposomes are then subjected to high-fre- 17.2.1 Transporter-mediated uptake
quency sound waves (sonic oscillation), which The existence of transporters can be revealed by
breaks them into smaller vesicles surrounded the kinetics of solute uptake. If one were to add a
by a single phospholipid bilayer resembling the solute (e.g., an amino acid or sugar) at different
lipid bilayer found in natural membranes. Then concentrations to a series of bacterial suspen-
the protein, which has been solubilized in deter- sions, and plot the initial rate of uptake into the
gent, is mixed with the sonicated phospholipids cell as a function of the external concentration
in the presence of detergent and buffer, and the of solute, a curve such as that shown in Fig. 17.2
suspension is diluted into buffer, which low- would be generated. Notice that the curve is a
ers the concentration of detergent. The protein hyperbola that approaches a maximum rate. The
leaves the detergent and becomes incorporated kinetics for transporter-mediated solute uptake
into the phospholipid bilayer. The protein and can be rationalized by assuming that the only sig-
lipid membrane vesicles that form are called nificant route of entry for the solute is on a lim-
proteoliposomes. ited number of transporters. That is, the solute
One can “load” the proteoliposomes with does not passively leak into the cell to any signifi-
ATP or other substrates by including these in cant extent. Therefore, the rate at which the sol-
the dilution buffer. When the proteoliposomes ute enters is directly proportional to the fraction
are incubated with solute, they catalyze uptake of transporters that are occupied with solute. As
of the solute into the vesicles, provided the the external concentration of solute increases, a
appropriate transporter has been incorporated. progressively larger fraction of the transporters
The vesicles can be reisolated (e.g., by centrifu- binds solute, and the rate of transport increases
gation or filtration), and the amount of solute to a maximum rate (Vmax), the point at which
taken into the vesicles can be quantified. Some there are no unloaded transporters.
transporters that have been reconstituted into The concentration of solute that produces
proteoliposomes and used to demonstrate cata- one-half the maximum initial rate of transport is
lyzed transport are the lactose permease from called the Km or sometimes the Kt. Because Km is
E. coli, the oxalate/formate antiporter from frequently assumed to be a measure of the affin-
Oxalobacter formigenes, the Na+/H+ antiporter ity of the solute for the transporter, it is called
from E. coli, and the histidine permease from the affinity constant. However, we will refer
www.ebook777.com
Fig. 17.3 The initial rate of uptake (Vi) of solute (S)

in the absence of a transporter. In the absence of a
transporter, the rate of solute entry is relatively slow
and does not approach a maximum even at very high
concentrations of [S]. Rather, it is proportional to the
Fig. 17.2 Kinetics of transport. Bacteria are incu- concentration gradient.
bated with different concentrations of solute (S), and
the initial rate of solute uptake is measured for each
solute concentration. The rate (v) approaches a max- many transport mechanisms catalyze the accu-
imum (Vm) as the fraction of transporter molecules mulation of solutes into the cell. The internal
bound to solute reaches a maximum. The concentra- concentration when the steady state is reached
tion of solute that gives 1/2Vm, the Km (sometimes can be several orders of magnitude higher than
called kt), is characteristic for each transporter. the external concentration. Of course, this
requires energy. The source of energy can be
either chemical, light, or electrochemical.
to this concentration simply as the solute con-
Bacterial transport systems are now divided
centration that gives 1/2Vmax. The value of the
into three main categories, primary trans-
Km is characteristic of the transporter and can
port systems, secondary transporter systems,
range from less than 1 μM to several hundred
and group translocation systems. Primary
micromolar. The kinetics shown in Fig. 17.2 is
transport systems, driven by an energy-pro-
described by the Michaelis–Menten equation
ducing (exergonic) metabolic event, include
for enzyme catalysis (see Section 7.2.1).
proton translocation drive by ATP, light, or
oxidation–reduction reactions (Sections 4.7.1,
17.2.2 Uptake in the absence of a 4.7.2, 4.8.4, 5.5), light-driven chloride trans-
transporter port (Section 4.9), sodium ion transporting
What if there were no transporter, and the nutri- decarboxylases (Section 4.8.1), and the uptake
ent simply diffused into the cell? Slow-growing of inorganic or organic solutes driven by ATP
mutants have been isolated that have no func- hydrolysis (described later in this chapter).
tional transporter for a particular nutrient. The Group translocation refers to transport cou-
kinetics of uptake in these mutants (Fig. 17.3) pled to a chemical modification of the solute so
reflects what one would expect in the absence of that it is released inside the cell in a chemically
a transporter. Note that the rates of uptake are modified form. An example of group transloca-
low and are proportional to the concentration tion is the uptake of sugars driven by phospho-
gradient, with no saturation at very high exter- ryl group transfer from phosphoenolpyruvate,
nal concentrations of solute. called the phosphotransferase system (described
later in this chapter). During transport by the
17.3 Energy-Dependent Transport phosphotransferase system, the sugar accu-
A transporter simply facilitates the entry and mulates inside the cell as the phosphorylated
exit of the solute across the membrane. At derivative.
equilibrium, it does not bias the transport in Secondary transport systems are usually
any direction and therefore cannot, on its own, driven by electrochemical gradients (e.g., pro-
cause the accumulation of solute against a con- ton and sodium ion gradients). During sec-
centration gradient. However, we know that ondary transport the solute moves “down” an
electrochemical ion gradient, usually of protons its electrochemical gradient. Antiport refers to
or sodium ions. (See note 7.) Both primary and the coupled movement of two solutes in oppo-
secondary transport systems also exist for the site directions. For example, an exchange of
efflux of drugs and other toxic substances, and H+ for Na+ on the same carrier is antiport. A
this is described in Section 17.5. major family of membrane transport proteins
Secondary transport is coupled to primary that catalyze secondary transport is called the
transport by the proton in the following way: major facilitator superfamily (MFS), and some
of these are reviewed in ref. 8, which includes
Primary transport structural studies.
+ +
H in + energy → H out
17.3.1 Secondary transport
Secondary transport A general description
+ + The transporter functions only when it trans-
H out + Sout → H in + Sin
ports both the ion and the solute in one direc-
The distinction between primary and secondary tion (symport) or the solute in one direction and
transport is an important one because it empha- the ion in the other (antiport) (Fig. 17.4).9 This
sizes a central feature of the chemiosmosis theis how solute transport is coupled to ion cur-
ory: that the coupling between energy-yielding rents. However, there are also transporters that
reactions and energy-requiring reactions in the move just an ion along its electrochemical gradi-
membrane is via ion currents. In the preceding ent (uniport). In all these cases, the transporter
example, the coupling between the energy-de- is part of the electrical circuit. This is analogous
pendent uptake of S and the primary energy- to saying that for an electrical current to make
yielding reaction is via the proton current, which a motor turn, the current must flow through the
is most commonly used. motor.
“Active transport” refers to primary trans- Most bacteria employ both proton and
port during which the solute is not chemically sodium symporters, but mainly the former. (See
modified (e.g., the ATP-dependent uptake of note 10 for more information.) (Sodium ion
histidine). Active transport is therefore a sub- symporters are common in bacteria for amino
class of primary transport. It should be pointed acid transporters.) However, certain bacteria
out that, before the chemiosmotic theory was (e.g., alkaliphiles, halophiles, marine bacteria)
developed, the distinction between primary rely more heavily on sodium symporters. The
and secondary transport was not made. In fact, reason for this becomes clear when one consid-
an older definition of active transport was any ers their ecological niches. The alkaliphiles live
transport that results in a concentration gradi- in environments with pH values in the range of
ent where the solute is accumulated in a chemi- 9 to 11. Because their cytoplasmic pH is below
cally unmodified form. However, the current 8.5, the ΔpH (pHin−pHout) is negative rather
definition restricts the use of “active transport” than positive. This decreases the Δp. These
to primary transport in which the solute does organisms therefore depend upon the sodium
not chemically change. ion and the membrane potential (ΔΨ) for sol-
Secondary transport is catalyzed by uniport- ute transport. Some of these bacteria also have
ers, symporters, and antiporters that use elec- flagellar motors powered by a sodium potential
trochemical ion gradients to accumulate solutes. rather than a proton potential.11,12 Halophilic
Examples are illustrated later (see Fig. 17.4). bacteria require high external NaCl concen-
Uniport refers to solute translocation down an trations (3–5 M) to grow. They use predomi-
electrochemical gradient in the absence of a cou- nantly Na+/solute symporters. Marine bacteria
pling ion. For example, the uptake of K+ down also live in high concentrations of Na+ (close
its electrochemical gradient is uniport. Symport to 0.5 M), and they rely heavily on Na+/solute
refers to solute uptake in which two solutes are symporters.13
carried on the carrier in the same direction. An The use of ionophores can aid in identifying
example of this would be the uptake of a solute the ion current that is responsible for second-
coupled to the uptake of one or more protons, ary transport. This is discussed in Section 4.4. It
or sodium ions, that drives the transport via might be added that in contrast to transporters
www.ebook777.com
driven by ATP described in Section 17.3.3, the into the cell. Note that 17.4 kJ does not refer to
characterized transporters driven by the Δp the energy required to move 103 moles of S to
(e.g., the Lac permease), are relatively simple in one side of the membrane to establish a ratio
composition, consisting of a single protein that of 103/1. Equation 17.1 can also be expressed
traverses the membrane in several loops. as an electrical potential in volts, since (RT/F)
(2.303) = 0.06 V, or 60 mV at 30 °C. We now
Energetics of transport discuss some examples of solute transport cou-
The free energy in joules required to move a pled to an electrochemical gradient (Fig. 17.4).
mole of solute from outside a cell to inside the
cell is given by eq. 17.1, where [S]in is the con- Some examples of solute transport coupled to
centration inside the cell and [S]out is the concen- an electrochemical gradient
tration outside the cell. 1. Symport of an uncharged solute with
protons (Fig. 17.4A)
ΔG = RT ln [S]in/[S]out J/mol
Many solutes are transported by symport.
= 60 log [S]in/[S]out mV at 30 oC (17.1) Assume that the stoichiometric ratio of H+/S is
y. At equilibrium, the force accompanying the
For example, if [S]in/[S]out = 103, then ΔG = 17.4
diffusion of the solute down its concentration
kJ/mol at 30 °C. (See note 14.) This means that
gradient (out of the cell) is –60 log [S]in/[S]out
at least 17.4 kJ of work must be applied against
mV and is balanced by the Δp drawing the sol-
the concentration gradient (at 30 °C) to move
ute in the opposite direction (into the cell), so
one mole of solute into the cell when the con-
that at 30 °C
centration ratio, in/out, is 103 and remains as
such. This would be the case in a steady state y(ΔΨ−60 ΔpH) = yΔp
situation, where Sin is used as fast as it is brought = –60 log[S]in/[S]out (17.2)
Fig. 17.4 Some examples of solute transport driven by the ΔΨ and the ΔpH. (A) Symport of protons with an
uncharged solute. (B) Symport of protons with a monovalent anion. (C) Antiport of protons with a monova-
lent cation. (D) Uniport of a monovalent cation. Another possibility, not shown, is symport of protons with
cations. The ratio of protons to solute is given by y, which is assumed to have a value of one or greater.
Alternatively, one can write the total driving uptake of cations (e.g., for K+ uptake). The
force, which in this case would be relationship between ΔΨ and the uptake of
cations is
yΔp + 60 log[S]in/[S]out
ΔΨ = –60 log([S+]in/[S+]out) (17.5)
and set this equal to 0 (equilibrium). Rear-
rangement of the equation then yields eq. 17.2. For a divalent cation, the driving force would
(A similar procedure produces eqs. 17.3–17.5, be 2ΔΨ, since there is twice as much charge per
given shortly.) ion. Equation 17.5 is one form of the Nernst
For example, if the ratio (y) of protons to solute equation.
transported is 1:1 and the concentration gradient
[S]in/[S]out is 103, then the minimum Δp required 17.3.2 Evidence for solute/proton or
to maintain that concentration gradient would be solute/sodium symport
–180 mV. Notice that [S]in/[S]out is an exponen- One way to demonstrate coupling of transport
tial function (logarithmic function) of yΔp. For to proton or sodium ion influx is to measure the
example, when y is increased to 2, the maximum alkalinization of the medium (decrease in pro-
concentration gradient attained at equilibrium is tons) or a decrease in the extracellular Na+ con-
squared. Theoretically, very large concentration centration when bacteria or membrane vesicles
gradients can be maintained by the Δp. are incubated with the appropriate solutes.15 The
2. Symport of a monovalent anion with changes in the external pH or Na+ concentra-
protons (Fig. 17.4B) tion occur as a result of symport with the solute
The Δp can also drive the uptake of anions. and can be measured with H+- or Na+-selective
However, whether the ΔΨ is part of the driv- electrodes. It is necessary to ensure that H+ or
ing force depends upon whether a net charge Na+ influx is electroneutral and a membrane
is transported. For example, for a monovalent potential, which would impede further influx
anion, the ratio H+/R– must be greater than one of cations, does not develop. Thus the experi-
ments are done in the presence of a permeant
if the ΔΨ is to be part of the driving force. The
anion (e.g., CNS–) that serves as a counterion
relative contributions of the ΔΨ and the ΔpH to
to protons or sodium ions, or K+ plus valinomy-
the driving force are as follows:
cin, whose efflux can exchange for the incom-
(y – 1)ΔΨ – y60 ΔpH ing protons or sodium ions. (This is explained
= –60 log([S–]in/[S–]out ) (17.3) in note 16.)
The experiments may be done so that the ion
Equation 17.3 is the same as eq. 4.25, whose
influx driven by the solute concentration (i.e.,
derivation can be found in Section 4.8.3. (See
symport) is demonstrated, rather than accumu-
also eq. 4.23 in Section 4.8.3 for a multivalent
lation of the solute against a concentration gra-
anion.)
dient. An example is the experiment reported by
3. Antiport of a monovalent cation West in 1970.17 West suspended E. coli in dilute
with H+ (Fig. 17.4C) buffer and added the sugar lactose, which pro-
The proton:sodium ion antiporter is widespread moted an increase in the external pH (Fig. 17.5).
among the bacteria. It is used for pumping This result implied that the lactose permease
sodium ions out of the cell. The equation is simi- had catalyzed a symport of lactose with pro-
lar to eq. 17.3 except that the sign is changed for tons. Other investigators have since performed
the expression for the electrochemical potential similar experiments, demonstrating that many
of S+ because it is moving out of the cell; that is, sugars and amino acids enter bacteria in sym-
[S+]in/[S+]out < 1: port with either protons or sodium ions.
(y – 1)Δ Ψ – y60 ΔpH
= +60 log([S+]in/[S+]out) (17.4) 17.3.3 Primary transport driven by ATP
The following transport systems are driven by
4. Electrogenic uniport of a ATP (or some phosphorylated derivative in
cation (Fig. 17.4D) equilibrium with ATP): (1) H+ transport [i.e., the
For some transporters, the membrane poten- ATP synthase (ATPase)], which uses the proton
tial alone can provide the driving force for the motive force (PMF) to make ATP; (2) K+ influx
www.ebook777.com
Fig. 17.5 Lactose-dependent proton uptake. Lactose was added to a suspension of Escherichia coli without
(a) and with (b) CNS–, and the pH was measured with a pH meter. Lactose caused the immediate uptake
of protons. Uptake was stimulated by CNS–, which prevented a membrane potential, inside positive, from
developing. Lactose uptake was inhibited by thiodigalactoside (TDG), which is a competitive inhibitor of the
lac permease (c). These data reflect symport between lactose and protons. Source: West, I. C. 1970. Lactose
transport coupled to proton movements in Escherichia coli. Biochem. Biophys. Res. Commun. 41:655–661.
Copyright 1970, with permission from Elsevier.
in E. coli; (3) some transport systems in gram- approximately 105, and Km values for uptake are
positive bacteria; and (4) transport systems in in the range of 0.1 to 1 μM. This means that they
gram-negative bacteria that use periplasmic can scavenge very low concentrations of solute
binding proteins and are also called shock-sen- and accumulate these to relatively high internal
sitive transport systems. (Review the discussion concentrations. The shock-sensitive systems
of the periplasm in Section 1.2.4.) generally utilize a membrane transporter that is
part of a superfamily of transporters found in
Shock-sensitive transport systems in both gram-negative and gram-positive bacteria,
gram-negative bacteria as well as in eukaryotes. We turn next to a dis-
The shock-sensitive systems rely on periplasmic cussion of the so-called ATP-binding cassette
binding proteins that combine with sugars and (ABC) transporters.
amino acids in the periplasm and transfer these
to the actual transporters in the cell membrane. ATP-binding cassette (ABC) transporters
They are distinguished from all other transport For reviews, see refs. 24 through 26. The ABC
systems in that they are not functional in cells transporters are found in a range of organisms
that have been osmotically shocked, a treat- from bacteria to humans; the name designates
ment that makes the outer envelope permeable their ATP-binding domains. The ABC transporter
and causes the release of periplasmic proteins. is a membrane complex consisting of four sub-
(The procedure for osmotic shock is described units (Fig. 17.6). Two subunits of the transporter
in note 18.) are nonidentical, hydrophobic transmembrane
Shock-sensitive transport systems, also called moieties (a and c in Fig. 17.6.) The other two
periplasmic permeases, are characteristic of subunits (b in Fig. 17.6) are identical and hydro-
gram-negative bacteria and are responsible for philic, each has a nucleotide-binding domain
the uptake of a wide range of solutes, including (NBD) that binds ATP on the cytoplasmic side of
sugars, amino acids, and ions.19–23 The shock- the membrane. (For the names of these subunits
sensitive transport systems are characterized in the histidine transport system, see note 27.)
by very high efficiencies. They are capable In addition, there is an ABC signature se-
of maintaining concentration gradients of quence of amino acids that indicates the presence
CM 2
cytoplasm
a 3
b
b
c
ADP, Pi
4
ATP
OM
5
periplasm
Fig. 17.7 Transport of histidine by reconstituted
proteoliposomes. The histidine carrier proteins plus
other membrane proteins were solubilized in deter-
Fig. 17.6 Model for periplasmic transport. (1) The gent and incorporated into proteoliposomes in the
solute enters the periplasm through a porin pore in presence or absence of ATP. Incubation mixtures
the outer membrane. (2) Inside the periplasm the contained the histidine-binding protein. Histidine
solute binds to a binding protein, which undergoes was taken up by the proteoliposomes that were
a conformational change when it binds the solute. loaded with ATP. Source: Adapted from Bishop, L.,
(3) The binding protein carrying the solute binds to et al. 1989. Reconstitution of a bacterial periplasmic
the transporters (a, b, c) located in the inner (cell) permease in proteoliposomes and demonstration of
membrane. (4) The transporter is thought to undergo ATP hydrolysis concomitant with transport. Proc.
a conformational change that may result in the open- Natl. Acad. Sci. USA 86:6953–6957.
ing of a pore through which the solute diffuses to the
cytoplasm. (5) The putative pore closes again when
The energy to actively transport solute across
the binding protein is released from the permease.
The transporter has a binding site for ATP, which has
the outer membrane is provided by the elec-
been demonstrated to be hydrolyzed during histidine trochemical gradient of the cell membrane.
transport, and presumably is hydrolyzed during the Once inside the periplasm, the solutes bind to
transport of other solutes. An alternative hypothesis a periplasmic solute-binding protein that deliv-
speculates that the binding protein triggers confor- ers the solute to the transporter in the cell mem-
mational changes in the permease that make a bind- brane. Thus the ABC transporters are more
ing site available for the solute. The solute is then complex than Δp-driven transporters, which
passed from one binding site to another on the per- consist of a single protein that crosses the cell
mease until it is released inside the cell, rather than membrane in several loops. Solute transport via
diffusing through a pore. the ABC transporter in gram-negative bacteria
can be visualized as occurring in several sequen-
tial steps, as illustrated in Fig. 17.6.
of the ATP-binding domains. Probably, the
binding and/or hydrolysis of ATP at both sites Step 1. The solute enters the periplasm through
causes a conformational change in the trans- an outer membrane pore (e.g., through a
porter that results in the translocation of the porin). (Consult Section 1.2.3 for a descrip-
substrate across the membrane. (For evidence tion of the outer membrane and of porins.)
that ATP drives uptake, read note 28 and see Step 2. The solute binds to a specific periplas-
Fig. 17.7.) mic binding protein to form a solute:binding
Smaller solutes, such as sugars, amino acids, protein complex. The binding protein under-
and small peptides, enter the periplasm by dif- goes a conformational change that allows it
fusing through a porin pore in the outer mem- to bind productively to the transporter in the
brane. Larger solutes, such as vitamin B12, are membrane.
actively transported across the outer membrane Step 3. The liganded binding protein binds to
into the periplasm by special transporters. the ABC transporter in the cell membrane,
www.ebook777.com
delivering the solute to the transporter. As Most of what we know about K+ transport is
noted earlier, the membrane-bound trans- derived from studies of E. coli.30–32
porter is a complex consisting of four pro- There are two transport systems. The major
teins, two of which are identical. route for K+ uptake occurs via the TrK system,
Step 4. The transporter complex translocates which is always present (constitutive) and oper-
the solute across the cell membrane, perhaps ates at a high rate, but relatively low affinity (high
through a channel that opens transiently. Km: i.e., 2 mM) for K+. This transporter requires
ATP hydrolysis, catalyzed by the transporter both a Δp and ATP to function, and the reasons
complex, occurs at this step. The binding for the dual requirement are not clear. (It has been
protein is essential not only for delivering the speculated that the energy source is actually the
solute, but also for stimulating the ATPase Δp and that ATP acts as a positive regulator.33)
activity. The ATP-binding domains of the When E. coli is grown in media in which the K+
transporter are exposed to the cytoplasm. concentrations are very low, the cells synthesize
Step 5. The transporter complex returns to its a second K+ transport system called the Kdp
unstimulated state. system, which serves to scavenge K+. However,
In some cases the membrane-bound transporter the induction depends upon the salt-dependent
interacts with more than one type of binding osmolarity of the medium. The threshold con-
protein and can transport more than one kind centration of K+ that prevents induction is higher
of solute. For example, the histidine permease, when the osmolarity is higher. This means that
besides transporting histidine, also transports the Kdp system can be induced by raising the
arginine via the arginine–ornithine binding osmolarity of the medium by using salts with-
protein. Several different binding proteins for out altering the external K+ concentrations, as
branched-chain amino acids use the same mem- long as the external K+ concentration is not too
brane transporter. high. This reflects the important role for K+ as
It is important to note that ABC transporters an intracellular osmolyte.34 (See Section 16.2.4
are part of a superfamily of structurally related for a discussion of K+ as an osmolyte.) The Kdp
proteins and are not confined to gram-negative system has a very low Km for K+ (2 μM) and uses
bacteria. They also occur in gram-positive bac- ATP as the energy source.
teria, where an equivalent of the periplasmic There are three structural proteins in the Kdp
binding protein that delivers the solute to the system, KdpA, KdpB, and KdpC, all of which
transporter is attached to the outside surface of are located in the cell membrane. The genes
the cell membrane or to the transporter, and in are encoded in the kdpABC operon. KdpA
eukaryotic cells.29 The ABC transporters can be is a membrane-spanning protein exposed to
used for either uptake or export, and because the periplasm. Since mutations in this protein
of this they are sometimes referred to as “traffic produce changes in the Km for K+ uptake, it is
ATPases.” ABC transporters also catalyze the believed that KdpA binds and translocates K+.35
export of surface components such as capsu- KdpA may form a channel through which the K+
lar polysaccharides and teichoic acids, as well moves across the membrane. KdpB, an integral
as drugs and proteins. See Section 17.5, which membrane protein, is an ATPase that couples
discusses drug-export systems; the subsection ATP hydrolysis with K+ uptake (Fig. 17.8). The
of Section 12.3.5 entitled Translocation via function of KdpC, a peripheral membrane pro-
ABC transporters; and the coverage of the type tein, is not known.
I pathway of extracellular protein secretion in There are two proteins that are required for
Section 18.5.1. the transcription of the kdpABC operon. These
are the KdpD protein, located in the cytoplasmic
ATP-driven K+ influx membrane with cytoplasmic domains, and the
Potassium ion is the principal cation in bacteria, KdpE protein, located in the cytosol. These two
and it not only plays a role in osmotic and pH proteins are part of a two-component signal-
homeostasis (Chapter 16) but is also a cofactor ing system that may sense intracellular signals
for many enzymes as well as ribosomes. Bacteria such as low intracellular K+ concentration or an
accumulate K+ to a level several orders of mag- increase in intracellular ionic strength, both of
nitude higher than the external concentrations. which are thought to stimulate transcription of
The phosphoryl group is transferred to the car-

bohydrate via a consecutive series of reactions
beginning with PEP as the initial donor. The
series, illustrated in Fig. 17.9, consists of the
following steps.
Step 1. In the soluble part of the cell, the phos-
phoryl group is transferred from phospho-
enolpyruvate to enzyme I (EI):
PEP + EI → P~EI + pyruvate
Step 2. Enzyme I transfers the phosphoryl group
to a small cytoplasmic protein called HPr:
P~EI + HPr → P~HPr + EI
Fig. 17.8 The Kdp system for K+ uptake. K+ is thought

Enzyme I and HPr are common to all the PTS
to bind to KdpA in the periplasm, after which it is carbohydrate uptake systems.
transported through the membrane. The energy for Steps 3a, 3b, 3c. The HPr then transfers the
K+ transport is derived from ATP hydrolysis medi-
phosphoryl group to a carbohydrate-specific
ated by KdpB. The function of KdpC (not shown) is
unknown. Source: Epstein, W., and L. Laimins. 1980.
permease complex in the membrane called
Potassium transport in Escherichia coli: diverse sys- enzyme II, which transfers the phosphoryl
tems with common control by osmotic forces. Trends group to the carbohydrate during carbohy-
Biochem. Sci. 5:21–23. drate uptake into the cell:
P~HPr + EII → HPr + P~EII
the kdpABC promoter.36–40 (See ref. 41 for a dis-
cussion of this point.) P~EII + [CH2O]out → EII + [CH2O−P]in
Enzyme II has three domains, A, B, and C. The
17.3.4 The phosphotransferase system
phosphoryl group is transferred from HPr to
The phosphotransferase system (PTS) differs
domain A, then to domain B, and finally to
from the systems driven by ATP or electrochemi-
the carbohydrate in a reaction that requires
cal gradients in that it catalyzes the accumulation
domain C, which is always an integral mem-
of carbohydrates (e.g., monosaccharides, disac-
brane protein:
charides, amino sugars, polyols) as the phospho-
rylated derivatives instead of as the free sugar.42–45 P~ HPr + EIIA → P~EIIA + HPr
The energy source and phosphoryl donor in P~EIIA + EIIB → EIIA + P~EIIB
these transport systems is phosphoenolpyruvate
P~EIIB + carbohydrateout →
E IIC
(PEP), an intermediate in glycolysis (Section 9.1).

EIIB + carbohydrate–Pin
Because the carbohydrate is modified (by phos-
phorylation), this type of transport is not referred How EIIC brings the carbohydrate into the cell
to as active transport but rather as group translo- is not understood.
cation. The phosphotransferase system is charac- Different carbohydrate uptake systems differ
teristic of anaerobic and facultatively anaerobic with respect to the number of separate proteins
bacteria, but it seems to be lacking in many aer- that constitute “enzyme II.” There can be from
obes. It also does not occur in eukaryotes and has one to four proteins, one of which (IIC), is always
not been found in archaea. The overall reaction of membrane bound and catalyzes the transport
the PTS for glucose transport is of the carbohydrate into the cell. For example,
enzyme II for mannitol uptake (EIIMtl in Fig. 17.9)
PEP + carbohydrateout
is a single membrane-bound protein with three
→ pyruvic acid + carbohydrate–Pin
domains, A, B, C. Domain C is in the membrane,
During the reaction, PEP donates a phosphoryl whereas domains A and B project into the cyto-
group to the carbohydrate, which accumulates plasm. However, in some transport systems the
inside the cell as the phosphorylated derivative. three domains are on separate enzyme II proteins.
www.ebook777.com
mannose mannitol
D C EIIC,DMan C EIICMtl
3c B
mannitol-1-P P EIIA,BMtl
3b A
B P S1 S1
P 3c
3b mannose-6-P
A
P 3a A
3a
EIIA,BMan
PEP EI P~HPr
S2 S2
1 2
pyruvate P~EI HPr 3a ATP
+ adenylate
A
3b P cyclase
EIIAGlc
3c cAMP
glucose-6-P B EIIBGlc
P
C EIICGlc
glucose
Fig. 17.9 The sugar–phosphotransferase system. The phosphoryl group is transferred via a series of proteins
to the sugar. The sugar-specific carrier proteins in the membrane, IIMan, IIMtl, and IIGlc, accept the phosphoryl
group from P–HPr and transfer it to the sugar-forming mannose-6-P, mannitol-1-P, and glucose-6-P. Enzyme
II can be a single protein with three domains, A, B, and C, as in the mannitol system, or separate proteins,
as in the glucose and mannose systems. The mannose carrier consists of two proteins, C and D. Note that
the phosphoryl group travels from HPr to IIA to IIB to the sugar, which is translocated by IIC into the cell
in an unknown manner. At some stage during translocation, the sugar becomes phosphorylated. However,
phosphorylation of the sugar need not take place during translocation per se. That is, the sugar may be phos-
phorylated on the inside surface of the cell membrane prior to its release into the cytoplasm. Also shown
is the stimulation of adenylate cyclase by P–EIIGlcA (formerly called P–IIIGlc) and the inhibition of non-PTS
sugar carriers (S1 and S2 carriers) by EIIGlcA. Source: Adapted from Postma, P. W., J. W. Lengeler, and G. R.
Jacobson, 1993. Phosphenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiol. Rev.
57:543–594.
Thus, enzyme II for glucose consists of two pro- a choice of glucose and another carbon source,
teins, IIA and IIBC (Fig. 17.9). In this case, IIBC is they will preferentially utilize the glucose
membrane bound, whereas IIA (formerly called and delay the use of the other carbon source
IIIGlc) is cytoplasmic. In a third case (EIIMan), EIIAB until the glucose has been depleted.46 This is
exists as a single cytoplasmic protein, whereas called glucose repression, or catabolite repres-
there are two membrane-bound EII proteins, sion, and is responsible for diauxic growth,
IIC and IID. In the cellobiose PTS in E. coli, the described in Section 2.2.4. The phenomenon is
enzymes II are three proteins: IIA and IIB in the not restricted to glucose, since many PTS car-
cytoplasm and IIC in the membrane. bohydrates are used in preference to other car-
bon sources. (See note 47.) For a more complete
Catabolite regulation by the PTS in discussion of how bacteria regulate the utiliza-
enteric bacteria tion of certain carbon sources over others, see
It has been known for many years that when Section 19.10 (Response to Carbon Sources:
bacteria using the phosphotransferase system Catabolite Repression, Inducer Expulsion,
for the transport of glucose are presented with Permease Synthesis).
1. The model sugars, and the decrease in P–IIAGlc would be

The widely accepted model for regulation by expected to inhibit the adenylate cyclase.
the phosphotransferase system in enteric bacte-
2. Rationale for the model
ria illustrated in Fig. 17.8 rests on the following
The phosphotransferase system is believed to be
postulates.
involved in the utilization of glucose-repressed
1. IIAGlc inhibits several enzymes required for carbon sources on the basis of the original find-
carbohydrate metabolism, including certain ing that mutants lacking HPr or enzyme I are
non-PTS sugar transporters such as the lac- unable to grow on glucose-repressed carbon
tose and melibiose transporter, the MalK sources. In E. coli, these carbon sources include
protein, which is essential for the malt- lactose, maltose, melibiose, glycerol, rhamnose,
ose transport system, and glycerol kinase. xylose, and citric acid cycle intermediates. The
P–IIAGlc is dephosphorylated to IIAGlc by reason for this is that the carbon sources can-
IICBGlc during glucose transport. Therefore, not phosphorylate IIAGlc, which stimulates the
according to the model, glucose transport production of cAMP by activating the adeny-
into the cell inhibits the above-mentioned late cyclase. The requirement for EI and Hpr for
enzymes. The inhibition by glucose of trans- growth on non-PTS sugars is not restricted to
port and metabolism of non-PTS carbohy- E. coli or other enterics, although it is less well
drates is called inducer exclusion. studied in other bacteria.
2. P–IIAGlc stimulates the membrane-bound Because adding glucose to wild-type cells has
enzyme adenylate cyclase, which makes the same repressive effect on growth due to non-
cyclic AMP (cAMP), which in turn stimu- PTS sugars and citric acid cycle intermediates
lates the transcription of many genes that as the presence of mutations in HPr or enzyme
code for catabolic enzymes. It is actually a I, it was suggested that P–IIAGlc is required for
complex of cAMP and the cAMP receptor growth on these substances. The reasoning is
protein (CRP) that binds to the polymerase that (1) a defect in enzyme I or HPr will result
and regulates transcription. (The CRP pro- in the inability to phosphorylate IIAGlc and (2)
tein is also called the catabolite activator pro- the addition of glucose to wild-type cells should
tein, or CAP.) During glucose uptake by the result in the dephosphorylation of P–IIAGlc.
PTS system, P–IIAGlc is dephosphorylated; Adenylate cyclase was implicated in the pro-
hence adenylate cyclase is no longer stimu- cess when it was discovered that the addition of
lated, and transcription of cAMP-dependent cAMP, the product of adenylate cyclase, could
genes is inhibited. Regardless of the carbon overcome the Hpr mutant phenotype with
source, however, there is always a basal respect to growth. Knowledge of the require-
level of cAMP synthesized. This is necessary ment for cAMP in the transcription of several
because the transcription of genes required genes led to a model postulating a stimulation
for the metabolism of many PTS sugars also by P–IIAGlc of the enzyme that synthesizes cAMP
requires cAMP. (adenylate cyclase) (Fig. 17.9). Mutations in
3. The model also explains how PTS carbohy- enzyme I or HPr, or the addition of glucose to
drates in addition to glucose can also depress wild-type cells, should lower the amounts of
the entry of non-PTS sugars and inhibit the P–IIAGlc and thus cause a decrease in the levels
expression of cAMP-dependent genes. The of cAMP.
uptake of PTS carbohydrates would be Of course, it is possible that the inhibition
expected to draw phosphoryl groups away of the adenylate cyclase is due to an increase
from P–HPr toward the sugars. This should in amounts of IIAGlc rather than to a decrease
decrease the phosphorylation state of IIAGlc, in levels of P–IIAGlc. However, mutations that
which is phosphorylated by P–HPr. Because result in lowered IIAGlc activity (crr mutants) do
the phosphorylation of IIAGlc by P–HPr is not relieve the inhibition of adenylate cylase by
reversible, phosphate would be expected to glucose. Also, it was reported in the mid-1970s
flow from P–IIAGlc to the PTS carbohydrates. that the addition of PTS carbohydrates to E. coli
The subsequent increase in IIAGlc would be cells that had been made permeable with tolu-
expected to inhibit the enzymes required ene inhibited adenylate cyclase activity, which
for uptake and metabolism of non-PTS according to the current model can be attributed
www.ebook777.com
to the lowering of the levels of P–IIAGlc by the cells of endogenous energy reserves whose
PTS carbohydrates.48,49 However, it should be metabolism produces ATP via substrate-
emphasized that the evidence for stimulation of level phosphorylation and (2) add an inhibi-
adenylate cyclase by P–IIAGlc is thus far indirect, tor of substrate-level phosphorylation (e.g.,
and based primarily on genetic evidence. What is arsenate). (See note 52.) Conversely, one
lacking is direct evidence that P–EIIGlc binds to the can increase the production of ATP from
adenylate cyclase and/or stimulates its activity. substrate-level phosphorylation in starved
Mutants defective in HPr are also defective cells by feeding them an energy source (e.g.,
in the uptake of certain non-PTS sugars. The glucose.) However, one must be careful.
model explains this by postulating that IIAGlc Glucose can also stimulate respiration, and
inhibits the carriers for these sugars. The evi- respiration can be a source of Δp. Therefore
dence for this is abundant. Uptake in HPr to stimulate substrate-level phosphoryla-
mutants is restored by a mutation in the gene tion with glucose without encouraging res-
for IIAGlc, which results in lowered activity of piration, respiration should be prevented by
IIAGlc (crr mutants). Also, IIAGlc is capable of using a respiratory inhibitor (e.g., cyanide)
binding to the lactose carrier (permease) and of or by incubating the cells anaerobically.
using liposomes made with the lactose permease 3. Perturbing the Δp How is the Δp manipu-
to inhibit the transport of the lactose analogue lated? One can use ionophores that collapse
thiomethylgalactoside (TMG).50,51 the proton potential. (However, see note
53.) (See Section 4.4 for a discussion of iono-
17.4 How to Determine the Source of phores.) It is also possible to increase elec-
trochemical potentials in de-energized cells
Energy for Transport
(e.g., starved cells, or cells in vesicles). For
Methods are available to distinguish whether example, if there is a high K+ concentration
bacterial transport is driven by the electrochem- in the cells, the addition of valinomycin will
ical proton potential (Δp) or by ATP. produce a potassium diffusion potential.
1. Inactivating the ATP synthase To determine (See Fig. 4.5.) Also, the addition of certain
whether the source of energy is ATP hydrolysis substrates that feed electrons directly into
or the Δp, it is necessary to isolate these sources the electron transport chain (e.g., D-lactate
of energy from each other and to perturb them or succinate) will produce a proton poten-
independently (i.e., to increase and decrease tial without increasing substrate-level
the ATP and Δp levels independently of each phosphorylation. A summary of some of
other). Since bacteria use ATP synthase to inter- these methods is presented in Table 17.1.
convert the proton potential and ATP, one can- However, caution must be observed in the
not vary the ATP levels and Δp independently
while the ATP synthase is functioning. To cir- Table 17.1 Summary of data that can distinguish
cumvent this problem, investigators either use between energy sources for transport using whole
mutants defective in the ATP synthase (“unc” cells
mutants) or add inhibitors of the ATP syn-
thase [e.g., N,N′-dicyclohexylcarbodiimide Effect on transport Δp ATP
(DCCD)]. Once this has been done, the ATP
Stimulated by
or the Δp can be decreased or increased inde- Glucose Noa Yes
pendently of the other. D-Lactate Yes No
2. Perturbing the intracellular levels of ATP Inhibited by
How are intracellular ATP levels manipu- DNPb Yes Noc
lated? Once the ATP synthase has been Arsenate No Yesd
inactivated, bacteria must rely completely a
Respiration must be inhibited.
b
on substrate-level phosphorylation to syn- One can use any combination of ionophores that abol-
thesize ATP, and the problem becomes one ishes both the ΔpH and the ΔΨ.
c
The ATP synthase must be inhibited.
of interfering with substrate-level phospho- d
Arsenate prevents ATP formation via substrate-level
rylation. To lower the levels of substrate- phosphorylation. This is the only source of ATP when the
level phosphorylation, one can (1) starve the ATP synthase is not operating.
interpretation of data derived from the use toxic substance coupled to influx of protons),
of inhibitors on whole cells because of indi- or in some cases sodium antiport, via the pro-
rect effects of the inhibitors.22,44 Also, ATP ton or sodium electrochemical potential. The
cannot be distinguished from other high- proton antiporter drug efflux systems are the
energy molecules in equilibrium with ATP major systems for exporting drugs in bacteria.
(e.g., PEP) when one is studying whole cells. However, sodium antiporters are also impor-
Ideally, one should incorporate the car- tant and have been discovered in several bac-
rier into proteoliposomes and directly test teria, including Escherichia coli, Neisseria
potential energy sources as was described in gonorrhoeae, N. meningitidis, Vibrio chol-
Section 17.3.3 for the histidine permease. erae, and V. parahaemolyticus. The sodium
antiporter drug efflux systems are called the
17.5 Drug-Export Systems MATE (multidrug and toxic compound extru-
For a review of drug-export systems, consult sion) systems. The primary transport systems,
refs. 54 through 57. Drug-export systems are which are energized by ATP hydrolysis, are
an important way by which bacteria and fungi present in prokaryotes and are the dominant
become resistant to antimicrobial agents, such drug efflux systems in eukaryotes, including
as antibiotics, dyes, detergents, disinfectants, pathogenic fungi and parasitic protozoa. The
and antiseptics. Pathogenic bacteria have a primary transport systems use an ATP-binding
variety of drug-export systems, and these are cassette (ABC) transporter, as described in
described in ref. 57. Other mechanisms of Section 17.3.3. All the transporters are inte-
resistance are described in note 58. Some drug- gral cell membrane proteins with several trans-
export systems (e.g., the system that exports membrane loops.
tetracycline antibiotics) are specific for the com- Drugs and certain other substances can be
pound transported. These are called dedicated exported without entering the periplasm by
drug-export systems. Most are not specific for means of a protein called the membrane fusion
the compound transported and export a range protein (MFP). The MFP, a periplasmic protein
of substances that are not necessarily structur- anchored to the cell membrane, links the cell
ally related. These are called multidrug resis- membrane-bound transporter with an outer
tance (MDR) transport systems. membrane protein that facilitates exit through
MDR transport systems exist in gram-neg- the outer membrane.60 It can be thought of as
ative and gram-positive bacteria, and in fungi, part of a channel bridging the periplasm. An
such as the pathogenic yeast Candida albicans. example of an MFP system is the type I protein
An example of an MDR system is the transporter secretion system diagrammed in Chapter 18
associated with antigen processing (TAP) sys- (see Fig. 18.4).
tem in Mycobacterium tuberculosis. The TAP As an example of a multidrug system that
system pumps out both aminoglycoside and uses an MFP, and is powered by the Δp, con-
tetracycline antibiotics. Other MDR systems sider the AcrAB–TolC system in Escherichia
pump out disinfectants and antiseptic com- coli, which is the major system for multiple
pounds as well as antibiotics. Here is an inter- drug resistance in E. coli.61 (The homologue in
esting question that the student might consider: Pseudomonas aeruginosa is MexAB–OprM.)
How is it that a single transporter can recognize The AcrAB–TolC system pumps out dyes,
and transport such a variety of chemically unre- detergents, bile salts, and antibiotics such as
lated compounds? This is discussed in ref. 59. tetracycline, chloramphenicol, β-lactams,
As one might imagine, bacteria with these MDR novobiocin, and erythromycin, all coupled
systems pose a severe problem in hospital set- to the influx of protons. AcrB is the cell mem-
tings and are responsible for nosocomial infec- brane transporter, AcrA is the periplasmic
tions (infections acquired in hospitals). MFP, and TolC is a transperiplasmic and outer
The two major classes of MDR systems, dis- membrane “channel” protein. Figure 17.10
tinguished from each other by the source of presents a model for how the AcrAB–TolC
energy for transport, are called primary and complex captures drugs laterally from the
secondary. The secondary transport systems outer leaflet of the inner membrane and pumps
are energized by proton antiport (efflux of them out of the cell.
www.ebook777.com
Amphiphilic
Drug
Porin OM Channel (TolC)

Outer
Lea et
(LPS)
OM Inner
Lea et
Peri- MFP (AcrA)

plasm
IM
RND Ef ux Transporter (AcrB)
Fig. 17.10 The AcrAB–TolC drug efflux system in Escherichia coli, where AcrB is the cell membrane trans-
porter, AcrA is the membrane fusion protein, and TolC is the outer membrane channel. The model shows
an amphiphilic drug entering the periplasm through a porin; the lipophilic portion is solid black. The drug
then partitions into the lipid phase of the inner membrane and from there travels into AcrB, from where it
is pumped out of the cell through TolC. Thus, the drug can be removed from the periplasm and need not
enter the cytoplasm. Abbreviations: OM, outer membrane; IM, inner membrane; MFP, membrane fusion
protein; LPS, lipopolysaccharide; RND, resistance–nodulation–division. Source: Yu, E. W., J. R. Aires, and
H. Nikaido. 2003. AcrB multidrug efflux pump of Escherichia coli: composite substrate-binding cavity of
exceptional flexibility generates its extremely wide substrate specificity. J. Bacteriol. 185:5657–5664.
(There is evidence that certain drugs, such 17.7 Summary

as aminoglycosides, can be captured from To bring solutes into or out of the cell, bacteria
the periplasm without entering the cell mem- employ several different types of energy-de-
brane.62) TolC is suggested to be a tunnel that pendent transport system. A single bacterium
extends through the outer membrane and at may have representatives of all the transport
least half the periplasmic space, where it is systems except the shock-sensitive systems,
thought to link up with the periplasmic domain which are present only in gram-negative bac-
of AcrAB. For more discussion about TolC, see teria. Transport systems may be classified as
the discussion of the type I pathway for protein being either primary or secondary, depending
secretion in Section 18.5.1. on how they are coupled to the energy source.
The difference is that primary transport
17.6 Bacterial Transport Systems in systems are directly coupled to an energy-
Summary generating reaction (e.g., ATP or PEP hydro-
Figure 17.11 summarizes the transport systems lysis), whereas secondary transport systems
used by bacteria. These include primary trans- are energized by an existing electrochemical
port systems (driven by chemical energy) and gradient (e.g., a proton or sodium ion gradi-
secondary transport systems (driven by elec- ent), itself having been produced by energy-
trochemical ion gradients). Note the role of the yielding reactions such as ATP hydrolysis or
proton circuit for secondary transport. Most redox reactions. That is, secondary transport
uptake transport is in symport with protons, is indirectly coupled to an energy-yielding
although several symporters use Na+ instead of chemical reaction by ion currents. (This is
H+. The return of the Na+ to the extracellular one of the predictions of the chemiosmotic
space requires a proton circuit in most bacteria. theory.) Active transport is defined as primary
ATP synthase (translocates protons), the Kdp

K+–ATPase in E. coli (K+ influx), and the mul-
tidrug export systems that use an ATP-binding
cassette (ABC) system.
The transport of carbohydrates via the
sugar–phosphotransferase system driven by
phosphoenolpyruvate is also a primary trans-
port system. It is widespread in bacteria, being
found in anaerobic genera and in facultative
anaerobes. It is absent from strict aerobes, how-
ever, and from eukaryotes. This is not an active
transport mode, since the sugar is phosphory-
lated during transport, and therefore the solute
accumulates in altered form. Sometimes this
type of transport is called group translocation.
Transport by the PTS results in the inhibition
of transport of non-PTS sugars and the inhibi-
tion of adenylate cyclase, leading to catabolite
repression by glucose. This can be explained in
terms of the regulatory roles of AGlc (IIIGlc) and
P–AGlc (P–IIIGlc).
There exist other catabolite repression
systems besides the one mediated by IIAGlc.
Fig. 17.11 Summary of transport systems found in
the Bacteria. Any single bacterium uses diverse trans-
Enteric bacteria have a cAMP-independent
port systems to take up sugars, amino acids, ions, catabolite repression system mediated by FruR,
vitamins, organic acids, and so on. Many of these which is also called Cra. FruR is also an acti-
systems are powered by electrochemical energy (ion vator of certain genes. Thus FruR represses
gradients). They catalyze symport, antiport, and enzymes required for sugar catabolism and
uniport. There are also transport systems that use activates enzymes required for gluconeogen-
chemical energy instead of electrochemical energy. esis. Glycolytic intermediates such as fructose-
An example of the latter is the PTS, which is specific 1,6-bisphosphate remove FruR from the FruR-
for carbohydrates, is driven by phosphoenolpyru- regulated operons, resulting in repression of
vate, and accumulates the sugar as the phosphory- gluconeogenic genes and stimulation of gly-
lated derivative. Other chemically driven transport
colytic genes. Gram-positive bacteria have
systems use ATP or a high-energy molecule in equi-
librium with ATP. These include shock-sensitive
a catabolite repression system mediated by
transport systems (periplasmic permeases) that are HPr(Ser-P), which combines with the protein
found in gram-negative bacteria, and other chemi- CcpA to repress catabolite-repressible genes.
cally driven uptake systems found in both gram-neg- Hpr(Ser-P) is a phosphorylated form of HPr
ative and gram-positive bacteria. made by a special ATP-dependent kinase that
is stimulated by fructose-1,6-bisphosphate. All
these catabolite repression systems involve the
transport in which the solute accumulates in a repression of genes by carbohydrates.
chemically unmodified form. There also exist catabolite repression sys-
ATP hydrolysis drives some primary trans- tems in which genes are repressed when cells
port systems, including osmotic shock-sensitive are grown on organic acids, rather than car-
(periplasmic permeases) systems, that are pres- bohydrates. As discussed in Section 2.2.5,
ent in gram-negative bacteria and transport a when several obligately aerobic bacteria (e.g.,
range of solutes similar to those systems driven Rhizobium) are given a mixture of glucose and
by electrochemical gradients. Primary transport the organic acid, they will grow first on organic
systems consist of a periplasmic binding protein acids. Clearly systems of catabolite repression
plus three or four membrane proteins. Primary mediated by intracellular signals remain to be
transport systems utilizing ATP include the discovered.
www.ebook777.com
Secondary transport systems require ion are also defective in the transport of some
symport, antiport, or uniport, and they make sugars that do not use the PTS?
use of electrochemical ion gradients to trans-
3. What are proteoliposomes, and how are
port a diverse range of solutes including sug-
they prepared?
ars, amino acids, and ions. The coupling ions
are either protons or sodium ions, depending 4. Solute transport might be driven by the Δp
upon the transporter. This type of transport is or by ATP. Describe some experiments that
found in bacteria, fungi, plants, protozoa, and could distinguish between the two sources
higher eukaryotes. It is the simplest of the trans- of energy. (Hint: You will have to manipu-
port systems and consists of a single membrane- late the ATP and Δp separately, and you
spanning protein. might want to use proteoliposomes for
The type of transport system used by a bacte- some of your experiments.)
rium depends upon the organism, not the solute
5. What is the procedure for inducing “osmotic
being transported. For example, Pseudomonas
shock”? What is the cellular location of the
(a strict aerobe) actively transports glucose by
proteins released by osmotic shock? What
means of symport with H+, whereas the fac-
are the functions of some of the proteins
ultative anaerobe E. coli transports the same
released by osmotic shock?
sugar via the phosphotransferase system.
Furthermore, a single bacterium can use several
different types of transport for the same class of
compound. For example, E. coli uses PTS for
some sugars but uses H+ symport, Na+ symport, 1. Brooker, R. J. 1990. The lactose permease of
or ATP for others. Escherichia coli. Res. Microbiol. 141:309–316.
Harold has speculated regarding the diversity 2. Racker, E., B. Violand, S. O’Neal, M. Alfonzo,
of transport systems in the bacteria, pointing and J. Telford. 1979. Reconstitution, a way of bio-
out that whereas transport systems coupled to chemical research: some new approaches to mem-
electrochemical gradients have the advantage of brane-bound enzymes. Arch. Biochem. Biophys.
198:470–477.
being simple in composition, they are limited by
the Δp with respect to the concentration gradients 3. Viitanen, P., M. Garcia, and H. Kaback. 1984.
Purified reconstituted lac carrier protein from
that can be attained.63 Furthermore, they oper- Escherichia coli is fully functional. Proc. Natl. Acad.
ate close to equilibrium and can theoretically be Sci. USA 81:1629–1633.
reversed if the Δp suffers a transient decrease (e.g.,
4. Maloney, P. C., V. Anantharam, and M. J. Allison.
during starvation). On the other hand, transport 1992. Measurement of the substrate dissociation
systems powered by ATP or PEP are structurally constant of a solubilized membrane carrier. J. Biol.
more complex but have the advantage of being Chem. 267:10531–10536.
driven unidirectionally by the relatively large free 5. Taglich, D., E. Padan, and S. Schuldiner. 1991.
energies available in ATP and PEP. Overproduction and purification of a functional Na+/
H+ antiporter coded by nhaA (ant) from Escherichia
coli. J. Biol. Chem. 266:11289–11294.
6. Bishop, L., R. Agbayani Jr., S. V. Ambudkar, P. C.
Study Questions Maloney, and G. F.-L. Ames. 1989. Reconstitution
of a bacterial periplasmic permease in proteolipo-
1. Suppose the transport of X is driven by somes and demonstration of ATP hydrolysis con-
symport with H+ in a 1:1 ratio. Assume that comitant with transport. Proc. Natl. Acad. Sci. USA
X is not charged, the ΔΨ is –120 mV, and 86:6953–6957.
the ΔpH is 1. What is the expected Xin/Xout? 7. The proton gradient is created by a primary trans-
What would be the answer if X were X–? If port system. Some bacteria can create sodium gradi-
ents with a primary transport system, but most create
X were X– and the ΔpH were 2? a sodium gradient by converting a proton gradient
ans. 1,000, 10, 100 into a sodium ion gradient by using a H+/Na+ anti-
porter.
2. What is the explanation of the curious fact 8. Law, C. J., P. C. Maloney, and Da-Neng Wang.
that mutants in the phosphotransferase sys- 2008. Ins and outs of major facilitator superfamily
tem that are defective in enzyme I or HPr antiporters. Annu. Rev. Microbiol. 62:289–305.
9. Wright, J. K., R. Seckler, and P. Overath. 1986. 21. Ames, G. F.-L. 1986. Bacterial periplasmic
Molecular aspects of sugar:ion cotransport. Annu. transport systems: structure, mechanism, and evolu-
Rev. Biochem. 55:225–248. tion. Annu. Rev. Biochem. 55:397–425.
10. Bacteria, yeast, and plants use primarily the pro- 22. Ames, G. F.-L., and A. K. Joshi. 1990. Energy
ton as the coupling ion for symport, whereas animals coupling in bacterial periplasmic permeases. J.
rely on the sodium ion. This may be because the cell Bacteriol. 172:4133–4137.
membranes of most bacteria, fungi, and plants create 23. Furlong, C. E. 1987. Osmotic-shock-sensitive
a proton potential by pumping protons out of the cell transport systems, pp. 768–796. In: Escherichia coli
via a proton-translocating ATPase, whereas the cell and Salmonella typhimurium: Cellular and Molecular
membranes of animal cells create a sodium potential Biology, Vol. 1. F. C. Neidhardt et al. (Eds.). ASM
by pumping sodium ions out of the cell via a Na+/ Press, Washington, DC.
K+-ATPase, which exchanges Na+ for K+.
24. Fath, M. J., and R. Kolter. 1993. ABC transport-
11. Hirota, N., and Y. Imae. 1983. Na+-driven fla- ers: bacterial exporters. Microbiol. Rev. 57:995–
gellar motors of an alkalophilic Bacillus strain YN-1, 1017.
J. Biol. Chem. 258:10577–10581.
25. Davidson, A. L., and J. Chen. 2004. ATP-
12. Imae, Y., and T. Atsumi. 1989. Na+-driven binding cassette transporters in bacteria. Annu. Rev.
bacterial flagellar motors. J. Bioenerg. Biomemb. Biochem. 73:241–268.
21:705–716.
26. Davidson, A. L., E. Dassa, C. Orelle, and J. Chen.
13. Maloy, S. R. 1990. Sodium-coupled cotrans- 2008. Structure, function, and evolution of bacterial
port, pp. 203–224. In: The Bacteria, Vol. XII. T. A. ATP-binding cassette systems. Microbiol. Mol. Biol.
Krulwich (Ed.). Academic Press, New York. Rev. 72:317–364.
14. R = 8.314 4 J K–1mol–1; T = 273.16 K + °C; to con- 27. In the specific case of the histidine permease
vert natural logarithms to log10, multiply by 2.303. system, the transporter consists of two hydropho-
bic proteins, HisQ and HisM, which span the mem-
15. Wilson, D. M., T. Tsuchiya, and T. H. Wilson. brane, and two identical hydrophilic proteins, HisP.
1986. Methods for the study of the melibiose car- The HisP protein may be a peripheral membrane pro-
rier of Escherichia coli, pp. 377–387. In: Methods tein bound to the inner surface of the membrane, or
in Enzymology. S. Fleischer and B. Fleischer (Eds.), it may span the membrane, being separated from the
Vol. 125. Academic Press, New York. hydrophobic lipids by the HisQ and HisM proteins.
16. The negative charge on SCN– is delocalized over The HisP protein binds ATP and is responsible for
the three atoms of the molecule, and this allows it to ATP hydrolysis. The periplasmic histidine binding
penetrate the lipid bilayer. protein is called HisJ.
28. Direct evidence that ATP is the source of energy
17. West, I. C. 1970. Lactose transport coupled to
for the transport of histidine via an ABC transporter
proton movements in Escherichia coli. Biochem.
was obtained by Bishop et al. (see ref. 6). It is instruc-
Biophys. Res. Commun. 41:655–661.
tive to learn how it was done. The investigators
18. Gram-negative cells are shocked in the following extracted the proteins, including the histidine carrier
way. They are first suspended in a hypertonic solution proteins, from the membranes of E. coli and incor-
of Tris buffer, EDTA, and sucrose. Such treatment porated these proteins into proteoliposomes. (See
removes much of the supply of divalent metal cations Section 17.1 for a description of proteoliposomes.)
that are holding the lipopolysaccharide together, The proteoliposome vesicles were sometimes loaded
along with the lipopolysaccharide, and plasmo- with ATP by including the ATP in the dilution step.
lyzes the cells. Then the cells are rapidly diluted into When the reconstituted proteoliposomes were incu-
water or dilute MgCl2 to neutralize the EDTA. This bated with the histidine-binding protein (HisJ) and
results in the release of the periplasmic proteins. The histidine, they transported histidine, but only when
treatment inhibits cellular functions that depend on ATP was present (Fig. 17.7). The transport of histi-
periplasmic binding proteins (e.g., ATP-dependent dine was not affected by the ionophores valinomycin
transport of sugars and amino acids). Other cellu- (and K+), nigericin, or the proton ionophore FCCP,
lar functions, including other transport systems, are all of which collapse the electrochemical potentials.
retained. (Ionophores and their physiological activities are
described in Section 4.4.) These experiments dem-
19. Ames, G. F.-L. 1988. Structure and mecha- onstrated that histidine uptake via its periplasmic
nism of bacterial periplasmic transport systems. J. binding protein is driven by ATP and not an electro-
Bioenerg. Biomembr. 20:1–18. chemical potential.
20. Ames, G. F.-L. 1990. Energetics of periplasmic 29. Boos, W., and J. M. Lucht. 1996. Periplasmic
transport systems, pp. 225–245. In: The Bacteria, binding protein-dependent ABC transporters. pp.
Vol. 12. T. A. Krulwich (Ed.). Academic Press, New 1175–1209. In: Escherichia coli and Salmonella:
York. Cellular and Molecular Biology, Vol. 1, 2nd ed. F.
www.ebook777.com
C. Neidhardt et al. (Eds.). ASM Press, Washington, pp. 273–299. In: The Bacteria, Vol. XII. T. A.
DC. Krulwich (Ed.). Academic Press, New York.
30. Epstein, W., and L. Laimins. 1980. Potassium 43. Meadow, N. D., D. K. Fox, and S. Roseman.
transport in Escherichia coli: diverse systems with 1990. The bacterial phosphoenolpyruvate:glycose
common control by osmotic forces. Trends Biochem. phosphotransferase system. Annu. Rev. Biochem.
Sci. 5:21–23. 59:497–542.
31. Rosen, B. 1987. ATP-coupled solute trans- 44. Postma, P. W., J. W. Lengeler, and G. R. Jacobson.
port systems, pp. 760–767. In: Escherichia coli and 1993. Phosphoenolpyruvate:carbohydrate phos-
Salmonella typhimurium: Cellular and Molecular photransferase systems of bacteria. Microbiol. Rev.
Biology, Vol. 1. F. C. Neidhardt et al. (Eds.). ASM 57:543–594.
Press, Washington, DC.
45. Deutscher, J., C. Francke, and P. W. Postma.
32. Wood, J. M. 1999. Osmosensing by bacteria: 2006. How phosphotransferase system–related
signals and membrane-based sensors. Microbiol. protein phosphorylation regulates carbohydrate
Mol. Biol. Rev. 63:230–262. metabolism in bacteria. Microbiol. Mol. Biol. Rev.
70:939–1031.
33. Stewart, L. M. D., E. P. Bakker, and I. R. Booth.
1985. Energy coupling to K+ uptake via the Trk 46. Saier, M. H., Jr. 1989. Protein phosphor-
system in Escherichia coli: the role of ATP. J. Gen. ylation and allosteric control of inducer exclu-
Microbiol. 131:77–85. sion and catabolite repression by the bacterial
phosphoenolpyruvate:sugar phosphotransferase
34. Epstein, W. 1986. Osmoregulation by potas-
system. Microbiol. Rev. 53:109–120.
sium transport in Escherichia coli. FEMS Microbiol.
Rev. 39:73–78. 47. Inhibition of transport and metabolism of non-
PTS carbohydrates such as lactose, melibiose, malt-
35. Buurman, E. T., K.-T., Kim, and W. Epstein.
ose, and glycerol (the class I PTS carbohydrates) by
1995. Genetic evidence for two sequentially occu-
PTS carbohydrates is enhanced in mutants of E. coli
pied K+ binding sites in the Kdp transport ATPase. J.
that have less EI activity (leaky ptsl strains).
Biol. Chem. 270:6678–6685.
48. Harwood, J. P., C. Gazdar, C. Prasad, A.
36. Walderhaug, M. O., J. W. Polarek, P. Voelkner,
Peterkofsky, S. J. Curtis, and W. Epstein. 1976.
J. M. Daniel, J. E. Hesse, K. Altendorf, and W.
Involvement of the glucose enzymes II of the sugar
Epstein. 1992. KdpD and KdpE, proteins that con-
phosphotransferase system in the regulation of ade-
trol expression of the kdpABC operon, are members
nylate cyclase by glucose in Escherichia coli. J. Biol.
of the two-component sensor-effector class of regu-
Chem. 251:2462–2468.
lators. J. Bacteriol. 174:2152–2159.
49. Peterkofsky, A., and C. Gazdar. 1975. Interaction
37. Voelkner, P., W. Puppe, and K. Altendorf.
of enzyme I of the phosphoenolpyruvate:sugar
1993. Characterization of the KdpD protein, the
phosphotransferase system with adenylate cyclase
sensor kinase of the K+-translocating Kdp system of
of Escherichia coli. Proc. Natl. Acad. Sci. USA
Escherichia coli. Eur. J. Biochem. 217:1019–1026.
72:2920–2924.
38. Nakashima, K., H. Sugiura, H. Momoi, and T.
50. Nelson, S. O., J. K. Wright, and P. W. Postma.
Mizuno. 1992. Phosphotransfer signal transduc-
1983. The mechanism of inducer exclusion.
tion between two regulatory factors involved in the
Direct interaction between purified IIIGlc of the
osmoregulated kdp operon in Escherichia coli. Mol.
Microbiol. 6:1777–1784.
system and the lactose carrier of Escherichia coli.
39. Sugiura, A., K. Hirokawa, K. Nakashima, and EMBO J. 2:715–720.
T. Mizuno. 1994. Signal-sensing mechanisms of the
51. Osumi, T., and M. H. Saier Jr. 1982.
putative osmosensor KdpD in Escherichia coli. Mol.
Regulation of lactose permease activity by the
40. Sugiura, A., K. Nakashima, K. T. Tanaka, and system: evidence for direct binding of the glucose-
T. Mizuno. 1992. Clarification of the structural and specific enzyme III to the lactose permease. Proc.
functional features of the osmoregulated kdp operon Natl. Acad. Sci. USA 79:1457–1461.
of Escherichia coli. Mol. Microbiol. 6:1769–1776.
52. Arsenate can substitute for inorganic phosphate
41. Heermann, R., A. Fohrmann, K. Altendorf, in the synthesis of 1,3-diphosphoglycerate. The acyl
and K. Jung. 2003. The transmembrane domains of arsenate is quickly chemically hydrolyzed to the car-
the sensor kinase KdpD of Escherichia coli are not boxylic acid, and ATP is not made.
essential for sensing K+ limitation. Mol. Microbiol.
53. As long as the ATP synthase is inhibited, the ATP
47:839–848.
levels should not be affected. However, if the ATP
42. Saier, M. J., Jr., and A. M. Chin. 1990. Energetics synthase is functioning, then collapsing the Δp will
of the bacterial phosphotransferase system in sugar shift the equilibrium of the ATP synthase in the direc-
transport and the regulation of carbon metabolism, tion of ATP hydrolysis.
54. Borges-Walmsley, M. I., and A. R. Walmsley. become resistant to a drug if the drug target is altered
2001. The structure and function of drug pumps. by chromosomal genes. For example, bacteria resis-
Trends Microbiol. 9:71–79. tant to the antibiotic erythromycin, which ordinarily
binds to the 50S ribosomal subunit and inhibits pro-
55. Nikaido, H. 1996. Multidrug efflux pumps of
tein synthesis, may make an altered form of the 50S
gram-negative bacteria. J. Bacteriol. 178:5853–
subunit that does not bind to erythromycin. Bacteria
5859.
resistant to the semisynthetic antibiotic rifampin are
56. Putman, M., H. W. van Veen, and W. N. known to synthesize an altered RNA polymerase that
Konings. 2000. Molecular properties of bacterial does not bind rifampin. Another example of a meta-
multidrug transporters. Microbiol. Mol. Biol. Rev. bolic change that confers resistance is the mechanism
64:672–693. of resistance to sulfonamides, drugs that inhibit the
synthesis of folic acid. Resistant bacteria develop a
57. Borges-Walmsley, M. I., K. S. McKeegan, and A.
pathway to take up preformed folic acid from the
R. Walmsley. 2003. Structure and function of efflux
medium. A third way for bacteria to become resistant
pumps that confer resistance to drugs. Biochem. J.
to a drug may entail the prevention, due to chromo-
376:313–338.
somal genes, of the entry of the drug into the cell. For
58. Bacteria become resistant to antimicrobial drugs example, increased resistance to penicillin may be
for a variety of reasons in addition to those due to due to alteration of outer membrane porin proteins,
drug-export systems. These include enzymatic alter- resulting in decreased entrance of penicillin.
ation of the drug. An example of enzymatic inactiva-
tion of drugs is the acylation or phosphorylation of 59. Adler, J., O. Lewinson, and E. Bibi. 2004. Role
aminoglycoside antibiotics that in the unaltered form of a conserved membrane-embedded acidic residue
bind to the 30S ribosomal subunit and inhibit protein in the multidrug transporter MdfA. Biochemistry
synthesis. For example, resistance to the aminoglyco- 43:518–525.
side streptomycin can be due to phosphorylation of 60. Zgurskaya, H. I., and H. Nikaido. 1999. AcrA is
a hydroxyl group on the antibiotic. The aminoglyco- a highly asymmetric protein capable of spanning the
side kanamycin has a free amino group and is inac- periplasm. J. Mol. Biol. 285:409–420.
tivated by N-acetylation. Similarly, resistance to the
antibiotic chloramphenicol can be due to enzymatic 61. Yu, E. W., J. R. Aires, and H. Nikaido. 2003. A
acylation of hydroxyl groups in the chloramphenicol multidrug efflux pump of Escherichia coli: compos-
that can no longer bind to the 50S ribosomal subunit, ite substrate-binding cavity of exceptional flexibility
with the result that protein synthesis is inhibited. generates its extremely wide substrate specificity. J.
Another example of an enzymatic alteration that Bacteriol. 185:5657–5664.
destroys the activity of drugs is the opening of the 62. Lomovskaya, O., and M. Totrov. 2005.
β-lactam ring in penicillin and cephalosporin antibi- Vacuuming the periplasm. J. Bacteriol 187:1879–
otics by β-lactamase. In all these cases, the bacterium 1883.
usually harbors a plasmid that encodes an enzyme
that inactivates the drug, although in some cases the 63. Harold, F. M. 1986. The Vital Force: A Study of
encoding gene is chromosomal. Bacteria also may Bioenergetics. W. H. Freeman, San Francisco.
www.ebook777.com
18
Protein Transport and Secretion
For reviews of protein transport and secretion as export. If the protein is transported to the
in bacteria, the student is referred to ref. 1 extracellular medium, into another cell, or to
through 4. Many hundreds of different pro- the bacterial cell surface, the process is called
teins synthesized by cytosolic ribosomes are secretion. [The extracellular transport of non-
destined for transport to various cellular loca- proteinaceous compounds (e.g., end products
tions, such as the cell membrane, periplasm, of fermentation) is called excretion.]
outer envelope, and cell wall, or secretion out What is it about the structure of a protein that
of the cell. The traffic in transporting these pro- determines whether it will be translocated at all,
teins to their correct locations is considerable, and if so, to what place? What mechanisms are
and mechanisms must exist to ensure that each responsible for the translocation of exportable
protein goes to its correct final destination. proteins through the cell membrane, which is
For example, cell membranes alone contain generally nonpermeable to proteins? What is
approximately 300 different proteins. In gram- the source of energy for protein translocation?
negative bacteria there may be 100 various These are important areas of research not only
proteins in the periplasm. The outer envelope in prokaryotes but in eukaryotes as well, where
of gram-negative bacteria is the site of perhaps proteins synthesized on cytosolic ribosomes are
50 different proteins. In addition, both gram- transported to different cell compartments such
negative and gram-positive bacteria transport as mitochondria and chloroplasts, or are secreted
proteins that are part of surface layers (glyco- out of the cell via the endoplasmic reticulum.
calyx) and appendages (flagella, fibrils, pili), as By far the majority of the research on bac-
well as extracellular hydrolytic enzymes (e.g., terial protein translocation through the cell
proteases, lipases, nucleases, saccharidases). membrane has been with Escherichia coli,
Pathogenic bacteria may also secrete protein which has served as a model system primarily
toxins that adversely affect host cells. The toxins because of the ease of genetic manipulation.
may be secreted into the extracellular medium Indeed, virtually all the proteins involved in
or directly into target cells. In addition to tox- E. coli protein translocation through the cell
ins, pathogenic bacteria may secrete hydrolytic membrane have been purified, and the genes
enzymes that degrade host connective tissue. have been cloned. This has allowed the for-
The degradation of host connective tissue pre- mulation of a model that describes the general
sumably facilitates the spread of the bacteria. features by which most proteins are translo-
We will begin with some definitions. The trans- cated through the cell membrane. It is called
port of proteins into or through a membrane the Sec system, or general secretory pathway
is referred to as translocation. If the protein (GSP). (Even though the Sec system is some-
is translocated into the periplasm of a gram- times referred to as the GSP system, we will
negative bacterium, the process is referred to see that it transports proteins into the cell
452
protein transport and secretion 453
membrane and periplasm, but not out of the chaperone protein, (3) a membrane-bound
cell.) Proteins to be translocated by the Sec complex of three proteins (called SecYEG),
system have an N-terminal signal peptide and (4) a cytoplasmic ATPase, called SecA,
that directs them to the membrane-bound Sec which is an ATP-driven motor that is bound
translocation apparatus, called SecYEG, and peripherally to SecYEG, which is a protein-
they are often bound to a chaperone protein, conducting channel in the membrane.5–11 The
for example, SecB, which prevents protein combination of SecYEG and SecA, referred
folding and brings the unfolded protein to to as the Sec translocase, is responsible for
the translocation apparatus. The proteins are moving the protein through the cytoplas-
translocated unfolded through the cell mem- mic membrane. The SecYE complex forms a
brane. As we shall see, the signal sequence is protein-conducting channel embedded in the
removed after translocation by a signal pepti- membrane sometimes referred to as the Sec
dase, which is a membrane-bound enzyme. translocon. SecG stimulates the transport.
A second general protein export pathway that Other membrane proteins that are involved
is Sec independent is called the twin-arginine in translocation (SecD, SecF, and yajC) play
translocation (Tat) protein export pathway a regulatory role as discussed in note 12 and
because the signal peptides on the proteins have described in refs. 13 through 15.
a highly conserved twin-arginine motif (Arg-Arg)
near the N–terminus. One of the significant differ- The leader peptide
ences between the Tat system and the Sec system Proteins that are translocated by the Sec sys-
is that the Tat system translocates folded proteins, tem are synthesized with a leader peptide (also
whereas the Sec system translocates unfolded pro- called leader sequence or signal sequence) at the
teins. The Tat system translocates cofactor-bound amino-terminal end that is removed during or
respiratory and photosynthetic electron transport after translocation. The leader peptide plays an
enzymes (redox proteins) to the periplasm, as important role in the Sec system. It is necessary
well as integral inner membrane proteins. The for attachment and insertion of the protein into
Sec system as well as the Tat system are found in the cell membrane. The protein with the leader
both gram-positive and gram-negative bacteria. peptide is called a precursor protein or a prepro-
We will also learn about the YidC translo- tein. If it is secreted to the outside, it is sometimes
case, which is also called an insertase because it called a presecretory protein. There is ample
inserts Sec-independent as well as Sec-dependent evidence that the leader peptide is necessary for
proteins into the cell membrane. the initial stages of translocation through the
There are several major secretion pathways membrane. For example, if the leader peptide is
that bacteria use to move proteins across the altered by amino acid substitutions, or deleted,
bacterial membranes, and these pathways are then translocation is impaired or does not occur
described in Section 18.5. The pathways that are at all.
discussed are concerned primarily with gram- So, what do we know about leader peptides?
negative bacteria, because the pathways describe Leader peptides have three regions: (1) a basic
how proteins are translocated across the inner region (positively charged near neutral pH) at the
membrane (cell membrane), the periplasm, extreme N-terminal end, called the N-domain
and the outer membrane. Some of these are Sec (1–3 positively charged amino acids), which
dependent, and some are Sec independent. attaches to the cytoplasmic side of the membrane,
When proteins are exported into the peri- perhaps to negatively charged phospholipids; (2)
plasm, they become folded there. Section 18.6 a central hydrophobic region (the core), called the
explains how this occurs. H-domain (10–15 amino acids), which inserts into
the membrane; and (3) an uncharged C-terminal
region, called the C-domain (3–7 amino acids),
18.1 The Sec System which contains a recognition site for a peptidase
18.1.1 The components that removes the leader peptide during or after
Before describing the model for protein export translocation (Fig. 18.1). The primary amino acid
by means of the Sec system (Section 18.1.2.), sequences of different leader peptides can vary sig-
we will introduce the four components of the nificantly. Although the leader peptide is impor-
system. These are (1) a leader peptide, (2) a tant for the initial stages of translocation through
www.ebook777.com
the membrane, it is clear that it does not determine A major chaperone protein in E. coli is SecB.
the final destination of the protein. Two reasons SecB binds to the mature domain (not the leader
for believing this are as follows. peptide) of the preprotein and prevents prema-
1. There are no obvious differences in ture folding in the cytoplasm and aggregation.
amino acid sequences between leader The SecB protein recognizes the unfolded topol-
peptides of periplasmic and outer mem- ogy of the proteins rather than specific amino
brane proteins. One would expect to acid sequences. The SecB protein has an addi-
find such differences if the leader peptide tional role: it binds to SecA at the membrane site
specified the final location of the protein. of translocation and delivers the preprotein to
2. Exchange of the leader peptide between SecA, which is a peripheral component of the
proteins destined for two different com- translocase. Although SecB is very important,
partments by means of recombinant DNA many proteins translocate by means of the Sec
technology does not influence the final des- system without requiring SecB; rather, they use
tination of the proteins. other chaperone proteins, described later.
Chaperone proteins The translocase: Sec A and SecYEG

When exportable proteins are synthesized, the The preprotein is transferred from SecB to SecA,
nascent preprotein binds to a soluble chaperone which is attached to the membrane-bound por-
protein upon leaving the ribosome. The chaper- tion of the translocase, SecYEG, which, as we shall
one protein brings the preprotein to the mem- see, is an integral membrane protein complex.16,17
brane and, very importantly, also prevents the It is not known exactly how the translocase moves
preprotein from assuming a tightly folded con- proteins through the membrane. For this discus-
figuration or aggregating into a complex that sion, we will assume that the translocase forms
cannot be translocated. The leader peptide is a hydrophilic channel for proteins. The forma-
very important in this regard because it retards tion of a hydrophilic channel can be rational-
the folding of the preprotein, giving the chaper- ized because many of the proteins secreted into
one protein time to bind. the external medium do not have significantly
Fig. 18.1 Leader peptides of exported proteins in E. coli. The leader sequence consists of a basic amino-terminal
end that has positively charged lysine and arginine residues, followed by a hydrophobic region. The boundary
between the hydrophilic basic region and the hydrophobic region is denoted with a vertical line. The cleavage
site for the leader peptide peptidase is at the carboxy-terminal end (∆). (a) Lipoprotein, located in outer mem-
brane. (b) Phage lambda receptor, located in outer membrane. (c) Maltose-binding protein located in periplasm.
(d) β-Lactamase, located in periplasm. (e) Arabinose-binding protein, located in periplasm. (f) fd Phage major
coat protein, located in cell membrane. (g) fd Phage minor coat protein, located in cell membrane. Note that
although the leader peptides share similar features with respect to charged and hydrophobic regions, the amino
acid sequences are not the same. Nor are there any differences that can distinguish between leader sequences of
outer membrane proteins, periplasmic proteins, and cell membrane proteins. Source: After Osborn, M. J., and H.
C. P. Wu. 1980. Proteins of the outer membrane of gram-negative bacteria. Annu. Rev. Microbiol. 34:369–422.
Reproduced, with permission, from the Annual Review of Microbiology, vol. 34, 1980, by Annual Reviews Inc.
hydrophobic regions that would facilitate their follows. The amino terminus of the leader
movement through the lipid bilayer. peptide leaves SecA and enters the mem-
The translocase consists of an integral membrane. When this happens, the positively
brane protein complex. The purified complex charged amino terminus binds to the nega-
is composed of three different polypeptides, tively charged phospholipid head groups
SecY, SecE, and SecG. The SecY protein spans or perhaps to the SecYEG complex and
the membrane several times and along with remains attached to the cytoplasmic side of
SecE is part of the protein-conducting channel. the membrane. The hydrophobic region of
(SecYE has been referred to as the Sec translo- the leader sequence spontaneously inserts
con because it is a protein-conducting channel.) into the membrane, forming a loop. The
Although SecG does not appear to be absolutely N terminus of the leader peptide remains
necessary for protein translocation, it is clearly on the cytoplasmic side of the membrane,
important. When SecG is incorporated into pro- possibly attached to the negatively charged
teoliposomes along with SecY and SecE it has phospholipid head groups, while the car-
a large stimulatory effect on protein transloca- boxy-terminal end “flips” into the lipid
tion, and antibodies against SecG inhibit protein bilayer as the preprotein enters the mem-
translocation in everted membrane vesicles.18 brane portion of the translocase (SecYEG).
It is envisaged that a channel opens up
18.1.2 A model for protein export within the translocase to accommodate
Protein export by the Sec system can be divided the protein. During these initial stages, a
into several stages, as illustrated in Fig. 18.2. For small segment of the preprotein (perhaps
a review of the mechanism of the Sec transport 20–30 amino acids) is translocated into
system, see ref. 17. As mentioned, before the the membrane. The precise mechanism by
protein is exported, it must bind to a chaperone which all this occurs is not known. Limited
protein, which brings the protein in an unfolded translocation occurs in vitro via nonhy-
state to the membrane-bound translocase. One drolyzable analogues of ATP, indicating
of the chaperone proteins is SecB, which brings that it is the binding of ATP rather than
many preproteins to the membrane for export its hydrolysis that provides the energy for
through the cytoplasmic membrane. translocation.22,23
Step 1. SecA binds to the lipid bilayer and to the Step 6. ATP is hydrolyzed, and this results in
protein trimer SecYEG in the cytoplasmic the release of the preprotein from SecA and
membrane. The combination of SecA and the release of SecA from the membrane.
SecYEG is called the translocase. After bind- Figure 18.2A indicates that both ADP and Pi
ing to SecYEG, SecA develops a high affinity are released from SecA as a consequence of the
for the preprotein–SecB complex. hydrolysis. However, it has been suggested
Step 2. SecB delivers the preprotein to SecA. The that only Pi is released and that ADP remains
SecB protein may be released at this step. bound and is exchanged for ATP during the
Step 3. After binding to the preprotein, SecA under- next cycle of SecA use. It is known that SecA
goes a conformational change that results in has two binding sites for ATP. See note 24.
the binding of ATP. Although not shown in Step 7. The rest of the protein is translocated,
Fig. 18.2, it may be that ATP exchanges for a driven by the ∆p. In agreement with this,
bound ADP that had been previously gener- translocation in respiring cells is immediately
ated by the ATPase activity of SecA. inhibited by uncouplers that dissipate the ∆p,
Steps 4 and 5. The binding of ATP causes trans- but the intracellular ATP levels do not change.
location of part of the preprotein through Furthermore, the direction of translocation
the translocase. It has been reported that a is reversed when the polarity of the ∆p is
portion of SecA also inserts into the mem- reversed in proteoliposomes translocating
brane at this time, and deinserts later upon an outer membrane protein (proOmpA).25
the hydrolysis of the ATP.19,20 (For a review However, it may also be that there occur
of the evidence for this, see note 21.) A fur- cycles of ATP-driven SecA binding and trans-
ther discussion of whether SecA becomes location followed by periods of ∆p-driven
inserted into the membrane can be found translocation after SecA is released. (See next
in ref. 17. The model for translocation is as subsection, entitled Recycling of SecA.)
www.ebook777.com
Fig. 18.2 Model for protein translocation in E. coli. (A) (1) The preprotein binds, either during or imme-
diately after its synthesis, to SecB, forming a preprotein–SecB complex, which moves to the SecA–SecY–E
membrane complex (the translocase) and binds to SecA. (2) The preprotein is transferred to SecA and SecB
is released. (3) SecA binds ATP. (4) The leader peptide leaves SecA and is inserted into the lipid bilayer as a
loop. (5) The carboxy terminus of the leader peptide “flips” to the periplasmic side, and the preprotein enters
the translocase channel. A short segment of the preprotein is translocated. (6) ATP is hydrolyzed and SecA is
released into the cytosol. (7) Translocation continues, but now driven by the ∆p alone. (8) The leader peptide
is cleaved, releasing the protein into the periplasm. (B) In the absence of a ∆p in vitro, there can be several
rounds of SecA binding and release, promoting ATP-dependent translocation: (a) binding SecA; (b) binding of
ATP to SecA accompanied by the translocation of a small segment; (c) hydrolysis of ATP and release of SecA;
(d) rebinding of SecA as the cycle continues (e, f). Symbols: A, SecA; B, SecB.
Step 8. The leader sequence is cleaved by a Recycling of SecA

leader peptidase, and the translocated The initial stages of translocation, which con-
protein is released into the periplasm. sist of the insertion of the leader peptide into the
The leader peptide must be removed from membrane and a limited amount of translocation
periplasmic and outer membrane proteins (steps 1–5), require SecA and ATP binding. ATP
for them to leave the membrane surface, but hydrolysis releases SecA (step 6), and the remain-
it need not be removed for translocation per der of translocation is driven by the Δp (step 7).
se to take place. According to this model, However, experiments have demonstrated, that
the initiation of translocation requires the in the absence of a Δp, ATP can drive transloca-
hydrolysis of ATP. The rest of translocation tion to completion in vitro. (Note 26 indicates
is driven by ∆p and does not require ATP. how this can be demonstrated.) This is because
However, see the discussion of the recycling SecA can rebind to the portion of the protein not
of SecA next. yet translocated and promote successive rounds
of ATP-dependent translocation (Fig. 18.2B, 18.2 The Translocation of Membrane-

steps a–d). For some proteins, it may be neces- Bound Proteins
sary to increase the SecA concentrations to dem-
18.2.1 Translocation of proteins whose
onstrate complete translocation in the absence of
final destination is the cytoplasmic
a Δp.27 The extent of the in vitro requirement for
membrane
the Δp apparently varies with the preprotein that
is used.28 Although SecA and ATP can be shown Thus far the description of protein transloca-
to drive translocation to completion in certain in tion has been confined to those proteins that
vitro experiments in which there is no Δp, the evi- pass through the cytoplasmic membrane (also
dence favors the conclusion that the major driving called inner membrane). What about pro-
force for proton translocation in vivo is the Δp, teins whose final destination is the cytoplas-
not ATP. (See the preceding description of step 7.) mic membrane? The export of several of the
The foregoing does not imply that SecA bind- cytoplasmic membrane proteins is also Sec
ing and rebinding may not be important beyond dependent and occurs in a similar fashion. The
the initiation stages of translocation. Cycles of question is, what keeps these proteins from
SecA rebinding and dissociation may, in fact, being translocated through the membrane into
occur in vivo, promoting limited translocation the periplasm?
in conjunction with translocation driven by the The answer lies in internal hydrophobic
∆p. One possible role for cycles of SecA bind- regions of the protein that stop translocation
ing and ATPase-dependent dissociation from and anchor the protein into the membrane
the preprotein as it is being translocated might because they bind to the lipid. Sometimes it is
be to unfold untranslocated regions that have the signal sequence itself that anchors the pro-
assumed a tight tertiary configuration that can- tein in the membrane (Fig. 18.3A, a). These sig-
not be threaded through the Sec translocase. nal sequences differ from those just discussed
One might expect such foldings to occur, since in not being recognized by the signal peptidase
protein secretion need not be coupled to trans- and are therefore not cleaved. At other times
lation, and the untranslocated regions of the it is an internal hydrophobic region called the
protein that are still in the cytosol may fold in a “signal–membrane anchor” or “stop-transfer”
way not suitable for translocation. sequence, that anchors the protein after the sig-
nal sequence has been cleaved (Fig. 18.3A, b).
Cotranslational translocation The “stop-transfer” regions have a stretch of 15
Polysomes translating presecretory proteins or more hydrophobic amino acids, which vary
associate with membranes, and it seems that from one protein to another. Proteins that span
some proteins are translocated while they are the membrane several times do so via a series
being translated, although in E. coli most proof alternating uncleaved signal sequences and
teins are exported post-translationally. The “stop-transfer” signals (Fig. 18.3A, c). (These
idea is that the ribosome translating a messen- include the membrane-bound solute transport-
ger RNA for a preprotein is so closely associ- ers discussed in Chapter 17.)
ated with the membrane that the polypeptide is
threaded through the Sec translocase (Sec YEG/ 18.2.2 Translocation of proteins whose
SecA) during translation. If the nascent poly- final destination is the gram-negative
peptide enters the cytosol before engaging the outer membrane (OM)
translocase, there might be a need for chaper- For a review of the assembly of the OM,
ones. Cotranslational translocation involves including the model for the insertion of outer
the signal recognition particle (SRP), which is membrane proteins (OMP) including porins
described later (Section 18.3). described next, see ref. 29. (Porins are described
in Chapter 1.) Most, if not all, OMPs are trans-
Protein translocation independent located across the inner membrane (IM) through
of the Sec proteins the SecYEG translocon. One model suggests
It would be misleading to imply that all proteins that after transport through the SecYEG trans-
translocated by E. coli require the Sec proteins. locon into the periplasm, the OMPs are bound
See the discussion of the Tat system, which to a periplasmic chaperone protein called Skp
translocates folded proteins, in Section 18.4. (seventeen-kilodalton protein) that prevents
www.ebook777.com
aggregation in the periplasm. The Skp/OMP of the OM. Some OMPs, such as porins, exist as
complex is then targeted to and bound to a pro- oligomers in the OM, and oligomerization may
tein complex called Omp85, which is in the lipid occur after insertion.
bilayer of the OM. Once the OMP becomes A model for the insertion of proteins that
bound to Omp85, Skp is released and the OMP span the membrane multiple times is shown
assumes proper folding, which is suggested to in Fig. 18.3B. A noncleavable signal sequence
be assisted by other periplasmic chaperone pro- initiates insertion into the membrane. At
teins. The folded OMP is then inserted via the some point, translocation stops as an inter-
Omp85 protein complex into the lipid bilayer nal hydrophobic “stop-transfer” signal enters
Fig. 18.3 Model for the Sec-dependent translocation of membrane proteins. (A) (a) Protein anchored to the
membrane by its leader peptide. (b) Protein anchored to the membrane by a “stop-transfer” signal. The leader
sequence has been proteolytically removed and is shown in the membrane. (c) Protein anchored to membrane
by alternating leader peptides and stop-transfer signals. (B) Insertion of a protein with four hydrophobic
domains. Domain 1 inserts into the membrane as a signal peptide. This opens the translocase channel and ini-
tiates translocation. When domain 2 (stop-transfer signal) enters the translocase, the putative channel opens
laterally, allowing domain 2 to diffuse laterally into the lipid matrix. The channel then closes. Domain 3 (sec-
ondary signal sequence) reinitiates translocation by inserting into the lipid bilayer and reopening the channel.
When domain 4 (stop-transfer signal) enters the translocase, it stops translocation and moves laterally out of
the translocase channel into the lipid matrix. Source: Adapted from Pugsley, A. P. 1993. The complete general
secretory pathway in gram-negative bacteria. Microbiol. Rev. 57:50–108.
the translocase. According to the model, the homologous to SRP and its receptor found in
translocase opens laterally when the “stop- eukaryotes.
transfer” signal enters, allowing the diffusion It has been speculated that the E. coli SRP
of the signal into the lipid bilayer. Then the functions in a manner similar to the function-
channel closes. A downstream secondary sig- ing of SRP in eukaryotes, and that protein
nal sequence reinitiates translocation, and the translocation and ribosomal translation take
process is repeated until the entire protein has place at the same time (cotranslational target-
been translocated into the membrane. This is ing). According to this model, SRP binds to the
not a well-understood process. hydrophobic N-terminal signal sequences as
An inner membrane protein called YidC has they emerge from the ribosome, and the com-
been suggested to play a role in the movement plex consisting of SRP and the ribosome-bound
of transmembrane segments of inner mem- nascent chain binds to the SRP receptor, FtsY,
brane proteins from the Sec translocase into which docks at the membrane and delivers
the lipid bilayer, as well as independently of the the nascent protein–ribosome complex to the
Sec translocase. YidC plays a role in the inser- SecYEG translocase. Translation continues
tion of a subset of membrane proteins.2,30,31 at the membrane. The newly synthesized pro-
Futhermore, as discussed in Section 18.4, some tein moves into the SecYEG translocase, and
integral membrane proteins are translocated from there into the membrane via the mem-
in a folded state by a Sec-independent system brane protein YidC. In some cases the pro-
called the Tat system. tein is delivered directly from SRP to YidC
without entering SecYEG. The SRP system
18.3 The E. coli SRP of cotranslational protein export is a means
of preventing the hydrophobic regions of the
The components required for the insertion newly synthesized membrane proteins from
of inner membrane proteins in E. coli are not self-aggregating in the aqueous cytoplasm. A
very well understood; more is known about the brief account of the role of SRP in eukaryotes
translocation of proteins into the periplasm. In appears in Box 18.1.
E. coli the placement of many of the inner mem-
brane proteins, especially those that do not have
extensive periplasmic loops, involves a “signal 18.4 Protein Translocation of Folded
recognition particle” (SRP) that delivers the Proteins: The Tat System
preprotein to SecYEG.32,33 Homologues of SRP There is a general protein export pathway
and its receptor are present in all prokaryotes that is Sec independent, post-translational,
thus far examined, and in all eukaryotes. The and driven by the ∆p; it translocates folded
E. coli SRP is a ribonucleoprotein particle that proteins, as opposed to the Sec system, which
consists of a 4.5S RNA and a 48 kDa protein translocates only unfolded proteins. It is called
subunit with GTPase activity. The term SRP the Tat (twin-arginine translocation) pro-
indicates that the particle recognizes a hydro- tein export pathway because the cleavable
phobic signal sequence in the protein that will N-terminal signal peptides on the proteins
be transferred into the membrane. These pro- have a highly conserved twin-arginine motif,
teins do not have a cleavable signal peptide, but which consists of a pair of consecutive argi-
instead have hydrophobic N-terminal signal nine residues (Ser-Arg-Arg-X-Phe-Leu-Lys:
sequences recognized by the SRP. The hydro- see note 35). The translocase is called the Tat
phobic regions become transmembrane helices translocase. It consists of integral membrane
in the membrane-integrated protein. A mem- proteins TatA, TatB, TatC, and TatE, and it
brane receptor for the SRP, FtsY, is also pres- translocates primarily cofactor-bound respi-
ent. Mutants lacking either SRP or FtsY are not ratory and photosynthetic electron transport
viable, although there is no profound effect on enzymes (redox proteins) to the periplasm, but
the export of proteins to the periplasm or outer also integral inner membrane proteins. Because
membrane. This is consistent with the conclu- these proteins often have cofactors such as
sion that SRP is primarily involved with the NADP+, FAD, molybdocofactors, iron–sulfur
assembly of certain inner membrane proteins. and iron–nickel clusters, or heme that are
(Reviewed in ref. 34.) Both SRP and FtsY are bound in the cytoplasm, they are folded in a
www.ebook777.com
BOX 18.1 SRP-DEPENDENT PROTEIN TRANSLOCATION ACROSS

THE ENDOPLASMIC RETICULUM MEMBRANE IN EUKARYOTES
In eukaryotic cells, proteins that are destined The translocase is a heterotrimeric ma-
for locations other than the cytosol, nucleus, chine called the Sec61 complex, which is
mitochondria, or chloroplasts are synthe- similar in some respects to the bacterial
sized on ribosomes attached to the endo- SecYEG. Meanwhile the SRP is released
plasmic reticulum (ER). The proteins are into the cytosol and can guide another
secreted into the lumen of the ER as they are ribosome–nascent polypeptide complex to
being translated; that is, secretion and trans- the ER.
lation are coupled. After being secreted into Eukaryotes also have a post-translational
the lumen of the ER, the proteins move to secretory pathway that is SRP independent.
the Golgi complex and from there to vari- It has been speculated that in eukaryotes the
ous destinations, such as the cell membrane, post-translational pathways come into play
secretory vesicles, or lysosomes. only if the SRP fails to bind to the nascent
It has been known since the 1970s that polypeptide. This can happen because
proteins destined to be translocated across the time of affinity of SRP for the signal
the ER membrane at their N-terminal ends sequence decreases as the polypeptide gets
have a signal sequence that is later removed longer, and thus there is only a short period
by a signal peptidase on the luminal side of during which SRP can bind effectively. As
the ER. This is analogous to the leader pep- in bacteria, the post-translational path-
tide found on the presecretory proteins in way appears to involve the use of chaper-
prokaryotes. (Some secreted or membrane ones analogous to SecB, which prevent the
proteins in eukaryotes, e.g., hen ovalbu- presecretory protein from folding into a
min, have a signal sequence that is internal conformation that prevents translocation.
rather than at the N-terminal end.) References 1 and 2 compare protein
Eukaryotes have a 16S ribonucleoprotein translocation in eukaryotes and prokary-
particle called the signal recognition particle otes (Bacteria and Archaea). The student is
(SRP), which binds to the signal sequence as encouraged to read these reviews for a more
the protein is being synthesized on cytosolic detailed account of the similarities and dif-
ribosomes and also binds to the ribosome ferences between the secretory pathways in
directly. The binding of SRP stops trans- eukaryotes and prokaryotes.
lation. The SRP–ribosome–nascent poly-
peptide complex then diffuses to the ER, REFERENCES
where the SRP binds to a receptor (docking
protein) and delivers the ribosome–nascent 1. Rapoport, T. A., B. Jungnickel, and U. Kutay.
1996. Protein transport across the eukaryotic
polypeptide complex to the translocase in
endoplasmic reticulum and bacterial inner mem-
the ER. Once the SRP has been removed branes. Annu. Rev. Biochem. 65:271–303.
from the ribosome, translation resumes,
2. Pholschröder, M., W. A. Prinz, E. Hartmann,
and the newly synthesized polypeptide is and J. Beckwith. 1997. Protein translocation in
translocated through the translocase as it is the three domains of life: variations on a theme.
being translated. Cell 91:563–566.
tertiary structure prior to translocation by the For a review of the Tat pathway, see refs. 36
Tat system. The evidence for this includes data through 39, Certain cofactor-containing pro-
demonstrating that mutants that fail to bind or teins, such as periplasmic c-type cytochromes,
synthesize a cofactor do not export the protein. are translocated unfolded by means of the Sec
system and acquire the cofactor after export. types), secreted by pathogenic strains of E.
However, most cofactor-containing proteins coli, proteases by Erwinia chrysanthemi and
use the Tat system. Pseudomonas aeruginosa, and leukotoxin
(toxin that damages white blood cells) by
18.5 Extracellular Protein Secretion Pasteurella haemolytica. These proteins (1) do
not have leader sequences, although the car-
A variety of different proteins are secreted out
boxyl terminus of the secreted proteins is rec-
of the cell, including hydrolytic enzymes such as
ognized by the secretion apparatus and (2) do
proteases, nucleases, lipases, and carbohydrases;
not require any of the sec genes for secretion. A
nonhydrolytic enzymes such as cholera toxin,
third characteristic of the system is that the pro-
diphtheria toxin, and pertussis toxin; structural
teins are secreted directly from the cytoplasm to
proteins such as pilin and flagellar proteins;
the outside of the cell without ever entering the
and virulence proteins secreted directly into
periplasm. This is done via a secretion appara-
host cells. The hydrolytic enzymes can generate
tus that spans the periplasm.
nutrient for the bacterium, and/or they can facil-
itate the invasion of plant or animal tissues. For
a description of what some of the nonhydrolytic The secretion apparatus
toxic proteins do to target cells, see note 40. The proteins required for the secretion appa-
There are at least eight secretion systems that ratus are encoded by secretory genes that have
have been reported for gram-negative bacteria. been identified by mutant analysis. The secre-
For a review see refs. 41 through 47. Seven of tory genes are usually contiguous with the struc-
these systems will be described here. Two of the tural gene for the protein that is secreted. Since
secretion systems are Sec independent, that is, the carboxyl terminus of the secreted protein is
independent of the GSP. These are the type I and required for secretion, it would appear that the
type III systems, and they will be described first. secretion apparatus recognizes this region of
Four of the secretion systems are Sec depen- the protein.
dent. These are the type II system, the type V The secretion apparatus consists of three
(also called autotransporter) system, and the secretory proteins. Two are inner membrane
closely related TPS system, and the chaperone/ proteins, and one is an outer membrane pro-
usher (CU) pathway, and these will be described tein. One of the inner membrane proteins is an
second. Another system, called the type IV sys- ATP-binding protein that belongs to the ABC
tem, can be Sec dependent or Sec independent, exporters (ATP-binding cassette) superfam-
depending upon the specific system, and this ily of proteins, which also catalyze extracellu-
will be described last. Very little is known about lar polysaccharide export, solute uptake, and
another system, called the extracellular nucle- drug export. (See Sections 12.3.1, 17.2.2, and
ation-precipitation (ENP) system, which is Sec 17.5.) This protein is sometimes referred to as
dependent, this system is not described, but it is the ABC transporter. ABC transporters actu-
reviewed in ref. 29. ally have two ATP-binding sites. (See the dis-
The Sec-dependent pathways work in two cussion of the structure of the ABC transporter
stages; stage I translocates proteins via the in Section 17.3.3.) The fact that the ABC trans-
Sec system across the cell membrane into the porter has ATP-binding sites indicates that ATP
periplasm, and stage II translocates the proteins binding and/or hydrolysis drives the secretion
from the periplasm across the outer envelope of the proteins, presumably by causing a confor-
to the outside. See Fig. 18.4 for an overview of mational change in the transporter. However, it
some of the secretion systems. Protein secretion is important to point out that the mechanism of
has been studied in gram-positive bacteria as translocation is not well understood.
well. These include penicillinase and the secre- The other two secretory proteins are called
tion of various proteases by Bacillus species. auxiliary proteins. One of these, the membrane
fusion protein (MFP), is anchored in the inner
18.5.1 The type I pathway membrane and has a periplasmic domain that
Examples of proteins secreted by the type I is believed to connect to the other auxiliary
system include α-hemolysin (HlyA, the toxin protein and to the outer membrane protein
that lyses red blood cells and kills other cell (OMP). The cellular location of the secretory
www.ebook777.com
proteins suggests that they may form a continu- In E. coli, the structural gene for hemolysin
ous channel through the inner and outer mem- is hlyA. It is located next to hlyB and hlyD,
brane, allowing proteins to move directly from the genes for the ABC transporter (HlyB) and
the cytoplasm to the external medium without the auxiliary inner membrane fusion protein
entering the periplasm. The three transporter (HlyD), respectively. The outer membrane
proteins in the different type I systems have dif- protein, TolC, is a channel that spans the
ferent names, but they are homologous proteins, outer membrane and extends deep into the
and the secretion systems can substitute for one periplasm; it is a narrow, tunnel-like structure,
another in mutants.48,49 There are actually fami- 100 Å long and 100 to 150 Å wide. (TolC is
lies of type I systems. Within a family there is reviewed in ref. 51.) TolC homologues are
sequence similarity between the components widespread among gram-negative bacteria and
of the secretion apparatus, and substitution play an important role in exporting proteins as
is possible. When components from different well as drugs and peptides. (See note 52 for a
families are interchanged, there is a significant brief summary.) For a model of TolC functions
drop in protein secretion. An illustrative examin the export of drugs, see Section 17.5 and Fig.
ple of a type I system, hemolysin secretion by 17.11. The TolC channel is open to the out-
Escherichia coli, is described next. side but is ordinarily closed at its periplasmic
entrance until a substrate to be transported is
Hemolysin secretion by E. coli brought to it by the translocation machinery.
Hemolysin toxin (HlyA) is an important viru- The role of the inner membrane fusion protein,
lence factor secreted by uropathogenic and HlyD, which has a periplasmic portion, is to
enterohemorrhagic E. coli. (EHEC). (See note provide contact between the ABC transporter
50 for a description of the diseases causd by in the inner membrane (HlyB) and the open-
EHEC.) HlyA inserts into eukaryotic cell mem- ing of the TolC channel in the periplasm. The
branes, creating pores that result in the leaking model shown in Fig. 18.4 shows that HlyD is
out of cytoplasmic contents and the death of the part of the transperiplasmic channel. Thus, the
cells. Reference to Fig. 18.4 will clarify the fol- ABC transporter (HlyB) is thought to transport
lowing discussion. the unfolded protein across the cell membrane
Fig. 18.4 Overview of types I, II, and III secretion systems. Type I is represented by hemolysin secretion by
E. coli. As pointed out in the text, TolC forms a channel in the outer membrane that projects as a tunnel into
at least 50% of the periplasmic space. Type II is represented by pulluninase secretion by Klebsiella oxytoca.
Type III is represented by Yop secretion by Yersinia. OM, outer membrane; PP, periplasm; IM, inner mem-
brane; CP, cytoplasm. SecB and Syc are cytoplasmic chaperone proteins. LspA is a periplasmic peptidase that
cleaves off the amino-terminal signal sequence. PulD and YscC (types II and III outer membrane proteins) are
homologous. Source: Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals
and plants. Microbiol. Mol. Biol. Rev. 62:379–433. Reproduced with permission from American Society for
Microbiology.
and into the periplasmic TolC tunnel via The secretion apparatus
HlyD. This is a very dynamic situation, and it The type III secretion apparatus is referred to as
has been suggested that the attachment of the an injectisome, and as discussed later is related
transporter apparatus (HlyD) to TolC takes to the flagellin secretion system. (See the sub-
place only when the substrate to be transported sections entitled Relationship to flagellin secre-
binds to the inner membrane ABC transporter tion apparatus.) The supramolecular structure
(HlyB); this in turn is believed to occur after the of the type III secretion apparatus is similar
membrane fusion protein (HlyD) and the ABC in the different bacteria and includes a hol-
transporter (HlyB) have interacted. A secretion low cylindrical basal body sometimes referred
signal in the C-terminal region of the hemolysin to as a syringe. The “syringe” is anchored in
recognizes the ABC transporter. It has been both the inner and outer membranes, and it
suggested that folding of the protein begins in spans the periplasm. (See the simplified draw-
the TolC tunnel. According to the model, the ing of the Yersinia type III system in Fig. 18.5.)
bridge between HylD and TolC collapses after Each “syringe” is attached to a needlelike
the protein has been secreted and is rebuilt for structure that protrudes from the bacterial cell
the next secretion event. Hydrolysis of ATP surface, and several of these are found on each
via the ABC transporter provides the energy to cell. This has been reported for Salmonella
drive the secretion of the protein. typhimurium, Shigella flexneri, and Yersinia
The gene (tolC) for the auxiliary outer mem- enterocolitica. A somewhat larger needlelike
brane secretory protein is not linked to the hly structure has been reported for enteropatho-
genes. This is an exception to the rule that all genic Escherichia coli (EPEC).55,62,63 When
three secretory genes are linked to the gene for the bacterium attaches to a target cell, it con-
the secreted protein, but TolC is a multifunc- structs a proteinaceous channel called a trans-
tional protein and is also used for the transport locon in the target cell plasma membrane.
of other molecules. Following recognition of The translocon is presumed to be the channel
the protein, the ABC transporter in the inner through which bacterial toxic proteins (called
membrane translocates the protein into the effectors) are injected into the cytosol of the
transperiplasmic channel, where folding might target cell. It is not known for certain whether
begin. the needle forms a continuous channel with the
translocon, or whether there is a discontinuity
18.5.2 The type III pathway between the tip of the needle and the translo-
Type III secretion systems (T3SSs) are Sec- con. (See ref. 59 and note 64 for a further dis-
independent systems in some gram-negative cussion of translocons and needles.)
animal pathogenic bacteria, including Yersinia
spp., Salmonella spp., Shigella spp., entero- Proteins secreted
pathogenic Escherichia coli (EPEC), Bordetella The secretion apparatus functions to secrete
pertussis, and Pseudomonas aeruginosa, and toxic proteins from the cytoplasm of the bac-
plant pathogens such as Pseudomonas syringae, terium across the inner and outer membranes
Ralstonia solanacearum, and Xanthomonas through the surface-protruding needle into the
campestris that form an apparatus used for the cytosol of target eukaryotic cells to which the
injection of virulence proteins directly into the pathogenic bacteria adhere. The toxic proteins
cytosol of eukaryotic host cells. The virulence secreted into the cytosol of the target cells are
proteins aid in the survival of the pathogen in called effector proteins, and there are several
the host. For a list of diseases caused by the ani- different types. Effector proteins interfere with
mal pathogens see note 53. For reviews, see refs. the defense mechanisms of the particular target
54 through 61. The archetype of the type III cell, including phagocytosis, the production of
systems is the Ysc (Yop secretion) system in toxic forms of oxygen, and the production of
Yersinia. Genes equivalent to the ysc genes are inflammatory cytokines, which are part of the
called the psc genes in Pseudomonas, the bsc host’s defense response to infection. The effec-
genes in Bordetella, the esc genes in EPECs, the tor proteins can also stimulate the entry into
hrc genes in plant pathogens, and the rhc genes nonphagocytic cells of certain pathogens, such
in Rhizobium. as Shigella and Salmonella, that can multiply
www.ebook777.com
phosphorylation YopO YopM
depolymerization YopE YopH dephosphorylation

of actin of signal transduction
protein
target cell contact
plasma membrane plasma membrane
outer membrane
inner membrane
bacterial cell
Fig. 18.5 Model for secretion and translocation into target cell of Yops proteins, the type III system. It is
suggested that the protein secretion apparatus in Yersinia, called Ysc, or the Ysc injectisome, forms a chan-
nel that crosses the inner and outer bacterial membranes. The entire secretion apparatus is very complicated
and consists of approximately 27 proteins of both the inner membrane and outer membrane varieties. When
there is no contact between Yersinia and the target cell, a plug, called YopN, closes the opening of the Ysc
channel. When the bacterium adheres to the target cell (but not necessarily directly to YopN), the channel is
no longer plugged. A hollow needle grows out from Ysc and penetrates the target cell. (For evidence of needle
formation, see ref. 63.) Certain Yop proteins (YopB, YopD, and LcrV) are injected through the needle into
the target cell plasma membrane, where they form a channel, called a translocon. Other injected Yop proteins
move via the needle through the translocon into the cytosol and cause several toxic effects. These include the
depolymerization of actin and inactivation of host cell signaling systems. Source: Adapted from Cornelis, G.
R., and H. Wolf-Watz. 1997. The Yersinia Yop virulon: a bacterial system for subverting eukaryotic cells.
Mol. Microbiol. 23:861–867.
intracellularly in host cells. (Not all pathogens and enteropathogenic Escherichia coli resem-
with type III systems are intracellular patho- bles that of purified flagella. This means that
gens. For example, enteropathogenic E. coli do the TTSS have a structure embedded in the
not enter the cells, but instead grow on the sur- inner and outer membrane resembling that of
face of the target cells.) For more information the flagellar basal body from which protrudes
about effector proteins, see note 65 and refs. 66 a “needle” that is analogous to the hook and
through 68. The entire apparatus, including the flagellar filament.
translocon, is a complex structure made from A model comparing flagella and the type
about 25 different proteins. III secretion system is shown in Fig. 18.6. An
additional point about the type III secretion
Relationship to flagellin secretion system is that one of the outer membrane pro-
apparatus teins of the type III secretion system is actually
Interestingly, it appears that perhaps the type III homologous to PulD, an outer membrane pro-
system evolved from the flagellin secretion sys- tein that is part of the type II secretion system.
tem. The reason for drawing this conclusion is (See Fig. 18.4.)
that most of the type III proteins that are located
in the bacterial cell membrane are homologues Summary of the common properties of the
to the flagellar biosynthetic proteins that are also type III systems
located in the cell membrane. In addition, elec- There is no leader sequence typical of the Sec
tron microscopic studies of the suprastructure pathway, and furthermore, there is no process-
of the type III systems in Salmonella typhimu- ing of the N-terminal end during secretion. There
rium, Shigella flexneri, Yersinia enterocolitica, are two hypotheses that attempt to explain the
Eukaryotic cell
YopB,
Plasma membrane
D
(d)
(b)
Needle/
pilus
Hook
Filament YopN
OM C
Basal body
F
D
IM FlhB U
P-R R-T
G FlhA LcrD
Q N
N I
(c)
(a) NTP
NTP
(RNA) Syc C
Flagellar biogenesis Type III secretion
Fig. 18.6 Models for flagellar and type III secretion systems. Many similarities between these systems have
been discovered by comparing the individual proteins, and also by examining the suprastructure of the type III
secretion apparatus. As discussed in Section 1.2.1, the flagellar basal body is anchored in the cell membrane,
and in the case of gram-negative bacteria it spans the periplasm as a hollow cylinder also anchored in the
outer membrane. The flagellin monomers are secreted through the central channel of the basal body into a
hollow flagellum filament that extends from the cell surface, and they are assembled into the flagellum at the
distal end. The suprastructure of the type III secretion systems has been visualized by electron microscopy.
A close resemblance to the basal body of the flagellum has been observed. Instead of a projecting flagellum,
however, the secretion apparatus has a hollow straight extension called a needle that projects from the cell
surface. Similar to the secretion of flagellin proteins, proteins secreted by the type III system are thought to
travel through the hollow needle extending from the cell surface. The needle is believed to insert into target
cells and deliver the proteins. In some cases (e.g., in Yersinia), a penetrating needle may form in certain media
only after the bacterium has made contact with the target cell. For a discussion of the conditions that allow
needle formation in Yersinia, see ref. 63. Source: Thanassi, D. G., and S. J. Hultgren. 2000. Multiple pathways
allow protein secretion across the bacterial outer membrane. Curr. Opin. Cell Biol. 12:420–430. Copyright
2000, with permission from Elsevier.
nature of the signal that directs the proteins to signal for secretion by the type III system
be secreted to the translocation machinery: (a) is actually in the 59 end of the mRNA that
the 59 end of the mRNA is the signal, and (b) the encodes the secreted protein, and that the 59
N-terminal 20 amino acids comprise the signal, mRNA targets the ribosome–RNA complex
and it functions with chaperone proteins that to the T3SS.69 On the other hand, it has been
bind to it and bring the presecretory proteins to pointed out that this would not apply to situ-
the T3SS. ations in which presynthesized proteins have
been known to be secreted.70
Hypothesis a. Because the amino acid sequences Hypothesis b. Research with Yersinia spp. has
at the N-terminal ends that determine secre- indicated that signals for secretion/transloca-
tion of proteins through the T3SS are not tion reside in the amino termini of secreted
very similar, some have suggested that the proteins. Gene fusion experiments have
www.ebook777.com
demonstrated that the N-terminal 15 to 17 Enteropathogenic E. coli (cells) have a similar

amino acids are able to direct the export of needlelike projection, except that it is a larger
heterologous reporter proteins through the filament.
T3SS into the culture supernate. However, We conclude our summary of the common
precisely how these signals operate is not properties of type III secretory systems by not-
understood, since there is relatively little ing that in the absence of binding of the bacte-
amino acid sequence similarity between the rium to the target cell, there is no secretion of
signals. As discussed next, there is evidence effector proteins. One of the reasons for this is
that chaperone proteins also take part in tar- that the channel (called an injectisome channel)
geting the presecretory proteins to the T3SS. is plugged, as described for Yersinia in the sub-
section that follows.
The following evidence that the chaperone
targets some presecretory proteins to the secre- Yersinia and its two type III secretion
tion apparatus has been presented by Lee and systems, Ysc and Ysa
Galán.71 There is a chaperone-binding domain The prototype type III secretion system is the
within approximately the first 140 amino acids Ysc system of pathogenic Yersinia spp., and we
of the presecretory protein, and in the absence turn now to some of the details of this particular
of these domains, the Salmonella typhimu- system.
rium SptP and SopE proteins are not secreted There are three pathogenic, or disease-caus-
through their cognate T3SS. Interestingly, ing, species of Yersinia. Yersinia pestis is the
they are instead secreted through the flagellar causative agent of plague. Both Y. enterocolitica
export pathway, which as described earlier and Y. pseudotuberculosis can cause severe gas-
is thought to be an ancestral precursor to the troenteritis, local abscess formation, and perito-
T3SS systems. nitis in humans. (See note 73 for a description
Is there a normally unused ancestral flagel- of these diseases.) All three species tend to grow
lar secretion signal in at least some presecretory primarily extracellularly in lymphatic tissue in
proteins destined for the T3SS? And does this the mammalian host, and they use the type III
signal operate to direct proteins to the flagel- system to secrete toxic proteins (the effector
lar secretion apparatus when the chaperone- proteins) into the cytosol of macrophages to pre-
binding domain is not present? The signals that vent phagocytosis and killing of the infectious
target the presecretory protein to the T3SS are bacteria by the macrophages. The bacteria thus
an important area of current research. Earlier evade the host defense mechanisms and multiply
publications using Yersinia have drawn differ- primarily extracellularly in lymphoid tissues.
ent conclusions from that just described for the However, they can also invade and replicate in
Salmonella SptP and SopE proteins; note 72 nonactivated macrophages. This is important
provide valuable further discussion of this very for the spread of the bacteria within the body, as
interesting area of research. described earlier (see note 72). The effector pro-
A second role for chaperone proteins has been teins secreted by type III systems into the target
proposed. It is suggested that they prevent pre- cells have cytotoxic effects. As an example, we
mature folding of the presecretory protein, thus will consider proteins of Yersinia’s outer mem-
keeping the protein in a nonglobular, secretion- brane, the Yops. (See refs. 74–76 for reviews.)
competent state that allows introduction of the The major type III system used by the patho-
protein into the secretion channel and move- genic Yersinia is called the Ysc (Yop secretion)
ment through the needle complex, as described system and is plasmid encoded. (For a recent
next. review of the Yop system, see ref. 77. Y. entero-
Proteins are secreted through both the inner colitica has recently been reported to possess a
and outer membranes, and into the cytosol of second contact-dependent type III secretion sys-
target cells. In pathogenic Yersinia spp., S. typh- tem, which associated with virulence. The sec-
imurium, and S. flexneri, a hollow cylindrical ond system, which is chromosomally encoded,
conduit resembling a needle is part of the secre- is called the Ysa system, and it secretes Ysps
tory apparatus and projects from the bacterial (Yersinia secreted proteins). The Ysa system is
surface. The needle inserts into the target cell. described in ref. 78.
Examine Fig. 18.6 again. It was reported that inflammation is important for the host defense
Y. enterocolitica forms needles that extend from against infections.) The student can refer to ref.
the secretion apparatus in the bacterial cell mem- 83 for a summary of the effects of the Yop pro-
brane, similar to those formed by S. typhimurium teins on host cells, as well as to the previously
and S. flexneri. It appears that the Yop proteins, cited review articles. A brief summary of these
including the translocon proteins (YopB, YopD, proteins and their functions is given in note 84.
and LcrV), are injected through the type III Note 85 discusses the mechanism of action of
apparatus into target cells through the needles, YopE and YopT.
which are made of the YscF protein. It has been
suggested that the needle is made upon adhesion 18.5.3 The type II pathway
of the bacterium to the target cell and penetrates The type II pathways appear to be the major
the target cell at that time, although needles route by which proteins are secreted by
can also be made when bacteria are growing in gram-negative bacteria, especially hydrolytic
certain media in the absence of target cells. See enzymes and toxins. Also, production of type
note 79 for experimental evidence in support of IV pili, which are responsible for adhesion and
this conclusion. When the needle inserts into the twitching motility, is dependent upon the type
plasma membrane of target cells, it encounters II pathway.
low Ca2+ levels that Lee et al. have suggested are
responsible for activating the injection of Yop Two stages
proteins.80 In extracellular host fluids the Ca2+ Type II pathways secrete proteins in two stages
levels are about 1.2 mM, but in intracellular flu- (Figs. 18.4 and 18.7). Stage 1 translocates the
ids they are only about 100 nM. protein into the periplasm through the SecYEG
Several proteins in Yersinia block secretion of translocase, and stage 2 secretes the protein
Yops prior to attachment to the target cell, even from the periplasm across the outer membrane
when calcium is present. HeLa cells can be used to the external medium. As mentioned, stage
to study this action. In Yersinia, one of the plugs 1 uses the general secretory pathway (the Sec-
is a protein called YopN, which is also translo- dependent pathway) to secrete the protein into
cated into HeLa cells when secretion takes place the periplasm. Once the protein has crossed the
(Fig. 18.5).81 Secretion after cell contact, includ- inner membrane (cell membrane), the leader
ing the unplugging of the injectisome channel sequence is proteolytically removed on the
to allow passage of proteins, is initiated upon external face of the inner membrane, and the
reception of an activation signal, which occurs protein enters the periplasm for the second step
when the bacterium adheres to the target cell. of secretion through the outer membrane to the
Yersinia can adhere via the binding of outer extracellular milieu. Although the type II path-
membrane proteins called adhesins to integrins, way is sometimes referred to as a GSP pathway,
which are proteins projecting from the eukary- the student should understand that this does not
otic cell. Alternatively, secretion can be initiated mean the GSP pathway in general, but merely
when Yersinia binds to phagocytic receptors, one of its terminal branches.
without the involvement of adhesins.
The first proteins introduced are probably Secretion apparatus for second stage
the Yop proteins (YopB, YopD, LcrV) that The translocation machinery and mechanism
form the channel (translocon) in the target cell of the second stage are not well understood.
membrane. After this, at least six Yops [YopE, After the proteins have been translocated in
YopH, YopJ (YopP), YopM, YopO (YpkA), and an unfolded state by the Sec system into the
YopT] are injected into the cytosol of the host periplasm in the first stage, the signal peptide
cell through the channel, producing a cytotoxic at the N-terminal end is removed. In the second
effect. (YopT is found only in Y. enterocolitica, stage, the proteins fold into a near-native confor-
whereas the other five Yops are found in all spe- mation and are translocated through the outer
cies of Yersinia.) These Yops inhibit phagocy- membrane. The secretory apparatus for the sec-
tosis and the production of proinflammatory ond stage of secretion is called the secreton, and
cytokines. (See note 82 for an explanation of it consists of 12 to 16 proteins. Surprisingly, the
how cytokines promote inflammation and why proteins of the secreton are located in both the
www.ebook777.com
OM D
folded protein polymerization

C C
depolymerization
IM pseudopilus (G-K)
Sec transport
E E
ATP
ADP + Pi
Fig. 18.7 Model of how type II systems might work. The protein to be secreted is transported in an unfolded
state via the Sec system across the inner membrane (IM). The N-terminal signal sequence is removed and the
protein folds in the periplasm. Protein D is an outer membrane (OM) complex called a secreton, consisting
of 12 to 16 subunits that form a pore. The folded protein is pushed out through the D-protein pore by what
is suggested to be a polymerizing pseudopilus. The pseudopilus comprises proteins G–K, with G being the
dominant protein. Proteins G–K show homology to type IV pili. Protein E is suggested to be an ATPase that
provides energy for the translocation and assembly of the pseudopilus. Protein C is suggested to be involved
in transferring energy for transport from the inner membrane to the protein D complex. Not shown are pro-
teins O and S. Protein O, an integral cell membrane protein, has two activities. It is a prepilin peptidase that
cleaves the N-terminal end of the pseudopilin proteins, and it also methylates the newly created N-terminal
amino acid. This all takes place on the cytoplasmic side of the membrane. Protein S is an outer membrane
lipoprotein that stabilizes the protein D complex. In addition, there are several other proteins that are present
in some type II pathways and not others. The latter proteins may be involved in specific requirements, such as
substrate recognition.
inner and outer membranes, with most being with a central channel in the outer membrane,
located in the inner membrane. Several of the in some ways similar to the TolC channel. PulS
inner membrane secreton proteins have exten- (protein S, not shown in Fig. 18.7) is an outer
sive periplasmic domains. membrane lipoprotein that is suggested to stabi-
The location of secreton proteins in the inner lize PulD. One model postulates that energy for
membrane is unexpected because the GSP sys- transport through the outer membrane channel
tem is responsible for translocating the proteins is transmitted by PulC to the outer membrane
across the inner membrane into the periplasm complex. This may be similar to how TonB
(stage 1). One possible role for secreton proteins energizes solute uptake. (See the discussion of
in the inner membrane may be related to trans- TonB in Section 1.2.4.) Interestingly, some of
ferring energy to the outer membrane for trans- the proteins (proteins G–K) anchored in the cell
port in stage 2. Each component of the type II membrane are believed to form a periplasmic
secreton has been designated Gsp, followed by pseudopilus that is required for the secretion
a letter corresponding to a homologous protein of proteins (Fig. 18.7). For more information
in the pul system for the secretion of pullanase. about the postulated roles of the proteins,
(See note 86 for a description of the pullula- including a role for the pseudopilus, see note 88
nases.) However, it has been argued that the and the legend to Fig. 18.7.
notation Gsp is inappropriate, and sometimes
the word “protein” is used instead. See note The Out system in Erwinia chrysanthemi
87 for a discussion of this point. According to and E. carotovora
Fig. 18.7, PulC (protein C) connects the inner There are several type II pathways, and each
membrane proteins with the PulD (protein D) can secrete more than one kind of protein. For
complex in the outer membrane. example, secretion machinery called the Out
PulD proteins are called secretins, and a popu- system exists in Erwinia chrysanthemi and E.
lar model proposes that the proteins form a ring carotovora, the causative agents of soft-rot
disease in plants. The enzymes secreted by the 18.5.4 The type V pathway:
Out system, which destroy plant tissue, are Autotransporters
pectate lyase, exo-poly-α-D-galacturonosidase, The type V pathway (see Fig. 18.8) is also called
pectin methylesterase, and cellulase. Mutants the autotransporter (AT) pathway because the
in the out genes have a defect in the secretion of proteins transport themselves across the outer
cellulases and pectinases and accumulate these membrane of gram-negative bacteria without
enzymes in the periplasm. the aid of an obvious secretion system. For a
review, see refs. 91 and 92. The secreted pro-
Cholera toxin and protease secretion by teins, referred to as autotransporters, include
Vibrio cholerae proteases, toxins, and adhesins. The type V
Other type II systems exist in other bacteria pathway is widespread among the gram-neg-
and are responsible for the secretion of differ- ative bacteria. A related secretion pathway
ent proteins. For example, the eps (extracel- that is similar to the type V pathway is called
lular protein secretion) genes are required for the TPS (two-partner secretion) system. It is
the secretion of cholera toxin and protease also widespread among gram-negative bacte-
by Vibrio cholerae.89 The eps genes are also ria and secretes large virulence proteins such
required for correct assembly of the proteins in as hemagglutinin by Bordetella pertussis and
the outer membrane. cytolysin by Serratia marcescens. The TPS sys-
tem is distinguished from the type V pathway
Specificity between each type II system in that the TPS system utilizes an accessory pro-
and the protein secreted tein for transport across the OM. The follow-
There is specificity between each type II system ing discussion focuses on the type V pathway.
and the proteins that it secretes. A resident type
II system need not secrete extracellular pro- Two stages
teins of related bacteria when the structural Translocation uses the Sec (GSP) pathway
genes are introduced, even though there may be to move the autotransporter through the cell
homology between the secretory proteins from membrane into the periplasm; but in contrast
the donor and recipient bacteria. For example, to the type II system, there is no requirement
there is homology in the proteins between the for a separate secretion apparatus to move the
out gene products from Erwinia and the pul protein from the periplasm through the outer
gene products from Klebsiella. Yet, when the membrane to the outside. So, how is this done?
pectate lyase gene from Erwinia is expressed in Autotransporters have three domains: an ami-
Klebsiella, the enzyme is not secreted. no-terminal domain recognized by the Sec sys-
There can even be specificity between Out tem, an internal domain called the α.-domain
systems from two different Erwinia species. or “passenger domain,” and a carboxyl ter-
For example, the E. chrysanthemi Out system minal domain (C-terminal domain) called the
does not secrete an extracellular pectate lyase β-domain. or “helper” domain, which inserts
encoded by a gene from a different species of into the outer membrane with 12 to 14 trans-
Erwinia, even though both species of Erwinia membrane β-strands that fold and interlink
use the Out system. with each other to form a pore, called a β-barrel,
What can we conclude from all of this? Each through the membrane. Once in the periplasm,
type II system not only distinguishes between the amino-terminal signal domain is cleaved by
different secreted proteins but also recognizes a peptidase. In one model the internal passenger
“self” proteins. The basis for this distinction is domain travels to the outside surface through
not understood. However, if the genes for a type the β-barrel.
II system as well as its cognate protein are trans- Depending upon the protein, the passenger
ferred to a different genus of bacterium, then domain can remain bound to the membrane
protein secretion can occur. For example, when via the helper domain (β-barrel, i.e., pore), or
the structural gene for pullulanase as well as the it can be severed by proteolytic cleavage and
adjoining secretion genes are transferred from released into the extracellular medium. Some
Klebsiella pneumoniae to E. coli, pullulanase is models hold that since the outer membrane
synthesized and secreted.90 pore is too narrow to accommodate large,
www.ebook777.com
N
pilus
passenger domain
exported
helper domain helper domain OM usher
forms a pore
growing end
of pilus
C
C
passenger domain P
N
C
helper domain C pilus subunit
chaperone
Sec IM Sec
presecretory protein presecretory pilus subunit

Type V (autotransporter) Chaperone/usher
Fig. 18.8 Type V and chaperone/usher secretion systems. The type V system is also called the autotransporter
system. The presecretory protein is exported via the Sec system to the periplasm (P), where the N-terminal sig-
nal sequence is cleaved by a peptidase. The protein has a C-terminal helper domain, followed by a passenger
domain. The C-terminal helper domain inserts into the outer membrane (OM) to form a secretion channel
through which the passenger domain is exported to the cell surface. The passenger domain can be released
into the extracellular medium by proteolysis or can remain attached to the outer membrane. The chaperone/
usher system is used for the formation of adhesive pili by pathogenic bacteria. The pilus subunit is translocated
through the inner membrane (IM) via the Sec system. In the periplasm, the amino-terminal signal sequences of
the pilus subunits are cleaved. When in the periplasm, these subunits bind at their C-terminal ends to chaper-
one proteins that allow the subunits to fold properly and also prevent them from binding to each other in the
periplasm. The chaperone brings the subunits to the outer membrane usher. The pilus subunits are assembled
into a linear fiber of folded subunits that threads through the usher. The pilus rod assumes its final helical
shape at the cell surface.
completely folded proteins, folding of the pas- protease by Neisseria gonorrhoeae, the caus-
senger domain is completed on the outer mem- ative agent of gonorrhea. The protease is synthe-
brane surface. However, it must be pointed out sized as a precursor protein with an N-terminal
that additional questions about mechanisms of leader sequence typical for proteins translocated
transport remain unanswered. For example, it by the Sec system. The leader sequence func-
is known that some folded passenger domains tions with the Sec machinery and is removed
are secreted via the autotransporter pathway. by the peptidase after translocation through
See ref. 91 for a review. the cell membrane. Up to this stage, the secre-
tion is similar to the secretion of proteins via the
Immunoglobulin A protease, an example of type II system. However, the immunoglobulin
an autotransporter A protease has information in its sequence that
As an example of an autotransporter, we will directs it through the outer membrane to the
consider the secretion of immunoglobulin A outside.
At the carboxy end of the protein, the helper these systems include the assembly and retrac-
domain aids the protein in traversing the outer tion of a conjugal pilus. T4SSs are also present
membrane, where it anchors the protein. The in gram-positive bacteria and are involved in
protein is released into the medium by prote- the conjugative transfer of plasmids between
olytic cleavage (autocatalytic) at the C-terminal cells (ref. 96). The targets of delivery of sub-
end, leaving the helper peptide embedded in the stances transported by type IV systems include
membrane. Some other proteins are secreted in a cells of bacteria, fungi, plants, and animals, and
similar fashion, although the proteolytic cleav- the extracellular medium. It is thus a ubiqui-
age may be catalyzed by a separate enzyme. tous and very important pathway. Substances
How the secreted proteins cross the periplasm secreted by type IV systems into the extracellular
and insert into the outer envelope is not known. medium, or transferred into host cells, include
In E. coli, the helper domain can promote trans- virulence factors such as pertussis toxin. The
location across the outer membrane of hybrid gram-negative bacterium Bordetella pertussis
proteins that would not normally release the secretes this toxin into the extracellular medium
passenger protein into the medium. It thus during infection, where it is taken up by cells
appears that the helper region may be sufficient of the upper respiratory tract, causing some of
for translocation through the outer envelope the symptoms of whooping cough (see earlier:
without the need for an accessory secretion note 40). Other virulence factors are T-DNA,
system. the oncogenic portion of the Ti plasmid that is
transferred from the gram-negative bacterium
18.5.5 The chaperone/usher Agrobacterium tumefaciens into the plant host
(CU) pathway cells and induces crown gall tumors; the conju-
Two stages gative IncN plasmid pKM101, transferred via
The chaperone/usher pathway (see Fig. 18.8) the Tra system between cells of A. tumefaciens,
uses the Sec system to secrete across the cell mem- and several other proteins and plasmids by a
brane to the periplasm such adhesive virulence variety of bacteria.
structures as the P and type 1 pili in uropatho- The type IV systems have different names. For
genic E. coli. The pilus subunits are exported example, the B. pertussis system is called the Ptl
individually across the cell membrane via the system, and the nine proteins for the secretion
Sec system, and their amino-terminal signal apparatus are encoded by the ptl locus. The
sequences are cleaved. Once in the periplasm, Agrobacterium system that transfers T-DNA is
each pilus subunit binds at its C-terminal end called the VirB system, and the 11 proteins for
to a periplasmic chaperone protein. The chap- the secretion are encoded by the virB locus. The
erone–pilus complex migrates to an outer mem- system for transferring the pKM101 plasmid is
brane “usher,” which forms a channel in the called the IncN system.
outer membrane. The chaperone binds to the
“usher” and transfers the pilus subunit to it. Secretion apparatus of the VirB system in
More pilus subunits are brought by chaperone Agrobacterium that translocates oncogenic
proteins to the “usher,” where they assemble at plasmid DNA into plant cells
the periplasmic surface onto a pilus fiber that Much of what is known about type IV secre-
is threaded through the “usher” channel to the tion comes from studying the VirB system of
outside. Agrobacterium tumefaciens, the prototypical
type IV system, which is responsible for secret-
18.5.6 The type IV pathway ing virulence factors that cause crown gall in
For a review of the type IV secretion systems infected plants. (See Fig. 18.9.) Export of onco-
(T4SSs), see refs. 45, 93, and 94; for a discus- genic plasmid DNA (T-DNA) into plant cells by
sion of the structure of the proteins and how the VirB system of Agrobacterium is Sec inde-
they interact, read ref. 95. Type IV systems pendent and takes place in a single step from the
(T4SSs) translocate DNA and/or protein across cytoplasm through a translocation channel that
the inner and outer membranes in gram-nega- spans the periplasm.
tive bacteria via a Sec-dependent (two-stage) Several proteins make up the secretory appa-
or a Sec-independent (one-stage) mechanism; ratus. The functions of the individual proteins
www.ebook777.com
are summarized in the legend to Fig. 18.9 and Example of a Sec-dependent system: Export
discussed in detail in refs. 97 and 98. Some of of protein toxins
the proteins form a complex attached to both Although type IV systems are generally Sec inde-
the inner and outer membranes and are thought pendent, there is at least one Sec-dependent sys-
to assemble as a translocation channel through tem, the secretion of pertussis toxin. Figure 18.9
the inner membrane, periplasm, and outer illustrates how pertussis toxin, from Bordetella
membrane; some form a pilus that transfers the pertussis, is exported in two stages. In the first
DNA from the translocation channel into the stage, which is Sec dependent, the protein enters
plant cell. As explained next, pertussis toxin is the periplasm and then moves into the translo-
secreted into the periplasm by the Sec system and cation channel. Pertussis toxin is not secreted
then enters the type IV translocation channel to into cells; rather, it is secreted into the extracel-
be transported into target eukaryotic cells. lular medium.
eukaryotic cell membrane

B2
pilus
B5
outer membrane
protein B6-B10 channel periplasm
B1
Sec inner membrane
B4, B11
DNA
Fig. 18.9 Type IV secretion system. The components of the secretion apparatus are labeled according to the
nomenclature for the VirB system (i.e., the prefix Vir is not used). The secretion of certain proteins, such as per-
tussis toxin, takes place in two stages. Stage 1 uses the Sec system to translocate the protein into the periplasm.
From there, the protein enters the translocation apparatus that spans the inner membrane, periplasm, and
outer membrane. Pertussis toxin is secreted into the extracellular medium during infections and causes the
symptoms of whooping cough. DNA secretion takes place in one stage, as the DNA enters the secretion appa-
ratus from the cytoplasm. VirB1 (B1 in the figure) is found in the periplasm. It has a motif seen in lytic transgly-
cosylases and may be important for hydrolysis of the peptidoglycan so that the transporter can be assembled.
A proteolytic product of VirB1 (not shown) is on the surface of the outer membrane and may be important for
contact between donors and recipient cells. It is not part of the pilus. The pilus is composed of a major protein
called VirB2 and a minor protein called VirB5. It is thought that the pilus is a tube through which proteins
and DNA are moved into target eukaryotic cells. The channel subunits are VirB6 through VirB10. VirB6 is
an inner membrane protein. VirB7 is a lipoprotein in the outer membrane that is complexed with VirB9. Both
VirB7 and VirB9 are thought to function in the assembly of the secretion apparatus in the outer membrane.
VirB10 is believed to provide a link between VirB components in the inner and outer membrane. VirB4 and
VirB11 are in the inner membrane and are believed to couple ATP hydrolysis with transport. (VirB11 proteins
may also be found in the cytoplasm.)
18.6 Folding of Periplasmic Proteins

18.6.1 The importance of intramolecular outer membrane
disulfide bonds for periplasmic proteins
For a review, see ref. 99. In E. coli, newly syn- SH
thesized proteins that are translocated across unfolded R R S folded
SH
the cell membrane into the periplasm often form
intramolecular disulfide bonds (S–S bonds)
when they arrive in the periplasm. (Actually, SH
periplasm
DsbA
disulfide bonds are very important for folding DsbA S
SH
and stability of extracytosolic proteins in all
organisms, not simply bacteria.) The S–S bonds
form between –SH groups in the side chains of SH
electron transport DsbB DsbB S cell membrane
nearby cysteine residues. These intramolecular chain SH
disulfide bonds are important for proper fold-
ing and three-dimensional stability of the poly-
peptide once it has been translocated into the Fig. 18.10 Formation of disulfide bonds in periplas-
periplasm. Such disulfide bonds are much less mic proteins. Periplasmic proteins (R) are secreted
common in cytoplasmic proteins because of the unfolded into the periplasm. Many of them form
existence of high concentrations in the cytosol intramolecular disulfide bonds, which are important
of reducing substances that break S–S bonds. for their folding and stability. This activity is cata-
As an example of disulfide bonds in periplasmic lyzed by a periplasmic protein called DsbA. Electrons
proteins, consider E. coli alkaline phosphatase. are transferred from the periplasmic proteins to DsbA
This is a periplasmic enzyme that in its active to a membrane protein called DsbB, and from there
form consists of two identical subunits, each of to the electron transport chain. The overall sum of
which has intrachain disulfide bonds that enable the reactions is the equivalent of transferring disul-
fide bonds between cysteine residues from DsbB to
the subunits to fold properly in the periplasm.
DsbA to R. The cysteine residues in DsbB are located
in periplasmic loops. Each loop has two pairs of
18.6.2 The role of thiol–redox enzymes cysteine residues, and electrons are transferred from
in catalyzing the formation of disulfide DsbA in sequence from one pair to the next before
bonds in the periplasm being transferred to the electron transport chain.
Disulfide bond formation is enzymatically cata-
lyzed in the periplasm by thiol–redox enzymes,
also called thiol-disulfide exchange proteins. Thiol–redox enzymes such as DsbA have pairs
This is diagrammed in Fig. 18.10 for E. coli. of cysteine residues separated by two amino
The periplasmic thiol–redox enzyme in E. coli acids, Cys-X1-X2-Cys. This is often referred
is DsbA (disulfide bond), which is required for to as a thioredoxin-like motif because it is also
disulfide bond formation in vivo and has been present in thioredoxin and is characteristic of
demonstrated to carry out this activity in vitro. disulfide oxidoreductases. (For a discussion
DsbA catalyzes the formation of disulfide bonds of the role of thioredoxin, see Section 16.6.3.)
in newly synthesized and exported periplasmic In DsbA, the thioredoxin-like motif is Cys30-
proteins, causing them to fold properly. In Pro31-His32-Cys33.
the process, disulfide groups in DsbA become
reduced to sulfhydryl groups. Reduced DsbA 18.6.3 Correcting the formation of
is reoxidized by donating electrons to DsbB, disulfide bonds between the wrong
a cytoplasmic membrane thiol–redox protein cysteine residues
with periplasmic loops. DsbB itself is reoxidized Occasionally, the wrong cysteine residues will
by ubiquinone or menaquinone in the electron take part in the formation of a disulfide bond,
transport chain, and the electrons eventually causing the polypeptide to fold incorrectly.
go to a terminal electron acceptor.100 For more E. coli has a mechanism to repair this, and it
information about the structure and function of involves the periplasmic thiol–redox protein
DsB, as well as its relationship to the electron called DsbC.102 DsbC binds to improperly
transport chain, see note 101. folded proteins, breaks (reduces) the disulfide
www.ebook777.com
bond, and catalyzes the formation of the correct Sec-dependent pathway for translocating
disulfide bond. DsbC is called an isomerase– proteins through the cell membrane into the
reductase, and it has a thioredoxin-like motif periplasm prior to secretion out of the cell, and
(Cys98-X1-X2-Cys101). others do not. These pathways secrete various
enzymes, pilus proteins, and flagellar proteins
into the extracellular space, as well as virulence
18.7 Summary proteins that aid infection into target cells. The
Bacteria translocate (export) proteins into the secretion pathways are type I, type II, type III,
cell membrane, periplasm, and outer envelope, type IV, type V (autotransporter), and the chap-
or secrete them into the extracellular medium. erone/usher pathway.
The translocation machinery and requirements Many periplasmic proteins fold properly by
for targeting to different destination sites are forming intrapeptide disulfide bonds between
being studied in several bacteria, especially in cysteine residues. This activity is catalyzed in
E. coli. E. coli by a periplasmic protein called DsbA.
The Sec system is necessary for the transloca- Improperly folded proteins are repaired by
tion of most proteins across the cell membrane. DsbC, another E. coli periplasmic protein.
A preprotein is made with a leader sequence at
the amino-terminal end. The role of the leader Study Questions
sequence is to initiate translocation by insert-
ing into the lipid bilayer. The leader sequence 1. What is the experimental basis for believ-
also aids in preventing premature folding of ing that the leader peptide is necessary
the newly synthesized protein. The preprotein for translocation? What are two reasons
binds to a chaperone protein in the cytosol. for believing that the leader peptide is not
An important chaperone protein is Sec B. necessary for determining the final desti-
Two roles of the chaperone protein are to pre- nation of the protein?
vent folding of the preprotein into a configura-
2. What are the postulated roles for chaper-
tion that disallows translocation and to prevent
one proteins in protein translocation?
the formation of protein aggregates. A third
role for the chaperone protein is to deliver the 3. What is the role of Sec A in protein trans-
preprotein to the translocase. location? What is the evidence that ATP
The preprotein–chaperone protein complex binding, and not hydrolysis, drives the
binds to the translocase on the inner mem- insertion of SecA into the membrane?
brane surface, and the translocation of the 4. What are the functions of the different
preprotein into the membrane is initiated. The domains of the leader sequence (i.e., the
early stages of translocation require ATP and basic N terminal, middle hydrophobic,
SecA, but the remainder of translocation can and carboxy terminus)?
be driven by the ∆p.
Once the preprotein has been translocated 5. What is the evidence, based upon muta-
through the membrane, the leader sequence is tional analysis, that there are two secre-
removed by a peptidase on the outer surface of tory systems for proteins secreted into the
the cell membrane. E. coli also has a signal recog- medium and that one is sec dependent and
nition particle (SRP) that may guide certain pro- one is sec independent.
teins destined to remain in the inner membrane. 6. How does the type III protein secretion
There exists a Sec-independent pathway for system differ from types I and II? How is it
translocating folded proteins with bound cofac- similar?
tors into the cell membrane and the periplasm.
7. What is the evidence that SecDFyajC plays
It is called the twin-arginine translocation (Tat)
a regulatory role in protein export via the
pathway. Respiratory and photosynthetic elec-
Sec system?
tron transport proteins are translocated via the
Tat pathway. 8. What is the evidence that in the absence of
There are six secretion pathways found a ∆p, ATP can drive Sec-dependent trans-
among gram-negative bacteria. Some use the location to completion?
9. What is the evidence that translocons location. In vitro studies using isolated membranes
form pores in target cell membranes? with depleted, wild-type, or enriched DF levels
indicate that SecDFyajC stabilizes the inserted form
10. How does the Tat system differ from the of SecA in the membrane and slows the movement
GSP system? of the preprotein, proOmpA. As a consequence,
translocation intermediates accumulate in the
membrane, and these are translocated by means of
REFERENCES AND NOTES the driving force of the ∆p. Mutants lacking SecDF
are severely defective in translocation and growth.
1. Juhas, M, D.W. Crook, and D. W. Hood. Overexpression of secDF suppresses mutations in
2008. Type IV secretion systems: tools of bacte- the leader peptide sequence; anti-SecD antibodies
rial horizontal gene transfer and virulence. Cell have been reported to block translocation in sphero-
Microbiol. 10:2377–2386. plasts. There appears to be only 10% as much
SecDFyajC as SecYEG in E. coli, suggesting that
2. Xie, K., and R. E. Dalbey. 2008. Inserting pro- either some SecYEG translocases operate without
teins into the bacterial cytoplasmic membrane using SecDFyajC or that SecDFyajC cycles between oper-
the Sec and YidC translocases. Nat. Rev. Microbiol. ating SecYEG translocases. For a discussion of the
6:234–244. role of SecDFyajC in protein secretion, see: Duong,
3. Driessen, A. J. M., and N. Nouwen. 2008. Protein F., and W. Wickner. 1997. The SecDFyajC domain
translocation across the bacterial cytoplasmic mem- of preprotein translocase controls preprotein move-
brane. Annu. Rev. Biochem. 77:643–667. ment by regulating SecA membrane cycling. EMBO
J. 16:4871–4879. Also, see: Driessen, A. J. M., and
4. Fisher, A. C., and M. P. DeLisa. 2004. A little help N. Nouwen. 2008. Protein translocation across
from my friends: quality control of presecretory pro- the bacterial cytoplasmic membrane. Annu. Rev.
teins in bacteria. J. Bacteriol. 186:7467–7473. Biochem. 77:643–667.
5. Driessen, A. J. M. 1992. Bacterial protein trans- 13. Gardel, C., K. Johnson, A. Jacq, and J. Beckwith.
location: kinetic and thermodynamic role of ATP 1987. secD, a new gene involved in protein export in
and the protonmotive force. Trends Biochem. Sci. Escherichia coli. J. Bacteriol. 169:1286–1290.
17:219–223.
14. Wickner, W., and M. R. Leonard. 1996.
6. Wickner, W., A. J. M. Driessen, and F.-U. Hartl. Escherichia coli preprotein translocase. J. Biol.
1991. The enzymology of protein translocation Chem. 271:29514–29516.
across the Escherichia coli plasma membrane. Annu.
Rev. Biochem. 60:101–124. 15. Duong, F., and W. Wickner. 1997. The
SecDFyajC domain of preprotein translocase con-
7. Saier, M. H., Jr., P. K. Werner, and M. Muller. trols preprotein movement by regulating SecA mem-
1989. Insertion of proteins into bacterial membrane cycling. EMBO J. 16:4871–4879.
branes: mechanism, characteristics, and compari-
sons with the eucaryotic process. Microbiol. Rev. 16. Manting, E. H., C. van der Does, and A. J. M.
53:333–366. Driessen. 1997. In vivo cross-linking of the SecA
and SecY subunits of the Escherichia coli preprotein
8. Pugsley, A. P. 1993. The complete general secre-
translocase. J. Bacteriol. 179:5699–5704.
tory pathway in gram-negative bacteria. Microbiol.
Rev. 57:50–108. 17. Manting, E. H., and A. J. M. Driessen. 2000.
Escherichia coli translocase: the unravelling of a
9. Murphy, C. K., and J. Beckwith. 1996. Export
molecular machine. Mol. Microbiol. 37:226–241.
of proteins to the cell envelope in Escherichia coli,
pp. 967–978. In: Escherichia coli and Salmonella: 18. Nishiyama, K.-I., S. Mizushima, and H. Tokuda.
Cellular and Molecular Biology. F. C. Neidhardt 1993. A novel membrane protein involved in protein
et al. (Eds.). ASM Press, Washington, DC. translocation across the cytoplasmic membrane of
Escherichia coli. EMBO J. 12:3409–3415.
10. Ito, K. 1992. SecY and integral membrane com-
ponents of the Escherichia coli protein translocation 19. Economou, A., and W. Wickner. 1994. SecA pro-
system. Mol. Microbiol. 6:2423–2428. motes preprotein translocation by undergoing ATP-
driven cycles of membrane insertion and deinsertion.
11. Rapoport, T. A., B. Jungnickel, and U. Kutay.
Cell 78:835–843.
1996. Protein transport across the eukaryotic endo-
plasmic reticulum and bacterial inner membranes. 20. Economou, A., J. A. Pogliano, J. Beckwith, D.
Annu. Rev. Biochem. 65:271–303. B. Oliver, and W. Wickner. 1995. SecA membrane
cycling at SecYEG is driven by distinct ATP binding
12. In E. coli the SecYEG translocase can be isolated
and hydrolysis events and is regulated by SecD and
from the membrane in a complex with three other
SecF. Cell 83:1171–1181.
proteins, SecD, SecF, and yajC. SecDFyajC plays a
regulatory role in protein secretion and facilitates 21. SecA insertion and deinsertion can be studied
the coordination between ATP and ∆p-driven trans- by means of vesicles prepared from the inner mem-
location, lowering the amount of ATP used for trans- brane. Instead of ATP, a nonhydrolyzable form of
www.ebook777.com
ATP, adenyl–imidodiphosphate (AMP–PNP), can be 5801–5812.

used for insertion of SecA. It is possible to determine 32. De Gier, J.-W. L., Q. A. Valent, G. V. Heijne, and
the portion of the protein that has inserted because J. Luirink. 1997. The E. coli SRP: preferences of a
it becomes resistant to protease, but is not so after targeting factor. FEBS Lett. 408:1–4.
detergent treatment of the vesicles. For more infor-
mation about these experiments, see ref. 20. 33. Keenan, R. J., D. M. Freymann, R. M. Stroud,
and P. Walter. 2001. The signal recognition particle.
22. Schiebel, A., A. J. M. Driessen, F.-U. Hartl, and Annu. Rev. Biochem. 70:755–775.
W. Wickner. 1991. ΔμH+ and ATP function at differ-
ent steps of the catalytic cycle of preprotein translo- 34. Duong, F., J. Eichler, A. Price, M. R. Leonard, and
case. Cell 64:927–939. W. Wickner. 1997. Biogenesis of the gram-negative
bacterial envelope. 1997. Cell 91:567–573.
23. The in vitro system consists of inverted Escherichia
coli inner membrane vesicles translocating the outer 35. This is the consensus sequence in Proteobacteria;
membrane protein proOmpA. the Ser, Phe, Leu, and Lys residues are present in a
frequency higher than 50%, and X is a polar amino
24. SecA has two nucleotide binding sites: NBS I acid.
(Kd,ADP = 0.1–0.3 μM) and NBS II (Kd,ADP = 300–500
μM). Both sites play a role in translocation. This is 36. Palmer, T., F. Sargent, and B. C. Berks. 2005.
further discussed in ref. 17. Export of complex cofactor-containing proteins
by the bacterial Tat pathway. Trends Microbiol.
25. Driessen, A. J. M. 1992. Precursor protein trans- 13:175–180.
location by the Escherichia coli translocase is directed
by the protonmotive force. EMBO J. 11:847–853. 37. Berks, B. C., T. Palmer, and F. Sargent. 2003.
The Tat protein translocation pathway and its role
26. There are several ways to separate effects of the in microbial physiology. Adv. Microbial Physiol.
∆p from those of SecA and ATP in vitro. For example, 47:187–254.
membrane vesicles or proteoliposomes will not have a
∆p unless one is imposed. Different methods can be used 38. Berks, B. C., F. Sargent, and T. Palmer. 2000.
to impose a ∆p. For example, the stimulation of elec- The Tat protein export pathway. Mol. Microbiol.
tron transport by incubating inner membrane vesicles 35:260–272.
with NADH has been used. Vesicles are prepared from 39. Palmer, T., and B. C. Berks. 2003. Moving
unc– cells so that the ∆p and ATP are not interconvert- folded proteins across the bacterial cell membrane.
ible. A ∆p has also been created by light in proteolipo- Microbiology 149:547–556.
somes in which bacteriorhodopsin was incorporated,
or by adding reduced cytochrome c to proteoliposomes 40. Cholera toxin (CT) is a protein produced by Vibrio
in which cytochrome c oxidase has been incorpo- cholerae and is responsible for the watery diarrhea
rated. Proton ionophores can be used to dissipate the associated with cholera. It is an enzyme that enters
∆p. SecA can be inactivated with anti-SecA antibody, intestinal cells and transfers ADP–ribose from NAD+ to
and ATP levels can be experimentally manipulated. a regulatory enzyme in the target cell (a process called
ADP-ribosylation); this results in the stimulation of the
27. Yamada, H., S. Matsuyama, H. Tokuda, and enzyme adenyl cyclase in the intestinal cells. Adenyl
S. Mizushima. 1989. A high concentration of SecA cyclase converts ATP to cAMP, and the increased
allows proton motive force–independent transloca- amounts of cAMP stimulate the secretion of Cl– and
tion of a model secretory protein into Escherichia inhibit the absorption of NaCl. As a consequence,
coli membrane vesicles. J. Biol. Chem. 264:18577– the osmolarity of the intestinal fluid is increased, and
18581. large amounts of water as well as bicarbonate and K+
28. Yamada, H., H. Tokuda, and S. Mizushima. enter the intestinal fluid and are lost from the body.
1989. Proton motive force–dependent and –indepen- Enterotoxigenic Escherichia coli (ETEC) produces a
dent protein translocation revealed by an efficient in similar toxin, called heat-labile toxin (LT). Pertussis
vitro assay system of Escherichia coli. J. Biol. Chem. toxin (PT) is produced by Bordetella pertussis and
264:1723–1728. is responsible for some of the symptoms of pertussis
(whooping cough). It is an enzyme that enters target
29. Bos, M.P. V. Robert, and J. Tommassen. 2007.
cells in the upper respiratory tract and ADP-ribosylates
Biogenesis of the gram-negative bacterial outer mem-
a regulatory protein, causing a rise in cellular cAMP.
brane. Annu. Rev. Microbiol. 61:91–214.
This is probably the cause of the increased secretions
30. Fröderberg, L., E. Houben, J. C. Samuelson, M. and mucus production in the upper respiratory tract
Chen, S.-K. Park, G. J. Phillips, R. Dalbey, J. Luirink, colonized by the bacteria. Diphtheria toxin is pro-
and J.-W. L. De Gier. 2003. Versatility of inner mem- duced by Corynebacterium diphtheriae, the causative
brane protein biogenesis in Escherichia coli. Mol. agent of diphtheria. The toxin is an enzyme that ADP-
Microbiol. 47:1015–1027. ribosylates elongation factor 2 in target cells and stops
31. van der Laan, M., M. L. Urbanus, C. Ten Hagen- protein synthesis. Many cell types, including those in
Jongman, N. Nouwen, B. Oudega, N. Harms, A. the heart and other organs, can be affected.
J. M. Driessen, and J. Luirink. 2003. A conserved 41. Lory, S. 1992. Determinants of extracellular pro-
function of YidC in the biogenesis of respiratory tein secretion in gram-negative bacteria. J. Bacteriol.
chain complexes. Proc. Natl. Acad. Sci. USA 100: 174:3423–3428.
42. Salmond, G. P. C., and P. J. Reeves. 1993. of TolC have been found among many gram-negative
Membrane traffic wardens and protein secretion bacteria and appear to be widespread.
in gram-negative bacteria. Trends Biochem. Sci.
53. The following are some animal pathogens that
18:7–12.
use the type III system to inject virulence proteins into
43. Wandersman, C. 1996. Secretion across the bac- eukaryotic cells, and the diseases they cause: Yersinia
terial outer membrane. pp. 955–966. In: Escherichia spp. (bubonic plague, pneumonic plague, enterocoli-
coli and Salmonella: Cellular and Molecular tis), Salmonella spp. (enteritis, septicemia, typhoid
Biology. F. C. Neidhardt et al. (Eds.). ASM Press, fever), Shigella spp. (gastroenteritis), Escherichia
Washington, DC. coli (enteropathogenic, enteroinvasive, and entero-
44. Thanassi, D. G., and S. J. Hultgren. 2000. hemorrhagic E. coli gastroenteritis), Bordetella
Multiple pathways allow protein secretion across pertussis (pertussis, also called whooping cough),
the bacterial outer membrane. Curr. Opin. Cell Biol. Pseudomonas aeruginosa (burn wound infections;
12:420–430. infections of the respiratory tract, urinary tract, ear,
and eye; gastroenteritis; bacteremia; endocarditis).
45. Henderson, I. R., F. Navarro-Garcia, M. Desvaux,
R. Fernandez, and D. Ala’Aldeen. 2004. Type V pro- 54. Rosqvist, R., S. Håkansson, Å. Forsberg, and H.
tein secretion pathway: the autotransporter story. Wolf-Watz. 1995. Functional conservation of the
Microbiol. Mol. Biol. Rev. 68:692–744. secretion and translocation machinery for virulence
proteins of yersiniae, salmonellae, and shigellae.
46. Ghosh, P. 2004. Process of protein transport by
EMBO J. 14:4187–4195.
the type III secretion system. Microbiol. Mol. Biol.
Rev. 68:771–795. 55. Kubori, T., Y. Matsushima, D. Nakamura, J.
47. Kostakioti, M., C. L. Newman, D.G. Thanassi, Uralil, M. Lara-Tejero, A. Sukhan, J. E. Galán, and
and C. Stathopoulos. 2005. Mechanisms of pro- S.-I. Aizawa. 1998. Supramolecular structure of the
tein export across the bacterial outer membrane. J. Salmonella typhimurium type III protein secretion
Bacteriol. 187:4306–4314. system. Science 280:602–605.
48. For example, colicin was exported through the 56. Wattiau, P., S. Woestyn, and G. R. Cornelis.
α-hemolysin export system (HlyBD system) in strains 1996. Customized secretion chaperones in patho-
of E. coli defective in the colicin export system. genic bacteria. Mol. Microbiol. 20:255–262.
49. Fath, M. J., R. C. Skvirsky, and R. Kolter. 1991. 57. Galán, J. E. 1996. Molecular genetic bases of
Functional complementation between bacterial Salmonella entry into host cells. Mol. Microbiol.
MDR-like export systems: colicin V, alpha-hemolysis, 20:263–271.
and Erwinia protease. J. Bacteriol. 173:7549– 58. Hueck, C. J. 1998. Type III protein secretion sys-
7556. tems in bacterial pathogens of animals and plants.
50. Enterohemorrhagic E. coli (EHEC) causes hemor- Microbiol. Mol. Biol. Rev. 62:379–433.
rhagic colitis, the symptoms of which are bloody diar-
rhea with little or no fever. EHEC produces a toxin 59. Büttner, D., and U. Bonas. 2002. Port of entry—
similar to Shiga toxin, produced by Shigella dysente- the type III secretion translocon. Trends Microbiol.
riae. An important EHEC serotype is 0157:H7. It is 10:186–192.
usually transmitted to humans from animals via con- 60. Cornelis, G. R., and F. Van Gijsegem. 2000.
taminated raw milk or undercooked meat. Outbreaks Assembly and function of type III secretory systems.
have occurred in fast-food restaurants. Hemorrhagic Annu. Rev. Microbiol. 54:735–774.
colitis can be very dangerous in children under 5
years of age because the disease sometimes develops 61. Tampakaki, A. P., V. E. Fadouloglou, A. D.
into hemolytic–uremic syndrome, in which the toxin Gazi, N. J. Panopoulos, and M. Kokkinidis. 2004.
causes kidney damage. E. coli is the most common Conserved features of type III secretion. Cell.
cause of urinary tract infections. The most important Microbiol. 6:805–816.
virulence factor is the ability to adhere to uroepithe- 62. Blocker, A., P. Gounon, E. Larquet, K. Niebuhr,
lial cells via pilus adhesins. V. Cabiaux, C. Parsot, and P. Sansonetti. 1999. The
51. Koronakis, V., V. K. J. Eswaran, and C. Hughes. tripartite type III secreton of Shigella flexneri inserts
2004. Structure and function of TolC: the bacte- IpaB and IpaC into host membranes. J. Cell Biol.
rial exit duct for proteins and drugs. Annu. Rev. 147:683–693.
Biochem. 73:467–489. 63. Hoiczyk, E., and G. Blobel. 2001. Polymerization
of a single protein of the pathogen Yersinia enteroco-
52. Besides participating in the export of proteins via
litica into needles punctures eukaryotic cells. Proc.
the type I system, TolC is used for the export of drugs
(Section 17.5). In addition, TolC is used for the export
of low molecular weight peptides such as heat-stable 64. Evidence suggests that when the bacteria adhere
enterotoxin and cationic antimicrobial peptides, to the target cell, some of the secreted proteins form
which are transported to the periplasm via the Sec a channel called a translocon in the target cell plasma
system or other transporter, whereupon they access membrane. Evidence for translocons includes the
the TolC channel from the periplasm. Homologues creation of pores by the insertion of putative trans-
www.ebook777.com
locon proteins into artificial lipid bilayers and the lengthwise. The protrusions envelop the bacterium
requirement of these proteins for the lysis of eryth- and pinch off an endocytic vesicle containing the bac-
rocytes by Yersinia spp., Pseudomonas aeruginosa, terium. The process resembles phagocytosis. The for-
Shigella flexneri, and enteropathogenic Escherichia mation of the protrusions and the uptake of Shigella
coli (EPEC). In addition, cells infected with mutant Y. are dependent upon localized actin polymerization,
enterocolitica that can make translocon proteins but which itself is stimulated by the effector protein IpaC
not Yop effector proteins have pores in their mem- secreted by the type III apparatus called the Mxi–Spa
branes. That is, if the infected cells are loaded with system, encoded by genes on a resident plasmid. (Ipa
a fluorescent dye before infection, they release the stands for invasion plasmid antigen, and Mxi stands
dye (but not cytosolic components) into the medium for membrane expression of invasion.) The engulf-
upon infection. This indicates that the infected cells ment into an endocytic vesicle by nonphagocytic
did not lyse; rather, the translocon proteins formed cells induced by Salmonella is due to the production
narrow pores in the membrane of the infected cells. of membrane protrusions that are part of a process
The needle may form a continuous passage with the called membrane ruffling or macropinocytosis.
translocon, or there may be a small space between the Salmonella effector proteins delivered by the type III
tip of the needle and the translocon. system after cell contact include four Sips (Salmonella
65. Bacteria generally produce two to eight differ- invasion proteins), proteins that are homologous to
ent effector molecules secreted by the type III sys- the Shigella Ipa proteins. The formation of protru-
tem. There have been more than 20 such molecules sions is stimulated by an effector protein SipC, which
described for various bacteria. For example, the binds to actin, nucleates its polymerization into sub-
Yersinia effectors delivered by the Yop apparatus membranous filaments (F-actin), and bundles the
are YopE, YopH, YopM, YopJ/P, YopO/YpkA, and filaments into cables. Once inside the cells, Shigella
YopT, and the Pseudomonas aeruginosa effectors and Salmonella replicate, although in different parts
delivered by the Psc apparatus are ExoS, ExoT, ExoU, of the cell. Shigella escapes from the vesicle and
and ExoY. The targets of the effector molecules and replicates in the cytoplasm, killing the cell. Shigella
the effects on the animal cell vary, but there are two spreads inter- and extracellularly to adjacent colonic
major targets: the host cell’s cytoskeleton system epithelial cells and to the underlying layers of con-
and the inflammatory response and cell signaling. nective tissue, causing tissue destruction. Salmonella
The text and the referenced notes describe the Yops replicates inside the endocytic vesicle until the cell
and how they block phagocytosis and interfere with lyses.
cell signaling and the inflammatory response. This 66. Nhieu, G. T. V., E. Caron, A. Hall, and P. J.
note gives more information about effector proteins Sansonetti. 1999. IpaC induces actin polymerization
produced by Shigella and Salmonella, and how they and filopodia formation during Shigella entry into
stimulate the uptake of the bacteria into nonphago- epithelial cells. EMBO J. 18:3249–3262.
cytic host cells.
As an example of the promotion of uptake, we will 67. Hayward, R. D., and V. Koronakis. 1999. Direct
consider Shigella and Salmonella. Shigella invades nucleation and bundling of actin by the SipC protein
epithelial cells in the colon, destroys the colonic of invasive Salmonella. EMBO J. 18:4926–4934.
mucosa, and causes bacillary dysentery. Salmonella 68. Zhou, D., M. S. Mooseker, and J. E. Galán. 1999.
invades nonphagocytic epithelial cells in the intesti- Role of the S. typhimurium actin-binding protein
nal tract and causes gastroenteritis in humans. Both SipA in bacterial internalization. Science 283:2092–
bacteria enter the nonphagocytic cells during infec- 2095.
tion by stimulating endocytosis with effector mol-
ecules secreted by their type III systems. Endocytosis 69. Anderson, D. M., and O. Schneewind. 1997. A
refers to the uptake by localized regions of the plasma mRNA signal for the type III secretion of Yop pro-
membrane that surrounds what is at the cell surface teins by Yersinia enterocolitica. Science 278:1140–
and pinches off an intracellular vesicle containing the 1143.
ingested material. For certain types of endocytosis, 70. Page, A.-L., and C. Parsot. 2002. Chaperones
such as the uptake of pathogens, membrane protru- of the type III secretion pathway: jacks of all trades.
sions surround the pathogen and fuse to form the Mol. Microbiol. 46:1–11.
vesicle. All forms of endocytosis require remodeling
71. Lee, S. H., and J. E. Galán. 2004. Salmonella type
of the cell cortex because shape changes must occur.
III secretion-associated chaperones confer secretion-
The actin cytoskeleton is an important part of the cell
pathway specificity. Mol. Microbiol. 51:483–495.
cortex, and it takes part in the remodeling, includ-
ing the formation of membrane protrusions required 72. San Ho Lee and J. E. Galán, in ref. 71, their pub-
for the uptake of pathogens. The mammalian cell lication showing the requirement for the chaperone-
contains a pool of actin monomers in the cytoplasm, binding domain as well as the N-terminal signal for
and these monomers can be recruited to form actin secretion of Salmonella proteins through the type III
filaments that can take part in shape change. Shigella system, cite earlier publications on Yersinia indicat-
binds to the surface of the epithelial cells and induces ing that about 20 amino-terminal acids will mediate
the formation of microspike-like protrusions from the secretion of some reporter proteins via the type
the cell surface, containing actin filaments aligned III system into the extracellular environment, and
that some Yersinia effector proteins can be secreted 75. Silhavy, T. J. 1997. Death by lethal injection.
in the absence of the chaperone-binding region or Science 278:1085–1086.
in the absence of these amino-terminal acids. They
76. Cornelis, G. R. 2002. Yersinia type III secretion:
suggest that perhaps that secretion through the fla-
send in the effectors. J. Cell Biol. 158:401–408.
gellar apparatus may have occurred in the Yersinia
studies. 77. Trosky, J. E., A. D. B. Liverman, and K. Orth.
2008. Yersinia outer proteins: YOPs. Cell. Microbiol.
73. Yersinia multiply outside the host cells and can 10:557–565.
also multiply within certain cell types, such as non-
activated macrophages in the host. Yersinia entero- 78. Young, B. M., and G. M. Young. 2002. Evidence
colitica causes a severe enterocolitis, especially in for targeting of Yop effectors by the chromosomally
young children. The symptoms include fever, diar- encoded Ysa type III secretion system of Yersinia
rhea, and abdominal pain. The disease reservoir is enterocolitica. J. Bacteriol. 184:5563–5571.
other animals (e.g., farm animals, cats, dogs). This 79. When Y. enterocolitica were grown in certain
enterocolitis is transmitted to humans via contami- media, they formed needles composed of the YscF
nated foods (e.g., milk, undercooked meat). The protein. When cells with preformed needles were
bacterium, which can also be found in well water incubated with HeLa cells in media that did not
and lakes, produces an enterotoxin that causes diar- allow the formation of needles, the HeLa cells did not
rhea. Y. enterocolitica grows in the lumen of the undergo cytotoxic damage. However, if the medium
bowel, but it can also adhere to and invade special- allowed needle formation, the HeLa cells were dam-
ized epithelial cells, called M cells, of the intestinal aged. Thus, contact with the HeLa cells was presumed
mucous membranes. The M cells overlie clusters of to trigger the formation of new needles, and as a con-
lymphoid tissue (T cells, B cells, macrophages) called sequence the HeLa cells were punctured and toxins
Peyer’s patches. Y. enterocolitica (and Y. pseudotu- were translocated. It was suggested that the polymer-
berculosis) invade the M cells but do not replicate ization of protein subunits into needles provides the
there. Instead, they pass through the M cells to the force to penetrate the target cell. See ref. 63.
underlying lymphoid tissue, enter macrophages, and
spread via the macrophages to mesenteric lymph 80. Lee, V. T., S. K. Mazmanian, and O. Schneewind.
nodes, where they replicate. Y. pseudotuberculosis 2001. A program of Yersinia enterocolitica type III
is primarily a pathogen of wild and domestic ani- secretion reactions is activated by specific signals. J.
mals and is transmitted to humans by the eating of Bacteriol. 183:4970–4978.
infected meat. The bacteria cause necrotic lesions in 81. Day, J. B., F. Ferracci, and G. V. Plano. 2003.
the liver, spleen, and lymph nodes. In humans there Translocation of YopE and YopN into eukaryotic
is swelling of the mesenteric lymph nodes, accompa- cells by Yersinia pestis yopN, tyeA, sycN, yscB, and
nied by severe abdominal pain and, usually, diarrhea IcreG deletion mutants measured using a phospho-
and fever. Yersinia pestis causes plague. The bacte- rylatable peptide tag and phosphospecific antibod-
ria live in wild rodents (rats, mice, ground squirrels, ies. Mol. Microbiol. 47:807–823.
chipmunks, prairie dogs) and are transmitted to
humans by fleas that have bitten these rodents and 82. The inflammatory response, which is important
then bite humans. The wild rodents are relatively for fighting infections, includes localized vasodilation
resistant to disease. The bacteria spread to the drain- and increased capillary permeability. These effects
ing lymph nodes, where they grow. A high fever are associated with the influx from the blood vessels
and swollen lymph nodes (buboes) develop, and of phagocytic cells into the infected tissue. Cytokines
the infection at this stage is called bubonic plague. are very important in the infection-fighting process.
If the infection is not treated, the bacteria spread to Cytokines are low molecular weight proteins secreted
the blood and cause a septicemia. Once in the blood, primarily by activated lymphocytes and activated
the bacteria spread throughout the body to all the macrophages, as well as by other cells in the body.
organs. Subcutaneous bleeding from ruptured blood They are cell-to-cell signaling molecules that acti-
cells produces black spots on the skin, and the dis- vate the recipient cell. Cytokines produced by white
ease at this stage has been called Black Death. If the blood cells such as lymphocytes (responsible for the
lungs become infected, the infection is called pneu- immune response), eosinophils, and basophils, as
monic plague. Mortality from pneumonic plague well as by macrophages (phagocytic cells found in the
approaches 100%. All three pathogenic species of tissues), are called interleukins. Interleukin 8 (IL-8)
Yersinia can prevent uptake by specialized phago- is secreted by activated macrophages, as well as by
cytic cells such as macrophages and polymorphonu- endothelial cells that compose the inner lining of
clear leukocytes. Hence they replicate extracellularly blood vessels. IL-8 targets the phagocytic white blood
and are found primarily outside the cells during an cells called neutrophils, inducing the adherence of
infection. Prevention of phagocytosis is due to the circulating neutrophils to vascular endothelial cells.
type III secretion system, as described in the text. This action promotes the exiting of the neutrophils
from the blood vessels into tissues spaces, where they
74. Cornelis, G. R., and H. Wolf-Watz. 1997. The phagocytize and kill infectious agents. IL-8 is also a
Yersinia Yop virulon: a bacterial system for subvert- chemoattractant for the neutrophils. Tumor necrosis
ing eukaryotic cells. Mol. Microbiol. 23:861–867. factor α (TNF-α) is a cytokine secreted by activated
www.ebook777.com
macrophages during an infection. Macrophages lulanases are used industrially as starch-debranching

infected by gram-negative bacteria can be stimulated enzymes as an early step in the conversion of starch to
to secrete increased amounts of TNF-α by bacterial saccharides. Further saccharification is catalyzed by
lipopolysaccharide (LPS). (The secretion of IL-8 is α- or β-amylases. There are four types of pullulanase,
LPS independent.) TNF-α stimulates the adherence and they differ in the specificities of the bonds that
of circulating neutrophils to the endothelial lining are cleaved. The pullulanase from Klebsiella pneu-
of blood vessels and their exit into the tissue spaces. moniae hydrolyzes α-1,6 linkages.
TNF-α also stimulates macrophages and endothelial
87. It has been pointed out that the use of Gsp to
cells to produce IL-8, which as mentioned, also pro-
denote secreton proteins can be misleading because
motes the exit of neutrophils into the tissue spaces
the secreton is not part of the GSP pathway.
to fight infections. Since certain Yop proteins, such
Furthermore, the Tat pathway can be used, as well
as YopJ, inhibit IL-8 and TNF-α secretion, they are
as the GSP pathway for the translocation of proteins
anti-inflammatory.
across the cell membrane prior to secretion by the
83. Viboud, G. I., S. S. Kin, M. B. Ryndak, and J. secreton. Because of this, the type II pathway is some-
B. Bliska. 2003. Proinflammatory signaling by the times referred to as the secreton-dependent pathway
type III translocation factor YopB is counteracted (SDP), and the proteins are denoted simply as “pro-
by multiple effectors in epithelial cells infected tein C,” “protein D,” and so on. For more discus-
with Yersinia pseudotuberculosis. Mol. Microbiol. sion of this point, see: Desvaux, M., N. J. Parham, A.
47:1305–1315. Scott-Tucker, and I. R. Henderson. 2004. The gen-
84. A summary of the functions of the Yop proteins: eral secretory pathway: a general misnomer? Trends
YopE, YopH, YopO (also known as YpkA), and Microbiol. 12:306–309.
YopT inhibit phagocytosis by macrophages and poly- 88. Protein D forms an outer membrane pore (secre-
morphonuclear leukocytes. YopE and YopT disrupt ton). Protein S is an outer membrane lipoprotein
the actin-containing microfilaments in the target cells thought to stabilize the protein D complex. Protein E
and in this way collapse the cytoskeleton, as well as appears to be an ATPase involved in providing energy
inhibiting phagocytosis. YopH is a protein tyrosine for the translocation and assembly of proteins G to
phosphatase. It dephosphorylates macrophage pro- K, which form a pseudopilus in the cytoplasm. The
teins, and in this way has been suggested to disrupt role for protein F is not known. Proteins G to K are
phosphate-dependent signal transduction neces- homologous to type IV pili and are suggested to form
sary for normal phagocytosis. YopH has also been a pseudopilus in the periplasm. Protein O is a prepilin
reported to dephosphorylate scaffolding proteins peptidase involved in processing and methylating the
that are important for the formation of focal adhe- N-terminal ends of proteins G to K. It is held to be
sions. When these proteins are dephosphorylated, unlikely that the pseudopilus provides a channel in
integrin-mediated phagocytosis of bacteria by host the periplasm because the secreted proteins are folded
cells is inhibited. YopH also prevents the production and the pilus is too narrow for the folded proteins.
of toxic forms of oxygen that are produced by mac- One model postulates that the pseudopilus pushes
rophages upon phagocytosis of pathogens. YopO the secreted protein through the outer membrane
(YpkA) binds to actin, but the mechanism by which it pore. The idea is that the cytoplasmic membrane-
inhibits phagocytosis is not clear. YopJ (also known bound pseudopilus would retract (depolymerize)
as YopP) inhibits the production of proinflammatory and extend (polymerize), and in doing so would push
cytokines such as TNF-α and IL-8 by infected cells the folded protein into the protein D pore in the outer
(e.g., macrophages). Thus, inflammation, which is a membrane, or open the gated channel. This might
way for the body to fight infections, is decreased. The be similar to the retraction and extension of type IV
intracellular action of YopM is unknown, although pili described in twitching motility for Pseudomonas
the protein does localize to the nucleus. (Chapter 21) and Myxococcus (Chapter 23). For a
85. YopE activates the GTPase activity of certain discussion of this model, as well as other aspects of
other proteins called RhoA, Rac, and Cdc42. These type II secretion, see: Sandkvist, M. 2001. Biology
proteins bind GTP and as a consequence activate of type II secretion. Mol. Microbiol. 40:271–283;
actin polymerization and phagocytosis. When YopE Camberg, J. L., and M. Sandkvist. 2005. Molecular
accelerates the hydrolysis of GTP by these proteins, analysis of the Vibrio cholerae type II secretion
it inactivates them with respect to actin polymeriza- ATPase EpsE. J. Bacteriol. 187:249–256.
tion. YopT inactivates RhoA, Rac, and Cdc42 but by 89. Sandkvist, M., L. O. Michel, L. P. Hough, V. M.
a different mechanism. It is a protease that detaches Morales, M. Bagdasarian, M. Koomey, V. J. Dirita,
these proteins from the membrane, and thus inacti- and M. Bagdasarian. 1997. General secretion path-
vates them. way (eps) genes required for toxin secretion and
86. Pullulan is a starchlike polysaccharide formed outer membrane biogenesis in Vibrio cholerae. J.
by the fungus Pullularia pullulans. Pullulanases are Bacteriol. 179:6994–7003.
enzymes that catalyze the cleavage of certain α-1,4 90. d’Enfert, C., A. Ryter, and A. P. Pugsley. 1987.
and/or α-1,6 linkages in starch and related oligosac- Cloning and expression in Escherichia coli of the
charides such as pullulan. These enzymes have been Klebsiella pneumoniae genes for production, surface
isolated from bacteria, archaea, and fungi. Some pul- localization, and secretion of the lipoprotein pullula-
nase. EMBO J. 6:3531–3538. chain of Escherichia coli. Proc. Natl. Acad. Sci. USA
97:10884–10889.
91. Dautin, N., and H. D. Bernstein. 2007. Protein
secretion in gram-negative bacteria via the autotrans- 101. During aerobic growth, reduced DsbB is oxi-
porter pathway. Annu. Rev. Microbiol. 61:89–112. dized by ubiquinone. Electrons are transferred
from ubiquinone to oxygen via cytochrome oxi-
92. Kim, D. S. H. , Y. Chao, and M. H. Saier Jr.
dase bd or bo. During anaerobic growth, reduced
2006. Protein-translocating trimeric autotransport-
DsbB is oxidized by menaquinone, and sometimes
ers of gram-negative bacteria. 2006. J. Bacteriol.
ubiquinone, depending upon the terminal electron
188:5655–5667.
acceptor. For a description of these two electron
93. Christie, P. J., and J. P. Vogel. 2000. Bacterial type carriers and the roles that they play in electron
IV secretion: conjugation systems adapted to deliver transport, see Section 5.7.1. As far as the struc-
effector molecules to host cells. Trends Microbiol. ture of DsbB is concerned, the DsbB protein has
8:354–360. four transmembrane and two periplasmic loops.
Each of the periplasmic loops has four cysteine
94. Christie, P. J., K. Atmakuri, V. Krishnamoorthy,
residues. These are Cys41 and Cys44 (Cys41-
S. Jakubowski, and E. Cascales. 2005. Biogenesis,
X1-X2-Cys44), which form a disulfide bond, and
architecture, and function of bacterial type IV secre-
Cys104 and Cys130, which also form a disulfide
tion systems. Annu. Rev. Microbiol. 59:451–485.
bond. One way to describe what is happening is
95. Yeo, H.-J., and G. Waksman. 2004. Unveiling in terms of transferring disulfide bonds. Thus, the
molecular scaffolds of the type IV secretion system. disulfide bond in DsbB between Cys41 and Cys44
J. Bacteriol. 186:1919–1926. is transferred to Cys104 and Cys130, and from
there to Cys30 and Cys33 in DsbA. DsbA transfers
96. Grohmann, E., G. Muth, and M. Espinosa. 2003.
its disulfide bond to the target periplasmic protein
Conjugative plasmid transfer in gram-positive bacte-
which then folds appropriately. The disulfide bond
ria. Microbiol. Mol. Biol. Rev. 67:277–301.
between Cys41 and Cys44 in DsbB is regenerated
97. Cascales, E., and P. J. Christie. 2004. Definition by transferring electrons to quinones in the elec-
of a bacterial type IV secretion pathway for a DNA tron transport chain. Another way of stating this
substrate. Science 304:1170–1173. is that DsbA reduces the disulfide bond at Cys104–
Cys130 in DsbB, and in so doing regenerates its
98. Jakubowski, S. J., V. Krishnamoorthy, E.
own disulfide bond that can accept electrons from
Cascales, and P. J. Christie. 2004. Agrobacterium
cysteine residues in the target periplasmic protein.
tumefaciens VirB6 domains direct the ordered export
The electrons are then transferred from the sulfhy-
of a DNA substrate through a type IV secretion sys-
dryl groups at Cys104 and Cys130 in DsbB to the
tem. J. Mol. Biol. 341:96–977.
disulfide bond at Cys41–Cys44. the resulting sulf-
99. Ritz, D., and J. Beckwith. 2001. Roles of thiol– hydryl groups at Cys41 and Cys44 are then reoxi-
redox pathways in bacteria. Annu. Rev. Microbiol. dized by the electron transport chain, regenerating
55:21–48. the disulfide bond.
100. Kadokura, H., M. Bader, H. Tian, J. C. A. 102. Sone, M., Y. Akiyama, and K. Ito. 1997.
Bardwell, and J. Beckwith. 2000. Roles of a con- Differential in vivo roles played by DsbA and DsbC
served arginine residue of DsbB in linking protein in the formation of protein disulfide bonds. J. Biol.
disulfide-bond-formation pathway to the respiratory Chem. 272:10349–10352.
www.ebook777.com
19
Responses to Environmental Cues
Bacteria may alter cell morphology, the genes are repressed. The activation or inactiva-
production of flagella, cell metabolism, gene tion of the transcription factor itself as a result
transcription, and cell behavior in response to of the signaling pathway usually occurs in one
environmental fluctuations such as the avail- of two ways: by covalent modification [e.g.,
ability of respiratory electron acceptors, the phosphorylation (activation) or dephosphory-
supply of carbon, nitrogen, or phosphate, lation (inactivation)] of the transcription factor
changes in the osmolarity and temperature of or by a conformational change of the transcrip-
the medium, and whether the microorganisms tion factor upon binding a coinducer.
are growing in liquid or on a solid surface. In Two-component systems, which will be
this way bacteria survive in and adapt to ever- described in Section 19.1, are the most com-
changing environmental conditions. mon type of signaling system. They include a
Indeed, bacteria can be found in essentially histidine kinase (HK) protein that autophos-
all environments where life is possible, and phorylates at a histidine residue in response to
from this point of view they can be considered a signal, then transfers the phosphoryl group
to be one of the most successful forms of life. to an aspartate residue in a response regulator
Underlying the adaptations and responses are (RR) protein that regulates gene transcription
sophisticated detection systems with which the in its phosphorylated form. There may be inter-
bacterium continuously monitors the environ- mediate protein carriers that transfer the phos-
ment and transmits signals across its cell mem- phoryl group from the HK to the RR. Among
brane to specific intracellular targets, which can the examples of two-component systems are the
be the transcriptional machinery, enzymes, or a following.
cellular component such as the flagellum motor. 1. The Arc system, which regulates gene tran-
This chapter describes some of these signaling scription in response to oxygen supply
circuits. 2. The Nar system, which regulates the tran-
There are many signaling pathways that scription of genes required for nitrate and
bacteria use to transmit environmental and nitrite metabolism
intercellular signals to a cell target, be it cellu- 3. The Che system, which regulates the rota-
lar proteins (e.g., flagellar motors, discussed in tion of flagellar motors (rather than gene
Chapter 20) or transcription factors that regu- transcription)
late gene expression. In most cases the signal 4. The RegB/RegA system in photosynthetic
causes the cells to activate or inactivate a cyto- bacteria, which regulates the transcription
plasmic transcription factor that regulates the of genes for the light-harvesting complex
transcription of a gene. Some genes are induced and the photoreaction center under anaero-
by activated transcription factors, and some bic conditions
482
responses to environmental cues 483
5. The Pho system, which regulates genes for cytoplasm cell membrane
phosphate assimilation when the supply of signal
1
phosphate is limiting
6. The EnvZ/OmpR system, which regulates
C HK N
the expression of genes for porin synthesis
7. The KdpABC system, which regulates
ATP
expression of genes for potassium ion 2
transport ADP
8. The Spo phosphorelay system for the regu-
lation of sporulation genes in Bacillus sub-
tilis P
9. The CtrA system, which regulates gene N
expression during the Caulobacter crescen- 3
RR
tus cell cycle
10. The Asg system in Myxococcus xanthus,
C
which regulates gene expression during P
multicellular development
11. The PhoQ/PhoP system, which regulates
4
gene expression important for the virulence
of Salmonella
12. The Bvg system, which regulates the
expression of virulence genes in Bordetella P
pertussis
13. The Agr system in Staphylococcus aureus, 5
P
which regulates virulence gene expression
14. The Vir system in Agrobacterium tume-
faciens, which regulates the expression of
genes necessary for the transfer of plasmid to target
DNA to plant cells
Fig. 19.1 Two-component regulatory systems. (1)
Also described is the FNR (fumarate nitrate A transmembrane histidine kinase (HK) is activated
reductase) system. It regulates the expression by a signal at its N-terminal domain. (2) The acti-
of genes required for fermentation and anaer- vated protein autophosphorylates in the C-terminal
obic respiration but is not a two-component domain. (3) The response regulator protein (RR)
system. binds to the C-terminal end of the histidine kinase
and the phosphoryl group is transferred from the
histidine kinase to the response regulator, thus acti-
19.1 Introduction to Two- vating the latter. (4) The activated response regulator
Component Signaling Systems leaves the histidine kinase and stimulates its target.
In several systems (both Bacteria and Archaea, Shaded and stippled areas of the histidine kinase and
as well as eukaryotes), a signal transduction response regulator represent conserved amino acid
pathway exists, called a two-component sys- sequences typical for the respective class of protein.
The change in shape of the proteins represents a pre-
tem. Each two-component system includes a
sumed conformational change. In some systems the
histidine kinase (HK) protein (often called a histidine kinase is cytoplasmic and detects signals
sensor kinase) that receives a signal, which can within the cytoplasm.
be light, temperature, nutrients, toxins, osmotic
stress, cell density, and misfolded proteins, and
transmits it to a partner response regulator (RR) Histidine kinases have two domains, an input
protein, sometimes via other proteins that take domain and a transmitter domain. Response
part in a phosphorelay. See refs. 1 and 2 for a regulators have two domains, a receiver domain
review. The response regulator protein in turn and an output domain. See note 3 for a further
transmits the signal to the target. This is sum- discussion of domains in sensor kinases and
marized in Fig. 19.1. response regulators. Specifically, the histidine
www.ebook777.com
kinase receives a signal at its input domain and proteins called chemoreceptors, or MCP
autophosphorylates (using ATP as the phos- proteins, are sensor proteins that respond to
phoryl donor) at a histidine residue in its trans- chemoeffectors and, as a consequence, change
mitter domain. The transmitter domain is the the activity of a cytoplasmic histidine kinase
carboxy-terminal region, comprising approxi- (CheA). Section 19.5 presents another exam-
mately 240 amino acids. The histidine kinase ple of an initial receiver of the signal that is
then transfers the phosphoryl group to an not the histidine kinase. It occurs in the PHO
aspartate residue in the receiver domain of the regulon control system, which is repressed
partner response regulator protein. The receiver by inorganic phosphate. (A regulon is a set of
domain is the amino-terminal region, compris- noncontiguous genes or operons controlled
ing about 120 amino acids. This activates the by the same transcription regulator.) The pro-
response regulator, which transmits the signal to teins that initially bind inorganic phosphate
its target via its output domain (Section 19.1.2 are in the phosphate transport system (Pts),
and Fig. 19.1). Most of the known phosphory- which is believed to bind inorganic phosphate
lated response regulators (RR-P) bind to DNA and then stimulate the enzymatic activity of
and stimulate or repress the transcription of the membrane-bound PhoR histidine kinase.
specific genes. Exceptions include P-CheB and A third (and more complicated) example is
P-CheY, which affect the chemotaxis machin- found in the Ntr regulon, which is repressed
ery (Chapter 20). Histidine kinases may reside by ammonia (Section 19.4). The ammonia
in the cell membrane (usually transmembrane) levels determine the concentrations of glu-
or in the cytoplasm, although they are often in tamine and α-ketoglutarate via the enzymes
the membrane. The response regulators are in glutamine synthetase and glutamate synthase,
the cytoplasm. respectively. (As discussed in Section 10.3.1,
The signaling pathway also includes a phos- these enzymes together function to incorpo-
phatase that dephosphorylates the response reg- rate ammonia into glutamate and glutamine,
ulator, returning it to the nonstimulated state, which donate amino groups to other molecules
where it once again can respond to the signal. during biosynthesis.) The α-ketoglutarate and
The phosphatase may be the histidine kinase glutamine in turn influence the activity of a
itself, the response regulator, or a separate pro- bifunctional enzyme, uridydyltransferase–uri-
tein. Even though signaling systems that consist dylyl-removing (UT-UR) enzyme, which mod-
of a histidine kinase and a response regulator ifies a signal transduction protein (PII), which
protein are called “two-component” systems, in turn regulates the activity of a cytoplasmic
often there are more than two proteins in the histidine kinase (NII). In this case the histidine
signal transduction pathway, since additional kinase is indeed far removed from the initial
proteins may exist that carry the phosphate signal, ammonia.
from the histidine kinase to the response regu- Two-component signaling systems have
lator protein. Proteins that make up the phos- been discovered in many bacteria, both gram-
phorelay pathway between the histidine kinase negative and gram-positive, and have been
and the partner response regulator protein are implicated in a wide range of physiological
called phosphotransferases. Two-component responses.6–8 These include nitrogen assimila-
signaling systems involving histidine kinases tion, outer membrane porin synthesis, chemo-
also occur in the archaea, fungi, plants, slime taxis in E. coli and S. typhimurium, nitrogen
molds, and presumably other eukaryotes.4,5 fixation in Klebsiella and Rhizobium, spo-
It must be emphasized that the histidine rulation in Bacillus, fruiting body formation
kinase need not be the first protein in the sig- in myxobacteria, oxygen regulation of gene
nal transduction pathway to respond to the expression in E. coli, the uptake of carboxylic
signal. In other words, it need not be the sen- acids in Rhizobium and Salmonella, the pro-
sor. In many systems, signals first interact with duction of virulence factors by Salmonella and
protein(s) other than the histidine kinase, and Bordetella, and bioluminescence in some Vibrio
the stimulus is relayed to the histidine kinase. species. It is clearly a widespread and important
For example, in the E. coli chemotaxis system signal transduction system that enables bacteria
described in Chapter 20, the transmembrane to adapt to changes in the external milieu. As
mentioned, there is also evidence to suggest that protein, hence in a suppression of a particular
similar systems occur in eukaryotes.9,10 response rather than an activation.)
In Fig. 19.1, the histidine kinase (HK) is
19.1.1 Components of two-component depicted as a transmembrane protein composed
signaling systems of three domains: an N-terminal domain that is
We noted at the outset that the “two-compo- presumed to be at the outer surface of the cell
nent” systems contain proteins of three types. membrane and to bind to an external signal-
ing ligand; a hydrophobic domain that is trans-
1. A histidine kinase (HK), sometimes called a membrane; and a conserved C-terminal domain
sensor kinase. The histidine kinase receives a that is cytoplasmic. Some histidine kinases are
signal at its input domain and autophospho- cytoplasmic proteins. (See the discussion of the
rylates at a histidine residue in its transmitter chemotaxis protein CheA in Chapter 20.) Signal
domain: transduction as depicted in Fig. 19.1 can be con-
HK + ATP ⎯⎯ → HK-P + ADD
signal veniently thought of as occurring in three steps.
2. A partner response regulator (RR), also Step 1. In response to a stimulus at the N-terminal
called a cognate response regulator, which is domain, the histidine kinase autophospho-
phosphorylated at an aspartate residue in its rylates at the C-terminal domain. The phos-
receiver domain by HK-P and sends a signal phoryl donor is ATP.
to its target (e.g., the genome or the flagella Step 2. The phosphoryl group is transferred from
motor) via its output domain: the histidine kinase to its partner response reg-
ulator protein. All the response regulator pro-
RR + HK-P → RR-Pactive + HK teins are related in having conserved amino
However, not all signals result in increased acid sequences in the N-terminal domain
synthesis of RR-P. See Section 19.1.2. (usually) that may bind to the conserved
Furthermore, as noted earlier, there may C-terminal region of the histidine kinases.
exist enzymes called phosphotransferases Step 3. After phosphorylation, the response
that carry the phosphate from the HK to the regulator becomes activated and changes the
RR in what is referred to as a phosphorelay activity of its target. The effect is usually the
pathway. stimulation or repression of gene transcrip-
tion. As we shall see, some response regula-
3. A phosphatase, which inactivates the RR-P:
tor proteins (e.g., NarL) do both, depending
RR-P + H2O → RR + Pi upon the target gene (Section 19.3). Other
response regulators may have targets other
The phosphatase may be the histidine kinase,
than the genome. For example, as described
the response regulator, or a separate protein.
for chemotaxis in Chapter 20, the phospho-
As mentioned, some signals stimulate phos-
rylated derivatives of the response regulators
phatase activity and thus act as inhibitors or
CheY and CheB change the rotational direc-
repressors rather than stimulators or induc-
tion of the flagellar motors, and the extent of
ers. We will see some examples of this later.
methylation of the chemoreceptor proteins
19.1.2 Signal transduction in in the membrane, respectively. The response
regulator proteins differ at their C-terminal
two-component systems
domains, and this probably confers specific-
In the simplified model of signal transduction
ity with regard to their targets and activities.
in two-component systems illustrated in Fig.
19.1, the signal stimulates phosphorylation of Not all two-component systems respond to
the response regulator protein. (The model will the signal by increasing the phosphorylation
be modified later to accommodate differences of the response regulator protein. Sometimes
among signaling systems, such as the inclusion the signal results in either dephosphorylation
of a separate sensor protein that receives the of the response regulator protein or inhibition
signal and transmits it to the histidine kinase, of the phosphorylation of the response regu-
and to show systems in which the signal actually lator protein. In either case, the level of RR-P
results in less phosphorylation of the regulatory falls rather than rises in response to the signal.
www.ebook777.com
The signal can result in the dephosphorylation response regulator, PhoB (Section 19.5). In the
of RR-P when the histidine kinases are bifunc- case of chemotaxis to an attractant signal, the
tional enzymes that can act either as kinases signal suppresses histidine kinase activity rather
or as phosphatases when stimulated, depend- than stimulating the activity. This results in the
ing upon the particular histidine kinase. For formation of less RR-P (i.e., CheY-P, the pro-
example, in the repression of the Ntr regulon by tein that causes the cells to tumble and swim
ammonia, the signal causes stimulation of the randomly: see Chapter 20), and consequently
phosphatase activity of the histidine kinase, and the cells swim smoothly toward the attractant.
as a consequence, represses gene transcription.
In the presence of excess ammonia, the histidine 19.1.3 Amino acid sequences define
kinase NRII acts as a phosphatase rather than histidine kinases and response
as a kinase and inactivates the response regula- regulator proteins
tor, NRI, which in its phosphorylated form is The histidine kinases are defined by a con-
a positive transcription factor (Section 19.4.1). served sequence of about 200 amino acids at
Another example may be the repression of the the C-terminal end. The C-terminal domain
PHO regulon by inorganic phosphate. The histi- is the site of the conserved histidine residue
dine kinase, PhoR, appears to respond to excess that becomes phosphorylated in response
inorganic phosphate by dephosphorylating the to a stimulus (Fig. 19.2). As indicated in Fig.
Fig. 19.2 Structures of the histidine protein kinases. The histidine protein kinases are part of the two-com-
ponent regulatory systems. They autophosphorylate at a histidine residue and then transfer the phosphate to
a response regulator protein. Most of the histidine kinases are believed to be transmembrane proteins with
an extracytoplasmic amino terminus that responds to stimuli. Exceptions are CheA, NRII, and FrzE, which
are cytoplasmic. Hydrophobic domains that presumably span the membrane are indicated by the solid boxes
at the amino-terminal end. Domain I, the cytoplasmic domain that includes the phosphorylated histidine
residue, is indicated by the solid hourglass symbol. Regions II and III (stippled and hatched boxes) represent
regions in the carboxy-terminal domain where certain amino acids appear with high frequency at specific
locations in the sequence. These are called conserved regions. For example, when the amino acid sequences
are lined up for comparison, position 43 may be a glycine in all the proteins (totally conserved), whereas posi-
tion 44 may be arginine in, say, 60% of the proteins, and so on (partially conserved). The open boxes at the
extreme carboxy ends are homologous to response regulator domains at their amino-terminal ends. Lengths
of proteins in amino acid residues, as predicted from nucleotide sequences, shown in right-most column.
Source: Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Protein phosphorylation and regulation of adaptive
responses in bacteria. Microbiol. Rev. 53:450–490. Reproduced with permission from American Society for
Microbiology.
19.1, most of the known histidine kinases are conserved histidine residue in the cytoplasmic
transmembrane. The C-terminal end is in the domain (Fig. 19.1).
cytoplasm, where it interacts with the response The response regulator proteins are defined
regulator protein. The N-terminal end may be by a conserved amino-terminal domain of
exposed on the extracellular membrane sur- about 100 amino acids (Fig. 19.3). The con-
face (the periplasm in gram-negative bacteria). served amino-terminal end of the response
As stated earlier, the amino acid sequence at regulator protein is thought to interact with the
the N-terminal domain varies with the dif- conserved carboxy end of the histidine kinase,
ferent histidine kinases, presumably because becoming phosphorylated at an aspartate resi-
they respond to different stimuli. The response due. The phosphate is eventually removed by a
to the stimulus within the N-terminal region phosphatase, which may be the histidine kinase,
causes the enzyme to autophosphorylate the the response regulator protein, or perhaps a
Fig. 19.3 Structures of the response regulator proteins. Response regulator proteins are cytoplasmic proteins
that are phosphorylated, or presumed to be phosphorylated, by the histidine kinase proteins. The phospho-
rylated regulator proteins transmit the signal to the genome or to some other cellular machinery (e.g., the
flagellar motor). The amino-terminal region of the response regulator protein has conserved amino acids
(open boxes). Approximately 20 to 30% of the amino acids are identical at corresponding posititions when
the sequences are aligned. This is the region thought to interact with the carboxy domain of the histidine
kinase, and to become phosphorylated. Other conserved regions are indicated by cross-hatched boxes. These
are in the carboxy-terminal domain, which is thought to interact with target molecules. For example, NRI,
DctD, and NifA share a homologous carboxy-terminal domain that is thought to interact with one of the E.
coli sigma factors, σ54. NRI and NifA have a common carboxy-terminal region that is thought to bind to DNA.
One of the proteins, ToxR, spans the membrane, and the hydrophobic region is indicated by the solid box.
Lengths of the protein in amino acid residues shown as in Fig. 19.2. Source: Stock, J. B., A. J. Ninfa, and A.
M. Stock. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev.
53:450–490. Reproduced with permission from American Society for Microbiology.
www.ebook777.com
third protein. Since the carboxy ends of the his- important changes take place (Fig. 19.4B).
tidine kinases and the amino ends of the differ- These include the replacement of fumarase A
ent response regulator proteins are conserved, and succinate dehydrogenase by fumarase B
it is theoretically possible that a histidine kinase and fumarate reductase, respectively, and the
in one signaling system may activate a response repression of the synthesis of α-ketoglutarate
regulator of a different system. This possible dehydrogenase. As described in Section 9.10,
interaction between different signaling systems the result of these enzymatic changes is the con-
would be an example of what has been termed version of the oxidative citric acid cycle into a
“cross-regulation,” the regulation of a response reductive noncyclic pathway. (Not all bacte-
regulator by a signal that comes from a source ria have a reductive citric acid pathway when
other than the cognate histidine kinase.11 Cross- they respire anaerobically. For example, when
regulation is discussed in Section 19.5.1. Pseudomonas stutzeri grows anerobically,
using nitrate as an electron acceptor, it oxi-
19.2 Responses by Facultative dizes glucose completely to CO2 and appears to
have an oxidative citric acid cycle.15 Similarly,
Anaerobes to Anaerobiosis
Paracoccus denitrificans has a complete citric
A shift from an aerobic to an anaerobic atmo- acid cycle when it grows anaerobically, with
sphere results in extensive changes in the nitrate as the electron acceptor.16)
metabolism of facultative anaerobes due to the Additionally, when E. coli is growing anaer-
repression of genes required for aerobic growth obically, acetyl–CoA is no longer made by
and the induction of genes necessary for anaero- pyruvate dehydrogenase but rather by pyru-
bic growth.12 These adaptive responses, which vate–formate lyase. This is an advantage under
have stimulated much interest, are described anaerobic conditions because it decreases the
next. Most of the discussion focuses on what amount of NADH that must be reoxidized. The
has been learned from studying E. coli and acetyl–CoA thus formed is converted to acetate
related bacteria. However, similar systems exist and ethanol, and the formate is converted to
in other bacteria. The metabolic changes will be hydrogen gas and carbon dioxide via formate–
discussed first, followed by a description of the hydrogen lyase. In E. coli there is also a decrease
regulation of the relevant genes. Many genes in synthesis of other enzymes used during aero-
that are responsive to anaerobiosis are globally bic growth (e.g., those of the glyoxylate cycle
regulated by two systems: the Arc (two-com- and fatty acid oxidation).
ponent) system, and the FNR (not two-com- Depending upon the presence of particular
ponent), system.13,14 Then we shall describe electron acceptors, major changes also take
the regulation of the formate–hydrogen lyase place in the respiratory pathway.17 Facultative
system, which also responds to anaerobiosis anaerobes such as E. coli carry out aerobic res-
but is regulated by formate and a transcrip- piration in the presence of oxygen; but in the
tion activator, FhlA. The RegB/RegA system absence of oxygen, they carry out anaerobic
will also be described; this two-component respiration by means of nitrate, fumarate, or
system stimulates the transcription of genes for some other electron acceptor (e.g., TMAO or
the light-harvesting complex and photoreac- DMSO, compounds that occur in nature and
tion center of certain photosynthetic bacteria presumably exist in the intestine, where E. coli
under anaerobic incubation conditions. A fifth grows).
system, the (NarL/NarP/NarX/ NarQ) system, There is a hierarchy of electron acceptors
is a two-component system that regulates the that are used: oxygen is the most preferred,
response to nitrate and nitrite as electron accep- with nitrate second, followed by fumarate and
tors under anaerobic conditions, is discussed in the other electron acceptors. The hierarchy
Section 19.3. parallels the maximum work that can be done
when electrons travel over the electrode poten-
19.2.1 Metabolic changes accompanying tial gradient to the terminal acceptor. The
the shift to anaerobiosis work is proportional to the difference in elec-
During aerobic respiration, the citric acid trode potential between the electron acceptor
pathway is cyclic and oxidative (Fig. 19.4A). and donor. For example, the maximum work
However, in the absence of oxygen several that can be done is greatest when oxygen is the
Fig. 19.4 Comparison of respiratory systems and carbon metabolism in E. coli. (A) An oxidative citric acid cycle feeds electrons into succinic dehydrogenase (SDH) and
NADH dehydrogenase (a primary dehydrogenase). (B) The citric acid pathway is noncyclic and reductive because the α-ketoglutarate dehydrogenase activity is absent
(or severely diminished). The electron transport chain has been altered so that electrons flow to acceptors other than oxygen. When nitrate is present, nitrate reductase
(NR) is synthesized. Nitrate represses the synthesis of fumarate reductase (FRD) and DMSO:TMNO reductase. In the absence of an electron acceptor for anaerobic
respiration, the major energy yielding pathways are fermentative. Source: Spiro, S., and J. R. Guest. 1991. Adaptive responses to oxygen limitation in Escherichia coli.
Trends Biochem. Sci. 16:310–314. Copyright 1991, with permission from Elsevier.
www.ebook777.com
electron acceptor (E9m= + 0.82 V), less when 19.2.2 Regulatory systems that govern
nitrate is the electron acceptor (E9m = + 0.42 V), gene expression accompanying the shift to
and least when fumarate is the electron accep- anaerobiosis
tor (E9m = + 0.03 V). As described in Section 5.4, Four systems for the regulation of gene expres-
the electrons are passed to these terminal elec- sion by oxygen or nitrate in E. coli will be
tron acceptors from reduced quinone. For described. Two of these are activated by anaer-
example, ubiquinone (E9m = 0.1 V) transfers obiosis (ArcA/B and FNR), the third by nitrate
electrons from the reductant to either the cyto- and nitrite (NarL/NarP/NarX/NarQ system);
chrome oxidase or the nitrate reductase mod- the fourth is activated by anaerobiosis and for-
ule. Menaquinone (E9m = –0.074 V), rather than mate and repressed by nitrate (the FhlA regu-
ubiquinone, transfers electrons to the fumarate lon). Photosynthetic bacteria have a regulatory
reductase complex. (Fumarate reductase is at system, called the RegA/RegB system, which is
too low a potential to accept electrons from activated under anaerobic conditions and con-
ubiquinone.) trols the induction of photosynthetic genes, as
E. coli determines which electron acceptors decribed in Section 19.6.1. A common way to
will be used, in part by regulating the transcrip- test the regulation of gene transcription is to
tion of genes coding for the electron acceptors. construct a fusion between the promoter of the
Thus, in the presence of oxygen, the genes for gene of interest and the gene for β-galactosidase
nitrate reductase, fumarate reductase, and (lacZ). Then strains harboring the fusion are
the other reductases are repressed, and there- tested under different conditions to see how
fore only aerobic respiration can take place. the test conditions affect the production of
(However, see the discussion in Section 5.7.2 β-galactosidase. (A more detailed explanation
regarding denitrifying enzymes not sensitive to of lacZ fusions is given later, in note 26.) The
oxygen in some facultative anaerobes.) In the four systems in E. coli for the regulation of gene
absence of oxygen but in the presence of nitrate, expression by oxygen or nitrate are as follows:
nitrate reductase genes are induced (by nitrate),
but the genes for fumarate reductase and the 1. The Arc (aerobic respiratory control, or
other reductases are repressed by nitrate. anoxic redox control) system. This two-
Therefore, nitrate respiration takes place. In component system represses under anaero-
the absence of both oxygen and nitrate, there is bic conditions the transcription of several
no longer any repression of fumarate reductase genes that are expressed only during aerobic
and so fumarate respiration takes place. growth. In addition, it stimulates the tran-
Changes also take place within the aerobic scription of a much smaller number of genes
respiratory pathway. When oxygen levels are that are expressed during microaerophilic or
high, E. coli uses cytochrome o, encoded by the anaerobic conditions. The Arc system func-
cyo operon, as the terminal oxidase. When oxy- tions during aerobic and anaerobic growth.
gen becomes limiting (and during stationary 2. The FNR (fumarate nitrate reductase) sys-
phase), cytochrome o is replaced by cytochrome tem. This system stimulates the transcrip-
d, encoded by the cyd operon. One advantage to tion of many genes that are required for
this is that cytochrome d has a higher affinity fermentation and anaerobic respiration, and
for oxygen (Km = 0.23–0.38 μM) than has cyto- represses the transcription of some genes
chrome o (Km = 1.4–2.9 μM). that function only during aerobic growth.
In summary, then, oxygen represses the syn- More than 75 genes are regulated by FNR in
thesis of the anaerobic reductases, ensuring E. coli. The FNR system is active only during
that oxygen is used as the electron acceptor anaerobic growth; it is not a two-component
in air. Nitrate induces the synthesis of nitrate system.
reductase and represses the synthesis of the 3. The NarL/NarP/NarX/NarQ system, also
other reductases, ensuring that nitrate is used as called the Nar system. This two-component
the electron acceptor in the presence of nitrate system stimulates the transcription of the
anaerobically. In the absence of any exoge- nitrate reductase and other genes required
nously supplied electron acceptor, E. coli relies for nitrate and nitrite metabolism, and
on fermentation as its major source of ATP. represses the transcription of the other
terminal reductase genes. The system is 1. Anaerobic repression of genes for

active only under anaerobic growth condi- Citric acid cycle enzymes
tions in the presence of nitrate or nitrite. The Glyoxylate cycle enzymes
Nar system is described in Section 19.3. Several dehydrogenases for aerobic
4. The FhlA regulon, also called the formate growth (e.g., pyruvate dehydrogenase)
regulon. This system consists of genes Fatty acid oxidation enzymes
repressed by oxygen and nitrate and induced Cytochrome o oxidase
by formate. The regulon includes several genes 2. Anaerobic induction of the gene for pyru-
including those for formate–hydrogen lyase, vate formate–lyase18,19
which is necessary to convert the fermenta- 3. Induction in low oxygen of the genes for
tion end product formate to H2 and CO2. cytochrome d oxidase and cobalamin syn-
thesis20 and (along with FNR), the pyruvate–
The Arc system formate lyase gene18
The Arc system consists of a transmembrane sen-
sor kinase (histidine kinase), ArcB, and a partner Interestingly, ArcA-P, but not ArcB, may also
response regulator, ArcA. Mutations in the arc activate the transcription of the mating system
genes cause derepression of several genes and genes, as explained in note 21.
repression of a relatively short list of others, indi- When oxygen levels are sufficiently low, the
cating that phosphorylated ArcA is both a repres- ArcB protein autophosphorylates, using ATP
sor and an activator of gene transcription under as the phosphoryl donor, and then transfers
anaerobic conditions. By analyzing these mutants, the phosphoryl group to ArcA (Fig. 19.5).22,23
it has been possible to conclude that the Arc sys- How does ArcB detect changes in the levels of
tem is responsible for the following activities: oxygen? Apparently, it is not oxygen itself that
Fig. 19.5 The ArcA/ArcB regulatory system in E. coli. ArcB is a membrane protein activated by anoxia, perhaps
by a reduced form of an electron carrier. The model postulates that the activated form of ArcB becomes phos-
phorylated and then phosphorylates ArcA, which then becomes a repressor of aerobically expressed enzymes
and an inducer of cytochrome d oxidase. It has been found that ArcA-P also activates the genes for cobalamin
synthesis (cob genes) in Salmonella typhimurium. (See: Andersson, D. I. 1992. Involvement of the Arc system
in redox regulation of the cob operon in Salmonella typhimurium. Mol. Microbiol. 6:1491–1494.) ArcA also
responds to CpxA, a membrane-bound sensor protein that is necessary for the synthesis of the F-pilus and
other sex factor functions. Source: Spiro, S., and J. R. Guest. 1991. Adaptive responses to oxygen limitation in
Escherichia coli. Trends Biochem. Sci. 16:310–314. Copyright 1991, with permission from Elsevier.
www.ebook777.com
is the signaling molecule. This conclusion has (e.g., a quinone) may signal ArcB, which in turn
been reached from two lines of evidence: (1) the autophosphorylates and then phosphorylates
level of expression of the sdh (succinate dehy- ArcA (Fig. 19.5). The control of the phospho-
drogenase) operon in cells grown with different rylating activity of ArcB also appears to involve
terminal electron acceptors and (2) the study of certain cellular metabolites, such as pyruvate,
mutants with deletions in the cytochrome oxi- acetate, and D-lactate, that increase when oxy-
dase genes. These experiments are described gen is lacking. These products increase the auto-
next. phosphorylation activity of ArcB in vitro. It has
The level of expression of the sdh operon, been demonstrated in vivo that D-lactate allos-
which is under ArcA/ArcB control, varies with terically enhances the autophosphorylation
the midpoint potential of the electron accep- activity of ArcB that has already been activated
tor. (The expression of the sdh operon is com- by anaerobiosis, rather than signaling inactive
plex and is controlled by factors in addition to Arcb.28 Other intermediates may have similar
ArcA/ ArcB. See note 24 for further comment.) effects. These and other aspects of the regula-
The level of expression is highest with oxygen, tion of the ArcA/ArcB system are discussed by
lower with nitrate, and lowest with fumarate. Iuchi et al.29 and by Lynch and Lin.30
This parallels the midpoint potential of the elec-
tron acceptor, which is most positive for oxy- The FNR system
gen, less positive for nitrate, and least positive When E. coli is shifted from aerobic to anaero-
for fumarate. Instead of responding to the ter- bic growth, a number of genes required to grow
minal electron acceptors fumarate, nitrate, and anaerobically are induced. At the same time,
oxygen per se, the ArcB protein may respond to genes required for aerobic growth are repressed
a reduced carbon compound in the cell, such as (Fig. 19.4). Part of the regulation is due to a
the reduced form of an electron transport car- protein called FNR, which is encoded by the fnr
rier (e.g., flavoprotein, quinone, cytochrome), gene. The FNR protein is a positive regulator of
or NADH, or some metabolic intermediate that transcription for many genes that are expressed
might accumulate anaerobically. only during anaerobic growth, and a repressor
One way to test this hypothesis is to delete for certain genes that are expressed only dur-
the cytochrome o and d genes and measure the ing aerobic growth (Fig. 19.6).31,32 There is no
expression of genes under the control of the evidence that FNR is phosphorylated or is part
Arc system. Deletion of the cytochrome oxi- of a two-component regulatory system, and
dase genes inhibits electron transport at the it is discussed here in the context of a protein
terminal step and would be expected to have involved in aerobic/anaerobic regulation of
extensive metabolic consequences, including, gene expression.
for example, an increase in the ratio of reduced Genes whose expression requires FNR
to oxidized forms of electron carriers, and the include those coding for the anaerobic respira-
accumulation of metabolites that are formed in tory enzymes fumarate reductase (frdABCD)
the absence of an exogenous terminal electron and nitrate reductase (narGHJI), as well as sev-
acceptor. If this is the case, then deletion of the eral other enzymes, including pyruvate formate
cytochrome oxidase genes would be expected lyase (pfl genes), formate dehydrogenase-N (the
to mimic the absence of oxygen. This proposi- respiratory formate dehydrogenase, fdnGHI),
tion was tested, using cyo–lacZ and cyd–lacZ aspartase (aspA), anaerobic fumarase B (fumB),
fusions as probes to monitor the expression and glycerol-3-phosphate dehydrogenase
of the cyo and cyd genes.25 (Gene fusions are (glpA).
explained in note 26.) When both the cyo and In agreement with the role of activator of
cyd genes were deleted and the cells were grown gene expression, mutations in the fnr gene
in air, cyo–lacZ expression was lowered and result in an inability to grow anaerobically on
cyd–lacZ expression increased, as if oxygen fumarate or nitrate as electron acceptors. FNR
were absent. (See note 27 for a description of is also a negative regulator for several aerobi-
the control experiments.) cally expressed genes, including those for cyto-
One interpretation of these experiments is that chrome o oxidase (cyoABCDE), cytochrome
the reduced form of an electron transport carrier d oxidase (cydAB), succinate dehydrogenase
Fig. 19.6 FNR is a transcription regulator protein during anaerobic growth. In the absence of oxygen, FNR
becomes activated to become an inducer for many anaerobically expressed genes and a repressor for certain
aerobically expressed genes. It has been speculated that FNR might become activated by the reduction of
bound ferric ion, causing a conformational change in the protein. This is not an example of a two-component
regulatory system.
(sdhCDAB), and superoxide dismutase (sodA). of active FNR increase anaerobically solely at
Thus, FNR is a global regulator of gene expres- the post-translational level, transcription of
sion during anaerobic growth. It should be fnr is strongly stimulated in B. subtilis during
pointed out, however, that many of the genes oxygen limitation. The anaerobic activation of
regulated by FNR are also regulated by the transcription of fnr in B. subtilis requires resDE,
ArcA/ArcB system and the Nar system. a two-component system (resD is a response
Although FNR regulates gene activity during regulator gene and resE is a histidine kinase
anaerobic growth, studies have shown that the gene).39 The activity of FNR in B. subtilis may
FNR protein in E. coli is present in comparable also be regulated by anaerobiosis in a fashion
amounts in both aerobically and anaerobi- similar to its regulation in E. coli.
cally growing cells. However, it is believed to An interesting parallel exists between the
be largely in an inactive state during aerobic FNR protein and another transcriptional regu-
growth.33,34 So, how does E. coli regulate the lator, the cAMP receptor protein, CRP (cyclic
activity of FNR? The answer entails an iron–sul- AMP receptor protein, also called the catabolite
fur cluster in FNR. Active FNR is a homodimer activator protein, CAP). The CRP protein is a
of an iron–sulfur protein with an oxygen-labile positive transcriptional regulator for catabolite-
[4Fe–4S] cluster in each monomer. Upon expo- sensitive genes (i.e., genes repressed by glucose).
sure to oxygen the [4Fe–4S] cluster is oxidized CRP, in response to binding to cAMP, binds to
and can even be lost from the protein. (See note specific sites on the promoter region of the tar-
35 for a further explanation.) When this hap- get gene to activate transcription. A compari-
pens a fraction of the FNR loses its iron clusters son between the nucleotide-derived amino acid
and becomes an apoprotein.36–38 Both the pro- sequences of FNR and CRP reveals that FNR
tein bearing an oxidized cluster and the apopro- and CRP are very similar in structure: both
tein are inactive. They bind with low affinity to have a DNA-binding domain that allows the
DNA and do not stimulate transcription. Both dimer to bind and a nucleotide-binding domain
the oxidation of the iron–sulfur clusters and (although numerous attempts to show specific
their loss are reversible, and under anaerobic binding of nucleotides to FNR have failed). This
conditions the Fe–S clusters are restored. has led to the suggestion that FNR and CRP are
However, FNR expression need not be examples of a family of proteins that regulate
regulated exactly the same in all bacteria. For transcription at the promoter region of target
example, whereas in E. coli the transcription of genes. Indeed, there have been several reports of
fnr seems to be similar under both aerobic and proteins resembling FNR in bacteria other than
anaerobic growth conditions, and the amounts the enterics, and these FNR-like proteins take
www.ebook777.com
part in the regulation of a variety of metabolic lyase to prevent the medium from becoming too
activities. (See note 40 and refs. 41 and 42.) acidic owing to the production of formic acid
In summary, ArcA and FNR are activated by from pyruvate (the pyruvate–formate lyase
anaerobiosis and regulate the enzymological reaction). (E. coli actually makes three distinct
changes that accompany the shift from aerobic to formate dehydrogenases and three different
anaerobic growth. For example, activated ArcA hydrogenases. See note 43 for a discussion of
(ArcA-P) primarily represses several aerobically this point.)
expressed genes, including pyruvate dehydroge- The genes for formate–hydrogen lyase are
nase, succinate dehydrogenase, α-ketoglutarate part of a regulon that has two names: the FhlA
dehydrogenase, and cytochrome o oxidase (Fig. regulon and the formate regulon. The regulon
19.5). It is also a positive regulator for some is controlled by a transcription factor called
genes expressed when oxygen is low or lack- FhlA and by formate, which is a coactivator of
ing, including the cytochrome d oxidase gene FhlA. (See ref. 44 for a review; see note 45 for a
and the pyruvate–formate lyase gene. Similarly, further description of the regulon.) Under aero-
activated FNR represses several genes normally bic conditions, formate levels are kept low, and
expressed during aerobic growth, including the because of this the regulon is not induced.
genes for cytochrome o oxidase, succinate dehy- Two different mechanisms keep the formate
drogenase, and pyruvate dehydrogenase, and is levels low under aerobic conditions. The synthe-
required for the induction of many genes during sis of formate from pyruvate is prevented under
anaerobic growth, including pyruvate–formate aerobic conditions because the pfl gene, which
lyase, fumarate reductase, and nitrate reductase codes for pyruvate formate–lyase, which in
(cf. Figs. 19.5 and 19.6. Also, see note 24.) turn synthesizes acetyl–CoA and formate from
It is important to point out that in several pyruvate and CoASH, requires active FNR as a
cases, FNR and ArcA regulate the same gene. positive regulator. Since oxygen prevents acti-
For example, they are both positive regulators vation of FNR, oxygen represses the pfl gene,
for the pyruvate–formate lyase gene. Sometimes thus preventing formate synthesis. Oxygen and
they have opposite effects on the same gene. For nitrate also lower the levels of formate in the cell
example, ArcA is a positive regulator for the because they induce the synthesis of the respi-
cytochrome d oxidase gene, whereas FNR is a ratory formate dehydrogenases FDH-O and
negative regulator for that gene. Multiple con- FDH-N, which oxidize formate to CO2.
trols of the same gene are a common theme in
bacterial physiology. However, a note of cau-
tion must be introduced when one is attempting 19.3 Response to Nitrate and Nitrite:
to extract from physiological studies of mutants The Nar Regulatory System
conclusions about coregulation by ArcA and E. coli can be grown anaerobically by using
–
FNR. This is because FNR has been reported nitrate (NO3) as the electron acceptor. The pro-
to stimulate anaerobic arcA expression when cess is called nitrate respiration and is reviewed
this activity was monitored via arcA–lacZ in Section 5.7.1. (See the subsection entitled
fusions.42 Anaerobic respiratory chains.) In fact, when E.
coli is given a choice of electron acceptors such
Regulation of the formate–hydrogen as nitrate, nitrite, or fumarate under anaerobic
lyase pathway conditions, it will utilize the nitrate. This prop-
When enterobacteria such as E. coli are grown erty may be advantageous because nitrate has
anaerobically, formic acid is converted to H2 a more positive redox potential than the alter-
and CO2 via the enzyme complex formate–hy- native electron acceptors, and therefore more
drogen lyase (Figs. 15.7 and 19.4B). Formate– energy is potentially available from electron
hydrogen lyase consists of two enzymes, formate transport when nitrate is the electron accep-
dehydrogenase H and hydrogenase 3, whose tor. Nitrate is preferentially used as an electron
genes are repressed by oxygen and nitrate, and acceptor during anaerobic respiration because
induced by formate. Induction also requires an it induces the transcription of genes resulting
acidic pH in the external medium, which proba- in the synthesis of a membrane-bound nitrate
bly means that the cells make formate–hydrogen reductase and represses the transcription of
genes encoding the other reductases (e.g., translocates one proton per electron from the
fumarate reductase, DMSO/TMNO reductase, cytoplasm to the periplasm during quinol oxi-
formate-dependent nitrite reductase). The dation and is therefore a coupling site. This
formate-dependent nitrite reductase operon topic, which was discussed in Section 5.7.1, is
nrf (nitrite reduction by formate) encodes a reviewed in ref. 48. The role of the cytoplas-
periplasmic nitrite reductase that catalyzes mic NADH-linked nitrite reductase, which is
nitrite reduction via formate as the electron encoded by the nirB operon (nirBDC), is to
donor. A ∆p is produced.46 (See note 47, which remove nitrite (which is toxic) produced from
lists the various genes in these operons.) When nitrate and to regenerate NAD+.
nitrite is the electron acceptor, it induces tran- As mentioned earlier, E. coli also makes a
scription of genes encoding both the cytoplas- periplasmic nitrite reductase involved in anaer-
mic and periplasmic nitrite reductases. obic respiration, cytochrome c552, encoded
After a brief summary of the pathway for by the nrfA operon (nrfABCDEFG). It is also
nitrate reduction (Section 19.3.1), we discuss called respiratory nitrite reductase and is a cou-
the enzymes involved (Section 19.3.2). A sim- pling site, but the mechanism is not understood.
plified overview of the Nar system and how it The enzyme is induced by nitrite and repressed
regulates the transcription of genes required by nitrate. For a review of nitrate and nitrite res-
to use nitrate and nitrite as electron acceptors piration, see ref. 49.
during anaerobic respiration (Section 19.3.3)
is followed by a more detailed account of the 19.3.3 The Nar system and gene
Nar regulatory system (Section 19.3.4). Then regulation, an overview
we consider the role of integration host factor For the following discussion, refer to Fig. 19.7.
(IHF) in the regulation of gene transcription by The responses to both nitrate and nitrite are
the Nar system (Section 19.3.5) and the bio- regulated by dual two-component signaling
chemical and genetic evidence for some of the systems that consist of two membrane-bound
conclusions regarding how the genes are regu- sensor kinases (NarX and NarQ) and two cyto-
lated by the Nar system (Section 19.3.6). plasmic response regulator proteins (NarL and
NarP), rather than one sensor kinase and one
19.3.1 Pathway of nitrate reduction response regulator protein.50–53 The systems
–
The nitrate (NO3) is first reduced to nitrite are NarX/NarL and NarQ/NarP. (Note 54
–
(NO2) which, in turn, is reduced to ammonia explains how the nar genes were discovered.)
+
(NH4). A total of eight electrons is required: The biochemical evidence for the roles of these
two to reduce nitrate to nitrite, and six to reduce systems is summarized in Section 19.3.6. The
nitrite to ammonia. The enzymes involved are result of this regulation is that in the presence of
described in Section 19.3.2. Part of the ammo- excess nitrate, there is preferential synthesis of
nia is assimilated into cellular nitrogenous com- enzymes that use nitrate as an electron accep-
pounds, and part is excreted into the medium. tor (i.e., nitrate reductase); and in the presence
Thus, under anaerobic conditions E. coli can of excess nitrite, there is preferential synthe-
use nitrate both as an electron acceptor during sis of enzymes that use nitrite as an electron
anaerobic respiration and as a source of cellular acceptor (i.e., respiratory nitrite reductase,
nitrogen. cytochrome c522).
Both nitrate and nitrite induce the forma-
19.3.2 The enzymes involved tion of a cytoplasmic nitrite reductase, and
The two major enzymes for the reduction of this enzyme probably serves to prevent the
nitrate to ammonia during nitrate respiration in accumulation of toxic nitrite in the cytoplasm.
E. coli are a membrane-bound nitrate reductase Nitrate stimulates transcription of the respira-
that reduces nitrate to nitrite (a two-electron tory nitrate reductase gene (and inhibits tran-
transfer) and a cytoplasmic NADH-linked scription of the respiratory nitrite reductase
nitrite reductase that reduces nitrite to ammo- gene) because nitrate causes increased amounts
nia (a six-electron transfer). Nitrate reductase of the response regulator NarL-P; and NarL-P
is also called respiratory nitrate reductase and stimulates the expression of narG, the operon
is encoded by the narG operon (narGHJI). It that encodes nitrate reductase, and inhibits
www.ebook777.com
expression of nrfA, the operon that encodes for the nrfA operon that has a high affinity for
respiratory nitrite reductase. On the other NarL-P.51
hand, nitrite stimulates expression of the respi-
ratory nitrite reductase operon and lowers the 19.3.4 The roles of NarX and NarQ
expression of the nitrate reductase operon. Evidence supporting this section’s descrip-
This is because nitrite lowers the amounts of tions of NarX and NarQ do is provided in
the response regulator NarL-P (hence depress- Section 19.3.6.
ing transcription of narG) and increases the
amounts of the response regulator NarP-P. NarX
The higher concentrations of NarP-P stimulate In the presence of nitrate, NarX phosphory-
expression of the respiratory nitrite reductase lates NarL. However, very importantly, in the
operon (nrfA). The lower concentrations of presence of nitrite, NarX dephosphorylates
NarL-P also stimulate expression of nrfA, and NarL-P (Fig. 19.7A, reaction 1). This is why
this has been explained by an activation site there is less NarL-P in the presence of excess
Fig. 19.7 Model for nitrate/nitrite control of anaerobic gene expression. The membrane-bound histidine
kinase sensor proteins are NarQ and NarX. The cytoplasmic response regulator proteins are NarL and NarP.
Nitrate stimulates the phosphorylation of NarL and NarP via NarX (A) and NarQ (B). Nitrite stimulates the
phosphorylation of NarL and NarP via NarQ, but causes the dephosphorylation of NarL-P by stimulating
the phosphatase activity of NarX. The result is that nitrate increases the amounts of NarL-P and nitrite lowers
the amounts of NarL-P. (C) Some of the genes regulated by the response regulator proteins. In the presence
of nitrate, the higher levels of NarL-P induce the narG operon, which encodes respiratory nitrate reductase,
and repress the nrfA operon, which encodes respiratory nitrite reductase. The nrfA operon is induced, how-
ever, in the presence of nitrite. This is because (1) nitrite stimulates the phosphorylation of NarP via NarQ,
and NarP-P stimulates transcription of the nrfA operon, and (2) the lower concentrations of NarLP caused
by nitrite actually induce the nrfA operon. The second effect has been explained by a second site for the nrfA
operon (an activation site) that has a high affinity for NarL-P, which stimulates transcription when bound
to this site. See: Darwin, A. J., K. L. Tyson, J. W. Busby, and V. Stewart. 1997. Differential regulation by the
homologous response regulators NarL and NarP of Escherichia coli K12 depends on the DNA binding site
arrangement. Mol. Microbiol. 25:583–595.
nitrite. As a consequence, the nitrate reductase nitrite stimulates the phosphatase activity), the
operon (narG) is stimulated by nitrate but not response of NarQ to either nitrate or nitrite, is
by nitrite, and the periplasmic nitrite reductase the same: that is, phosphorylation of NarL and
operon (nrfA) is repressed by nitrate but not NarP (Fig. 19.7B). Therefore, it is the NarX/
by nitrite. The levels of NarP-P, on the other NarL couple that is responsible for the differen-
hand, are the same for both nitrate and nitrite, tial response to nitrate and nitrite.
and therefore the differential effects of nitrate
and nitrite on gene transcription depend upon 19.3.5 The regulatory role of IHF
NarL-P and not NarP-P. There are several other We know that IHF (integration host factor)
examples of a signaling molecule lowering the activates as well as represses genes. See the dis-
levels of a phosphorylated response regulator cussion of the nucleoid in Section 1.2.6 for a
protein by stimulating its dephosphorylation. description of IHF and its role in DNA bend-
Another example of a signal that does this is PII, ing and the regulation of gene activity, as well
which represses the glnALG operon by stimu- as Sections 11.2.5, 19.4.2, and 19.7.2 for more
lating the dephosphorylation of NRI–P by NRII discussion of IHF. The transcription of the
(Fig. 19.8). nitrate reductase gene (narG) and of the nitrite
extrusion protein gene (narK), which catalyzes
NarQ the electrogenic excretion of nitrite, also is stim-
In contrast to the differential effect of nitrate ulated by integration host factor.55 This indi-
and nitrite on the sensor kinase, in NarX activ- cates that IHF brings a loop of DNA to which
ity with respect to the response regulator NarL NARL-P is attached close to the promoter of
(nitrate stimulates the kinase activity, whereas these genes.
Ap1 Ap2 A Lp L G
+
GSII NRII NRI
ATP NRI Pi
NRII NRII
+ PII
ADP NRI –P H2O
B UTP PII UMP

glutamine UT UT/UR UR + glutamine
PPi PII –UMP H2O
Fig. 19.8 Model for transcriptional regulation of glnALG. The protein products of the operon are glutamine
synthetase (GS), a histidine protein kinase/phosphatase, nitrogen regulator II (NRII), and a response regulator
(NRI, also called NtrC). (A) Under ammonia-limiting conditions, NRII phosphorylates NRI, forming NRI–P.
The increased levels of NRI–P stimulate a large increase in transcription rate from glnAp2 (labeled Ap2),
which uses σ 54 RNA polymerase. As the levels of NRI–P increase, other Ntr operons are also stimulated. Under
conditions of excess ammonia, PII levels rise, and NRII is stimulated by PII to decrease phosphorylation of NRI
(not shown), and to dephosphorylate NRI–P. As a consequence, transcription from glnAp2 stops and σ 70 RNA
polymerase transcribes from promoters glnAp1 and glnLp (labeled Ap1, Lp) at a low rate. Transcription from
these promoters is further regulated by repression by NRI. Not shown is the effect of α-ketoglutarate, which
is to inhibit the activity of PII on NRII, resulting in more NRI–P. (B) The levels of PII are increased by glutamine
(signaling high ammonia). The bifunctional enzyme uridylyltransferase/uridydyl removal (UT/UR) enzyme,
which can both add and remove UMP from PII, is inhibited by glutamine in adding UMP to PII, and is stimu-
lated by glutamine in removing UMP from PII–UMP. The net result is that more PII is generated. For a descrip-
tion model, see: Ninfa, A. J., and P. Jiang. 2005. PII signal transduction proteins: sensors of α-ketoglutarate
that regulate nitrogen metabolism. Curr. Opin. Microbiol. 8:168–173.
www.ebook777.com
19.3.6 Some biochemical and genetic Glutamate dehydrogenase

evidence for roles of NarX, NarQ, NarL,
α-Ketoglutarate + NADPH + H+ + NH3
and NarP
There exist biochemical data in support of the → L-glutamate + NADP+ + H2O
model that NarX and NarQ are kinases that A third reaction, which will become important
autophosphorylate and transfer the phospho- for the discussion that follows, is the conversion
ryl group to the response regulator proteins. It of α-ketoglutarate to L-glutamate by the enzyme
has been demonstrated with purified proteins glutamate synthase.
that NarX and NarQ can autophosphorylate,
using ATP as the phosphoryl donor, and trans- Glutamate synthase
fer the phosphoryl group to NarL.56,57 NarX α-Ketoglutarate + glutamine + NADPH + H+
and NarQ can also negatively regulate the
→ 2 L-glutamate + NADP+
NarL protein by dephosphorylating NarL-P,
although NarX is more active than NarQ in The three foregoing reactions, which were
this respect.58 described in Section 10.3.1, are crucial because
NarX and NarQ have two transmembrane glutamate supplies nitrogen to approximately
regions near the N-terminal end and a region in 85% of the nitrogen-containing molecules in the
between, exposed to the periplasm. Mutations cell, and glutamine supplies the remaining 15%.
in the periplasmic region of NarX or NarQ When the supply of ammonia is restricted,
result in mutants that either have lost the abil- glutamine synthetase is the main reaction
ity to respond to nitrate and nitrite or behave for the assimilation of ammonia because the
as if nitrate or nitrite were present even in its enzyme has a low Km for ammonia. On the
absence.54,59 The mutants were analyzed by other hand, glutamate dehydrogenase has a
means of narG–lacZ and frdA–lacZ fusions. high Km for ammonia (about 1 mM) and oper-
(Note 26 gave a description of gene fusions.) ates only under conditions of high external
From these data it has been concluded that ammonia concentrations. When the ammonia
nitrate and nitrite are detected in the periplas- supply is adequate, then ammonia is the pre-
mic region and that the signal is transduced ferred source of nitrogen, and it represses genes
across the cell membrane to the response regu- required for the assimilation of other nitrog-
lator proteins. Both NarL and NarP have been enous compounds. However, when ammonia
demonstrated to bind to specific sites in target in the growth medium becomes limiting, genes
operons.51 required for ammonia transport and ammonia
production from external nitrogen sources are
19.4 Response to Nitrogen induced. Furthermore, under limiting ammonia
Supply: The Ntr Regulon conditions, glutamine synthetase becomes more
active and the transcription of the gene for glu-
Bacteria use inorganic and organic nitrogenous
tamine synthetase is also stimulated, resulting
compounds as a source of nitrogen for growth.
in more glutamine synthetase to scavenge the
The inorganic nitrogen is in the form of ammo-
small amounts of ammonia for the synthesis of
nia, nitrate, or dinitrogen gas, and the organic
glutamine and glutamate.
nitrogen is in the form of amino acids, urea, and
The genes that are regulated by the ammo-
other nitrogenous compounds. Ultimately, all
nia supply belong to the Ntr regulon.60–63 As
these nitrogenous compounds are either reduced
mentioned previously, a regulon is a set of non-
or catabolized by the bacteria to ammonia. The
contiguous operons or genes controlled by a
enteric bacteria have two enzymatic reactions
common regulator, and for the Ntr regulon, the
for the assimilation of ammonia; the glutamine
regulator is a response regulator protein, called
synthetase reaction and the glutamate dehydro-
NRI (the nitrogen regulatory protein; also called
genase reaction.
the nitrogen regulatory protein C, or NtrC),
Glutamine synthetase which must be phosphorylated to activate gene
transcription.
L-Glutamate + ATP + NH3 The Ntr regulon consists of many genes
→ L-glutamine + ADP + Pi in several operons. (See note 64 for a further
explanation.) These include the genes for glu- below 1 mM, whereupon PII, which inhibits
tamine synthetase, as well as NRI (NtrC), and expression of the Ntr regulon, is inactivated
NRII (also called NtrB), a histidine kinase that and, as a result, the Ntr regulon is activated.
phosphorylates NtrI, all of which are in an This inactivation/activation sequence leads to
operon that is called glnALG in Escherichia coli the synthesis of proteins required for nitrogen
and glnABC in Salmonella typhimurium. assimilation, described earlier. The reason for
Other genes in the Ntr regulon are the gene this is that active PII stimulates the dephospho-
that encodes another positive transcription fac- rylation (and thus inactivation) of NRI–P, which
tor, Nac (see note 64), the genes for the nitroge- is the positive transcription regulator for the
nase system in Klebsiella (nif genes), the genes Ntr regulon. In summary, then, high levels of
for the degradation of urea in Klebsiella, the ammonia cause the activation PII, thus repress-
genes for periplasmic binding proteins required ing the Ntr regulon, and low levels of ammonia
for amino acid transport [e.g., for glutamine cause the inactivation PII, thus inducing the Ntr
(glnH)], arginine, lysine, ornithine (argT), regulon. However, as described next, the effects
and histidine (hisY) in Salmonella, the genes of ammonia are not direct. Rather, they are
for histidine degradation (hut) in Klebsiella, mediated via glutamine and α-ketoglutarate,
and the genes for proline degradation (put) in whose levels reflect the ammonia levels.
Salmonella.65 It should not be concluded auto- When the environmental levels of ammonia
matically that if a gene system is under Ntr regu- go sharply from high to low, there is a decrease
lation in one genus of bacterium, it is also under in the intracellular concentration of glutamine,
Ntr regulation in all bacteria. This is clearly and an increase in the intracellular concentra-
not the case. For example, the hut genes are tion of α-ketoglutarate. Glutamine drops when
not regulated by ammonia levels in Salmonella ammonia levels fall sharply because while the
typhimurium, although they are so regulated in cells are growing in high ammonia, they have
Klebsiella pneumoniae. Along the same lines, low amounts of glutamine synthetase, the
it should be emphasized that the following dis- enzyme that synthesizes glutamine from ammo-
cussion regarding the Ntr system applies only to nia and glutamic acid (Chapter10). Thus, when
the enteric bacteria, in which the regulation of the cells experience a sharp drop in external
nitrogen metabolism has been studied in detail. ammonia levels, there are low amounts of glu-
There is much less known about the regulation tamine synthetase, hence of glutamine. The
of nitrogen metabolism in other prokaryotes, increase in α-ketoglutarate occurs because the
and one should be careful about extrapolating conversion of α-ketoglutarate to glutamate
from the enterics to other prokaryotes. requires glutamine or ammonia to donate the
amino group. The low glutamine and high
19.4.1 Model for the regulation of the α-ketoglutarate levels signal the cells that
Ntr regulon ammonia levels are low; the result is increased
Before describing the regulation of the Ntr regu- transcription of the Ntr regulon, which encodes
lon, it will be helpful to introduce a regulatory enzymes required to increase ammonia assimi-
protein called PII. As will be made clear shortly, lation. Glutamine inhibits the Ntr regulon
PII has two important roles: (1) it represses tran- because glutamine stimulates the production of
scription of the Ntr regulon, and (2) it inhibits PII, which as was noted, inhibits the regulon by
the activity of glutamine synthetase. In both causing the dephosphorylation of the positive
cases PII acts indirectly, as will be explained response regulator, NRI–P. It has been proposed
later. This is a complex signaling pathway. See that α-ketoglutarate stimulates the Ntr regulon
ref. 66 for a review. Refer to Fig. 19.8 in read- because α-ketoglutarate inhibits the activity of
ing the following set of explanations; the ver- PII (rather than its production). This line of rea-
sion with fewer details precedes the subsection soning has been reviewed in ref. 67.
providing more complete coverage. As stated earlier, PII influences transcription
indirectly, as shown in Fig. 19.8. An increase
The less detailed explanation in the amount of PII inhibits transcription of
Transcription of the Ntr regulon is induced the Ntr regulon because PII activates the phos-
when the extracellular ammonia levels fall phatase activity of the histidine kinase, NRII,
www.ebook777.com
and the phosphatase activity of NRII inactivates

the response regulator (positive transcription
factor), NRI–P. When the levels of PII decrease,
NRII acts as a kinase, rather than as a phos-
phatase, and phosphorylates NRI, which then
activates transcription.
Thus, the sequence of events is as follows: (1)
low ammonia results in an increase in the con-
centrations of α-ketoglutarate because its con-
version to glutamine and glutamate is slowed;
(2) the α-ketoglutarate lowers the concentra-
tion of PII by stimulating its conversion to PII–
UMP; (3) when the PII levels fall, NRII no longer
acts as a phosphatase but rather as a kinase and
phosphorylates NRI; and (4) NRI–P stimulates
transcription. When ammonia levels are high,
transcription is repressed because glutamine
levels rise and the glutamine stimulates the con-
version of PII–UMP to PII, which stimulates the
phosphatase activity of NRII, which results in a
decrease in NRI–P. These events are described
in more detail next.
The more detailed explanation

The central operon involved in the regulation of
the Ntr regulon, the glnALG operon, encodes
three genes (Fig. 19.8A): glnA, the gene for
glutamine synthetase; glnL, the gene for NRII,
which is a bifunctional histidine kinase/phos-
phatase whose substrate is NRI; and glnG,
the gene for NRI, a response regulator. Under Fig. 19.9 Activation by NRI–P and NifA plus IHF.
nitrogen excess (high ammonia), the operon is Promoters read by σ54 polymerase require an activa-
transcribed at a low level from the glnAp1 and tor that binds at an upstream site, sometimes over
glnLp promoters by σ 70 RNA polymerase. The more than 100 base pairs away from the promoter.
small amount of transcription from glnAp1 is According to one model for how the activator con-
sufficient to guarantee the synthesis of enough tacts the polymerase, the DNA bends. (A) Activation
glutamine synthetase to meet the cell’s needs for of glnA promoter by NRI–P. Because the binding site
glutamine when the ammonia concentrations are for NRI–P has an inverted repeat, each one capable of
binding NRI–P, it has been proposed that two NRI–P
high. Part of the reason that only a small amount
molecules bind. (B) Activation of promoter by NifA.
of transcription takes place from these promot- It is known that IHF is necessary for the transcrip-
ers is that transcription from them is repressed tional activation by NifA of most NifA-dependent
by NRI (Fig. 19.9A).68 Under nitrogen-limiting genes. The model supposes that IHF binds at a site
conditions, the glnALG operon is transcribed between the NifA-binding site and the polymerase-
at a high frequency from the glnAp2 promoter binding site and bends the DNA, bringing the acti-
by the σ54 RNA polymerase. (See note 69 for a vator to the RNA polymerase. See: Magasanik, B.
discussion of the σ54 polymerase and more on 1996. Regulation of nitrogen utilization, pp. 1344–
transcriptional regulation.) Transcription from 1356. In: Escherichia coli and Salmonella, Cellular
the glnAp2 promoter requires the phosphory- and Molecular Biology, Vol. 1. F. C. Neidhardt et al.
lated form of the response regulator NRI (i.e., (Eds.). ASM Press, Washington, DC.
NRI–P), reflecting the requirement of σ54 RNA
polymerases for the binding of a transcription
regulator upstream to the promoter to form
an open complex. Since NRI–P binds more of the histidine kinase/phosphatase (NRII).
than 100 base pairs upstream of the promoter Alternatively, one might consider glutamine
region, the DNA must bend for NRI–P to make synthetase to be the sensor because it responds
contact with the RNA polymerase (Fig. 19.9A). to the ammonia supply and determines the lev-
(See note 70 for a model of how NRI–P stimu- els of glutamine and α-ketoglutarate, which sig-
lates transcription.) nal the UT/UR.
The protein kinase, NRII, is itself regulated by
another protein, called PII (product of the glnB Regulation of the other operons in
gene). When PII levels are high, NRII acts like a the Ntr regulon
phosphatase rather than a kinase, and the levels It has been estimated that the level of NRI in
of NRI–P drop. Hence, PII inhibits transcription ammonia-repressed cells is about five molecules
of the glnALG operon. per cell, which when phosphorylated under
α-Ketoglutarate inhibits the activity of PII. conditions of ammonia starvation, will acti-
Under conditions of excess ammonia, PII is vate transcription of glnALG. When the levels
generated from PII–UMP because glutamine of NRI–P become sufficiently high (around 70
stimulates the removal of UMP from PII–UMP molecules per cell) as a result of increased tran-
catalyzed by the bifunctional enzyme uridylyl- scription of glnALG, transcription of the other
transferase–uridylyl removal (UT/UR) (product operons in the Ntr regulon system is activated,
of the glnD gene) (Fig. 19.8B). When ammonia as well.
levels fall, the PII is inactivated by being con- NRI–P directly stimulates some of the Ntr
verted to PII–UMP in a reaction catalyzed by UT/ operons [e.g., glnALG and operons encoding
UR and inhibited by glutamine. Under these cir- proteins for the uptake of glutamine (glnHPQ)
cumstances, NRII acts as a kinase rather than as in E. coli, as well as histidine (hisJQMP) and
a phosphatase and transcription is stimulated. the gene for arginine uptake (argT) in S. typh-
In summary then, we have four steps. imurium]. However, in other instances, NRI–P
activates the transcription of a gene encoding
1. High ammonia leads to high PII because glu-
a second positive regulator, which activates
tamine stimulates the removal of UMP from
transcription of the target promoters. For
PII–UMP.
example, in Klebsiella pneumoniae, NR1–P
2. High PII converts NRII from a kinase to a
activates the transcription of nifAL. (See ref.
phosphatase.
71 for a review.) The product of nifA is NifA,
3. The phosphatase activity of NRII leads to the
which is a positive regulator of the nitrogen
dephosphorylation of NRI–P.
fixation (nif) genes. NifA binds to upstream
4. Dephosphorylation of NRI–P leads to a low-
activator sequences (UAS) and makes contact
ering of transcription of the glnALG operon,
with σ54-RNA polymerase, which is bound
since NRI–P is a positive transcription
downstream. The contact between NifA and
factor.
the polymerase is facilitated by a bending of
Thus, high ammonia leads to the dephospho- the intervening DNA, as described later. (See
rylation of the response regulator, hence to Section 19.4.2, and Fig. 19.9B.) [In K. pneu-
repression of the glnALG operon, As we shall moniae, the product of nifL inactivates the nifA
see later, a similar situation exists for the regu- product in response to oxygen and ammonia,
lation of the PHO regulon. In this case a high thus making certain that the nif genes (except
concentration of inorganic phosphate leads to for nifAL) are expressed only during anaero-
the dephosphorylation of the response regula- biosis and ammonia starvation.] In Klebsiella
tor and a consequent repression of the PHO species, but not in Salmonella typhimurium,
regulon (Section 19.5.1). P–NR1 also activates the transcription of
A key enzyme in the signal transduction path- nac, which encodes a positive regulator of
way is the bifunctional enzyme UT/UR, which the hut operons that encode enzymes for the
either adds or removes UMP from PII. It can be catabolism of histidine as a source of nitro-
considered to be a sensor protein that responds gen. The hut genes use a σ70 RNA polymerase
to the α-ketoglutarate/glutamine ratio and (the major polymerase), rather than the σ54
modifies PII, which in turn regulates the activity polymerase.
www.ebook777.com
19.4.2 Regulatory Role of IHF adenylylated and inactivated in the presence

The student should review the discussion of of high amounts of glutamine (gln) and PII, and
the nucleoid in Section 1.2.6, where the DNA- activated (deadenylylated) in the presence of
binding protein known as integrating host fac- PII–UMP. Levels of PII rise when ammonia lev-
tor is described. See ref. 72 for a review of IHF, a els are low. The adenylylated enzyme is further
protein that is known to bend DNA. (Any bent inactivated because it is susceptible to feedback
DNA, including DNA to which IHF is bound inhibition by a variety of nitrogenous com-
in vitro, can be analyzed via gel electrophoresis pounds that depend upon glutamine for their
because bent DNA fragments travel at different synthesis (e.g., glucosamine-6-phosphate, car-
rates.) IHF activates certain genes and represses bamoyl phosphate, CTP, AMP, tryptophan,
others. In some genes it binds at or near the histidine). Thus PII turns out to be an important
promoter region, and it may interfere with the protein that signals the ammonia levels to both
binding or activity of RNA polymerase. It also the transcription machinery and glutamine
binds to specific locations on the DNA between synthetase.
the promoters and NifA-binding sites of genes Table 19.1 presents an overview of the regu-
activated by NifA (with the exception of one lation of transcription and glutamine synthetase
NifA activated gene, nifB). IHF is also required activities by ammonia and PII. While the post-
for the activation of one NRI-dependent pro- translational control of glutamine synthetase by
moter, glnHp2, but not the others. According to adenylylation occurs in E. coli and related bacte-
the model, IHF bends the DNA, bringing bound ria, as well as in species of Vibrio, Thiobacillus,
activator such as NifA or NRI–P to the pro- and Streptomyces, it does not occur in all bac-
moter, where the activator stimulates transcrip- teria (e.g., Bacillus and Clostridium) (Reviewed
tion (Fig. 19.9B). (This material is reviewed in in ref. 62.) This is another reason for the need to
ref. 5. See Sections 20.7.2 and 19.11.5 for other use caution in extrapolating results from some
examples of IHF regulatory activity.) bacteria to all bacteria, especially with regard to
the regulation of metabolism.
19.4.3 Effect of PII–UMP and PII on
glutamine synthetase activity Table 19.1 Regulation of transcription of glnALG
When ammonia levels are low, not only is tran- and glutamine synthetase
scription of the gene for glutamine synthetase Glutamine
(GS) stimulated, but the activity of the exist- NH3 PII/PII–UMP Transcriptiona synthetaseb
ing enzyme is stimulated, ensuring that the low
ammonia levels do not preclude incorporation High High Low Inactive
of ammonia into glutamine and glutamic acid. Low Low High Active
A model for the regulation of GS is illustrated a
Indirectly repressed by PII.
in Fig. 19.10 and explained in the legend. b
Indirectly inactivated by PII; indirectly activated by PII–
Basically, what happens is that GS becomes UMP.
Pi GS-AMP [inactive] PPi
gln + PII − − PII + gln

AR AR/AT AT
PII –UMP + + PII –UMP
(2) (1)
ADP GS [active] ATP
Fig. 19.10 Adenylylation of glutamine synthetase (GS) inhibits its activity, and deadenylylation increases
its activity. The reactions are catalyzed by AT/AR, the bifunctional adenylyltransferase/adenylyl removal
enzyme. In this model, glutamine (gln) and PII act together to inactivate GS by stimulating its adenylylation (1)
and by inhibiting its deadenylation (2). PII–UMP activates GS by inhibiting its adenylylation (1) and stimulat-
ing its deadenylylation (2). Furthermore, the levels of PII are increased by glutamine, as described in Fig. 19.8.
For a discussion of this model, see: Ninfa, A. J., and P. Jiang. 2005. PII signal transduction proteins: sensors of
α-ketoglutarate that regulate nitrogen metabolism. Curr. Opin. Microbiol. 8:168–173.
19.5 Response to Inorganic The components required for regulation are (1)
Phosphate Supply: The PHO Regulon PstS, a periplasmic Pi binding protein, (2) PstA,
PstB, and PstC, integral membrane proteins
Bacteria have evolved a signaling system to required for Pi uptake (PstB is the permease),
induce the formation of phosphate assimila- (3) PhoU, (4) PhoR, and (5) PhoB. PhoR detects
tion pathways when the supply of phosphate Pi. However, it is not known whether Pi inter-
becomes limiting. The following discussion per- acts with PhoR indirectly by binding to the Pst
tains specifically to E. coli and related bacteria, system or directly by binding to PhoR. What
although homologues of the histidine kinase is known is that under conditions of Pi excess
(PhoR) and response regulator (PhoB) have (about 4 μM), repression of the PHO regulon
been found in other bacteria including Bacillus, requires the Pst system, PhoU, and a form of
Pseudomonas, Rhizobium, Agrobacterium, PhoR called PhoRR. When Pi becomes limiting,
and Synechococcus. (For references to homo- the PHO regulon is induced and this requires
logues of PhoR and PhoB, see ref. 73.) Under a form of PhoR called PhoRA, as well as PhoB,
low inorganic phosphate conditions, E. coli which is the response regulator.
stimulates the transcription of at least 38 genes One model proposes that Pi binds to PstS and
(most of them in operons) involved in phos- that the Pi-bound PstS binds to PstABC, which
phate assimilation, including genes encoding a is a member of the ABC transporter family.
periplasmic alkaline phosphatase that can gen- (See note 78 for further comment on Pst and its
erate phosphate from organic phosphate esters relationship to the ABC-type transporters.) A
(phoA); an outer membrane porin channel for “repressor complex” is postulated that consists
anions, including phosphate (phoE); a high-af- of Pi bound to PstS, PstABC, PhoU, and PhoR.
finity inner membrane phosphate uptake system PhoR is thought to be a histidine kinase/phos-
called the Pst system and a protein required for phatase bifunctional enzyme. (See note 79.) In
phosphate repression (pstSCAB–phoU); a his- the “repressor complex” PhoR is suggested to
tidine kinase and response regulator (phoBR), be a monomer with phosphatase activity and to
14 genes for phosphonate74 uptake and break- maintain PhoB in its dephosphorylated (inac-
down (phnCDEFGHIJKLMNOP); genes for tive) state. (This would be analogous to the Ntr
glyceraldehyde-3-phosphate uptake; and a gene regulon, where PII stimulates the phosphatase
encoding a phosphodiesterase that hydrolyzes activity of NRII. See Section 19.4.1.)
glycerophosphoryl diesters (deacylated phos- The model further proposes that when
pholipids) (ugpBAECQ). All these genes (except phosphate becomes limiting (<4 μM), PhoR
for pit, which functions in the presence of excess is released from the “repressor complex”
phosphate) are repressed by phosphate and are in and functions as a histidine kinase, which is
the PHO regulon. The PHO regulon is controlled depicted in Fig. 19.11 as a dimer, although that
by PhoR, which appears to be a histidine kinase, has not been demonstrated. The dimer form of
and a response regulator, PhoB. The phospho- PhoR autophosphorylates and activates PhoB
rylated form of PhoB (i.e., PhoB-P) activates the by transferring to it the phosphoryl group. In
transcription of the genes in the PHO regulon. agreement with this model, Pst and PhoU are
The pho promoters use the σ70 RNA polymerase required for repression of the PHO regulon,
(encoded by rpoD). (See note 75 for a description but not for its activation. However, it must be
of how PhoR activates transcription.) added that the interaction of the Pst system and
When inorganic phosphate is in excess, it is PhoU with PhoR to cause Pi repression is not
transported into the cell by a low-affinity trans- really understood. In mutants that do not have
porter called Pit. (See ref. 76 for a comprehen- PhoR, the PHO regulon is expressed constitu-
sive review of phosphate transport.) Pit consists tively; that is, the genes are not repressed by
of a single transmembrane protein that is driven inorganic phosphate. This is because PhoB is
by the ∆p. (See the discussion of secondary always phosphorylated. This may be due to the
transport in Section 17.3.1.) postulated phenomenon of cross-regulation,
in which a response regulator of one two-com-
19.5.1 The signal transduction pathway ponent system might be controlled by a differ-
A model for the regulation of the PHO regu- ent regulatory system (e.g., a related histidine
lon in E. coli by Pi is shown in Fig. 19.11.73,77 kinase) or by acetyl phosphate, whose levels
www.ebook777.com
S
Pi
1 [Pi]
2
S periplasm
Pit A B C
U cytoplasm
R R R
Pi Pi
B B B
unshaded:
Pit low affinity Pi transporter 3
Pst high affinity Pi-specific transporter:
S Pibinding protein Pi Pi
A,C integral membrane proteins
B Pi permease
U negative regulator PhoU
B B
shaded:
B response regulator PhoB
R Pi sensor PhoR
Fig. 19.11 Model for the regulation of the PHO regulon. Inorganic phosphate in the periplasm binds to
the phosphate-binding protein, PstS. The Pi–PstS binds to the phosphate transporter, PstABC, in the cell
membrane. A “repressor complex” forms between Pi–PstS, PstABC, PhoU, and PhoR. In the repressor com-
plex, PhoR acts as a phosphatase and maintains PhoB in the dephosphorylated state. When Pi becomes lim-
iting, PhoR is released from the repressor complex, autophosphorylates, and then phosphorylates PhoB.
Phosphorylated PhoB is a positive transcription regulator for the PHO regulon. PhoR as a phosphatase is
depicted as a monomer and PhoR as a kinase is drawn as a dimer, although this has not been demonstrated.
Source: Wanner, B. L. 1993. Gene regulation by phosphate in enteric bacteria. J. Cell. Biochem. 51:47–54.
may reflect growth conditions (Section 19.9).73 of the photosynthetic genes. One of these is a
In this case, the different regulatory systems that repressor system (the CrtJ system) activated
may phosphorylate PhoB are CreC (formerly by oxygen. A second is an inducer system (the
called PhoM), which is induced by growth on RegB/RegA system) activated by anaerobiosis.
glucose and is not regulated by phosphate, and And the third is an inducer system (the HvrA
the phosphoryl donor acetyl phosphate pro- system) activated by low light. These systems
duced during growth on pyruvate.73 The evi- are described next.
dence for acetyl phosphate as a global signal
for response regulator proteins is discussed in 19.6.1 Response to oxygen
Section 19.9. The student should review the description of
the light-harvesting complex in photosynthetic
19.6 Effect of Oxygen and Light purple bacteria presented in Section 6.6.1 and
Fig. 6.9. The photosynthetic genes in the pur-
on the Expression of Photosynthetic
ple photosynthetic bacteria such as R. capsula-
Genes in the Purple Photosynthetic tus exist in several operons collectively known
Bacterium Rhodobacter capsulatus as the photosynthetic gene cluster. These genes
Rhodobacter capsulatus is a purple nonsulfur code for proteins required for the biosynthesis
photosynthetic bacterium that can grow het- of the photosynthetic pigments (bacteriochlo-
erotrophically in the dark, deriving its energy rophyll and carotenoids), as well as proteins
from aerobic respiration, or photoheterotroph- that make up part of the reaction center and
ically in the light without oxygen. The synthesis light-harvesting complexes. Although not
of the photosynthetic apparatus is under the discussed here, some of the regulators of pho-
control of light and oxygen with regulation at tosynthetic genes expression described next
the level of transcription. There are at least three (RegB/RegA, Aer, CrtJ, HvrA) also regulate the
regulatory systems that control transcription expression of terminal oxidases in the electron
transport chain. For more information, refer growing R. capsulatus and R. sphaeroides that
to ref. 80. is analogous to that of the ArcB/ArcA system
in E. coli. RegB autophosphorylates, activat-
Aerobic repression ing the RegB/RegA system, in the absence of
Oxygen represses the synthesis of photosyn- oxygen. It may be that the signal preventing
thetic pigments, ensuring that photosynthesis autophosphorylation originates with the redox
occurs only under anoxic conditions. (The regu- state of a component of the respiratory chain
lation of photosynthetic genes is reviewed in ref. rather than with oxygen itself. Recall that ArcB
81.) One of the genes in the photosynthetic gene also autophosphorylates under anaerobic con-
cluster encodes carotenoid, an aerobic transcrip- ditions. (See Fig. 19.5.) See the discussion in
tion repressor of genes for bacteriochlorophyll, Section 19.2.2 of how ArcB might sense anaer-
as well as the genes for synthesis of light-harvest- obiosis. Homologues of RegA and RegB have
ing II complex (puc genes). The aerobic repres- been found in a wide variety of photosynthetic
sor is called CrtJ (or PpsR), and it is activated in and nonphotosynthetic bacteria, where they
the presence of oxygen. When CrtJ is exposed are presumed to exert redox control over gene
to oxygen, an intramolecular disulfide bond expression (cited in ref. 83.)
forms, causing a conformational change in CrtJ
that results in a significantly increased ability to 19.6.2 Response to light
bind to its target promoters. (This is cited in ref. Photosynthetic pigment synthesis is increased
82.) Review the discussion of another redox- anaerobically when the light intensity is
sensitive transcription regulator, the FNR pro- decreased, allowing the cells to capture more
tein, which is inactivated by oxygen rather than light energy despite the lower light intensities.
being activated: see Section 19.2.2 and note 35. R. capsulatus has a gene, called the hvrA gene,
A second aerobic repressor of photosynthetic whose product (HvrA) is required for the stim-
genes, Aer, was described in 1996.81 Some of ulation of transcription of some genes in the
the genes repressed by CrtJ are also repressed by photosynthetic gene cluster by low light inten-
Aer; that is, they are co-repressed. Other genes sity. (See note 86 for more information about
are repressed by CrtJ or Aer, but not by both. HvrA.) This conclusion is supported by the
finding that mutations in hvrA cause a loss in
Anaerobic induction low-intensity light stimulation of transcription
The photosynthetic bacterium Rhodobacter of target operons. It has been shown that HvrA
capsulatus uses a two-component system binds to target promoters in the photosynthetic
(RegB/RegA) to stimulate the transcription of gene cluster, and it has been demonstrated that
genes for its light-harvesting and photoreaction high light reduces the intracellular levels of
center complexes under anaerobic incubation HvrA. However, not all the low-light-inducible
conditions.83 (For a review of the RegB/RegA genes require HvrA for transcription, and thus
global regulatory system, see ref. 84.) Under there is clearly another light-sensitive signaling
anaerobic conditions, a signal is sent to RegB, system.
causing it to autophosphorylate. RegB-P then In addition to transcriptional regulation by
phosphorylates RegA.82 DNA-binding studies light, there appears to be post-transcriptional
indicate that RegA-P is a transcriptional activa- regulation.87 This was reported for the synthesis
tor.85 It has been concluded that the RegB/RegA of the peptides of the light-harvesting complex
system is a conserved global regulatory system I (B800–850 peptides), which increase around
involved in redox control of transcription of fourfold in parallel with the number of B800–
a variety of the genes in R. capsulatus and R. 850–bacteriochlorophyll complexes when
sphaeroides that are related to energy-gener- the light intensity is diminished. It was found,
ating and energy-utilizing processes, including however, that low-light-grown cells had only
expression of genes for CO2 fixation, nitrogen one-fourth the mRNA for the B800–850 pep-
fixation, hydrogen oxidation, denitrification, tides. We therefore have more protein present
electron transport, and aerotaxis.83 under conditions of less mRNA. Perhaps there
Thus, RegB/RegA is a global regulator for is decreased turnover of the protein under low
many genes that serve a function in anaerobically light conditions. This is discussed in ref. 86.
www.ebook777.com
19.7 Response to Osmotic Pressure media would prevent the phosphorylation of

and Temperature: Regulation of Porin OmpR by EnvZ.90 The model further proposes
that high levels of P-OmpR repress ompF and
Synthesis
stimulate ompC by binding to low-affinity
When E. coli is growing in higher osmolarity or sites upstream from the respective promoters,
at high temperature, the synthesis of the bacte- and that low levels of P-OmpR stimulate tran-
rium’s slightly smaller porin channel, OmpC, scription of ompF by binding to high-affinity
increases relative to the larger OmpF channel. sites upstream from the ompF promoter. Thus,
For example, when E. coli is grown at 30 °C and when the external osmotic pressure is raised, the
in the presence of 1% NaCl (the temperature levels of OmpR-P should increase and repress
and approximate osmolarity of the intestine), ompF while stimulating ompC. Several lines of
only OmpC is made.88 When E. coli is grown evidence support the model: (1) it has been dem-
at lower temperatures and at osmolarities that onstrated in vitro that EnvZ can accept a phos-
approximate the conditions in lakes and streams, phoryl group from ATP and in turn transfer it to
OmpF is preferentially made. The changes in OmpR, forming OmpR-P; (2) it is also known
porin composition of the outer membrane do that when E. coli is shifted to a medium of high
not change the intracellular osmotic pressure, osmolarity, the levels of OmpR-P increase; (3)
and therefore they are not part of a homeostatic in a mutant of E. coli that has elevated levels
response to osmotic pressure changes. of OmpR-P, ompF is repressed and ompC is
One can rationalize the change in porins by expressed all the time (constitutively); and (4)
assuming that the smaller OmpC channel is OmpR-P stimulates transcription from ompF
advantageous in the intestinal tract because it and ompC promoters.91–93
may retard the inflow of toxic substances such
as bile salts, whereas in lower osmolarity envi-
ronments, such as would be experienced out-
19.7.2 Repression of transcription of
side the body, the larger OmpF channel might
ompF and ompC by IHF
be an advantage to increase inward diffusion of Also repressing ompF and ompC is integration
dilute nutrients. host factor. IHF binds to nucleotide sequences
in the promoter regions of the ompF and ompC
19.7.1 Regulation of expression of genes in Escherichia coli and represses tran-
ompF and ompC by a two-component scription. (See Section 19.4.2 for a discussion
system, EnvZ/OmpR of IHF.) This was demonstrated in vitro for
EnvZ is an inner membrane histidine protein ompF, and the investigators also showed that
kinase that has been postulated to function OmpR reverses the inhibition, suggesting that
also as an osmotic sensor.89 However, very IHF might somehow interfere with the activity
little is known concerning the signals to which of OmpR at the promoter site.94
EnvZ responds when it phosphorylates or
dephosphorylates OmpR. EnvZ is a transmem- 19.7.3 Inhibition of translation of ompF
brane protein, part of which is exposed to the mRNA by micF RNA, an antisense RNA
periplasm (a loop near the N-terminal end) and There appears to be an additional mode of inhi-
part to the cytoplasm (the C-terminal domain). bition of ompF expression, particularly at levels
It is thought that the domain that senses osmo- of osmolarity that are not very high. Another
larity signals resides either in the periplasm or regulatory gene, micF, is also stimulated by
in the membrane. The response regulator is a high concentrations of OmpR-P and produces
cytoplasmic protein called OmpR. an RNA molecule that inhibits ompF mRNA
One model proposes that increased external translation by binding to the mRNA (Fig.
osmolarity activates EnvZ so that it phosphory- 19.12).95 One example of an antisense RNA
lates OmpR (Fig. 19.12). However, mutations is micF. Other examples of antisense RNAs
in EnvZ result in a high-osmolarity phenotype are known that function in regulating gene
(constitutive OmpF–OmpC+ phenotype), indi- activity.96,97 Support for the conclusion that
cating that perhaps EnvZ detects a signal in low- MicF inhibits ompF mRNA translation derives
osmolarity media rather than in high-osmolarity from the observation that when multiple copies
media. The putative signal in low-osmolarity of micF are introduced on a high-copy-number
Fig. 19.12 Model to explain the regulation of porin synthesis. EnvZ is a transmembrane histidine kinase/
phosphatase whose substrate is the cytoplasmic response regulator OmpR. To explain how a shift to high-
osmolarity media represses ompF and stimulates ompC, it is assumed that high osmotic pressures activate
(not necessarily directly) the kinase activity of EnvZ. The resulting high levels of OmpR-P repress transcrip-
tion of ompF and stimulate transcription of ompC. OmpR-P can effect these changes by binding to DNA
sequences upstream of the respective promoters. A third gene, micF, is transcribed from the ompC promoter
but in the opposite direction. Its transcription is also activated by OmpR-P. The gene micF codes for an RNA
transcript that is complementary to the 59 end of the ompF mRNA (i.e., the region where protein translation is
initiated) and blocks translation. micF RNA is therefore “antisense” RNA because it binds to the ompF RNA
transcript that is translated into protein (i.e., the “sense” RNA). It may be that an additional control of ompF
expression is the inhibition of translation of ompF mRNA by micF RNA. When the external osmotic pres-
sures are lower, the concentrations of OmpR-P are too low to bind to the DNA sites for repression of ompF
and stimulation of ompC. However, low levels of OmpR-P can stimulate the transcription of ompF because
of high-affinity binding sites for OmpR-P upstream of the ompF promoter. Source: Stock, J. B., A. J. Ninfa,
and A. M. Stock. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol.
Rev. 53:450–490. Reproduced with permission from American Society for Microbiology.
plasmid, there is inhibition of in vivo synthesis K+ are growth limiting, the bacterium responds
of OmpF. by synthesizing a high-affinity K+ uptake system
to correct the situation. The uptake system,
19.8 Response to Potassium Ion and called the KdpABC transporter, is described in
External Osmolarity: Stimulation of Section 17.3.3. (See subsection entitled ATP-
driven K+ influx.) For a discussion of the role
Transcription of the kdpABC
of K+ uptake in pH and osmotic homeostasis,
Operon by a Two-Component the student is referred to Sections 16.1.3 and
Regulatory System 16.2.2.
When E. coli is placed in a medium that is high in There are two regulatory genes for the tran-
osmolarity due to salts, or if the concentrations of scription of the kdpABC operon, kdpD and
www.ebook777.com
kdpE, both of which are required for expres- Acetate kinase makes acetyl phosphate directly
sion of the kdpABC genes. The KdpD and from acetate and ATP. This is also a reversible
KdpE proteins are part of a two-component reaction:
regulatory system, with KdpD being a mem-
Acetate kinase
brane sensor kinase that autophosphorylates
when stimulated, and KdpE being a cytoplas- Acetate + ATP → acetyl–P + ADP
mic response regulator protein that is phospho-
Mutants in ackA (acetate kinase gene) have
rylated by KdpD. The phosphorylated form
increased levels of acetyl phosphate when
of KdpE activates the kdpABC promoter.98–100
growing on carbon sources such as glucose or
What signal does KdpD sense? It appears that
pyruvate that yield acetyl–CoA (via pyruvate
KdpD senses both a decrease in intracellular K+
dehydrogenase), since they can use the phos-
concentration and an increase in intracellular
photransacetylase to form acetyl phosphate
ionic strength.101
from acetyl–CoA but cannot convert the acetyl
phosphate to acetate without the acetate kinase.
19.9 Acetyl Phosphate Is a Possible The same result might be achieved by grow-
Global Signal in Certain Two- ing pta (phosphotransacetylase) mutants on
Component Systems acetate, since these should be able to synthesize
It has been proposed that in the absence of the acetyl phosphate by using the acetate kinase but
cognate HK proteins, that is, in mutants lack- could not convert it to acetyl–CoA. On the other
ing specific histidine kinases, acetyl phosphate hand, mutants lacking both pta and ackA genes
donates the phosphoryl group to the cognate cannot make acetyl phosphate from pyruvate or
response regulator protein, thus counteract- acetate. Such mutants show the regulation that
ing the effect of the mutation. (Reviewed in ref. would be expected if acetyl phosphate were able
102.) The extent to which acetyl phosphate to phosphorylate response regulator proteins in
does this in vivo, especially in wild-type cells, mutants lacking the cognate histidine kinases.
is not known. Nevertheless, it is an interesting The choice of carbon source also influences
phenomenon and potentially very significant. the levels of acetyl phosphate. Acetyl phosphate
The following discussion pertains to the role pools are lowest in cells grown on glycerol
that acetyl phosphate might play as a global (<0.04 mM), higher in cells grown on glucose
regulator, and to the methods used to investi- (0.3 mM), and highest in cells grown on acetate
gate this role. (1.5 mM).101
Sometimes increasing the copy number of
19.9.1 Changing the intracellular levels ackA, the gene that codes for acetate kinase, can
of acetyl phosphate to investigate its indicate whether acetyl phosphate can phospho-
possible role rylate the regulatory protein. For example, one
To study the role of acetyl phosphate in of the genes of the PHO regulon, phoA (codes
regulation in vivo, the investigator must be for alkaline phosphatase), could be expressed in
able to change its intracellular concentra- the absence of the cognate HK proteins (mutants
tions. Mutants can be used for this purpose. lacking both PhoR and CreC) by increasing the
Mutants lacking phosphotransacetylase copy number of ackA.
(encoded by the pta gene) or acetate kinase
(encoded by the ackA gene) have higher levels In vitro phosphorylation of a response
of acetyl phosphate, whereas mutants lack- regulator protein by acetyl phosphate
ing both enzymes have lower levels of acetyl Some investigators have demonstrated in vitro
phosphate. To understand why this is so, phosphorylation of a response regulator pro-
we must first recall how E. coli makes acetyl tein by acetyl phosphate. For example, NRI (the
phosphate. (See Section 8.3.2 for a review.) response regulator protein that regulates tran-
Phosphotransacetylase reversibly converts scription of the Ntr regulon) has been phospho-
acetyl–CoA to acetyl phosphate: rylated by acetyl phosphate in vitro.103 Mutants
lacking NRII, the histidine kinase/phosphatase
Phosphotransacetylase
for the Ntr regulon, have residual regulation
Acetyl–CoASH + Pi → acetyl–P + CoASH of expression of glnA (codes for glutamine
synthetase), which is part of the Ntr regulon. gene encodes acetate kinase, and in the absence
This suggests the presence of an alternative of that enzyme, the cells use the phospho-
donor of the phosphoryl group for the response transacetylase to make acetyl phosphate from
regulator, NRI. The alternative donor may be acetyl–CoA but cannot metabolize the acetyl
acetyl phosphate. phosphate further. Hence the pool size of acetyl
phosphate increases. All this is consistent with
the hypothesis that acetyl phosphate represses
19.9.2 Regulatory role for OmpR-P and
the flhDC operon.103
acetyl phosphate in the repression of
transcription of flagellar biosynthesis How the involvement of OmpR in the
genes in E. coli repression of flagellar genes is detected
In addition to regulating transcription of the The involvement of OmpR in the repression of
porin genes (Section 19.7.1), OmpR regulates flhDC–lacZ by acetyl phosphate was shown by
the transcription of other genes, including fla- using ompR– strains. OmpR– mutants showed
gellar genes. When E. coli is grown under stress- an enhanced expression of flhDC–lacZ, indicat-
ful conditions (e.g., high osmolarity, elevated ing that OmpR is a negative regulator of flhDC–
temperatures), flagellar synthesis is inhibited. lacZ, and this enhanced expression was not
The repression of the flagellar genes by high reduced in ackA mutants that had higher levels
osmolarity is apparently due to a repressive of acetyl phosphate, indicating that repression
effect of OmpR-P, which is known to regulate by acetyl phosphate requires OmpR.
other genes in addition to OmpF and OmpC. Acetyl phosphate has been shown to phos-
Under conditions of high osmolarity, OmpR phorylate OmpR in vitro, and presumably it also
is phosphorylated by the sensor kinase EnvZ. does so in vivo. Since OmpR is phosphorylated
However, acetyl phosphate can also phospho- by EnvZ when cells are grown at high osmo-
rylate OmpR, even in wild-type cells, leading to larities (see Section 19.7.1), it might be expected
the repression of flagellar genes.104 that flhDC expression would be decreased in
bacteria grown at higher osmolarities, and this
How the effect of acetyl phosphate on the in fact is the case in wild-type cells but not in
transcription of flagellar genes is measured ompR– cells.
To explain how the effect of acetyl phosphate Finally, it was demonstrated by a DNA
on the transcription of flagellar genes can be mobility shift assay that phosphorylated OmpR
measured, it is necessary to introduce the master binds to the flhDC promoter. The student
operon for flagellar genes and to describe how should recall that in other instances in which it
the transcription of the operon can be moni- was inferred that acetyl phosphate had caused
tored. The master operon for flagellar biosyn- the phosphorylation of a response regulator in
thesis is flhDC, which codes for FlhD and FlhC, vivo, the effects were seen in mutants lacking the
two DNA-binding proteins that are required cognate histidine kinase. However, acetyl phos-
for transcription of all the genes in the flagel- phate appears to phosphorylate OmpR even in
lar regulon. The use of flhDC–lacZ fusions can cells that have the kinase, which is EnvZ.
be used to monitor the activity of the flhDC
promoter by doing β-galactosidase assays. (See Possible explanation of why E. coli produces
note 26 for a description of lacZ fusions.) It was fewer flagella at higher temperatures
shown that the expression of flhDC is substan- It has been known that E. coli grown at elevated
tially increased in pta ack double mutants or temperatures produces fewer flagella. This can
pta mutants compared with wild-type cells.104 be explained by the repression of the flagellar
These mutants can convert the carbon source to operon by acetyl phosphate. The acetyl phos-
acetyl–CoA; but without phosphotransacety- phate levels in cells grown at higher tempera-
lase (encoded by pta), they cannot convert the tures is elevated because the acetate kinase is
acetyl–CoA to acetyl phosphate. Hence the less active. Therefore, the inhibition of flagel-
conclusion that in wild-type cells acetyl phos- lar synthesis at higher temperatures might be
phate represses the operon. due to the increased levels of acetyl phosphate,
On the other hand, expression of the flhDC which may phosphorylate OmpR, which may
operon was decreased in ack mutants. The ack then repress the activity of flhDC.
www.ebook777.com
19.10 Response to Carbon Sources: cAMP. It has been called the FruR (fructose
Catabolite Repression, Inducer repressor) system and more recently the Cra
(catabolite repressor/activator) system.106,107
Expulsion, and Permease Synthesis
Catabolite repression refers to the preferen- Cra is a transcriptional regulator
tial use of one carbon source for growth over Transcriptional regulators exist that induce
another when bacteria are grown in the pres- certain genes and repress others. Cra is such a
ence of both carbon sources. It results in diauxic regulator. It represses genes coding for enzymes
growth as described in Section 2.2.4. There is required for growth on sugars, that is, enzymes
more than one mechanism responsible for catab- of the Embden–Meyerhof–Parnas (EMP, or
olite repression. One mechanism was discussed glycolytic) pathway and the Entner–Doudoroff
in Section 17.3.4, which describes catabolite (ED) pathway, while inducing genes encoding
repression by glucose as a consequence of glu- enzymes required for growth on organic acids
cose transport by the phosphotransferase system and amino acids, that is, enzymes of the citric
(PTS) in E. coli and related bacteria. Here the acid cycle and the glyoxylate cycle, and gluco-
uptake of glucose by the PTS lowers the cAMP neogenic enzymes.
pools, and this slows the transcription of cAMP- The reason for believing that Cra is an acti-
dependent genes required to metabolize alterna- vator for genes required for growth on organic
tive carbon sources. Section 17.3.4 also points acids is that cra mutants cannot grow on lac-
out that glucose uptake by the PTS inhibits per- tate, pyruvate, and citric acid cycle intermedi-
meases required for the uptake of other carbo- ates and show lower activities of enzymes of the
hydrate carbon sources (inducer exclusion). citric acid cycle, glyoxylate cycle, as well as glu-
However, inducer exclusion by the PTS and coneogenic enzymes.108 Recall that growth on
adjustment of cAMP pools are not the only these carbon sources requires gluconeogenesis,
means of catabolite repression. Sections 19.10.1 the citric acid cycle, and the glyoxylate pathway
and 19.10.2 describe two more catabolite repres- (acetate) (Sections 9.1, 9.8.3, and 9.12). The rea-
sion systems. The Cra system, in E. coli, does not son for believing that Cra is a repressor of genes
involve either PTS or cAMP. The second system, required for growth on sugars is that mutations
the CRE system, is found in the gram-positive in cra lead to higher levels of enzymes required
bacterium B. subtilis. The following discussion for the uptake and catabolism of sugars such as
is restricted to catabolite repression by sugars. glucose and fructose. (For more background
However, there exist many bacteria that prefer- information on Cra, see note 109.)
entially use carboxylic compounds rather than
carbohydrates as a source of carbon and will
Model for how Cra activates certain genes
show diauxic growth with the appropriate mix-
and represses others
tures. In these instances it is the carboxylic acid
One model proposes that Cra activates certain
that represses the expression of genes required
genes and inhibits others depending on where
to metabolize the carbohydrate. Some of these
it binds to the DNA in relationship to the RNA
systems have been recently reviewed.105
polymerase binding site. If it sits upstream of the
In addition to catabolite repression, bacteria
binding site, then it is postulated to be an acti-
may use other means to favor growth on particu-
vator; if it sits downstream of the binding site,
lar carbon sources. One of these is called inducer
it is suggested to be a transcription inhibitor
expulsion (Section 19.10.3), which is character-
(Fig. 19.13). This model is in fact supported by
ized by glucose inhibition of the accumulation
sequence analysis studies of Cra-binding sites
of alternative carbon sources. A second is the
that place these sites downstream of negatively
induction of a permease that brings certain car-
controlled promoters and upstream of positively
bon sources into the cell (Section 19.10.4).
controlled promoters. (Reviewed in ref. 110.)
19.10.1 Catabolite repression in E. coli Glucose and fructose remove Cra from DNA,
that does not involve cAMP or CRP: The explaining catabolite repression
Cra system When glucose or fructose is in the growth medium,
E. coli and Salmonella have an additional catab- Cra no longer binds to the promoter regions of
olite repressor system that does not involve the target genes. Thus growth on the sugars can
Cra
RNAP
+
RNAP
pckA pykF
0 –35 –10 –35 –10 0
Cra-activated transcription Cra-inhibited transcription

sugar (out)
PTS
catabolite (in)
C
C C C
C
C C RNAP C C
RNAP
pckA pykF
0 –35 –10 –35 –10 0
catabolite repression catabolite activation

(deactivation) (derepression)
Fig. 19.13 Model for the regulation of gene transcription by Cra. It is proposed that Cra activates gene tran-
scription if it binds upstream of the RNA polymerase-binding site and inhibits gene transcription if it binds
downstream of this site. Catabolites derived from glucose or fructose are proposed to bind to Cra, preventing
it from binding to DNA. Thus, these catabolites would repress genes whose transcription is stimulated by Cra
(gluconeogenic genes) and activate genes whose transcription is repressed by Cra (glycolytic genes). See text
for details. Source: Saier, M. H. Jr., 1996. Cyclic AMP-independent catabolite repression in bacteria. FEMS
Microbiol. Lett. 138:97–103.
occur because the repressor (Cra) for the genes of Model for how glucose and fructose remove
the EMP and ED pathways is removed. However, Cra from DNA
the presence of glucose or fructose prevents Two catabolites derived from glucose or
growth on organic acids, such as acetate or suc- fructose, fructose-1-phosphate (F1B) and
cinate, or amino acids because the activator (Cra) fructose-1,6-bisphosphate (FBP), are thought
of genes that encode enzymes required for growth to bind to Cra and remove it from the DNA,
on these carbon sources is removed. therefore diminishing the effect of Cra. Thus,
www.ebook777.com
these catabolites would repress genes acti- Repression of a nitrite reductase gene by Cra
vated by Cra (gluconeogenic genes) and acti- and catabolite activation
vate genes inhibited by Cra (glycolytic genes). As discussed in Section 19.3, E. coli makes two
The evidence in favor of this model includes different nitrite reductases under anaerobic
the finding that Cra does bind to Cra-regulated conditions regulated by the Nar system. These
operons in vitro and that fructose-1-phos- are the formate-dependent periplasmic nitrite
phate or fructose-1,6-bisphosphate displaces reductase encoded by the nrf operon and the
Cra from DNA. NADH-linked cytoplasmic nitrite reductase
encoded by the nir operon. Both nitrate and
Comparison of the Cra system and nitrite stimulate transcription of nir, whose
the cAMP system product is required for detoxifying the nitrite
The Cra system is similar to the cAMP system formed from nitrate during nitrate respiration.
in that the sugar prevents catabolite-repressed Transcription of nrf is induced by nitrite but
genes from being activated. In one system repressed by nitrate. In addition to regulation
the activator is cAMP–CRP, whose levels are by the Nar system, these operons are induced
lowered by glucose because the levels of the anaerobically by FNR. (See Section 19.3.2.)
adenylyl cyclase activator IIAGlc–P are lowered In addition to stimulation by FNR, nitrate,
when incoming glucose is phosphorylated by and nitrite, the nir operon is repressed by Cra
IIAGlc–P, whereas in the other system the acti- and activated by catabolites.111 (Transcription
vator of gluconeogenic genes is Cra, which is from the nrf operon is not regulated by Cra and
removed from the DNA by F1P or FBP, lead- is catabolite repressed in an unknown man-
ing to catabolite repression of these genes. The ner.) These experiments were done in E. coli
Cra system differs from the cAMP system in strains carrying nir–lacZ and nrf–lacZ tran-
that Cra is also a negative transcription regu- scriptional fusions. Thus, the nir and nrf pro-
lator of genes required for sugar catabolism moters could be monitored by β-galactosidase
and provides for the activation of these genes assays. Transcription of the nir operon was
by glucose and fructose because F1P and FBP repressed in poor medium but increased in rich
remove Cra from these genes. Hundreds of medium containing glucose. Furthermore,
genes are regulated by cAMP and Cra in the transcription of the nir gene was increased
enteric bacteria. in a cra– mutant (Fig. 19.14). Thus it was
Fig. 19.14 Repression of nir (NADH-dependent nitrite reductase) by Cra. E. coli was grown anaerobically in
rich [Luria broth (LB) plus glucose] or poor (minimal salts plus glycerol and fumarate) medium without nitrite
(solid bars), with 2.5 mM nitrite (lined bars), or with 20 mM nitrate (open bars). The first two sets of bars show
the results with the wild-type strain; the results using a cra– mutant are shown on the right. The expression of
the nir–lacZ fusion was monitored via β-galactosidase assays. The expression of the nir operon was enhanced
in rich medium or by a mutation in the cra gene. This is consistent with the hypothesis that Cra represses the
operon and that a catabolite derived from glucose reverses the repression. See text for further details. Source:
Adapted from Tyson, K., S. Busby, and J. Cole. 1997. Catabolite regulation of two Escherichia coli operons
encoding nitrite reductases: role of the Cra protein. Arch. Microbiol. 168:240–244.
concluded that Cra inhibits transcription of the synthesize detectable cAMP or CRP.112,113) It
nir operon and that the inhibition is relieved is called the CcpA (catabolite control protein
by glucose. This is similar to the inhibition by A) system. CcpA is a transcription inhibitor
Cra of the genes responsible for the uptake of genes that are catabolite repressed by glu-
and catabolism of sugars such as glucose and cose. A second transcription repressor, CcpB,
fructose, and the release of the inhibition by has been identified in B. subtilis and proposed
fructose-1-phosphate (F1B) and fructose-1,6- to act in parallel with CcpA via the same
bisphosphate (FBP) derived from glucose. (See mechanism.114
earlier subsection entitled Model for how Cra
activates certain genes and represses others The model
and Fig. 19.13.) The binding of Cra to the nir A proposed model is drawn in Fig. 19.15. When
promoter and the release of Cra from the DNA cells are grown on glucose, the HPr protein that
by fructose-1-phosphate was demonstrated in is also part of the phosphotransferase system
vitro.110 becomes phosphorylated at a serine residue,
Hpr(Ser-P), and binds to the transcription regu-
19.10.2 Catabolite repression in gram- lator CcpA. The Hpr–P–CcpA complex binds
positive bacteria via the CcpA system to a nucleotide sequence called CRE (catabo-
A third system of catabolite repression has been lite responsive elements) (possibly as a ternary
discovered in low-GC (30–40%) gram-posi- complex with fructose-1,6-bisphosphate) that
tive bacteria such as Bacillus, Staphylococcus, is proximal to the promoter of genes susceptible
Streptococcus, Enterococcus, Lactococcus, to catabolite repression and inhibits transcrip-
and Lactobacillus. (These bacteria do not tion. (See note 115.) Thus the key event with
kinase
(inactive) HPr
FBP ATP Pi
kinase
(active) phosphatase
FBP H2O
ADP
HPr P CcpA FBP
DNA
HPr P HPr P
FBP CcpA FBP
HPr(Ser-P)-CcpA-inhibited
transcription
(catabolite repression)
Fig. 19.15 Catabolite repression in B. subtilis. An ATP-dependent Hpr kinase is activated by fructose-1,6-
bisphosphate (FBP). Phosphorylated Hpr (Hpr-P) binds to the transcription regulator CcpA, and possibly
also to FBP, to form a ternary complex. The ternary complex binds to the CRE region near the promoters of
catabolite-repressible genes and represses transcription. Source: Adapted from Saier, M. H., Jr., S. Chauvaux,
J. Deutscher, J. Reizer, and J.-J. Ye. 1995. Protein phosphorylation and regulation of carbon metabolism in
gram-negative versus gram-positive bacteria. Trends Biochem. Sci. 20:267–269.
www.ebook777.com
respect to catabolite repression is that glycolytic catabolite-repressible genes (i.e., cAMP–CRP

intermediates stimulate the phosphorylation or Cra), catabolite repression in B. subtilis is
of Hpr and therefore the repression of CcpA- due to an increase in a negative regulator factor
sensitive operons. (i.e., CcpA–Hpr-Ser-P or CcpB–Hpr-Ser-P).
The key regulatory event, stimulated Does the CcpA-dependent catabolite

by glycolytic intermediates, is the repression system exist in gram-negative
phosphorylation of Hpr bacteria?
It is important to understand that the phospho- A BLAST (Basic Local Alignment Search Tool)
rylation of Hpr is not due to enzyme I, the PEP- search of the Protein Data Bank revealed that
dependent kinase that phosphorylates a histidine homologues of the Hpr kinase are present in
residue, resulting in Hpr(His-P), which functions some gram-negative bacteria.118 Future research
in the PTS. Rather, the phosphorylation is due to might reveal that CcpA-dependent catabolite
an ATP-dependent kinase that phosphorylates a repression occurs in gram-negative bacteria as
serine residue in Hpr, forming Hpr(Ser-P). See well as in gram-positive bacteria.
note 116 for a further explanation.
Hpr is phosphorylated by a specific ATP- 19.10.3 Two mechanisms leading to
dependent kinase that is activated by the gly- the prevention by glucose of the
colytic intermediates fructose-1,6-bisphosphate intracellular accumulation of other
and 2-phosphoglycerate, as well as by gluconate- sugars in some gram-positive
6-P. Thus, when the cells are growing on glucose bacteria: Uncoupling from the proton
and the levels of glycolytic intermediates rise, gradient and inducer expulsion
Hpr becomes phosphorylated and represses the Glucose prevents the accumulation of other
transcription of genes having the CRE sequences. sugars in some gram-positive bacteria, but the
A phosphatase removes the phosphate from mechanism is not via the inhibition of sugar
Hpr-P under starvation conditions. The model uptake by IIAGlc, as is the case with E. coli. Two
also provides a role for inorganic phosphate (Pi). known mechanisms lead to the equilibration of
According to the model, inorganic phosphate the sugar across the cell membrane, hence no
would prevent catabolite repression by inhibit- intracellular accumulation. One of these is the
ing the kinase and by stimulating a phosphatase uncoupling of sugar transport from the proton
that dephosphorylates HPr(Ser-P). motive force, and the second is the dephospho-
rylation of the incoming sugar–P. Consider
Genetic evidence for the model the situation in Lactobrevis, where Hpr(Ser-P)
Mutations in CRE, or the genes coding for CcpA appears to bind to the lactose/proton symporter
or CcpB, or mutations that lead to a failure to and uncouples sugar transport from proton
phosphorylate Ser-46 in Hpr, will cause failure symport.119 The consequence of uncoupling the
of catabolite repression, and such failures com- symporter from the proton gradient is that the
prise genetic evidence in support of the model. symporter catalyzes facilitated diffusion rather
(See note 117 for a general description of how than active transport. The result is that lactose
some of these mutants can be isolated.) Also, equilibrates across the membrane and cannot
the ATP-dependent kinase that phosphorylates be accumulated. In some gram-positive bacteria
HPr at Ser-46 has been purified and shown to (streptococci, lactococci, and enterococci), but
be stimulated by FBP and inhibited by inorganic not others (bacilli, staphylococci), there exists
phosphate. A Pi-activated phosphatase that a membrane-associated sugar–P phosphatase
dephosphorylates HPr-Ser-P has been isolated activated by HPr(Ser-P).120 It is suggested that
from several gram-positive bacteria. when the sugar–Ps are accumulated in the cell
via the PTS, they are dephosphorylated and
Comparison to E. coli then exit along their concentration gradient.
A comparison of the catabolite repression This phenomenon, called inducer expulsion,
mechanism of E. coli with that of B. subtilis occurs upon the addition of glucose. The puri-
shows that whereas catabolite repression in E. fied phosphatase has broad substrate specific-
coli is due to a decrease in a positive regulator of ity and dephosphorylates a long list of sugar–P
molecules, including glucose-1-P, glucose-6-P, to the dicarboxylic acids only when integrated
and fructose-6-P. Presumably only certain into the membrane.
sugar–Ps are attacked because the enzyme’s
location on the membrane determines its acces- 19.11 Virulence Factors: Synthesis
sibility to specific sugar–Ps being transported
via sugar transporters into the cell. in Response to Temperature, pH,
Nutrient, Osmolarity, and Quorum
19.10.4 Response to carbon source: Sensors
Induction of a permease for dicarboxylic This section discusses some of the regulatory
acids in Rhizobium meliloti systems that transduce environmental signals to
R. (Sinorhizobium) meliloti are nitrogen-fixing, control the transcription of genes that encode
gram-negative α-proteobacteria that live in the virulence factors. Virulence factors are struc-
soil or symbiotically in the root nodules of legu- tures or substances that aid pathogenic bacte-
minous plants (e.g., the agriculturally important ria in the infection process and/or contribute
alfalfa plants). (See Section 20.8 for a discussion toward the symptoms of disease (reviewed in
of chemotaxis in R. meliloti.) The bacteria grow refs. 122 and 123). Virulence factors are gen-
most rapidly on C4 dicarboxylic acids such as erally exported to the bacterial cell surface or
succinate, malate, and fumarate. R. meliloti beyond. They include pili for adsorption to host
has a well-characterized two-component sys- tissue, toxins, flagella to aid the bacterium in
tem that regulates the production of a permease arriving at the site of colonization, and extra-
(transporter) that brings dicarboxylic acids into cellular enzymes such as proteases (see, e.g.,
the cell. When R. meliloti is cultured with C4 ref. 124.). As discussed in Section 18.5.2, some
dicarboxylic acids, the gene for the dicarboxylic virulence factors produced by Yersinia spp. are
acid permease (dctA) is induced. actually injected into target cells.
Virulence factors are not necessarily consti-
The model tutively produced, and it appears that several
A transmembrane histidine kinase (sensor are made only during the course of an infection,
kinase) called DctB autophosphorylates at a his- not when the pathogen is growing outside the
tidine residue when the cells are incubated with host. Certainly this is the case with laboratory
dicarboxylic acids; subsequently, it transfers the cultures, where only under specific conditions
phosphoryl group to an aspartate residue in a of temperature, pH, nutrient composition,
cytoplasmic response regulator protein, DctD. iron availability, and osmolarity are certain
DctD-P then activates the transcription of dctA, virulence factors made.125 How bacteria sense
the dicarboxylic acid permease gene. environmental factors such as temperature and
osmolarity is not well understood. For a review
Support for the model of information about a regulatory system influ-
Support for this model includes demonstration enced by osmotic pressure and temperature,
in vitro of autophosphorylation of the purified see Section 19.7, entitled Response to osmotic
pressure and temperature: Regulation of porin
DctB protein, along with the transfer of the
synthesis. The following discussion focuses on
phosphoryl group to DctD and the binding of
the effect of environmental factors, including
phosphorylated DctD to DNA.121
extracellular ones (quorum sensors), on the
The dicarboxylic acids themselves are not expression of virulence genes.
sufficient to activate the sensor kinase
The inclusion of dicarboxylic acids in the incu- 19.11.1 ToxR and cholera
bation mixture does not affect the autophos- ToxR is a transcription regulatory protein pro-
phorylation reaction, indicating that there duced by Vibrio cholerae, the bacterium that
may be another component (perhaps in the causes cholera. It stimulates the transcription of
periplasm) that binds to the dicarboxylic acid virulence genes that are in the ToxR regulon.
and transmits the signal to the histidine kinase, (For reviews, see refs. 126 and 127.) To under-
or that the histidine kinase responds directly stand the importance of ToxR, it is necessary to
briefly describe cholera pathogenesis.
www.ebook777.com
Cholera bifunctional protein that can sense environ-

Vibrio cholerae bacteria grow in the small intes- mental signals and subsequently activate the
tine, attached to the intestinal epithelial layer. transcription of 30 to 40 genes in the ToxR
The microorganisms swim through the mucus regulon. The genes activated by ToxR are called
layer and attach to intestinal cells via type IV the ToxR-activated, or tag, genes, and these are
pili called TCP pili (toxin-coregulated pilus). required for efficient colonization of the intes-
The symptoms of cholera include an exten- tine, toxin production, and survival within the
sive watery diarrhea, which can result in the host.
loss of as much as 10 to 20 liters of fluid a day, The tag genes have been identified by screen-
and if untreated results in dehydration, loss of ing TnphoA fusions whose transcriptional
electrolytes, and death in approximately 60% activity under different growth conditions par-
of infected people. The diarrhea is caused by allels that of cholera toxin production (See note
cholera toxin (CT), a potent enterotoxin that is 26 for an explanation of gene fusions and note
secreted by V. cholerae. (Cholera toxin consists 131 for how such fusions were used to detect
of one A subunit and five identical B subunits and ToxR-regulated genes.) Genes whose transcrip-
is encoded by the ctxAB operon, which resides tion is increased by ToxR include the ctxAB
in the CTX lysogenic phage. See note 128 for operon, which encodes the A and B subunits of
a description of how cholera toxin works.) V. cholera toxin; tcpA, which encodes the major
cholerae is spread via the oral–fecal route, and subunit of the V. cholerae pilus; several genes
people become ill with cholera when they ingest required for pilus assembly and transport; and
food or water contaminated with feces contain- other genes, including toxT. As mentioned, the
ing the bacteria. pilus is required for V. cholerae to bind to the
intestinal epithelium and for subsequent colo-
Response to temperature, pH, osmolarity, nization. Thus, it can be considered to be a viru-
and amino acids lence factor. As described next, except for the
Exactly which environmental stimuli in the transcription of ctxAB, the genes in the ToxR
intestine influence the activity of ToxR within regulon are not activated directly by ToxR.
infected animals is a subject of study.124,129 Toxin
production is lowered in pure culture when the How ToxR activates gene expression
temperature is increased to 37 °C, which is For study purposes, the toxR gene was cloned
opposite to what one would intuitively expect. in E. coli, where it activates the transcription of
Additionally, low pH in culture stimulates chol- a ctx–lacZ fusion. (See note 26.) ToxR binds to
era toxin production, but the pH in the intestine the promoter region of the ctxAB operon, result-
where V. cholerae grows is believed to be alka- ing in increased transcription, but its activation
line. This suggests that factors other than tem- of the other genes in the ToxR regulon is indi-
perature and pH increase the activity of ToxR rect. That is, it appears that ToxR increases the
during growth in the intestine. Two of these transcription of a second regulatory gene, toxT,
factors might be osmolarity and the presence of whose product directly activates the other genes
certain amino acids, since toxin production in in the ToxR regulon, including ctxAB and and
pure culture is increased under osmolar condi- the toxin-coregulated pilus (TCP) operon.132
tions similar to those in mucus, as well as in the Figure 19.16 shows one model of the regulatory
presence of amino acids likely to be present in pathways.
mucosal secretions. However, the expression of
the ToxR regulon in response to environmental The mystery of how ToxR transmits
cues can differ with the strain of V. cholerae. environmental signals to the ToxR regulon
See note 130. It is unclear exactly how ToxR receives and
transmits environmental signals to the ToxR
The role of ToxR in pathogenesis regulon. It has been concluded, based in part
The importance of ToxR for pathogenesis is upon amino acid sequence data and the con-
clear, since mutants in which the toxR gene has struction of ToxR–PhoA fusions that have
been deleted do not colonize the intestine in alkaline phosphatase activity and also regulate
human volunteers. ToxR is an inner membrane the cholera toxin gene, that ToxR spans the cell
environmental signals periplasmic domain that may actually receive

environmental signals and activate ToxR by
TcpPH ToxRS cell membrane interacting with its periplasmic domain. The
cytoplasmic domain of ToxR would then acti-
vate the toxT gene, whose product in turn would
+ activate the genes in the ToxR regulon (Fig.
toxT
19.16). However, other signaling proteins are
also involved. Two of these, TcpP and TcpH,
ToxT are proteins encoded by the TcpPH operon, as
+ + + described next.
TCP operon CTX operon
The toxT gene is also positively regulated

pilus production ToxT cholera toxin by TcpP
The toxT gene is also positively regulated by
Fig. 19.16 Model showing regulatory pathways for another membrane-bound transcriptional acti-
ToxR regulon. TcpPH and ToxRS represent pairs of vator, TcpP, which itself may be activated by
inner membrane proteins (TcpP/TcpH and ToxR/ a membrane-associated protein with a large
ToxS) that respond to environmental signals and periplasmic domain, TcpH.134 Optimal expres-
coactivate the transcription of toxT. ToxT, along sion of toxT requires both ToxR and TcpP. As
with ToxR, activates the transcription of the cholera shown in Fig. 19.16, both ToxR and TcpP are
toxin (CTX) operon, encoded on a lysogenic phage involved in sensing environmental and cyto-
(CTX phage). ToxT also activates the transcription plasmic signals and cooperatively activate the
of the toxin-coregulated pilus (TCP) operon, which is toxT gene.
responsible for the biogenesis of the pilus required to
colonize the intestine. The TCP operon contains the
gene for the main pilin subunit (TcpA, not shown), as 19.11.2 The PhoP/PhoQ two-component
well as genes required for the regulation of pilin gene regulatory system
expression, pilin secretion, and the assembly of the A two-component system called the PhoP/
pilus. The toxT gene is also within the TCP operon, PhoQ system exists in many nonpathogenic
and thus more ToxT is produced when the operon is and pathogenic gram-negative bacteria, includ-
induced. ing E. coli, P. aeruginosa, and Salmonella spp.
PhoQ is a transmembrane sensor histidine
kinase, and PhoP is a cytoplasmic response
membrane with a periplasmic C-terminal region regulator. For a review of the PhoP/PhoQ sys-
and an N-terminal cytoplasmic region that has tem, see refs. 133 and 135. The PhoP/PhoQ
a DNA-binding domain. (See ref. 133 and refer- system is not related to the two-component
ences therein and note 26 for an explanation of PhoB/PhoR system discussed in Section 19.6.1.
why alkaline phosphatase activity from a toxR– The gene phoP was originally discovered as a
phoA fusion can indicate a C-terminal periplas- positive regulator for the synthesis of an acid
mic domain.) The C-terminal domain may be phosphatase, the product of the phoN gene.
sensitive to certain environmental stimuli, dis- See note 136 for a description of how phoP was
cussed next. The N-terminal domain is thought discovered.
to bind to promoter sequences in the DNA of
target genes, and it may also respond to intrac- The PhoP/PhoQ system is activated by low
ellular signals. Mg2+ concentrations and inactivated by high
Mg2+ concentrations
A signal-transducing cascade The PhoP/PhoQ system is stimulated by low
How the ToxR protein senses the environmental extracellular Mg2+ and is required to adapt to
signals, and whether other components besides low extracellular Mg2+ concentrations (micro-
ToxR are involved in sensing these signals, molar range).137,138 Two of the genes activated
are open questions. It has been suggested that by PhoP/PhoQ encode proteins for two of the
another protein, ToxS, encoded in the toxRS three Mg2+ uptake systems. The model shown
operon, is a membrane-bound protein with a in Fig. 19.17 proposes that when the Mg2+
www.ebook777.com
PhoP-P
H2O
Pi (+) PhoQ Mg++ A
PhoP
ATP
low Mg++
ADP
PhoP P PhoQ B
target (+) PhoQ

PhoP-P
genes
PmrD PrmB
target (+)
(+) PmrA-P
genes ATP
Fe++ C
ADP
PmrA
P PrmB
Fig. 19.17 The regulation of gene transcription by PhoP/PhoQ and PmrA/PmrB. (A) All genes activated by
PhoP are repressed by high concentrations of Mg2+ because the sensor histidine kinase PhoQ dephosphory-
lates PhoP-P when the Mg2+ concentrations are high. (B) PhoQ autophosphorylates when the Mg2+ concentra-
tions are low and transfers the phosphoryl group to PhoP. PhoP-P activates the transcription of many genes,
including pmrD. In a way that is not understood, PmrD increases the levels of PmrA-P, which in turn stimu-
lates transcription of target genes. (C) Fe2+ binds to PmrB and activates its autophosphorylation activity, thus
stimulating transcription of PmrA-dependent genes independently of PhoP/PhoQ.
concentrations are sufficiently low, PhoQ auto- Salmonella pathogenesis

phosphorylates at a conserved histidine residue We will begin with a brief summary of Salmonella
in its cytoplasmic domain and then acts as a pathogenesis followed by the role of PhoP/
kinase and phosphorylates PhoP at a conserved PhoQ in regulating the expression of virulence
aspartate residue. Phosphorylated PhoP then genes. Salmonella spp. cause gastroenteritis,
interacts with promoters on target genes and enteric fever, and septicemias (blood infections).
activates transcription. However, high Mg2+ con- During the course of the infections, the bacteria
centrations repress genes regulated by PhoP-P are ingested by phagocytic cells such as mac-
because PhoQ is a Mg2+ sensor, and when the rophages and epithelial cells and must survive
periplasmic domain of PhoQ binds Mg2+, PhoQ intracellularly. Indeed, the intestinal epithelial
becomes a phosphatase and dephosphorylates cells must be invaded to allow the spread of the
PhoP-P. In addition to high Mg2+, Ca2+ binds to bacteria from the intestinal mucosal surface to
PhoQ and represses the transcription of genes underlying tissue and the development of gas-
activated by PhoP, and these two cations should troenteritis, enteric fever, and septicemia.
be viewed as being physiological signals con-
trolling the PhoQ/ PhoP system. See ref. 139 for The PhoP/PhoQ system regulates
a review. This discussion will focus on what has virulence genes
been learned from studying Salmonella because There are at least 40 genes regulated by the
this genus has been investigated the most. PhoP/PhoQ system in Salmonella, and they are
in the PhoP/PhoQ regulon. (Recall that a regu- pmrD gene, and PmrD stimulates the PmrA/
lon is a set of noncontiguous operons or genes PmrB system at the post-transcriptional level
controlled by a common regulator. The regu- so that more PmrA becomes phosphorylated.143
lator in this case is PhoP.) Some of the genes Exactly how this is done is not yet known, but
regulated by the PhoP/PhoQ system govern either increased phosphorylation or decreased
virulence. For example, PhoQ and PhoP are dephosphorylation may be involved. The
responsible for the induction in Salmonella PmrA/PmrB system can also be activated inde-
spp. of the five pag genes (PhoP-activated pendently of PhoP/PhoQ by growth in high Fe2+
genes), which are important for the survival concentrations. This stimulates the synthesis
of Salmonella inside the acid bactericidal envi- of proteins that protect the bacterium from
ronment of phagolysosomes of macrophages the toxic effects of Fe2+. Thus, the PmrA/PmrB
(they are not induced inside epithelial cells) system in Salmonella is activated by low Mg2+
and for the repression of several other genes concentrations via PhoP/PhoQ, or directly by
called prg (PhoP-repressed genes).140 One of high Fe2+ concentrations sensed by the histidine
the pag genes, namely pagC, is responsible kinase sensor protein PmrB. This is outlined in
for the synthesis of an envelope protein that Fig. 19.17. The pmrD gene is also present in
supports virulence as well as survival inside several gram-negative bacteria, indicating that
macrophages. control of PmrA/PmrB by PhoP/PhoQ is not
Mutants in phoQ/phoP are less virulent restricted to Salmonella spp.
and have decreased survivability inside mac-
rophages. They are also less resistant to cationic 19.11.4 The bvg genes and pertussis
antimicrobial peptides such as defensins, which There are several other regulatory genes that
the bacteria can encounter in the intestinal lumen respond to environmental stimuli and promote
and within phagolysosomes of neutrophiles virulence. (Reviewed in ref. 124.) These include
and macrophages. (See Section 19.11.3.) Some the bvgAS genes (Bordetella virulence genes),
of the genes activated by PhoP-P modify the which are required for the expression of viru-
lipopolysaccharide, whereas others are respon- lence genes (the vir regulon) in Bordetella per-
sible for the synthesis of extracellular proteases tussis, the causative agent of whooping cough
that cleave the antimicrobial peptides. (pertussis).144 (The bvgAS locus was formerly
called the vir locus.) Growth at 37 °C as opposed
19.11.3 PmrA/PmrB: A two-component to 25 °C results in increased expression of viru-
system that interacts with the lence genes.
PhoP/PhoQ system
In addition to regulating the activity of several Pertussis
genes directly, PhoP is part of a signaling hier- Whooping cough is primarily a childhood dis-
archy that indirectly regulates the expression ease in which the bacteria grow attached to the
of several Salmonella genes by activating a sec- ciliated epithelial cells of the bronchi and tra-
ond two-component signaling pathway called chea. (Virulence is dependent upon bacterial
the PmrA/PmrB system, where PmrB is the his- surface adhesins that bind the bacteria to the
tidine kinase and PmrA is the response regu- cilia.) Episodes of severe, spasmodic coughing
lator. The PmrA/PmrB system in Salmonella are followed by a “whoop” during inspiration.
spp. and Pseudomonas aeruginosa activates The organism is transmitted via respiratory dis-
genes conferring resistance to the antibiotic charges. Children are now vaccinated with the
polymyxin, a cationic antimicrobial peptide DPT (diphtheria, pertussis, tetanus) vaccine.
(CAP).141 See note 142 for an explanation of However, prior to the vaccine pertussis was a
CAPs and related antimicrobial compounds major childhood killer.
called defensins.
Resistance to polymyxin is due at least in part The vir regulon
to a modification of the lipopolysaccharide, Genes induced by the bvg genes (i.e., the
to which polymyxin and other CAPs bind. In genes in the vir regulon) include the pertussis
Salmonella, PhoP-P activates the PmrA/ PmrB toxin operon ptxA-E, the adenylate cyclase
system by promoting the transcription of the toxin–hemolysis gene cyaA, the filamentous
www.ebook777.com
hemagglutinin locus fhaB, the fimbrial subunit 1. hns (virR): Response to temperature
genes fim, and the bvg operon. Shigella has a chromosomal regulatory gene
called hns (virR) that encodes a repressor for
The bvg genes encode a two-component the virulence genes. The significance of hns for
signaling system that regulates the vir this discussion is that the regulation of viru-
regulon lence genes by hns is controlled by temperature.
According to deduced amino acid sequences, Shigella strains grown at 30 °C are avirulent
BvgS is a transmembrane sensor kinase with a (i.e., the virulence genes are repressed by hns),
periplasmic N-terminus region and a cytoplas- whereas those grown at 37 °C are virulent in
mic C-terminus region, Similarly, BvgA is pro- tissue culture assays and in animal assays (i.e.,
posed to be a response regulator protein that is they cause invasion of tissue culture cells and
phosphorylated by BvgS-P and then activates production of keratoconjunctivitis in guinea
transcription of the virulence genes, perhaps pigs). Consistent with the role of hns as a repres-
in some cases by activating other genes that sor gene at 30 °C, mutations in this gene allow
are transcripitional activators. (In E. coli, the expression of the invasive phenotype at both 30
cloned bvgAS locus will activate the fhaB and and 37 °C. Presumably, turning off virulence
bvgA promoters but not the ptxA-E or cyaA pro- genes is an advantage to the bacteria when they
moters, suggesting that perhaps a signal relay are growing outside a host. The product of the
is required for the latter promoters.) Increased hns gene is a histonelike protein called H-NS
expression of virulence genes at 37 °C has also (H1), and binding of H-NS near target genes
been studied in Shigella, as discussed next. leads to transcriptional repression, perhaps by
interfering with the binding of RNA polymerase.
19.11.5 Virulence genes and bacillary (See the discussion of H-NS and the nucleoid in
dysentery Section 1.2.6, and the discussion of the control
of rRNA synthesis in Section 2.2.2.)
Bacillary dysentery
Shigella spp., including S. dysenteriae and S. 2. virF and virB
flexneri, cause bacillary dysentery (shigellosis), The virulence genes are also positively regulated.
which can be manifested simply as a watery One of the regulatory genes, virF, encodes a pos-
diarrhea or, in more serious cases, as a severe itive regulator (VirF) of the virulence gene regu-
diarrheal disease of the colon accompanied by lon. It activates some of the vir genes directly,
passage of bloody, mucoid stools. In the severe and others indirectly, by activating the tran-
cases, the bacteria invade and multiply within scription of another positive regulatory gene,
the mucosal epithelial cells, causing degenera- virB. It is thought that H-NS (H1), the product
tion of the epithelium and intense inflammation of the hns gene, represses virF expression at 30
of the colon (colitis). (Shigella are facultative °C and that this accounts for the repression of
intracellular pathogens.) Most Shigella species the other virulence genes. The virulence genes
produce a toxin, and the toxin produced by are also repressed when the cells are grown in
Shigella dysenteriae is shiga toxin, which is at low-osmolarity media at 37 °C, and this was
least partly responsible for damaging the blood also shown to be due to repression by H-NS in
vessels, for inflammation, and probably for the S. flexneri.147
watery diarrhea.
3. IHF: Responses to temperature and
Virulence genes osmolarity
Virulence of Shigella is due primarily to a plas- When Shigella grow outside the host in nature,
mid that encodes the virulence genes. Virulence they are generally found in media of low osmo-
genes are required for adherence and invasion larity as well as at temperatures lower than 37
(via endocytosis) of host cells, spreading (via °C. In addition to being repressed by H-NS
polymerization of actin filaments), and synthe- at 30 °C, the expression of virulence genes
sis of toxin. (For a review of virulence genes in is positively stimulated upon a temperature
Shigella and information on how the bacteria upshift. The positive stimulation appears to be
invade host cells and spread intercellularly, see due to another DNA-binding protein, integra-
ref. 145. Also, see note 146.) tion host factor (IHF).148 (See the discussion
of DNA-binding proteins in the nucleoid in the Ti (tumor-inducing) plasmid. The process

Section 1.2.6.) This was determined by examin- of transfer of the T-DNA from the bacterium
ing the expression of a lacZ fusion to a virulence into the plant resembles bacterial conjugation
gene upon a temperature upshift in wild-type and takes place via a type IV secretory system
cells and ihf:Tn10 mutants. described in Section 18.5.6. The T-DNA inte-
Thus, DNA binding proteins H-NS (a repres- grates into the plant genome, and a tumor
sor) and IHF (an activator) appear to take part results. The tumor forms because T-DNA
in signaling pathways that detect environmen- encodes plant growth hormones (i.e., auxin and
tal changes (i.e., temperature and osmolarity) cytokinin), which stimulate growth and divi-
that result in alterations in the expression of sion of the plant cells.
virulence genes.
When the T-DNA integrates into the plant
19.11.6 Virulence gene expression genome, the plant makes nutrients called
in Staphylococcus: Stimulation of a opines, which feed the bacteria
two-component system by a peptide The crown gall tumor actually feeds the bacte-
pheromone ria. There are genes in the T-DNA that encode
Staphyloccus aureus secretes several extracel- enzymes to make organic molecules called
lular protein virulence factors. For example, S. opines, which when released by the plant cell can
aureus secretes enterotoxin B and toxic shock be used as a source of nutrient by the bacteria.
syndrome toxin 1, as well as other toxins in the Opines are small organic molecules covalently
postexponential phase of growth. The tran- bonded to amino acids. For example, one opine,
scription of the toxin genes is regulated by a called octopine, is N2-(1,3-dicarboxyethyl)-L-
two-component system consisting of a mem- arginine. The genes that encode enzymes to uti-
brane histidine kinase called AgrC, a response lize the opines are in the region of the Ti plasmid
regulator protein called AgrA, and a small pep- that is not transferred. Thus, when the T-DNA
tide secreted by the cells into the medium.149 The becomes incorporated into the plant genome,
peptide is a quorum-sensing signal because the the plant makes the opines and the bacteria use
amount secreted increases with cell density; and these opines for growth.
when the concentration of the peptide reaches Interestingly, the genes from the Ti plasmid
a threshold concentration, it activates AgrC, that are expressed in the plant have eukaryotic-
which leads to increased transcription of viru- type promoters and eukaryotic-type translation
lence genes. initiation regions, whereas the plasmid genes
expressed in the bacterium have prokaryotic-
19.11.7 Virulence gene expression in type promoters and prokaryotic-type transla-
Agrobacterium tumefaciens: Stimulation of tion initiation regions. The transfer of T-DNA
a two-component system by phenolic plant from the Ti plasmid into the plant cell does
exudates not result in loss of T-DNA from the plasmid.
One of the best-characterized systems for the This is because only one strand of the T-DNA
regulation of expression of virulence genes in is transferred to the plant cell, and a new com-
bacterial–plant interactions is the VirA–VirG plementary strand is synthesized in the donor
two-component system in Agrobacterium bacterium. The process is very much like the
tumefaciens, the bacterium that causes crown transfer of a single plasmid strand during bacte-
gall tumors in plants. (For reviews, see refs. 124 rial conjugation.
and 150.)
The plant signals the bacteria to transfer
Stages in the formation of crown gall tumors T-DNA into the plant
The bacteria gain entrance to the plant tissue For T-DNA to be transferred from the bacte-
via a wound site on the plant and grow extra- rium into the plant, virulence genes (vir genes)
cellularly at the site of infection. Bacteria that on the plasmid must be activated. The vir gene
are directly attached to plant cells transfer into products include an endonuclease (VirD) that
the plant cell nucleus a small piece of oncogenic nicks the plasmid DNA, a DNA-binding pro-
DNA, called T-DNA, from a plasmid, called tein (VirE) that attaches to the single strand of
www.ebook777.com
DNA produced by the endonuclease and brings by the bacteria within the tumor. TraR also stim-
the DNA into the plant cell, and a bacterial ulates the genes for plasmid replication. Thus,
membrane protein (VirB) that functions as a the plant opines stimulate plasmid replication
bridge in transporting the DNA across the bac- and therefore tumorigenesis, as well as conjuga-
terial membrane into the plant cell. [The VirB tion and plasmid transfer. For more information
protein(s) are actually part of a type IV secre- about how all this occurs, see note 151.
tion system. See Section 18.5.6 for a discussion
of this point.] 19.12 Summary
The plant activates the expression of the vir Bacteria can sense environmental and cytoplas-
genes in the bacterium. What happens is that mic signals and transmit information to the
phenolic signaling compounds (e.g., p-hydroxy- genome or to other parts of the cell to elicit a
benzoic acid) produced by the wounded plant response. Such signaling systems detect and
tissue activate in the bacterium a two-com- transmit signals when there are changes in con-
ponent system called the VirA–VirG system. ditions such as pH, osmolarity, temperature,
VirA is a transmembrane sensor kinase whose starvation, and nutrient source, as well as the
periplasmic domain is thought to interact with absence or presence of specific inorganic ions,
the phenolic compounds. VirA is then proposed such as phosphate. Often global transcription
to autophosphorylate and transfer the phos- regulators are involved. These are transcription
phoryl group to VirG, a cytoplasmic regulator regulators that control transcription in noncon-
protein. The phosphorylated form of VirG is tiguous operons. A set of noncontiguous oper-
believed to be a positive transcription factor for ons or genes regulated by the same transcription
the vir genes, including virA and virG, which regulator is called a regulon.
are also plasmid encoded. The result is transfer Many of the signaling systems are called
of T-DNA into the plant cell. “two-component” signaling systems because
they have a histidine kinase (component 1) that
Role of plant opines and a bacterial quorum- causes the phosphorylation (transphosphory-
sensing system in stimulating plasmid lation) of a response regulator (component 2).
transfer between bacteria in the tumor However, other proteins may also be involved
As described in Section 18.5.6, the Ti plasmid in signal transduction. (The histidine kinase is
is transferred between Agrobacterium cells in a also called a histidine/sensor kinase.)
conjugation process that requires the expression All the “two-component” systems have
of tra genes on the plasmid. When the density of four functional components: (1) a sensor that
the cell population is sufficiently high, the trans- receives the signal, (2) an autophosphory-
fer is stimulated by a bacterial quorum-sensing lating histidine protein kinase, (3) a partner
system called the TraR–TraI system, which is response regulator that is the substrate for the
homologous to the Lux system in luminescent histidine kinase, and (4) a phosphatase that
bacteria. TraI is the acyl–HSL synthase, and removes the phosphate from the regulator. As
it synthesizes the signaling molecule 3-oxo- discussed shortly, some of the proteins can be
C8-HSL. The acyl–HSL binds to and activates bifunctional, hence able to carry out two of the
TraR. Active TraR then stimulates the expres- activities. During signaling, phosphate (actu-
sion of genes required for conjugation, as well ally the phosphoryl group) travels from ATP
as the transcription of traI, the gene responsible to a histidine residue in the histidine kinase to
for making the acyl–HSL. The plant opines an aspartate residue in the response regulator
stimulate the transcription of traR, and there- to water (the phosphatase reaction). The phos-
fore the production of the acyl–HSL, since TraR phorylated form of the response regulator is
stimulates the transcription of traI. Because of the active state and transmits the signal to the
this, the plant opines stimulate plasmid transfer genome (stimulation or inhibition of transcrip-
between bacterial cells in the tumor. tion), to the flagellar motor (reversal of turning
Since the genes to metabolize the opines are direction), or to an enzyme. Although many
part of the Ti plasmid, the spread of the plasmid signals cause an increase in phosphorylation
between the bacteria distributes the ability to of the response regulator, some (e.g., chemoat-
catabolize the opines, which are used as nutrient tractants, excess inorganic phosphate, excess
ammonia) cause a decrease in phosphorylation serve as a global metabolic signal that affects
(which in some cases is due to dephosphoryla- the activity of several different genes by phos-
tion of the response regulator) and therefore an phorylating response regulator proteins.
inactivation of the response regulator. Responses that use a two-component regula-
All the histidine protein kinases have con- tory system include the Che system in chemot-
served domains (transmitter domains) that bear axis (discussed in Chapter20), the ArcA/ArcB
a startling similarity to one another and are the system for oxygen regulation of gene expression,
basis for the classification of these proteins. the NarL/X/Q system for nitrate regulation, the
These conserved domains, lying in the carboxy PHO system for phosphate assimilation, the
terminus of the protein, are the site of the con- Ntr system for nitrogen assimilation, the EnvZ
served histidine residue that is phosphorylated. system for porin gene expression, the KdpABC
The response regulators have conserved amino- system for K+ uptake, the RegB/RegA system
terminal (receiver) domains that are believed for the regulation of expression of the genes for
to interact with the carboxy terminus of the the light-harvesting and photoreaction center
kinase. For several of the response regulators, proteins in Rhodobacter capsulatus, some bio-
it has been shown that the phosphoryl group is luminescence signaling systems in Vibrio, and
transferred from the histidine to the conserved the dicarboxylic acid permease system in R.
aspartate residue in the amino terminus of the meliloti. Two-component systems also detect
response regulator. environmental signals and regulate the tran-
Although the two-component systems are reg- scription of virulence genes in certain bacteria
ulated primarily by the activities of their respec- (e.g., Agrobacterium, Salmonella, Bordetella).
tive histidine kinases, it has been suggested that There are several other important regula-
acetyl phosphate may be a phosphoryl donor to tory systems in bacterial pathogens that control
different response regulator proteins under cer- the expression of virulence genes in response
tain physiological conditions. This has usually to environmental signals. These may involve
been observed in mutants that lack the cognate a transmembrane sensor, which need not be
histidine kinase. In these mutants, acetyl phos- part of a two-component signaling system, and
phate is able to donate the phosphoryl group which responds to environmental signals such
to the response regulator. The extent to which as temperature, osmolarity, and pH. Such sys-
this might occur in wild-type cells that have tems include the ToxR/S system in Vibrio chol-
the cognate histidine kinase is generally not yet erae. Shigella spp. possess a regulatory gene
known. However, analysis of mutants in acetyl called the virR gene that encodes a protein that
phosphate synthesis or utilization in which the is part of a system that regulates the tempera-
levels of acetyl phosphate would be expected ture-dependent expression of virulence genes.
to be higher or lower than wild-type cells sug- The FNR system, which regulates anaerobic
gests that increased levels of acetyl phosphate gene expression, is not a two-component regu-
can inhibit flagellar biosynthesis in E. coli wild- latory system. Neither is there evidence that the
type cells by phosphorylating OmpR, which in formate regulon, which is induced by formate
its phosphorylated form is a negative transcrip- under anaerobic conditions, is controlled by a
tion regulator of regulatory genes required for two-component system. It is clear, moreover,
flagellar biosynthesis. If this sequence in fact that the Arc and the FNR systems represent two
occurs, it could account for why E. coli grown systems of gene regulation with which E. coli
at higher temperatures have fewer flagella. The senses anaerobiosis and activates the transcrip-
possibility is reasonable, since in cells grown tion regulators ArcA and FNR.
at higher temperatures, the acetyl phosphate The situation is often complex, with mul-
levels are elevated because the enzyme that tiple regulators controlling the same operon.
converts acetyl phosphate to acetate (acetate In the control of the cydAB operon in E. coli,
kinase) is less active at the higher temperatures. for example, the operon encodes cytochrome
Presumably this set of conditions leads to a d oxidase and is regulated by FNR and ArcA,
phosphorylation of OmpR and a consequent and interestingly by Cra as well. This was dem-
repression of the flagellar regulatory genes. All onstrated by using cyd–lacZ fusion strains that
these data suggest that acetyl phosphate may have null mutations in either cra, arcA, or fnr
www.ebook777.com
genes. Catabolite repression refers to the repres- 2. What criteria must be established to char-
sion by particular carbon sources of expression acterize a protein as a kinase or response
of genes required for growth on other carbon regulator protein in a two-component
sources. The result is diauxic growth. For system?
example, in many bacteria glucose represses
3. What is FNR, and what is its role? Why is
genes required for the uptake and/or catabolism
it not believed to be part of a two-compo-
of other carbon sources. Some bacteria utilize
nent regulatory system?
carboxylic acids in preference to glucose, and
it is the carboxylic acids that repress the genes 4. In the Arc and Nar systems, which pro-
required for glucose catabolism. teins are thought to be the sensor/kinase
The mechanisms that underlie catabolite proteins and which ones the regulator
repression vary. These include the cAMP– proteins? What is the evidence for this?
CRP-dependent system for glucose repression
5. What is the phenotype of an Fnr– mutant
that has been well studied in E. coli. This sys-
in E. coli? An arc– mutant?
tem involves the PTS sugar uptake system. E.
coli also has a cAMP–CRP-independent sys- 6. Describe the role that PII plays in tran-
tem, the Cra system. Cra is a global regulator scriptional regulation of the glnALG
that influences the direction of carbon flow in operon and in the regulation of activity of
either the glycolytic or gluconeogenic direction glutamine synthetase. How are the levels
by controlling the expression of genes encoding of PII and PII–UMP regulated?
enzymes required for the relevant pathways. Cra
7. In what way is catabolite repression by the
is an activator for genes required for gluconeo-
Cra system similar to the same mechanism
genesis and a repressor for genes required for
in the cAMP system? How does it differ
glucose catabolism. When the cells are growing
from catabolite repression by the CcpA
on glucose or fructose, it is proposed that the
system in B. subtilis?
levels of fructose-1-phosphate and fructose-1-
,6-bisphosphate rise and bind to Cra, removing 8. How does B. subtilis partition Hpr-P
it from the DNA. Thus, gluconeogenic genes between the phosphotransferase sys-
required for growth on pyruvate, lactate, ace- tem and the CcpA catabolite repression
tate, and citric acid cycle intermediates would system?
be repressed and glycolytic genes required for
9. What is the evidence that acetyl phosphate
growth on carbohydrates would be activated.
may be a global signal?
Catabolite repression in low-GC (percentage
of cytosine and guanine in DNA), gram-posi- 10. Discuss the role that a two-component
tive bacteria involves yet a different system, the phosphorelay system might play in the pro-
CcpA system. When these bacteria are growing duction of virulence factors by pathogenic
on glucose, the levels of the glycolytic interme- bacteria. Give an example of such a two-
diates fructose-1,6-bisphosphate and 2-phos- component system. Describe some experi-
phoglycerate rise and activate a special kinase ments to test which environmental factors
that phosphorylates Hpr, forming Hpr(Ser-P). might provide the necessary signals.
Phosphorylated Hpr may form a ternary com-
plex with the CcpA DNA-binding protein and
fructose-1,6-bisphosphate, and this complex
can bind to the CRE sequence in target genes, 1. Kirby, J. R. 2009. Chemotaxis-like regulatory
repressing transcription. systems: unique roles in diverse bacteria. Annu. Rev.
2. Hoch, J. A., and T. J. Silhavy (Eds.). 1995.
Study Questions Two-Component Signal Transduction. ASM Press,
Washington, DC.
1. Describe the components of two-compo-
3. The transmitter domain of the sensor kinase,
nent regulatory systems and how they work. which is at the carboxyl terminus of the protein, is
In your answer, explain what is meant by generally attached to an input domain (e.g., a region
“cross talk” or “cross-regulation.” that might bind a signaling molecule, such as a
ligand-binding site). Thus, the sensor kinase would stutzeri. Aerobic and nitrate respiration routes of
be N–input domain–transmitter domain–C. The carbohydrate catabolism. J. Bacteriol. 91:245–250.
input domain would be expected to be exposed to the
16. Reviewed in: Berks, B. C., S. J. Ferguson, J. W.
signal; for example, it might be in the periplasm if
B. Moir, and D. J. Richardson. 1995. Enzymes and
the sensor kinase were a transmembrane protein in a
associated electron transport systems that catalyze
gram-negative bacterium. Presumably, after binding
the respiratory reduction of nitrogen oxides and
the signal, the sensor kinase undergoes a conforma-
oxyanions. Biochim. Biophys. Acta 1232:97–173.
tional change that results in its autophosphorylation
in the transmitter domain and subsequent phosphory- 17. Gunsalus, R. P. 1992. Control of electron flow in
lation of the response regulator protein in its receiver Escherichia coli: coordinated transcription of respi-
domain. As expected, input domains vary, depend- ratory pathway genes. J. Bacteriol. 174:7069–7074.
ing upon the sensor kinase. The receiver domain of
the response regulator protein is usually attached to 18. Sawers, G., and B. Suppman. 1992. Anaerobic
an output domain at the amino terminus such as a induction of pyruvate formate–lyase gene expres-
DNA-binding region. Thus, the response regulator sion is mediated by the ArcA and FNR proteins. J.
would be N–receiver domain–output domain–C. Bacteriol. 174:3474–3478.
The sequences of the DNA-binding domains are 19. Sawers, G. 1993. Specific transcriptional
divided into classes that place the response regulator requirements for positive regulation of the anaerobi-
proteins into families (e.g., the OmpR family). Not cally inducible pfl operon by ArcA and FNR. Mol.
all response regulator proteins have output domains. Microbiol. 10:737–747.
For example, CheY and Spo0F do not.
20. Anderson, D. I. 1992. Involvement of the Arc
4. Hughes, D. A. 1994. Histidine kinases hog the system in redox regulation of the cob operon in
limelight. Nature 369:187–188. Salmonella typhimurium. Mol. Microbiol. 6:1491–
5. Wurgler-Murphy, S. M., and H. Saito. 1997. 1494.
Two-component signal transducers and MAPK cas- 21. Mutations in arcA produce a defect in the syn-
cades. Trends Biochem. Sci. 22:172–176. thesis of F pili, which results in resistance to male-
6. Saier, M. H., Jr. 1993. Introduction: protein specific bacteriophages that attach to the F pilus. (See
phosphorylation and signal transduction in bacteria. Fig. 19.5.) ArcA also responds to a second sensor
J. Cell. Biochem. 51:1–6. kinase protein (CpxA), which is necessary for the
production of the F pilus in donor strains of E. coli.
7. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. The cpxA gene was originally discovered as a muta-
Protein phosphorylation and regulation of adaptive tion that reduced the efficiency of DNA transfer as a
responses in bacteria. Microbiol. Rev. 53:450–490. consequence of reduced F-plasmid tra gene expres-
8. Parkinson, J. S., and E. C. Kofoid. 1992. sion. It is now known that the CpxA protein is an
Communication modules in bacterial signaling pro- inner membrane protein whose amino acid sequence
teins. Annu. Rev. Genet. 26:71–112. places it in the class of sensor kinase proteins. There
also exists a cognate response regulator (CpxR) for
9. Ota, I. M., and A. Varshavsky. 1993. A yeast pro- CpxA, shown in Fig. 19.5. The complex relationship
tein similar to bacterial two-component regulators. between these two regulatory systems is discussed in
Science 262:566–569. the following papers: Iuchi, S., D. Furlong, and E.
10. Chang, C., S. F. Kwok, A. B. Bleecker, and E. C. C. Lin. 1989. Differentiation of arcA, arcB, and
M. Meyerwitz. 1993. Arabidopsis ethylene-response cpxA mutant phenotypes of Escherichia coli by sex
gene etr1: similarity of product to two-component pilus formation and enzyme regulation. J. Bacteriol.
regulators. Science 262:539–544. 171:2889–2893. Dong, J.-M., S. Iuchi, H.-S. Kwan,
Z. Lu, and E. C. C. Lin. 1993. The deduced amino
11. Wanner, B. L. 1992. Is cross-regulation by acid sequence of the cloned cpxR gene suggests the
phosphorylation of two-component response regu- protein is the cognate regulator for the membrane
lator proteins important in bacteria? J. Bacteriol. sensor, CpxA, in a two-component signal transduc-
174:2053–2058. tional system of Escherichia coli. Gene 136:227–
12. Reviewed in Antonie van Leeuwenhoek. 1994. 230.
Vol. 66, Nos. 1–3. 22. Iuchi, S., and E. C. C. Lin. 1992. Purification and
13. Spiro, S., and J. R. Guest. 1991. Adaptive phosphorylation of the Arc regulatory components
responses to oxygen limitation in Escherichia coli. of Escherichia coli. J. Bacteriol. 174:5617–5623.
Trends Biochem. Sci. 16:310–314. 23. Georgellis, D., A. S. Lynch, and E. C. C. Lin.
14. Gunsalus, R. P., and S.-J. Park. 1994. Aerobic– 1997. In vitro phosphorylation study of the Arc two-
anaerobic gene regulation in Escherichia coli: con- component signal transduction system of Escherichia
trol by the ArcAB and FNR regulons. Res. Microbiol. coli. J. Bacteriol. 179:5429–5435.
145:437–449.
24. With regard to the genes for succinic dehydro-
15. Spangler, W. J., and C. M. Gilmour. 1966. genase (sdhCDAB), it should be pointed out that the
Biochemistry of nitrate respiration in Pseudomonas aerobic/anaerobic regulation of transcription of this
www.ebook777.com
operon involves more than simply the ArcA/ArcB have the transposon inserted into the host bacterial
system. Not only are the levels of succinic dehydro- genes in the proper orientation and frame, so that
genase suppressed during anaerobic growth, but they β-galactosidase production is under the control of
are also lowered in E. coli by growth on glucose, even the promoter of the interrupted gene. Insertion of the
when the cells are grown aerobically and raised by transposon into a gene interrupts the gene so that the
growth on acetate. Furthermore, mutations in FNR normal gene product is not made. The gene is identi-
increase the levels of transcription of an sdhC–lacZ fied by mutant analysis.
fusion, suggesting that FNR represses the succinate Cloned genes fused to the E. coli periplasmic alka-
dehydrogenase operon under anaerobic conditions. line phosphatase gene, phoA, are also used. These
The mechanism by which glucose lowers the expres- produce a hybrid protein consisting of the N-terminal
sion of the sdh operon is not understood. It appears region of the gene whose transcription is being ana-
to be independent of the cAMP–CAP system. See: lyzed and the C-terminal end of alkaline phosphatase.
Park, S.-J., C.-P. Tseng, and R. P. Gunsalus. 1995. The hybrid protein is missing the N-terminal alkaline
Regulation of succinate dehydrogenase (sdhCDAB) phosphatase signal sequence for protein export and
operon expression in Escherichia coli in response to relies on the protein export signal sequence of the
carbon supply and anaerobiosis: role of ArcA and target gene for export. When the C-terminal region
FNR. Mol. Microbiol. 15:473–482. of the hybrid protein containing the alkaline phos-
phatase region reaches the periplasm, the alkaline
25. Iuchi, S., V. Chepuri, H.-A. Fu, R. B. Gennis,
phosphatase activity can be measured. Such fusions
and E. C. C. Lin. 1990. Requirement for terminal
can be used to examine the transcription of genes that
cytochromes in generation of the aerobic signal for
encode secreted proteins such as virulence factors.
the arc regulatory system in Escherichia coli: study
utilizing deletions and lac fusions of cyo and cyd. J. 27. The control experiments were done to show
Bacteriol. 172:6020–6025. that, in an arc+ cyo+ cyd+ background, anaerobiosis
repressed cyo–lacZ and induced cyd–lacZ expres-
26. Gene fusions are valuable probes to monitor
sion, and that a mutation in the arc genes prevented
the expression of genes of interest. Genes are fused
the repression of cyo–lacZ and lowered the expres-
to reporter genes whose products are easy to iden-
sion of cyd–lacZ. These experiments showed that
tify. For example, a reporter gene might be lacZ
the fusion genes are regulated by the availability of
(β-galactosidase), lux (luciferase), cat (chloram-
oxygen and that the Arc system is responsible for the
phenicol acetyltransferase), or phoA (alkaline phos-
regulation.
phatase). Consider a lacZ fusion. The fused gene
has the promoter region of the target gene but not 28. Rodriguez, C., O. Kwon, and D. Georgellis.
the promoter for the lacZ gene. Expression of the 2004. Effect of D-lactate on the physiological activ-
fused gene is therefore under control of the pro- ity of the ArcB sensor kinase in Escherichia coli. J.
moter region of the target gene. If the translational Bacteriol. 186:2085–2090.
initiating region (ribosome-binding site) is not
29. Discussed in: Iuchi, S., A. Artistarkhov, J. M.
present in the reporter gene, a translational fusion
Dong, J. S. Taylor, and E. C. C. Lin. 1994. Effects
results. The fusions produce a hybrid protein whose
of nitrate respiration on expression of the Arc-
amino-terminal end is derived from the target gene
controlled operons encoding succinate dehydroge-
and its carboxy-terminal end from β-galactosidase.
nase and flavin-linked L-lactate dehydrogenase. J.
The hybrid protein has β-galactosidase activity.
Bacteriol. 176:1695–1701.
Therefore, an assay for β-galactosidase is a measure
of the expression of the target gene. Thus, one can 30. Lynch, A. S., and E. C. C. Lin. 1996. Responses
measure the expression of virtually any gene simply to molecular oxygen, pp. 1526–1538. In: Escherichia
by constructing the proper gene fusion and perform- coli and Salmonella: Cellular and Molecular Biology,
ing an assay for β-galactosidase. One can construct Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press,
gene fusions in vitro or in vivo. In vitro construction Washington, DC.
involves using restriction endonucleases to cut from
31. Spiro, S., and J. R. Guest. 1990. FNR and its role
a plasmid containing the cloned DNA a portion of
in oxygen-regulated gene expression in Escherichia
the gene with its promoter region and ligating it to a
coli. FEMS Microbiol. Rev. 75:399–428.
lacZ gene, without its promoter or ribosome-binding
site, in a second plasmid. The plasmid containing the 32. Unden, G., S. Becker, J. Bongaerts, G. Holighaus,
fused gene is then introduced into the bacterium and J. Schirawski, and S. Six. 1995. O2-sensing and
transformants are selected on the basis of their resis- O2-dependent gene regulation in facultatively anaer-
tance to an antibiotic-resistant marker on the plas- obic bacteria. Arch. Microbiol. 164:81–90.
mid, and their production of β-galactosidase. In vivo
33. Jones, H. M., and R. P. Gunsalus. 1987.
construction of gene fusions can also be performed.
Regulation of Escherichia coli fumarate reductase
In this case, one uses Tn5 transposons fused to a pro-
(frdABCD) operon expression by respiratory elec-
moterless lacZ gene. The transposon is introduced
tron acceptors and the fnr gene product. J. Bacteriol.
into the bacterium, where it can recombine with the
169:3340–3349.
bacterial chromosome. Cells harboring the trans-
poson are selected by means of the antibiotic-resis- 34. Unden, G., and J. R. Guest. 1985. Isolation and
tance marker on the transposon. Many of the strains characterization of the FNR protein, the transcrip-
tional regulator of anaerobic electron transport in genases to distinguish them from FDH-H, which
Escherichia coli. Eur. J. Biochem. 146:193–199. functions only during fermentation. (FDH-O is also
called formate oxidase.) Since FDH-O is also present
35. It appears that inactivation of FNR is via the
in cells grown with nitrate as the electron acceptor,
reversible oxidation of the [4Fe–4S]2+ clusters. The
it may also be able to reduce nitrate. FDH-H is part
oxidation of the cluster results in the inactivation of
of the formate–hydrogen lyase enzyme complex that
the protein under conditions of prolonged incubation
oxidizes formate to CO2 and H2 during fermentation.
in oxygen by the disassembly and loss of the clusters
E. coli also synthesizes three hydrogenases: Hyd-1,
from the protein. This leads to the release of sulfide
Hyd-2, and Hyd-3. Of these, Hyd-1 and Hyd-2
(S2–) and ferric ion (Fe3+). Reassembly of the clusters
(uptake hydrogenases) oxidize H2 (e.g., during the
from sulfide and Fe ions (Fe2+ and Fe3+) has been
reduction of fumarate or nitrate). Hyd-3 is part of
demonstrated in vitro under anaerobic conditions
the formate–hydrogen lyase and is responsible for the
and requires a special protein, the NifS protein from
reduction of protons generating hydrogen gas.
Azotobacter, which is required for the incorporation
of Fe–S clusters into the Azotobacter nitrogenase. E. 44. Reviewed in: Unden, G., S. Becker, J. Bongaerts,
coli has a similar protein. See ref. 36. J. Schirawski, and S. Six. 1994. Oxygen-regulated
gene expression in facultatively anaerobic bacteria.
36. Unden, G., and J. Schirawski. 1997. The oxy-
Antonie van Leeuwenhoek 66:3–23.
gen-responsive transcriptional regulator FNR of
Escherichia coli: the search for signals and reactions. 45. The regulon consists of the following genes: the
Mol. Microbiol. 25:205–210. hyc operon (hydrogenase 3), the hyp operon (uptake
hydrogenases 1 and 2), and fdhF (formate dehydro-
37. Roualt, T. A., and R. D. Klausner. 1996. Iron–
genase). The fhlA gene is encoded in the hyp operon.
sulfur clusters as biosensors of oxidants and iron.
Trends Biochem. Sci. 21:174–177. 46. Motteram, P. A. S., J. E. G. McCarthy, S. J.
Ferguson, J. B. Jackson, and J. A. Cole. 1981. Energy
38. Jordan, P. A., A. J. Thomson, E. T. Ralph, J. R.
conservation during the formate-dependent reduc-
Guest, and J. Green. 1997. FNR is a direct oxygen
tion of nitrite by Escherichia coli. FEMS Microbiol.
sensor having a biphasic response curve. FEBS Lett.
Lett. 12:317–320.
416:349–352.
47. In E. coli, the membrane-bound nitrate reductase
39. Nakano, M. M., P. Zuber, P. Glaser, A. Danchin,
is encoded by the narGHJI operon. The fumarate
and F. M. Hulett. 1996. Two-component regulatory
reductase is encoded by the frdABCD operon. The
proteins ResD–ResE are required for transcriptional
DMSO/TMAO reductase is encoded by the dmsABC
activation of fnr upon oxygen limitation in subtilis. J.
operon. The periplasmic formate-dependent nitrite
Bacteriol. 178:3796–3802.
reductase is encoded by the nrfABCDEFG operon.
40. These proteins more closely resemble FNR than The cytoplasmic nitrate reductase is encoded by the
CRP on the basis of one or more of the following nirBDC operon.
characteristics: DNA-binding specificity, sequence
48. Unden, G., and J. Bongaerts. 1997. Alternative
identity, failure to bind cAMP, complementation
respiratory pathways of Escherichia coli: energetics
of an E. coli fnr mutant, or being activated under
and transcriptional regulation in response to electron
conditions of oxygen limitation. They are involved
acceptors. Biochim. Biophys. Acta 1320:217–234.
in the regulation of various physiological reactions
such as luminescence (Vibrio fischeri), denitrifica- 49. Gennis, R. B., and V. Stewart. 1996. Respiration,
tion (Pseudomonas aeruginosa), nitrogen fixation pp. 217–261. In: Escherichia coli and Salmonella:
(Rhizobium species), and hemolysin biosynthesis Cellular and Molecular Biology, Vol. 1. F. C.
(Bordetella pertussis). Neidhardt et al. (Eds.). ASM Press, Washington,
DC.
41. Spiro, S. 1994. The FNR family of transcriptional
regulators. Antonie van Leeuwenhoek 66:23–36. 50. Rabin, R. S., and V. Stewart. 1993. Dual
response regulators (NarL and NarP) interact with
42. Compan, I., and D. Touati. 1994. Anaerobic
dual sensors (NarX and NarQ) to control nitrate and
activation of arcA transcription in Escherichia coli:
nitrite-regulated gene expression in Escherichia coli
roles of Fnr and ArcA. Mol. Microbiol. 11:955–964.
K-12. J. Bacteriol. 175:3259–3268.
43. E. coli actually synthesizes three different for-
51. Darwin, A. J., K. L. Tyson, S. J. W. Busby, and
mate dehydrogenases (FDH) and three distinct
V. Stewart. 1997. Differential regulation by the
hydrogenases (Hyd). FDH-O is synthesized during
homologous response regulators NarL and NarP of
aerobic growth and also anaerobically when nitrate
Escherichia coli K-12 depends on DNA binding site
is the electron acceptor. FDH-N is made anaerobi-
arrangement. Mol. Microbiol. 25:583–595.
cally when nitrate is present, and FDH-H is made
only under fermentation conditions. The FDH-O 52. Williams, S. B., and V. Stewart. 1997.
and FDH-N enzymes oxidize formate and transfer Discrimination between structurally related ligands
the electrons via quinones to oxygen and nitrate, nitrate and nitrite controls autokinase activity of
respectively, during respiration. Accordingly, these the NarX transmembrane signal transducer of
are sometimes called respiratory formate dehydro- Escherichia coli K-12. Mol. Microbiol. 26:911–925.
www.ebook777.com
53. Chiang, R. C., R. Cavicchioli, and R. P. Gunsalus. and S. Kustu. 2000. Nitrogen regulatory protein
1997. “Locked-on” and “locked-off” signal trans- C–controlled genes of Escherichia coli: scavenging
duction mutations in the periplasmic domain of as a defense against nitrogen limitation. Proc. Natl.
the Echerichia coli NarQ and NarX sensors affect Acad. Sci. USA 97:14674–14679.
nitrate- and nitrite-dependent regulation by NarL 64. It has been estimated that NtrI (NitrC) activates
and NarP. Mol. Microbiol. 24:1049–1060. approximately 75 genes in 25 operons in E. coli.
54. The discovery of the nar genes unfolded as fol- Most of these are transcribed by σ54-holoenzyme
lows. Originally mutants that were defective in nitrate and activated directly. One of the activated genes is
repression of the fumarate reductase gene, frdA, were nac, which encodes the nitrogen assimilation control
isolated. (The actual screen was to look for mutant (Nac) protein that binds to DNA and activates the
colonies in which nitrate did not repress the expres- transcription by σ70-holoenzyme of about 25 of the
sion of the fusion gene frdA–lacZ.) The mutations 75 genes (about 9 operons). For a list of the genes
were found to be in two genes, called narL and narX. regulated by NitrC and Nac, see ref. 63.
Whereas narL was definitely required for nitrate
65. Magasanik, B. 1982. Genetic control of nitro-
repression or induction of the nitrate-regulated
gen assimilation in bacteria. Annu. Rev. Gen. 16:
genes, mutations in narX caused only a partial loss
135–168.
of nitrate regulation. It therefore appeared that narX
was dispensable. Another gene was later discovered, 66. Reviewed in: Magasanik, B. 1996. Regulation of
narQ, that was also a sensor for nitrate reductase. E. nitrogen utilization, pp. 1344–1356. In: Escherichia
coli can use either NarX or NarQ as the sensor when coli and Salmonella, Cellular and Molecular Biology,
one or the other is inactivated. To observe the loss of Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press,
nitrate-dependent repression of fumarate reductase Washington, DC.
and loss of induction of nitrate reductase, cells must 67. Ninfa, A. J., and P. Jiang. 2005. PII signal trans-
be mutated in both narX and narQ. duction proteins: sensors of α-ketoglutarate that reg-
55. Stewart, V. 1993. Nitrate regulation of anaero- ulate nitrogen metabolism. Curr. Opin. Microbiol.
bic respiratory gene expression in Escherichia coli. 8:168–173.
Mol. Microbiol. 9:425–434. 68. MacNeil, T., G. P. Roberts, D. MacNeil, and
56. Schröder, I., C. D. Wolin, R. Cavicchioli, and R. B. Tyler. 1982. The products of gln L and gln G are
P. Gunsalus. 1994. Phosphorylation and dephospho- bifunctional regulatory proteins. Mol. Gen. Genet.
rylation of the NarQ, NarX, and NarL proteins of the 188:325–333.
nitrate-dependent two-component regulatory system 69. Sigma 54 (encoded by the rpoN gene in E. coli)
of Escherichia coli. J. Bacteriol. 176:4985–4992. polymerases are widely distributed among bacteria,
57. Cavicchioli, R., I. Schröder, M. Constanti, and not simply the enterics, and they bind to promoters
R. P. Gunsalus. 1995. The NarX and NarQ sensor– that are unrelated to σ70 promoters. They transcribe
transmitter proteins of Escherichia coli each require many different genes (i.e., not merely genes in the Ntr
two conserved histidines for nitrate-dependent signal regulon). Sigma 54 polymerases can bind to promoter
transduction to NarL. J. Bacteriol. 177:2416–2424. regions of target genes but, unlike σ70 polymerases,
58. Cavicchioli, R., I. Schröder, M. Constanti, and are unable to produce an open complex without the
R. P. Gunsalus. 1995. The Escherichia coli NarQ and aid of an activator protein which usually binds 100
NarX regulatory proteins contain two conserved his- to 150 base pairs upstream of the promoter. (See
tidines that are required for nitrate dependent signal Fig. 19.9.) Binding of the activator protein at such
transduction to NarL. J. Bacteriol. 177:2416–2424. long distances from the start site of transcription is
required for the looping mechanism of activation. In
59. Cavicchioli, R., R. C. Chiang, L. V. Kalman, and the looping mechanism the intervening DNA loops
R. P. Gunsalus. 1996. Role of the periplasmic domain out, allowing the activator to interact with the poly-
of the Escherichia coli NarX sensor–transmitter pro- merase. For some promoters, the looping out of the
tein in nitrate-dependent signal transduction and DNA is facilitated by DNA-bending proteins such as
gene regulation. Mol. Microbiol. 21:901–911. integration host factor (IHF). (For σ70 polymerases,
60. Magasanik, B. 1993. The regulation of nitro- the transcription activators bind adjacent to the poly-
gen utilization in enteric bacteria. J. Cell. Biochem. merase site and interact with the polymerase with-
51:34–40. out looping of the DNA.) The activator proteins are
ATP dependent and, in a way that is not understood,
61. Magasanik, B. 1988. Reversible phosphoryla- the hydrolysis of ATP provides the energy to convert
tion of an enhancer binding protein regulates the the closed complex to the open complex, thus initi-
transcription of bacterial nitrogen utilization genes. ating transcription. For a review of σ54, see: Buck,
Trends Biochem. Sci. 13:475–479. M., M.-T. Gallegos, D. J. Studholme, Y. Guo, and
62. Merrick M. J., and R. A. Edwards. 1995. J. D. Gralla. 2000. The bacterial enhancer-depen-
Nitrogen control in bacteria. Microbiol. Rev. dent σ54 (σN) transcription factor. J. Bacteriol. 182:
59:604–622. 4129–4136.
63. Zimmer, D. P., E. Soupene, H. L. Lee, V. F. 70. The reason for the requirement of NRI–P for tran-
Wendisch, A. Khodursky, B. J. Peter, R. A. Bender, scription is that the σ54 polymerase binds to the glnAp2
promoter but cannot by itself initiate transcription. It group to PhoB, but it has not been demonstrated that
is unable to do this because it cannot form an open it dephosphorylates PhoB-P (see ref. 73).
complex (i.e., it cannot “melt” the DNA double helix
80. Swem, D. L., and C. E. Bauer. 2002. Coordination
around the promoter site to gain access to the tran-
of ubiquinol oxidase and cytochrome cbb3 oxidase
scription start site). However, NRI–P binds to sites
expression by multiple regulators in Rhodobacter
on the DNA base pairs upstream from the promoter
capsulatus. J. Bacteriol. 184:2815–2820.
region (called “enhancer sites,” containing 100–130
bp) and interacts with the RNA polymerase to form 81. Bauer, C. E., and T. H. Bird. 1996. Regulatory
the open complex and thus initiate transcription. circuits controlling photosynthesis gene expression.
Cell 85:5–8.
71. Dixon, R. 1998. The oxygen-responsive NIFL–
NIFA complex: a novel two-component regula- 82. Dong, C., S. Elsen, L. R. Swem, and C. E. Bauer.
tory system controlling nitrogenase synthesis in 2002. Aer, a second aerobic repressor of photosyn-
γ-proteobacteria. Arch. Microbiol. 169:371–380. thetic gene expression in Rhodobacter capsulatus. J.
Bacteriol. 184:2805–2814.
72. Friedman, D. I. 1988. Integration host factor: a
protein for all reasons. Cell 55:545–554. 83. Mosely, C., J. Y. Suzuki, and C. E. Bauer. 1994.
Identification and molecular genetic characteriza-
73. Wanner, B. L. 1993. Gene regulation by
tion of a sensor kinase responsible for coordinately
phosphate in enteric bacteria. J. Cell. Biochem.
regulating light harvesting and reaction center gene
51:47–54.
expression in response to anaerobiosis. J. Bacteriol.
74. Phosphonates are organophosphates in which 176:7566–7573.
there is a direct carbon–phosphorus bond rather
84. Elsen, S., L. R. Swem, D. L. Swem, and C. E.
than a phosphate ester linkage.
Bauer. 2004. RegB/RegA, a highly conserved redox-
75. The promoters in the PHO regulon are unusual responding global two-component regulatory sys-
in that they have –10 sequences but no consensus tem. Microbiol. Mol. Biol. Rev. 68:263–279.
–35 sequences for σ70-RNA polymerase. Thus, in the
85. Du, S., T. H. Bird, and C. E. Bauer. 1998. DNA
absence of PhoB, RNA polymerase does not bind to
binding characteristics of RegA. J. Biol. Chem.
the promoter. The promoters, however, have one to
273:18509–18513.
three (tandemly repeated) 18-base-pair sequences
called pho boxes, which are 10 base pairs upstream 86. HvrA is an H-NS-like protein. Such proteins are
of the –10 consensus sequence. The phosphorylated widespread in gram-negative bacteria. They bind to
PhoB protein binds to the pho boxes, and this enables DNA and modulate gene expression. See the subsec-
the RNA polymerase to bind to the promoter. It has tion of Section 2.2.2 entitled 3. Other factors that
been postulated that PhoB interacts with the sigma control rRNA synthesis: Fis, H-NS. HvrA has been
subunit of the RNA polymerase, enabling the poly- reported to be necessary for the ammonium-depen-
merase to bind to the promoter. See Makino, K., M. dent inhibition of the transcription of certain of the
Amemura, S.-K. Kim, A. Nakata, and H. Shinagawa. nitrogenase genes (nif genes) in the nonsulfur purple
1993. Role of the σ70 subunit of RNA polymerase in photosynthetic bacterium Rhodobacter capsulatus,
transcriptional activation by activator protein PhoB as well as being a low-light activator of genes for the
in Escherichia coli. Genes Dev. 7:149–160. photosynthetic apparatus. Raabe, K., T. Drepper,
K.-U. Riedel, B. Masepohl, and W. Klipp. 2002. The
76. van Veen, H. W. 1997. Phosphate transport in
H-NS-like HvrA modulates expression of nitrogen
prokaryotes: molecules, mediators and mechanisms.
fixation genes in the phototrophic purple bacterium
Antonie van Leeuwenhoek. 72:299–315.
Rhodobacter capsulatus by binding to selected nif
77. Wanner, B. L. 1996. Phosphorus assimilation promoters. FEMS Microbiol. Lett. 216:151–158.
and control of the phosphate regulon. pp. 1357–
87. Zucconi, A. P., and J. T. Beatty. 1988.
1381 in: Escherichia coli and Salmonella, Cellular
Posttranscriptional regulation by light of the steady-
and Molecular Biology. F. C. Neidhardt et al. (Eds.).
state levels of mature B800–850 light-harvesing
complexes in Rhodobacter capsulatus. J. Bacteriol.
78. The Pst system is similar to the histidine uptake 170:877–882.
system described in Section 17.3.3 and belongs to the
88. Nikaido, H., and M. Vaara. 1987. Outer mem-
superfamily of ABC (ATP-binding cassette) transport-
brane, pp. 7–22. In: Escherichia coli and Salmonella
ers. These have four membrane-bound proteins, two of
typhimurium, Cellular and Molecular Biology,
which are transmembrane and probably form a chan-
nel, whereas the other two have ATP-binding sites. In
Washington, DC.
addition, there is a periplasmic binding protein. Pst A
and Pst C are thought to form a transmembrane chan- 89. Forst, S. A., and D. L. Roberts. 1994. Signal
nel; PstB is the ATPase, which probably exists as a transduction by the EnvZ–OmpR phosphotransfer
dimer; and PtsS is the periplasmic binding protein. system in bacteria. Res. Microbiol. 145:363–373.
79. PhoR autophosphorylates in vitro using ATP as 90. Reviewed in: Pratt, L. A., and T. J. Silhavy. 1995.
the phosphoryl donor and transfers the phosphoryl Porin regulon of Escherichia coli, pp. 105–127.
www.ebook777.com
In: Two-Component Signal Transduction. J. A. Hoch of phosphorylated metabolic intermediates in the

and T. J. Silhavy (Eds.). ASM Press, Washington, regulation of glutamine synthetase synthesis in
DC. Escherichia coli. J. Bacteriol. 174:6061–6070.
91. Waukau, J., and S. Forst. 1992. Molecular 104. Shin, S., and C. Park. 1995. Modulation of fla-
analysis of the signaling pathway between EnvZ and gellar expression in Escherichia coli by acetyl phos-
OmpR in Escherichia coli. J. Bacteriol. 174:1522– phate and the osmoregulator OmpR. J. Bacteriol.
1527. 177:4696–4702.
92. Aiba, H., T. Mizuno, and S. Mizushima. 1989. 105. For a review of catabolite repression by car-
Transfer of phosphoryl group between two regula- boxylic acids, see: Research in Microbiology, 1996,
tory proteins involved in osmoregulatory expression Vol. 147, No. 6–7.
of the ompF and ompC genes in Escherichia coli. J.
106. Saier, M. H., Jr. 1996. Cyclic AMP-independent
Biol. Chem. 264:8563–8567.
catabolite repression in bacteria. FEMS Microbiol.
93. Aiba, H., F. Nakasai, S. Mizushima, and T. Lett. 138:97–103.
Mizuno. 1989. Evidence for the physiological impor-
107. Saier, M. H., Jr., T. M. Ramseier, and J. Reizer.
tance of the phosphotransfer between two regulatory
1996. Regulation of carbon utilization, pp. 1325–
components, EnvZ and OmpR, in osmoregulation in
1343. In: Escherichia coli and Salmonella typhimu-
Escherichia coli. J. Biol. Chem. 264:14090–14094.
rium: Molecular and Cellular Biology, Vol. 1. F. C.
94. Ramani, N., L. Huang, and M. Freundlich. 1992. Neidhardt et al. (Eds.). ASM Press, Washington,
In vitro interactions of integration host factor with DC.
the ompF promoter-regulatory region of Escherichia
108. Chin, M. A., B. U. Feucht, and M. H. Saier Jr.
coli. Mol. Gen. Genet. 231:248–255.
1987. Evidence for regulation of gluconeogenesis by
95. Mizuno, T., M.-Y. Chou, and M. Inouye. 1984. the fructose phosphotransferase system in Salmonella
A unique mechanism regulating gene expression: typhimurium. J. Bacteriol. 169:897–899.
translational inhibition by a complementary RNA
109. Mutants in cra show increased levels of
transcript (micRNA). Proc. Natl. Acad. Sci. USA
enzymes such as the following, which are involved in
81:1966–1970.
the uptake and catabolism of glucose: enzyme I and
96. Reviewed in: Delihas, N. 1995. Regulation of HPr of the phosphotransferase system, phospho-
gene expression by trans-encoded antisense RNAs. fructokinase, pyruvate kinase, 6-phosphogluconate
Mol. Microbiol. 15:411–414. dehydratase, and KDPG aldolase. Cra is also a repres-
sor of the fruBKA operon, which consists of genes
97. Ramani, N., H. Hedeshian, and M. Freundlich.
required for the fructose-specific PTS. The fruBKA
1994. micF Antisense RNA has a major role in osmo-
operon codes for three proteins. FruB is an HPr-like
regulation of OmpF in Escherichi coli. J. Bacteriol.
protein called Fpr, which differs structurally from
176:5005–5010.
HPr used in the regular PTS but can substitute for
98. Sugiura, A., K. Nakashima, K. T. Tanaka, and HPr in mutants unable to synthesize HPr. In fact, cra
T. Mizuno. 1992. Clarification of the structural and mutants were originally found as suppressor mutants
functional features of the osmoregulated kdp operon that allowed mutants defective in HPr synthesis (ptsH
in Escherichia coli. Mol. Microbiol. 6:1769–1776. mutants) to grow on PTS carbohydrates. In addition
to a C-terminal HPr domain, Fpr has an N-terminal
99. Nakashima, K., H. Sugiura, H. Momoi, and T.
IIAFru domain. FruK is a fructose-1-phosphate kinase
Mizuno. 1992. Phosphotransfer signal transduc-
that is necessary to convert the incoming fructose-1-
tion between two regulatory factors involved in the
phosphate to fructose-1,6-bisphosphate. FruA is the
osmoregulated kdp operon in Escherichia coli. Mol.
permease. These fructose-specific genes are required
Microbiol. 6:1777–1784.
for fructose transport via the fructose–PTS system,
100. Sugiura, A., K. Hirokawa, K. Nakashima, and and the conversion of the incoming fructose-1-phos-
T. Mizuno. 1994. Signal-sensing mechanisms of the phate to fructose-1,6-bisphosphate.
putative osmosensor KdpD in Escherichia coli. Mol.
110. Ramseier, T. M. 1996. Cra and the control of
carbon flux via metabolic pathways. Res. Microbiol.
101. Heermann, R., A. Fohrmann, K. Altendorf, 147:489–493.
and K. Jung. 2003. The transmembrane domains of
111. Tyson, K., S. Busby, and J. Cole. 1997.
the sensor kinase KdpD of Escherichia coli are not
Catabolite regulation of two Escherichia coli oper-
essential for sensing K+ limitation. Mol. Microbiol.
ons encoding nitrite reductases: role of the Cra pro-
47:839–848.
tein. Arch. Microbiol. 168:240–244.
102. McCleary, W. R., J. B. Stock, and A. J.
112. Saier, M. H., Jr., S. Chauvaux, J. Deutscher, J.
Ninfa. 1993. Is acetyl phosphate a global signal in
Reizer, and J.-J. Ye. 1995. Protein phosphorylation
Escherichia coli? J. Bacteriol. 175:2793–2798.
and regulation of carbon metabolism in gram-nega-
103. Feng, J., M. R. Atkinson, W. McCleary, J. B. tive versus gram-positive bacteria. Trends Biochem.
Stock, B. L. Wanner, and A. J. Ninfa. 1992. Role Sci. 20:267–271.
113. Hueck, C. J., and W. Hillen. 1995. Catabolite 120. Ye, J.-J., and M. H. Saier Jr. 1995. Purification
repression in subtilis: a global regulatory mecha- and characterization of a small membrane-as-
nism for the gram-positive bacteria? Mol. Microbiol. sociated sugar–phosphate phosphatase that is
15:395–401. allosterically activated by HPr(Ser-P) of the phos-
photransferase system in Lactococcus lactis. J. Biol.
114. Chauvaux, S., I. T. Paulsen, and M. H. Saier Chem. 270:16740–16744.
Jr. 1997. CcpB, a novel transcription factor impli-
cated in catabolite repression in subtilis. J. Bacteriol. 121. Giblin, L., B. Boesten, S. Turk, P. Hooykaas,
180:491–497. and F. O’Gara. 1995. Signal transduction in the
Rhizobium meliloti dicarboxylic acid transport sys-
115. Depending upon the gene, CREs may overlap tem. FEMS Microbiol. Lett. 126:25–30.
promoters or be located within the reading frame.
One suspects that if the CRE overlapped the pro- 122. Finlay, B. B., and S. Falkow. 1997. Common
moter, binding of a regulatory protein would prevent themes in microbial pathogenicity revisited.
initiation of transcription; and if the CRE were in the Microbiol. Mol. Biol. Rev. 61:136–169.
reading frame, then binding of a regulatory protein
123. Falkow, S. 1997. What is a pathogen? ASM
would inhibit transcription elongation. The protein
News 63:359–365.
that binds to the CREs has not been unambiguously
identified. 124. Beier, D., G. Spohn, R. Rappuoli, and V.
Scarlato. 1997. Identification and characterization
116. Also present in B. subtilis is the PEP-dependent
of an operon of Helicobacter pylori that is involved
kinase, which can also phosphorylate Hpr and which
in motility and stress adaptation. J. Bacteriol.
operates during phosphotransferase-mediated glu-
179:4676–4683.
cose uptake. (Although the phosphotransferase
system is characteristic of anaerobes and facultative 125. Miller, J. F., J. J. Mekalanos, and S. Falkow.
anaerobes, it does occur in some aerobes, e.g., B. sub- 1989. Coordinate regulation and sensory transduc-
tilis.) However, the PEP-dependent kinase is not part tion in the control of bacterial virulence. Science
of the catabolite-repressing system. This is because 243:916–922.
the phosphorylation of Hpr by the ATP-dependent
kinase is at a serine residue (Ser-46), whereas the 126. Skorupski, K., and R. K. Taylor. 1997. Control
phosphorylation of Hpr by the PEP-dependent kinase of the ToxR virulence regulon in Vibrio cholerae by
(EI) is at a histidine residue (His-15), and Hpr(His-P) environmental stimuli. Mol. Microbiol. 25:1003–
is not active in catabolite repression. 1009.
127. Raskin, D., J. Bina, and J. Mekalanos. 2004.
117. One method of isolating mutations in genes
Genomic and genetic analysis of Vibrio cholerae.
required for catabolite repression by glucose is to
ASM News 70:57–62.
mutagenize a strain carrying an operon transla-
tional fusion of the gene of interest to lacZ. One can 128. Cholera toxin consists of one A subunit and
then screen for blue colonies on media containing five identical B subunits. The pentamer of B subunits
glucose plus the alternative carbon source and the binds to GM1 ganglioside (a glycolipid) on the sur-
indicator X-Gal (5-bromo-4-chloro-3-indoyl-β-D- face of host intestinal epithelial cells. The A subunit is
galactoside). X-Gal is an analogue of lactose. It is separated from the B pentamer and enters the cytosol,
colorless, but when it is cleaved by β-galactosidase where it is dissociated into two fragments, A1 and A2,
a blue color is produced. Mutagenesis can be per- by the reduction of a disulfide bond that links these
formed by using transposons. A transposon (Tn) is subunits together. The A1 subunit activates adeny-
a DNA sequence that moves from one DNA mole- late cyclase. The activation of adenylate cyclase leads
cule to another or from one part of a DNA molecule to an increase in cAMP, which leads to secretion of
to another part. When the transposon inserts in a chloride and water. Activation of adenylate cyclase
gene of interest, it can produce a mutation in that is a multistep process. The A1 subunit catalyzes the
gene because of the disruption in the gene’s DNA transfer of ADP-ribose from NAD+ to a GTP-binding
sequence. The transposon can be introduced into the protein (Gs) that is associated with adenylate cyclase.
bacterial cell on a plasmid via transformation or con- ADP-ribosylation of the GTP-binding protein inhib-
jugation. Transposons can carry antibiotic resistance its its GTPase activity, locking the adenylate cyclase
genes that may be used to select for transformants into the “on” mode. To understand this, it is neces-
during the isolation of mutants as well as for selec- sary to know how adenylate cyclase is regulated.
tion of transformants during cloning of the gene. Adenylate cyclase is a membrane-bound protein that
exists in a complex with Gs. When Gs binds ATP, it
118. Warner, J. B., and J. S. Lolkema. 2003. CcpA-
becomes activated and activates adneylate cyclase.
dependent carbon catabolite repression in bacteria.
But, Gs is a GTPase, and ordinarily the GTP is hydro-
Microbiol. Mol. Biol. Rev. 67:475–490.
lyzed so that Gs has bound GDP and is not active.
119. Ye, J.-J., and M. H. Saier Jr. 1995. Cooperative Hormones such as glucagon or epinephrine bind to a
binding of lactose and HPr(Ser-P) to the lactose:H+ hormone receptor, which then binds to the Gs protein
permease of Lactobrevis. Proc. Natl. Acad. Sci. USA and triggers the exchange of GTP for bound GDP on
92:417–421. the Gs protein. This activates the Gs protein (hence
www.ebook777.com
adenylate cyclase) until it hydrolyzes the bound GTP. constructed, see: Manoil, C., and J. Beckwith. 1985.
ADP-ribosylation prevents GTP hydrolysis; hence Gs TnphoA: a transponson probe for protein export
continually activates adenyl cyclase in the presence of signals. Proc. Natl. Acad. Sci. USA 82:8129–8133.
the A1 subunit of cholera toxin. To screen for virulence genes that encode exported
virulence proteins, Peterson and Mekalanos intro-
129. Miller, V. L., and J. J. Mekalanos. 1988. A novel
duced transposon TnphoA on a plasmid (pRT291)
suicide vector and its use in construction of insertion
into V. cholerae. The plasmid was introduced into
mutations: osmoregulation of outer membrane pro-
a streptomycin-resistant strain of V. cholerae via
teins and virulence determinants in Vibrio cholerae
conjugation, using a donor E. coli strain carrying the
requires toxR. J. Bacteriol. 170:2575–2583.
plasmid containing TnphoA. Transconjugates were
130. There are two major disease-causing strains of selected on the basis of resistance to streptomycin
V. cholerae: the classical strain (V. cholerae O1) and and kanamycin. The antibiotic-resistant transconju-
the EI Tor strain. Higher expression of the ToxR reg- gates were then mated with E. coli carrying a plas-
ulon in the classical strain occurs at pH 6.5 and 30 °C mid (pPH1JI) that was not compatible with pRT291.
and lower expression at pH 8.5 and 37 °C. The con- The plasmid pPH1JI carries a gentamicin-resistance
ditions that favor optimal expression in the classical gene. V. cholerae colonies that were resistant to gen-
strain are not the same conditions that favor optimal tamicin, kanamycin, and streptomycin, and also
expression in the EI Tor strain. This is reviewed in produced alkaline phosphatase, were selected on XP
Skorupski, K., and R. K. Taylor. 1997. Control of the agar containing gentamicin, kanamycin, and strepto-
ToxR virulence regulon in Vibrio cholerae by envi- mycin. These carried pPH1JI (conferring resistance
ronmental stimuli. Mol. Microbiol. 25:1003–1009. to gentamicin) and TnphoA (conferring resistance
to kanamycin). Because pRT291 and pPH1JI are
131. To understand how to find and monitor expres-
incompatible, the kanamycin-resistant colonies
sion of virulence genes by means of TnphoA fusions,
carried random insertions of the transposon in the
it is necessary to know something about transposons.
chromosome. In other words, TnphoA transposed
Transposons are DNA elements that move from one
(moved) from the plasmid to the bacterial chromo-
part of a DNA molecule to another, or from one
some. The PhoA-positive cells (detected with the
DNA molecule to another. Most transposons move
XP) were grown in the absence of gentamicin to pro-
almost randomly. The movement is called transposi-
mote the loss of pPH1JI. To find the genes carrying
tion, and it is catalyzed by enzymes called transpos-
TnphoA insertions, which are regulated by ToxR,
ases, which are encoded in the transposon. Bacteria
the production of PhoA in the different isolates was
have transposons of many different kinds, some of
monitored under conditions known to produce
which carry antibiotic resistance genes in addition
high or low levels of cholera toxin and TcpA. Proof
to the transposase genes. The presence of the genes
that these genes were indeed positively regulated
for antibiotic resistance can be used to select for
by ToxR was obtained by testing the production of
cells carrying the transposon. TnphoA is a deriva-
PhoA in toxR– strains. See: Peterson, K. M., and J.
tive of the transposon Tn5 carrying a kanamycin-
J. Mekalanos. 1988. Characterization of the Vibrio
resistance gene that can be used to isolate random
cholerae ToxR regulon: identification of novel genes
gene fusions. The transposon contains an inserted
involved in intestinal colonization. Infect. Immun.
reporter gene, phoA, which encodes the C-terminal
56:2822–2829.
end of periplasmic alkaline phosphatase but does
not encode the promoter, the translational start site, 132. DiRita, V. J., C. Parsot, G. Jander, and J. J.
or the signal sequence (for export of the protein) Mekalanos. 1991. Regulatory cascade controls viru-
of the alkaline phosphatase gene. When TnphoA lence in Vibrio cholerae. Proc. Natl. Acad. Sci. USA
inserts into a gene in the bacterial chromosome in 88:5403–5407.
the right reading frame, a fusion protein will be syn-
133. Groisman, E. A. 2001. The pleiotropic two-
thesized. The fusion protein has the C-terminal end
component regulatory system PhoP–PhoQ. J.
of alkaline phosphatase and the N-terminal end of
Bacteriol. 183:1835–1842.
the protein encoded by the gene into which TnphoA
has inserted. The fusion protein will have alkaline 134. Häse, C. C., and J. J. Mekalanos. 1998. TcpP
phosphatase activity only if TnphoA inserts into a protein is a positive regulator of virulence gene
gene that encodes a protein to be exported, so that expression in Vibrio cholerae. Proc. Natl. Acad. Sci.
the N-terminal end of the fusion protein has a signal USA 95:730–734.
sequence for export. This ensures that the C-terminal
135. Groisman, E. A., and F. Heffron. 1995.
end of the fusion protein enters the periplasm, where
Regulation of Salmonella virulence by two-com-
it can exhibit alkaline phosphatase activity. It is pos-
ponent regulatory systems, pp. 319–332. In: Two-
sible to screen the colonies for active alkaline phos-
Component Signal Transduction. J. A. Hoch and T.
phatase. The agar includes an indicator, XP, which
J. Silhavy (Eds.). ASM Press, Washington, DC.
is 5-bromo-4-chloro-3-indolylphosphate. When the
phosphate is removed, the XP turns blue. If active 136. An acid phosphatase, the product of the phoN
PhoA is present, it cleaves XP, producing the char- gene, is induced under certain growth conditions
acteristic blue color. Genes detected in this way can such as phosphate limitation. A mutant hunt revealed
be mapped and cloned. To learn how TnphoA is phoP, which when expressed constitutively results in
high levels of phosphatase synthesis. The reason for 145. Hale, T. L. 1991. Genetic basis of virulence in
this is that PhoP is a positive regulator of the phoN Shigella species. Microbiol. Rev. 55:206–224.
gene. It was later discovered that Tn10-generated
146. Shigella spp. as well as Listeria monocytogenes
mutants that were very sensitive to killing by mac-
invade eukaryotic cells and make a cell surface pro-
rophages mapped to the phoP gene identified earlier
tein that elicits the formation of a tail of actin at one
as being required for alkaline phosphatase synthesis.
cell pole when the bacteria are growing intracellu-
See the review by Groisman and Heffron in ref. 135.
larly. As a consequence, the bacteria are propelled
137. Véscovi, E. G., F. C. Soncini, and E. A. through the cytoplasm of the host cell. The actin
Groisman. 1996. Mg2+ as an extracellular signal: tail also aids in intercellular spread of the bacteria.
environmental regulation of Salmonella virulence. See: Bernardini, M. L., J. Mounier, H. D’Hauteville,
Cell 84:166–174. M. Coquis-Rondon, and P. J. Sansonetti. 1989.
Identification of icsA, a plasmid locus of Shigella
138. Soncini, F. C., E. G. Véscovi, F. Solomon, and
flexneri that governs bacterial intra- and intercellular
E. A. Groisman. Molecular basis of the magnesium
spread through interaction with F-actin. Proc. Natl.
deprivation response in Salmonella typhimurium:
Acad. Sci. USA 86:3867–3871.
identification of PhoP-regulated genes. J. Bacteriol.
178:5092–5099. 147. Porter, M. E., and C. J. Dorman. 1994. A role
for H-NS in the thermo-osmotic regulation of viru-
139. Groisman, E. A. 2001. The pleiotropic
lence gene expression in Shigella flexneri. J. Bacteriol.
two-component regulatory system. J. Bacteriol.
176:4187–4191.
183:1835–1842.
148. Porter, M. E., and C. J. Dorman. 1997. Positive
140. Behlau, I., and S. I. Miller. 1993. A Pho-P
regulation of Shigella flexneri virulence genes by inte-
repressed gene promotes Salmonella typhimurium
gration host factor. J. Bacteriol. 179:6537–6550.
invasion of epithelial cells. J. Bacteriol. 175:4475–4484.
149. Otto, M., R. Süssmuth, G. Jung, and F. Götz.
141. Moskowitz, S. M., R. K. Ernst, and S. I. Miller.
1998. Structure of the pheromone peptide of the
2004. PmrAB, a two-component regulatory system of
Staphylococcus epidermidis agr system. FEBS Lett.
Pseudomonas aeruginosa that modulates resistance to
424:89–94.
cationic antimicrobial peptides and addition of amin-
oarabinose to lipid A. J. Bacteriol. 186:575–579. 150. Zhu, J., P. M. Oger, B. Schrammeijer, P. J. J.
Hooykaas, S. K. Farrand, and S. C. Winans. 2000.
142. Cationic antimicrobial peptides (CAPs) are
The bases of crown gall tumorigenesis. J. Bacteriol.
detergent-like peptides that kill gram-positive and
182:3885–3895.
gram-negative bacteria by collapsing the membrane
potential of the cytoplasmic membrane. In gram- 151. TraR is a LuxR-type transcription factor that
negative bacteria, CAPs bind to the lipopolysaccha- binds to the A. tumefaciens acyl–HSL and stimulates
ride and traverse the periplasm. Pathogens can be the transcription of genes required for plasmid trans-
exposed to CAPs at epithelial cell surfaces, such as fer (tra genes), the gene whose product synthesizes the
in the lungs. Similar peptides, called defensins, are acyl–HSL (tral), and genes responsible for plasmid
present in human neutrophils and macrophages and replication (rep genes). The increased replication of
are part of the body’s defense against infection. The the plasmid results in increased tumorigenesis. The
antibiotic polymyxin, produced by Polymyxa, is an transcription of traR is stimulated by plant opines.
acylated cyclic CAP. As noted in the text, polymyxin- The opines bind a receptor protein (either AccR or
resistant mutants of Pseudomonas aeruginosa have OccR, depending upon the strain) in the bacterium,
been mapped to the PmrA/PmrB two-component sig- and the complex initiates the transcription of traR.
nal transduction system. This ensures that conjugal transfer of the Ti plasmid
occurs only when the opines are present. (AccR is a
143. Kox, L., F. F. Wösten, and E. A. Groisman.
repressor. The AccR/opine combination derepresses
2000. A small protein that mediates the activation of
traR, and the OccR/opine combination induces
a two-component system by another two-component
traR.) Additional regulation involves the TraM
system. EMBO J. 19:1861–1872.
protein. TraM binds to TraR and inhibits its activ-
144. Miller, J. F., S. A. Johnson, W. J. Black, D. ity. For more information about TraM, see: Chen,
T. Beattie, J. J. Mekalanos, and S. Falkow. 1992. G., J. W. Malenkos, M.-R. Cha, C. Fuqua, and L.
Constitutive sensory transduction mutations in Chen. 2004. Quorum-sensing antiactivator TraM
the Bordetella pertussis bvgS gene. J. Bacteriol. forms a dimer that dissociates to inhibit TraR. Mol.
174:970–979. Microbiol. 52:1641–1651.
www.ebook777.com
20
CHEMOTAXIS, PHOTORESPONSES,
AEROTAXIS
The question asked in this chapter is, “Why do term “receptor–transducer protein” is used.
bacteria move in the direction that they move?” Adaptation plays an important role in chemo-
The answer is complex. Some bacterial move- taxis, and this will also be discussed.
ments are influenced by the chemical envi- Most of what is known about chemotaxis is
ronment, a process called chemotaxis. Some derived from the original studies with the enteric
bacteria move in response to the light in their bacteria Escherichia and Salmonella, and most
environment, a process called photoresponses. of what follows is derived from research with
And some bacteria move in response to cer- these organisms. Systems that differ from the
tain terminal electron acceptors in the electron enteric bacteria are described in Section 20.8.
transport chain, such as oxygen (aerotaxis) and
nitrate (nitrate taxis). This chapter summarizes
some of these behavioral responses. We will 20.1 Bacteria Measure Changes in
begin with a definition of chemotaxis. Concentration over Time
Chemotaxis refers to the ability of bacteria One might think that bacteria swimming along
to move along a concentration gradient toward a chemical concentration gradient are detecting
a chemical attractant (positive chemotaxis) the spatial gradient itself, but this is not the case.
or away from a chemical repellent (negative Calculations indicate that because of the small
chemotaxis). For reviews see refs. 1 through 5. size of bacteria, the difference in concentration
The chemical attractants and repellents are of a chemoeffector between the ends of the cell
called chemoeffectors. As we shall see, the would be too small to be measured accurately.6
chemotaxis signaling system is a two-compo- What the bacterium actually measures is the
nent system that includes a histidine kinase absolute concentration of chemoeffector, which
(CheA) and a response regulator (CheY) that it then compares with the concentration previ-
affects the switch mechanism in the flagellar ously measured. In other words, bacteria mea-
motor. Histidine kinases and response regula- sure changes in concentration over time. They
tors are discussed in Chapter 19. Reference to the actually “remember” the previous concentra-
description of bacterial flagella in Section 1.2.1 tion. If a bacterium finds that it is in a higher
while reading Section 20.1 may be useful. concentration of an attractant or a lower con-
The histidine kinase that responds to chemot- centration of repellent than at a previous time,
axis is cytoplasmic. Importantly. the signal is it will continue to move in that direction. If the
first sensed by a transmembrane sensor pro- bacterium finds that it is in a lower concentra-
tein. The transmembrane sensor protein is tion of attractant or a higher concentration of
usually referred to as the methyl-accepting repellent than at a previous time, it will move in
chemotaxis protein (MCP), but sometimes the a randomly different direction.
534
chemotaxis, photoresponses, aerotaxis 535
The propensity to swim randomly is fun- place for a fraction of a second. This is called
damental to chemotaxis. Some bacteria swim tumbling. Then the flagella move CCW again
randomly because they periodically tumble, as to re-form the bundle and the cell “runs” again,
discussed next, whereas other bacteria swim but in a randomly different direction.
randomly without tumbling. Chemotaxis As mentioned, repellents cause the flagellar
in bacteria that tumble is understood best. motor to reverse and the cells to tumble, whereas
Random swimming without tumbling is dis- attractants decrease the frequency of reversal
cussed in Section 20.8. and promote smooth swimming. The tumbling
and smooth swimming responses to changes
in chemoeffector concentrations can be seen
20.2 Tumbling by observing the swimming of individual cells
E. coli and many other bacteria swim along a under a microscope. It is also possible to moni-
path for a few seconds or less and then tumble. tor the change in direction of flagellar rotation
Very importantly, when they recover from the by tethering the bacteria to a glass slide by a fla-
tumble, they swim in a randomly different direc- gellum. One way to do this is to use an antibody
tion. If nothing affected the frequency of tum- to the flagellum that sticks to both the flagellum
bles and swimming, bacteria would just swim and the glass slide.8 The cells are grown in mini-
in random directions. This is the situation when mal media with glucose as the carbon source.
they are not responding to chemoeffectors. Under these conditions, the average number of
However, attractants decrease the frequency flagella per cell is reduced and cells tethered by a
of tumbles, whereas repellents increase the fre- single flagellum can be observed. When the fla-
quency of tumbles.6,7 This is illustrated in Fig. gellar motor rotates, the cell rotates because the
20.1. If a bacterium moves into an area in which flagellum is fixed to the slide. It is then possible
the concentration of chemoattractant is higher to infer that the addition of attractants causes
than at a previous moment, the cell detects the the flagella to rotate CCW and the addition of
higher concentration (because of increased bind- repellents causes CW rotation.9
ing to chemoreceptors), tumbles less frequently,
and continues to move in the same direction.
Likewise, if the cell swims into an area in which
the concentration of repellent is higher, tum-
bling increases and the cell changes its direction,
moving away from the repellent. But why does
the cell tumble? This is explained next.
E. coli and S. typhimurium are peritrichously
flagellated bacteria in which helical flagella (an
average of four in E. coli) protruding randomly
from the cell sides are wrapped around one
another in a helical bundle that extends several
cell lengths at the rear of the swimming cell (Fig.
20.2). The flagella are usually in the conforma-
tion of a left-handed helix. When the flagella
rotate counterclockwise (CCW), as viewed Fig. 20.1 Chemoeffectors bias random swimming.
from the tip of the flagellum toward the cell, (A) The swimming pattern is smooth swimming
they remain in a bundle as a helical wave moves for a short period interspersed with brief periods of
from the proximal to the distal portion of the tumbling. After tumbling, the bacterium resumes
flagellum, pushing the bacterium forward. The swimming in a randomly different direction. (B) A
chemoattractant decreases the frequency of tum-
forward movement is called a run. Such “runs”
bling, thus prolonging the periods of smooth swim-
by E. coli last about one second. However, when ming. The result is that the bacterium swims in the
one or more flagella rotate clockwise (CW), they direction of higher concentrations of chemoattrac-
unwind and leave the filament bundle. The result tant. A chemorepellent increases the frequency of
is that there is no longer any coordinated flagel- tumbling, causing the bacterium to swim away from
lar movement, and the cell moves erratically in the repellent.
www.ebook777.com
Fig. 20.2 Flagella fly apart during tumbling. In E. coli the flagella are peritrichously located and form a trail-
ing bundle while the cell is swimming. (A) The filaments are in a left-handed helix. When the motors rotate
counterclockwise, a helical wave travels proximal to distal (outward from the cell), pushing the cell forward.
(B) When the motors rotate clockwise, the filaments undergo a transition to a right-handed waveform, caus-
ing them to fly apart and the cells to tumble. Source: Adapted from MacNab, R. M. 1987. Motility and
chemotaxis, pp. 732–759. In: Escherichia coli and Salmonella typhimurium. F. C. Neidhardt et al. (Eds.).
20.3 Adaptation given repellent, hence would be trapped in the

An important aspect of the chemotactic area of that repellent.
response is adaptation. Bacteria adapt (i.e.,
become desensitized) to a chemoeffector, and 20.4 Proteins Required for
thus after a short period of time (seconds to Chemotaxis
minutes), they no longer respond to it at the
Chemotaxis employs complex regulatory cir-
initial concentration but may respond to a
cuits involving both cytoplasmic and membrane
higher concentration. In fact, one could say
proteins. In E. coli, these proteins (MCPs, CheA,
that adaptation is the way the bacterium
CheW, CheR, CheZ) are all located as a cluster
“remembers” the previous concentration of
at the cell poles.10–12 A description of the proteins
chemoeffector.
and their functions will be given first, followed
Why is an adaptation circuit built into the sig-
by an explanation of the regulatory circuits. As
naling pathway? Adaptation makes good sense.
we shall see, homologous proteins are also part
When a bacterium adapts to a chemoattractant,
of the signaling transduction pathways that
it resumes tumbling at the nonstimulated fre-
operate during phototaxis and aerotaxis.
quency and will stay in the area; but if it encoun-
ters an increasing concentration gradient of
attractant, tumbling will again be suppressed. 20.4.1 Cytoplasmic proteins involved
Consequently, the bacterium will swim toward in chemotaxis
and remain in the area of the higher concentra- Mutants of E. coli and Salmonella typhimurium
tion of chemoattractant. Another way of saying have been isolated that fail to show chemotaxis.
this is that adaptation allows the bacterium to These mutants have resulted in the identifica-
respond to any local gradients it may encoun- tion of six genes required for chemotaxis. The
ter while randomly swimming. Adaptation to genes are cheA, cheB, cheR, cheW, cheY, and
repellents is also easily rationalized. If a bacte- cheZ. Deletion of any one of these genes pre-
rium did not adapt to repellents, it would con- vents chemotaxis without affecting motility.
tinue to tumble at the stimulated rate even when The Che proteins are cytoplasmic proteins
swimming toward the lower concentration of a that are part of a signal transduction pathway
between the attractant or repellent and the fla- first bind to periplasmic binding proteins, also
gellar motor switch.13 (See note 14 for a further called primary chemotaxis receptors, and the
description of the flagellar switch proteins and sugar-bound periplasmic binding proteins
the Mot proteins.) CheA belongs to the class interact with specific receptor–transducers in
of signaling proteins called histidine kinases. the membrane. Under these circumstances,
It has conserved amino acid sequences at the the receptor–transducer (MCP) acts as a sec-
carboxy-terminal domain similar to the other ondary chemoreceptor rather than a primary
histidine kinases, is phosphorylated at a his- chemoreceptor. These are the same periplasmic
tidine residue, and transfers the phosphoryl binding proteins that bind sugars and inter-
group to a response regulator protein.15 (See act with membrane-bound transporters in the
Chapter 19, Fig. 19.1.) CheY and CheB are shock-sensitive permease systems described in
response regulator proteins. They have amino- Section 17.3.3. They therefore function in both
terminal domains similar to the other response chemotaxis and solute transport. However, the
regulators and undergo phosphorylation and majority of periplasmic binding proteins in E.
dephosphorylation. coli function only in transport, not in chemot-
axis. (See note 16 for a more complete discus-
sion of the MCP proteins.)
20.4.2 Membrane proteins involved
in chemotaxis
Chemoreceptor proteins for the chemoeffec- 20.5 A Model for Chemotaxis
tors are built into the bacterial cell membrane. 20.5.1 Stimulation
The chemoreceptor proteins are also called The cytoplasmic domains of the receptor–
receptor–transducer proteins, or simply trans- transducer proteins (Tar, Tsr, Trg, Tap) control
ducer proteins. The chemoreceptor proteins are the autophosphorylation activity of CheA in an
thought to loop across the membrane with a activity that requires CheW (Fig. 20.3):
periplasmic domain separating two hydropho-
CheA + ATP → CheA-P + ADP + Pi
bic membrane-spanning regions that connect
to cytoplasmic domains. This is similar to the CheA-P is a protein kinase that phosphorylates
configuration illustrated in Chapter 18 (Fig. CheY:
18.3A, B) for a protein anchored to the mem-
CheA-P + CheY → CheA + CheY-P
brane by a stop-transfer signal and a leader pep-
tide, which for the case of the chemoreceptor CheY-P has autophosphatase activity. CheZ
proteins would be uncleaved. The cytoplasmic binds to CheY-P, and as a consequence phos-
domain interacts with cytoplasmic components phatase activity is increased. It has been sug-
such as the CheW and CheA proteins, whereas gested that CheZ is an allosteric effector that
the periplasmic domain interacts with chemo- enhances CheY-P autodephosphorlation17:
effectors or periplasmic proteins to which the CheZ
chemoeffectors are bound. The cytoplasmic
CheY-P + H2O › CheY + Pi
domain of the chemoreceptor proteins also con- However, recent evidence indicates that CheZ
tains four or five glutamate residues that become itself can catalyze the dephosphorylation of
methylated and demethylated as part of the CheY-P.18,19 Thus, phosphatase activity is due
adaptation response described in Section 20.3. to both CheY-P autodephosphorylation and
Therefore, the chemoreceptor proteins are also CheZ-catalyzed dephosphorylation. In the non-
called methyl-accepting chemotaxis proteins stimulated state there is a steady state level of
(MCPs). CheY-P that is determined by the rate of phos-
E. coli has four different MCPs. These are phorylation of CheY and the rate of dephos-
the Tsr protein (taxis to serine and away from phorylation of CheY-P.
repellents), Tar (taxis to aspartate and malt- CheY-P is a response regulator protein. It
ose and away from repellents), Trg (taxis to binds to the flagellar motor switch, FliM, FliG,
ribose, glucose, and galactose), and Tap (taxis FliN (primarily FliM), and initiates a reversal
to dipeptides). Salmonella does not have Tap, of the motor (Mot) so that it rotates in a clock-
but it does have Tcp, which senses citrate. wise direction, causing the cell to tumble.20
Some chemoeffectors, primarily certain sugars, In the nonstimulated cell, there is a certain
www.ebook777.com
intermediate level of CheY-P that supports nor-

mal run–tumble behavior.
When an attractant or repellent stimulates
a chemoreceptor protein, a signal is sent to the
cytoplasmic region of the receptor–transducer
protein. The signal is presumably a conforma-
tional change in the periplasmic region of the
chemoreceptor protein that is propagated to
the cytoplasmic region of the chemoreceptor
protein, which interacts with CheW/CheA.
An attractant decreases the rate of autophos-
phorylation of CheA, and a repellent increases
the rate of autophosphorylation. Thus, attrac-
tants promote counterclockwise rotation and
smooth swimming, whereas repellents promote
clockwise rotation and tumbling. (Chemotaxis
in the gram-positive subtilis differs from that in
the enterics in several ways. For example, the
attractant causes an increase in CheY phos-
phorylation, and CheY-P causes the flagel-
lar motor to rotate CCW resulting in smooth
swimming.21)
Fig. 20.3 Model for signal transduction dur-
ing chemotaxis in E. coli and S. typhimurium. 20.5.2 Signal gain
Attractants bind to the receptor–transducer pro- The interesting subject of signal gain is reviewed
teins (Tar, Tsr, etc.) in the membrane. The recep- in ref. 22. Chemotactic signals are amplified.
tors also mediate the response to certain repellents This means that a very small change in recep-
(leucine, indole, acetate, Co2+, Ni2+). This activates tor occupancy to a particular chemotaxis ligand
a CheW-dependent change in the rate of autophos-
can result in a large change in the CheA kinase
phorylation of CheA and the proteins that are phos-
activity, and therefore a relatively large change
phorylated by CheA (CheB and CheY). Attractants
reduce the rate of autophosphorylation, and repel- in motor bias and the probability of tumbling.
lents increase the rate of autophosphorylation. The Signal gain is defined as the “ratio of the frac-
phosphoryl group is transferred from CheA-P to tional change in motor bias to the fractional
CheY and CheB. CheY-P interacts with the switch change in receptor occupancy.”23 Somehow, a
proteins (FliM, FliN, FliG) to cause clockwise rota- small change in receptor occupancy results in a
tion of the motor and tumbling. Thus repellents change in the activity of many CheA molecules.
increase tumbling because they increase the level of How might this occur?
CheY-P, and attractants reduce tumbling because One clue, previously noted, is that the five
they reduce the level of CheY-P. CheB-P is a meth- different chemotaxis receptors in E. coli (Tsr,
ylesterase whose activity results in demethylation of
Tar, Tap, Trg, and Aer) are clustered together
the receptor–transducer proteins. As CheB-P goes
at the cell poles. A model proposed on the basis
up, methylation goes down. Thus, repellents reduce
the level of methylation because they increase the of the physical interactions between the differ-
level of CheB-P and attractants increase methylation ent receptor molecules in a cluster suggests that
because they reduce the levels of CheB-P. CheB-P
autodephosphorylates. CheR is a methyltransferase removes the phosphate from CheY-P. The rela-
that methylates the receptor–transducer proteins. tive activities of CheA-P and CheZ determine the
The more highly methylated receptor–transducer level of CheY-P, hence the frequency of tumbling.
protein does not transmit the chemoattractant sig- Theoretically, changes in the activity of CheZ could also
nal to CheA. Hence adaptation to a chemoattractant alter the frequency of tumbling. Source: Stock, J. B., A.
is due to increased methylation of the receptor– J. Ninfa, and A. M. Stock. 1989. Protein phosphoryla-
transducer proteins. Adaptation to a repellent is tion and regulation of adaptive responses in bacteria.
due to undermethylation of the receptor–transducer Microbiol. Rev. 53:450–490. Reproduced with per-
protein. CheZ is thought to be a phosphatase that mission from American Society for Microbiology.
the binding of a chemoattractant to one recep- protein by the chemoattractant, the levels of
tor molecules induces conformational changes CheB-P also decrease because there is less of the
in other receptor molecules, not necessarily of phosphoryl donor, CheA-P. When autophos-
the same type, producing a large signal gain. In phorylation of CheA, hence phosphorylation of
other words, the receptor cluster behaves like CheB, is inhibited by binding of the attractant
an integrated signaling unit affecting the activ- to the chemoreceptor protein, the removal of
ity of CheA. methyl groups is slowed relative to the addi-
tion of methyl groups, and all the chemorecep-
20.5.3 Acetyl phosphate can tor proteins become more highly methylated.
phosphorylate CheY (There are four or five glutamate residues that
It has been suggested that acetyl phosphate can be methylated.)
can donate its phosphoryl group in vivo to The more highly methylated chemoreceptor
CheY in the absence of the cognate histidine protein stimulates tumbling, hence adapta-
kinase (CheA). The initial observation was tion to that concentration of attractant. There
that mutants of E. coli that have none of the is ample evidence, both in vivo and in vitro,
cytoplasmic chemotaxis proteins other than that the methylated forms of the chemorecep-
CheY swim smoothly except when incubated tor proteins stimulate the activity of CheA,
with acetate. Then they tumble. The response hence the phosphoryation of CheY, clock-
to acetate requires acetate kinase, the enzyme wise rotation, and tumbling.26,27 See note 28
that synthesizes acetyl phosphate from acetate for experimental evidence. (Another way of
and ATP. This suggests that acetyl phosphate saying this is that methylation of the chemore-
can phosphorylate CheY, and indeed this has ceptor protein cancels the attractant-induced
been demonstrated in vitro (as discussed in signal.) As a result of the methylation, the
ref. 24). Acetate has no effect on chemotaxis to prestimulus levels of CheY-P and CheB-P are
aspartate or serine in wild-type cells, indicating restored, and the cell resumes the prestimulus
that the levels of CheY-P are not significantly run–tumble behavior, despite the presence of
influenced by acetyl phosphate in wild-type bound attractant. As long as the attractant is
cells and are regulated entirely by CheA and bound to the receptor–transducer protein, it
CheZ in response to the appropriate chemoef- remains highly methylated, and the cell is said
fectors.24 Thus, the physiological function of to be adapted to that particular concentration
acetate regarding CheY phosphorylation and of attractant.
chemotaxis in general is not clear at all. Perhaps However, the other (nonoccupied) chemore-
it plays a subtle regulatory role in chemotaxis, ceptor proteins quickly lose their extra methyl
somehow connecting chemotaxis to metabo- groups so that the cell is adapted only to the
lism. It was reported in 1998 that acetyl–AMP specific attractant. If the chemoreceptor pro-
can acetylate CheY with results on flagellar tein is not fully saturated with bound attrac-
rotation similar to those of the phosphorylation tant, the cells will respond to the attractant
of CheY, and therefore it appears that there are when it swims into areas of higher concentra-
two pathways for stimulating CheY by acetate tion of attractant. However, as the chemore-
in E. coli.25 ceptor protein binds more chemoattractant,
the average number of methyl groups per
20.5.4 Model for adaptation transducer is increased two- to fourfold, thus
CheA-P also phosphorylates a second response balancing the increased binding of attractant
regulator protein, CheB: and resulting in adaptation to the higher con-
centration of chemoattractant. (See note 29 for
CheA-P + CheB → CheA + CheB-P an explanation of how this might occur.) When
CheB-P is the active form of a methylesterase the chemoattractant is removed, the overmeth-
that removes methyl groups from glutamate ylated chemoreceptor protein stimulates CheA
residues in the receptor–transducer proteins. autophosphorylation. This results in increased
There is also a methyltransferase, CheR, that CheY-P and consequent tumbling.30 However,
adds methyl groups to these residues. As a the methylesterase CheB-P also increases
result of the stimulation of the chemoreceptor and returns the chemoreceptor protein to the
www.ebook777.com
methylation level of the prestimulus state and chemoreceptors stimulate CheY phosphoryla-
normal run–tumble behavior. tion and tumbling.31–33
Adaptation to a repellent differs from adap-
tation to an attractant in that during repellent 20.5.5 Adaptation mechanism that does
adaptation the activated CheB-P demethylates not rely on methylation
the chemoreceptor protein that responds to the If cells are experimentally exposed to a large
repellent. In summary, then, binding of the and abrupt change in the concentration of a
attractant to the chemoreceptor inhibits the stimulus (e.g., the addition of an attractant),
kinase activity of CheA, which leads to a fall then adaptation, which can take minutes, is
in the levels of CheY-P, hence to smooth swim- certainly correlated with an increased level of
ming. However, binding of the attractant to receptor methylation, as described earlier. On
the chemoreceptor also promotes the methy- the other hand, it has been pointed out that the
lation of the chemoreceptor. The methylated methylation-dependent adaptation system may
chemoreceptor stimulates the kinase activity be too slow for adaptation by cells swimming
of CheA, which leads to a rise in the levels of in gradients, and in fact a methylation-indepen-
CheY-P and tumbling; thus adaptation takes dent system may exist.34 During chemotaxis
place. along an attractant concentration gradient, the
cells encounter very small changes in stimulus
Some evidence in favor of the model concentrations (as opposed to the experiments
The phenotypes of chemotaxis mutants support in which large concentrations of attractant are
the model. This is summarized in Table 20.1. added to a cell suspension and the behavior of
For example, cheW, cheA, and cheY mutants the cells is monitored), and adaptation takes
do not tumble, whereas cheB and cheZ mutants only about one second. It has been argued
tumble constantly. This is in agreement with the that the increase in methylation of the MCP
model, which stipulates that CheW and CheA proteins in a second is too small to account
promote the phosphorylation of CheY and that for adaptation to attractants in gradients.
CheY-P causes the motor to reverse and tum- Furthermore, cheR–cheB– double mutants that
bling to take place. CheB competes with CheY show no changes in methylation are neverthe-
for the phosphoryl group on CheA-P and would less chemotactic in swarm plates and also show
therefore be expected to promote smooth swim- adaptation to the addition of very small concen-
ming. Hence a mutation in cheB should make trations of attractant when CWW rotation of
the cells tumble. CheZ dephosphorylates CheY, tethered cells is measured.35–38 (See note 39 for
and therefore a mutation in cheZ should result a description of chemotaxis assays and how to
in high levels of CheY-P and therefore increased measure responses to attractants and repellents
tumbling. using tethered cells.)
Biochemical data also support the model. Apparently there exists a methylation-in-
When an attractant binds to the extracel- dependent system that allows cells to respond
lular domain of the chemoreceptor protein, chemotactically to small concentrations of
several events occur: the rate of phosphoryla- stimulus and to be chemotactic in gradients.
tion of CheY decreases; the rate of demethyla- Furthermore, since chemotaxis to sugars trans-
tion is reduced; and the rate of methylation is ported by the PTS does not involve MCP pro-
increased. There is also a great deal of evidence teins, adaptation cannot be via methylation of
indicating that in their methylated form, the these proteins. (See Section 20.7.)
Table 20.1 Effects of chemotaxis mutants
Gene Mutant phenotype Rationale for mutant phenotype
cheY– Smooth swimming CheY-P is required for tumbling

cheA–, cheW– Smooth swimming CheA-P and CheW phosphorylate CheY
cheB– Tumbling CheB competes with CheY for phosphoryl group from CheA-P
cheZ– Tumbling CheZ dephosphorylates
20.6 Mechanism of Repellent Action activities of the receptor–transducer proteins

Repellents for E. coli and S. typhimurium Tsr and Tap by lowering the intracellular pH.
include indole, glycerol, ethylene glycol, phenol
(for S. typhimurium only), organic acids (e.g., 20.7 Chemotaxis That Does Not
formic, acetic, benzoic, salicylic), alcohols, Co2+ Use MCPs: The Phosphotransferase
and Ni2+ (for E. coli only), and hydrophobic System Is Involved in Chemotaxis
amino acids (leucine, isoleucine, tryptophan, toward PTS Sugars
valine). In general, the mechanism of action
of organic repellents is not well understood, In addition to functioning in sugar uptake,
except that the response requires the receptor– the PTS (see Section 17.3.4) is important for
transducer (MCP) proteins. Yet, there is no chemotaxis toward PTS sugars.41 Some of the
direct evidence for specific binding between an evidence for this is as follows:
organic repellent and a specific MCP. In some 1. Mutants of E. coli defective in EI and HPr are
cases (e.g., glycerol, ethylene glycol, aliphatic usually defective in chemotaxis toward PTS
alcohols) there is not even specificity regarding sugars.
which receptor–transducer protein is required. 2. EII mutants defective in chemotaxis toward
This has been learned by using mutants lacking the sugar specifically recognized by EII also
one or more of the receptor–transducer pro- exist.
teins and showing that any of the remaining
receptor–transducer proteins can mediate the Comparison of PTS-mediated and MCP-
repellent response. dependent chemotaxis
However, in other instances there is speci- E. coli mutants lacking CheB, CheR, or the MCP
ficity. For example, the response to leucine, proteins display normal chemotaxis to PTS sug-
isoleucine, and valine requires the Tsr protein, ars. This means that methylation-dependent
and the response to the cations Co2+ and Ni2+ adaptation does not occur in the system, and
requires Tar. Because much higher concentra- there must be another adaptation mechanism.
tions of organic repellents (in the millimolar However, CheA, ChW, and CheY are required
range) are usually required than for attrac- for chemotaxis toward PTS sugars.42 One model
tants (in the micromolar range), it has been postulates that EI–P can cause the phosphoryla-
suggested that the organic repellents (which tion of CheA as well as the incoming PTS sugar.
all have some hydrophobic character as well Thus, as the sugar becomes phosphorylated
as polar groups) do not actually bind to the during its uptake, the levels of CheA-P fall,
receptor–transducer proteins but act indirectly causing lower levels of CheY-P, hence smooth
by perturbing the membrane at the site of the swimming. (Reviewed in ref. 43.)
receptor–transducer proteins. For example,
they might make the membrane more fluid,
which could result in the alteration of recep-
20.8 Chemotaxis That Is Not Identical
tor–transducer activity. with the Model Proposed for the
In 1990 a study was carried out to investigate Enteric Bacteria
the action of repellents on the membrane fluid- Chemotaxis in many bacteria differs in sev-
ity of E. coli.40 It was concluded that changes in eral respects from chemotaxis in E. coli. For a
membrane fluidity do not account for repellent review of this subject, see refs. 44 and 45. For
activity, suggesting that repellents may indeed example, some bacteria do not reverse the rota-
bind to receptor–transducer proteins, however, tion of their flagella and do not tumble. Other
with low affinity, and in some cases low specific- bacteria do reverse the direction of rotation of
ity. In summary, the response to repellents does their flagella, but this causes them to back up,
involve the receptor–transducer proteins (MCP not tumble. Additionally, some bacteria must
proteins), but the mechanism of how they affect metabolize the chemoattractant to generate
the activities of the receptor–transducer pro- the chemoattractant signal; that is, the initial
teins is not understood. However, there is more signal does not come from the binding of the
known about the repellent activity of weak chemoattractant to an MCP-like protein. How
organic acids. These are believed to affect the such bacteria adapt is not known. Of course,
www.ebook777.com
as discussed earlier, sometimes even E. coli be necessary for chemotaxis, because if metab-
does not use the MCP proteins (e.g., when it olism is blocked (e.g., by a specific inhibitor),
responds to PTS sugars). Several bacteria have chemotaxis is also inhibited.
multiple che genes, especially cheY. The follow- R. sphaeroides has homologues to all the E.
ing is a description of chemotaxis in bacteria coli che genes except cheZ. 48,49 (See note 50 for
other than E. coli in which the process has been a description of how the che genes were detected
studied in detail. in R. sphaeroides.) A cytoplasmic methyl-ac-
cepting chemotaxis protein called TlpA (trans-
20.8.1 Bacteria that do not tumble ducer-like protein A), with some similarity to the
Not all bacteria tumble, yet some of those enteric MCPs, exists.51 (See note 52 for a further
lacking this ability are capable of chemotaxis. description of how this was determined.) Recall
For example, Rhodobacter sphaeroides and that metabolism is required for chemotaxis in
Rhizobium (Sinorhizobium) meliloti rotate their R. sphaeroides. It has been suggested that TlpA
flagella only in a clockwise direction and do not responds to a cytoplasmic metabolic intermedi-
tumble. (See Sections 20.8.2 and 20.8.3.) Yet ate and transfers the signal to CheA, which takes
they swim randomly and exhibit chemotaxis. part in signaling the flagellar motor. In addition,
In some bacteria with polar flagella (fla- R. sphaeroides is reported to have a cluster of
gella at one pole or both poles, either single or transmembrane chemosensory proteins (MCP
in bundles) the reversal of the flagellar motor proteins) at the cell poles, and these also signal
causes the cells to reverse the direction of swim- the flagellar motor. It has been postulated that
ming rather than tumble. Examples include attractants are detected by the MCP proteins
Caulobacter and certain photosynthetic bacte- at the cell poles in a fashion similar to E. coli.
ria that change direction by reversing the direc- If either the cytoplasmic or polar chemotaxis
tion of flagellar rotation. Because they do not proteins are missing, then chemotaxis does not
back up in an absolutely straight line, and/or do occur. Thus, the cytoplasmic and membrane
not swim forward in a straight line, the direction systems interact to enable chemotaxis to occur.
of forward swimming is random with respect For a more complete description of the complex
to the original forward direction. Some bacte- interacting membrane and cytoplasmic chemot-
ria whose swimming behavior and chemotaxis axis systems in R. sphaeroides,, see ref. 46.
pathway are not exactly like those of E. coli are Since R. sphaeroides changes its direction as
described next. a result of stopping rather than tumbling, it is
reasonable to suppose that an increased con-
20.8.2 Rhodobacter sphaeroides centration of chemoattractant might decrease
For a review, see ref. 46. R. sphaeroides is a pho- the frequency of stops (the equivalent of the
tosynthetic bacterium that can live aerobically suppression of tumbling), thus ensuring that
heterotrophically in the dark or anaerobically the cells continue to swim in the direction of the
as a photoheterotroph in the light. (See subsec- chemoattractant. Likewise, a decreased con-
tion of Section 6.1.2 entitled: Purple nonsulfur centration of chemoattractant might increase
phototrophs.) It has a single medially (subpo- stopping frequency, hence promote changes in
lar) located flagellum that rotates in one direc- the direction of swimming. However, it does
tion only (clockwise), pushing the cell forward. not happen exactly this way.
The cells stop swimming intermittently, and When anaerobically grown cells tethered to
when swimming resumes, the cells swim in a glass via anti-flagellin antibody were exam-
randomly different direction. It appears that ined, no change in rotational behavior due to
while the bacterium is stopped the flagellum is an increase in attractant concentration was
coiled against the cell body and rotates slowly, observed; but when the attractant concentra-
resulting in reorientation of the cell. (Reviewed tion was decreased, the cells transiently stopped
in ref. 47.) R. sphaeroides is attracted to weak rotating. (Reviewed in ref. 53.) It appears that cell
organic acids, sugars and polyols, glutamate, accumulation in response to a chemoattractant
ammonia, and certain cations (e.g., K+ and gradient by R. sphaeroides is dependent upon
Rb+). Unlike the case of the enteric bacteria, sensing a reduction, not an increase, in the con-
metabolism of the chemoattractant appears to centration of chemoattractant. This is opposite
to the situation in E. coli, which responds to an copies), cheA, cheW, cheR, and cheB.58 (There
increase in attractant concentration. is no cheZ.) Mutations in the che genes make
Some metabolites, especially weak organic the cells defective in chemotaxis. R. meliloti
acids, cause an increase in the speed of swim- possesses a tlpA gene, which as in R. sphaeroi-
ming by R. sphaeroides, a response called des encodes a cytoplasmic MCP-like protein.
chemokinesis.54 Many compounds that cause Other MCP-like proteins have been detected, as
chemokinesis also cause chemotaxis. However, reviewed in ref. 53. As noted earlier, R. meliloti
the two phenomena are separable because (1) flagella turn in only one direction (clockwise),
most amino acids and sugars that are chemoat- and the cell does not seem to stop momentarily
tractants do not cause chemokinesis, and (2) before it turns in a new direction.
metabolism is required for chemotaxis but not R. meliloti responds to attractants by increas-
for chemokinesis. (It appears that transport ing its swimming speed (chemokinesis). This is
across the cell membrane may be involved in equivalent to decreasing the frequency of tum-
chemokinetic signaling.) The increased rate of bles in E. coli. It has been speculated that the rate
swimming in response to certain metabolites of flagellar rotation is determined by CheY-P,
lasts for a long time; that is, there is no rapid which perhaps slows the flagellar motor. Thus
adaptation. It has been pointed out that in the the presence of the attractant might inhibit the
absence of adaptation, chemokinesis should autophosphorylation of CheA, which would
result in the spreading of the population rather decrease the levels of CheY-P and increase
than its accumulation.55 swimming speed. This sequence would result
in chemotaxis toward the attractant if the cells
were also capable of adaptation.
20.8.3 Rhizobium meliloti
Rhizobium meliloti (also called Sinorhizobium 20.8.4 Azospirillum brasilense
meliloti) is an aerobic bacterium that can live
Azospirillum brasilense lacks a methylation-
freely in soil, in the rhizosphere of host plants,
dependent pathway for chemotaxis toward
or eventually symbiotically with leguminous
organic compounds.59,60 In Azospirillum brasi-
plants in root nodules, where it fixes nitrogen.
lense, it appears that the strongest chemoattrac-
The bacteria enter the plants via plant root-
tants are electron donors to the aerobic redox
lets, to which they are attracted by root exu-
chain. This, plus the fact that chemoattraction
dates that include amino acids, carbohydrates,
toward oxygen is so strong in this organism
and flavones (cyclized isoprenoid compounds
that it masks responses to other chemoattrac-
made from phenylalanine and malonyl–CoA).
tants, has led to the suggestion that the signal-
Chemotaxis studies with R. meliloti employ
ing system involves the respiratory chain. It is
L-amino acids and D-mannitol as attractants.
reasonable to suggest that the signaling system
R. meliloti has 5 to 10 peritrichously located
in these cases (and in phototaxis and aerotaxis,
flagella that form a bundle when the cell swims.
as discussed next) may involve the redox level of
Swimming consists of straight runs interrupted
one of the electron carriers or the proton motive
by very quick turns. Unlike, for example, R.
force.
sphaeroides, the cells do not seem to stop before
turning. It has been suggested that the flagella
never actually stop rotating. When the cell is 20.9 Photoresponses
swimming in one direction, the flagella rotate as As we shall see, tactic responses to light as well as
a single bundle. The cell turns when the flagellar to electron acceptors such as oxygen and nitrate
motors slow down and rotate at different rates, also exist among the prokaryotes. Consult ref.
whereupon the bundle flies apart. (Reviewed 61, which is a review article that discusses vari-
in ref. 53.) Cells tethered to a microscope slide ous photosensors, including a family of photo-
by a single flagellum (the other flagella were sensors called photoactive yellow protein (PYP),
mechanically sheared off the cell) rotated only which appears to be widespread among the bac-
clockwise with very brief (<0.1 s), intermittent teria. It seems that in several of these cases the
stops.56 signaling pathway to the flagellum motor uses
R. meliloti has homologues to the enteric che the Che proteins that function in the chemot-
genes.57 Present in R. meliloti are cheY (two axis signaling pathway. Thus, one might view
www.ebook777.com
various sensory signaling pathways as differing can damage DNA. Attracting light retards
in the initial sensory receptors but converging at the process of switching; hence the bacterium
the level of the Che proteins (e.g., CheA). swims in the direction of the light. Repulsive
Taxis toward electron acceptors or toward light increases the switching frequency; hence
light has been referred to as “energy taxis” the bacteria swim away from repulsive light.
and is reviewed in ref. 62. Photoresponses in This behavior is not really phototaxis in that
halobacteria and photosynthetic bacteria will the bacteria do not perceive a gradient of light
be discussed, followed by a description of taxis and are simply responding to changes in inten-
toward oxygen, that is, aerotaxis, as well as taxis sity. In this way, the phenomenon is similar to
toward nitrate under anaerobic conditions. chemotaxis.
Signal transfer to flagellar motor

20.10 Halobacteria The photoreceptors are rhodopsins similar
It will be helpful to review the description of the to bacteriorhodopsin.64–66 As a consequence
halophilic archaea in Sections 1.1.1 and 4.8.4, of absorbing light, there is a conformational
and the discussion of the role of light and retinal change in the photoreceptor, which results in
pigments in establishing a Δp and in Cl– trans- a signal (probably a conformational change) to
port in Sections 4.8.4 and 4.9. One of the pig- a closely associated, membrane-bound MCP-
ments is a rhodopsin called bacteriorhodopsin, like protein called Htr. It has been suggested
which couples the absorption of light energy that the Htr signals proteins homologous to the
to pumping protons across the cell membrane Che proteins, which then alter the frequency
from inside to outside, thus creating a Δp. A sec- of flagellar reversal. From this perspective, the
ond rhodopsin that is present is halorhodopsin, signaling pathway for photosensing is simi-
a light-driven pump used to pump Cl– into the lar to chemotaxis except that in photosensing,
cell to maintain osmotic stability. the sensor (Htr) responds to a light-activated
photoreceptor, whereas the sensor (MCP) in
20.10.1 Swimming behavior chemotaxis responds to the binding of a chemo-
Halobacteria have a single flagellum consist- effector or a periplasmic protein to which the
ing of 5 to 10 filaments in a right-handed helix chemoeffector is bound. The signal in both
at one pole of the cell. (Reviewed in ref. 63.) cases would be transferred via phosphorelay to
(Actually, the cells show monopolar flagellation the flagellar motor via the Che proteins. Exactly
only during exponential growth. In the station- how the absorption of light by the photorecep-
ary phase their flagellation is mostly bipolar.) tor activates the appropriate protein, Htr I, is
When the flagellum motor turns clockwise, not known, but see note 67 for a more complete
the cells swim forward, and when the motor discussion of photosensing.
turns counterclockwise, the swimming direc-
tion is reversed. (The flagellar filaments remain 20.11 Photosynthetic Bacteria
together even when the direction of rotation is 20.11.1 Avoidance of the dark
reversed.) During a brief period between CW Photosynthetic bacteria such as Chromatium,
and CCW rotation, the flagellum is not rotat- Thiospirillum, and Rhodospirillum that are
ing and the bacterium changes its orientation, swimming in a cell suspension will reverse their
perhaps as a result of Brownian motion. When direction of swimming when they swim out of
swimming resumes it is in a direction random to a light area into a dark area, whereupon they
the original direction. accumulate in the light. This is easily observed
microscopically by shutting the condenser so
20.10.2 Photoresponse that there is a small circle of light in an otherwise
The cells are capable of both positive and dark field. For example, when Rhodospirillum
negative photoresponses depending upon the rubrum, a photoheterotroph with polar flagel-
wavelength of light. The bacteria are attracted lation, swims from an area of light into an area
toward orange/red light (500–600 nm), which of darkness, the polar flagellar bundles reverse
is where bacteriorhodopsin and halorhodopsin their direction of rotation and the cells back up
absorb, and are repelled by UV/blue light, which into the light. Such behavior has been called
scotophobia because it is an avoidance of the evidence for this is derived from the use of the
dark rather than a seeking of light. proton ionophore FCCP, which stimulates elec-
Experiments with a horizontal beam of light tron transport while it collapses the Δp. (See
in a microscope field indicate that free-swim- Section 4.4.1 for an explanation of uncouplers.)
ming photosynthetic bacteria do not respond The experiments were done with cells tethered
to the direction of propagation of the light but to glass by anti-flagellin antibody. Under these
rather to a decrease in intensity.68 If the cells can conditions the cells rotated and the frequency of
reverse their direction of swimming by revers- stops was measured. The tethered cells showed
ing the rotation of the flagella, then they tend no response to the addition of FCCP, although
to back up whenever they start to swim across the ΔΨ decreased to the same extent caused by a
the light/dark boundary. They become trapped step-down in light intensity. This indicates that
throughout the beam of light, not simply near the cells respond to a decrease in photosynthetic
its source, indicating that they are not respond- electron transport rather than to a decrease in the
ing to the direction from which the light is com- Δp upon a step-down in light intensity. Perhaps
ing but rather to the decrease in intensity as they when the rate of electron transport is decreased,
cross into the dark. a signal is sent to an Aer-like protein, which
Not all photosynthetic bacteria can reverse results in an increase in CheY-P that stops the
the direction of flagellar rotation, and there- flagellar motor. This, however, is speculation.
fore some do not back up when crossing a light/ (Aer proteins are MCP homologues required
dark boundary. In R. sphaeroides, for example, for response to oxygen gradients.)
the flagellum rotates in only one direction, and
therefore when the light intensity is reduced,
20.11.3 Phototaxis of colonies
there is an increase in the frequency of stopping.
Individual cells of Rhodospirillum centenum in
When the cells stop swimming, their orienta-
liquid culture show the scotophobic response
tion changes, leading to a change in swimming
described earlier for other photosynthetic bac-
direction when flagellar rotation resumes.
teria. Cells grown in liquid media have a single
Interestingly, a narrow beam of light aimed hor-
polar flagellum and change the direction of
izontally across a microscope field causes bac-
swimming by reversing the rotation of the fla-
teria to accumulate in the dark, rather than in
gellum. However, R. centenum colonies move
the light. This is because the cells do not reverse
and are capable of phototaxis in gradients
when crossing the light/dark boundary but stop
of light.69 This is a true phototaxis in that the
and swim off in a randomly different direction,
colonies move along a light gradient toward the
which more often than not brings them into the
source of light down an intensity gradient when
dark.
a converging beam of light is used. When two
attractant beams of light at 90° to each other
20.11.2 Role of photosynthetic electron are used, the colonies move in a direction that
transport is 45° between the beams, suggesting that they
Mutants in the photosynthetic reaction center are integrating the signal from the light beams.
do not respond to a decrease in light intensity, The colonies move toward a source of infrared
suggesting that the response involves moni- light and away from a source of visible light.
toring changes in photosynthetic electron The cells have lateral flagella when grown on
transport, the Δp, or both. (A decrease in light agar, and they swarm, or move cooperatively,
intensity in wild-type cells causes a decrease in on the solid agar.
Δp and in photosynthetic electron transport.) Mutants in the photoresponses have been
This conclusion is supported by the finding that isolated.70 Some of the mutants that are not
inhibitors of electron transport prevent photo- capable of photoresponses have defects in the
responses even though the cells swim normally. synthesis of reaction center components or elec-
The evidence gathered for R. sphaeroides sug- tron carriers that function in photosynthetic
gests that it is photosynthetic electron transport electron transport. This indicates that the pho-
(probably the redox level of one of the carriers) toresponses are in some way related to changes
rather than the Δp that signals a step-down in in photosynthetic electron transport as the cells
the light intensity. Some of the experimental move along the light gradient. Other mutants
www.ebook777.com
have defects in the che genes, indicating that 20.13 Summary

the signal cascade proceeds from the photore- Chemotaxis refers to the process of bacteria
ceptor through the Che proteins to the flagellar moving toward a chemical attractant that can
motors. be a source of nutrient for them, or away from a
chemical repellent that may be toxic to them (e.g.,
20.12 Aerotaxis aromatic compounds such as benzoate or phe-
Many aerobic bacteria, including E. coli and nol with respect to E. coli). The attractants and
Salmonella typhimurium, swim toward a repellents are referred to as chemoeffectors and
higher concentration of oxygen. The response are sensed via a two-component system consist-
to oxygen requires electron transport as well as ing of a histidine kinase (CheA) and a response
a portion of the chemotaxis pathway, includ- regulator (CheY). CheY affects the switch mech-
ing CheW, CheA, and CheY. This conclusion anism in the flagellar motor. Part of the regula-
was based on the results of experiments with tory machinery includes a transmembrane sensor
mutants lacking the terminal cytochrome oxi- protein called the methyl-accepting chemotaxis
dases o and d or the che genes. When wild-type protein (MCP). Chemotaxis involves (1) bacteria
E. coli is grown anaerobically in the presence of measuring changes in concentration of effectors
nitrate so that cytochrome o is absent and nitrate over time and (2) adaptation. Exactly how the
reductase is present, the cells are attracted to bacteria change their direction of swimming var-
nitrate but not to oxygen. This indicates that ies with the particular bacterium and can include
electron transport to oxygen generates the oxy- tumbling, or reversal of the direction of flagella
gen signal and that electron transport to nitrate motility at cell poles, or simply stopping rotation
generates the nitrate signal. When both cyto- of flagellar allowing random swimming. This
chrome o and nitrate reductase are present, the chapter also describes tactic responses to light
cells are attracted to both oxygen and nitrate as well as to electron acceptors such as oxygen
in a competitive manner. Inhibitors of electron and nitrate. Photoresponses are described for
transport prevent taxis toward the electron halobacteria as well as photosynthetic bacteria.
acceptors. The primary signal might be a change Many aerobic bacteria swim toward a higher con-
in the Δp, or as suggested next, in the redox state centration of oxygen, and this is also described
of one of the electron carriers. (aerotaxis). Some of these bacteria swim toward
A protein called Aer has been referred to as nitrate when it serves as the sole or major electron
an MCP homologue and seems to be required, acceptor in the electron transport chain.
inasmuch as mutants in aer show reduced
responses to oxygen gradients.71 Unlike MCPs,
Study Questions
Aer does not have a periplasmic sensing domain.
However, it does have a hydrophobic domain
1. Describe the signaling pathway for chemot-
that anchors it in the cell membrane, as well as
axis in E. coli.
cytoplasmic domains that interact with CheA/
CheW. One can suppose that Aer senses the 2. What causes E. coli to swim randomly? Is
change in the redox state of a component of the this the mechanism for all bacteria? Explain
electron transport chain and transmits the sig- the differences. How is random swimming
nal to CheA. related to chemotaxis?
For example, suppose that increased levels
3. What is the relationship between adapta-
of oxidation of the electron carriers signaled
tion and methylation of MCPs?
Aer to lower the levels of CheA-P. This would
result in lower amounts of CheY-P, hence 4. What is bacteriorhodopsin, and what is its
smooth swimming and accumulation of cells role in halobacteria? How does it create a Δp?
in areas of higher oxygen concentration. Aer
5. What is halorhodopsin and its function?
noncovalently binds cytoplasmic FAD, and
perhaps the redox state of the FAD, itself 6. What is the evidence that aerobic bacteria
altered by a component of the electron trans- that swim toward oxygen require both a
port chain, alters the configuration and thus functioning electron transport pathway
the activity of Aer. and chemotaxis proteins?
REFERENCES AND NOTES proteins do not make flagella (Fla– phenotype) even
though these proteins are not part of the basal body,
1. Macnab, R. M. 1987. Motility and chemotaxis, hook, or filament. Missense mutations in these genes
pp. 723–759. In: Escherichia coli and Salmonella cause a Fla– phenotype, a Che– phenotype (CW or
typhimurium: Cellular and Molecular Biology, CCW biased), or paralyzed flagella (Mot phenotype),
Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press, depending upon the mutation. It therefore appears
Washington, DC. that these proteins are necessary for basal body
synthesis and switching, and perhaps are involved
2. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. in energy coupling. Two other proteins should be
Protein phosphorylation and regulation of adaptive mentioned, even though they are not chemotaxis
responses in bacteria. Microbiol. Rev. 53:450–490. proteins. The MotA and MotB proteins are neces-
3. Manson, M. D 1992. Bacterial motility and sary for motor function. Mutations in the motA and
chemotaxis, pp. 277–346. In: Advances in Microbial motB genes cause paralyzed flagella. The proteins are
Physiology, Vol. 33. A. H. Rose (Ed.). Academic integral membrane proteins that are thought to form
Press, New York. a ring around the S and P rings of the basal body,
coupling the proton potential to flagellar rotation.
4. Blair, D. F. 1995. How bacteria sense and swim. The MotA protein is a proton channel and the MotB
Annu. Rev. Microbiol. 49:489–522. protein may mediate the interaction of MotA with
5. Szurmant, H., and G. W. Ordal. 2004. Diversity the basal body.
in chemotaxis mechanisms among Bacteria and 15. The histidine kinase, CheA is a dimer. One of
Archaea. Microbiol. Mol. Biol. Rev. 68:301–319. the monomers acts as a kinase and phosphorylates
6. Macnab, R. M., and D. E. Koshland Jr. 1972. The a histidine residue on the second monomer to form
gradient-sensing mechanism in bacterial chemotaxis. CheA-P.
Proc. Natl. Acad. Sci. USA 69:2509–2512. 16. The MCP proteins are truly remarkable with
7. Berg, H. C., and D. A. Brown. 1972. Chemotaxis respect to the different classes of signals that they
in Escherichia coli analysed by three-dimensional transduce. Tsr responds to the chemoattractant
tracking. Nature 239:500–504. L-serine. It also responds to the repellents L-leu-
cine and indole, and to low external pH. Tsr also
8. Silverman, M., and M. Simon. 1974. Flagellar responds to weak organic acids such as acetate that
rotation and the mechanism of bacterial motility. act as repellents because they lower the internal pH.
Nature 249:73–74. In addition, the Tsr protein is a thermoreceptor medi-
9. Larsen, S. H., R. W. Reader, E. N. Kort, W.-W. ating taxis toward warmer temperatures up to 37 °C.
Tso, and J. Adler. 1974. Changes in direction of fla- The Tar protein in E. coli responds to the chemoat-
gellar rotation is the basis of the chemotactic response tractant L-aspartate as well as to maltose bound to
in Escherichia coli. Nature 249:74–77. the periplasmic maltose-binding protein. Tar also
detects the repellents Co2+ and Ni2+. In mutants
10. Maddock, J. R., and L. Shapiro. 1993. Polar loca- lacking Tsr, the Tar protein mediates taxis toward
tion of the chemoreceptor complex in the Escherichia higher temperatures. The Trg protein responds to
coli cell. Science 259:1717–1723. the chemoattractants ribose, glucose, and galactose
11. Sourjik, V., and H. C. Berg. 2000. Localization when these are bound to their respective periplasmic
of components of the chemotaxis machinery of binding proteins. The Tap protein (present in E. coli
Escherichia coli using fluorescent protein fusions. but not S. typhimurium) mediates taxis toward a
Mol. Microbiol. 37:740–751. variety of dipepetides when they are bound to DPP, a
periplasmic dipeptide-binding protein. Trg and Tap
12. Shiomi, D., I. B. Zhulin, M. Homma, and I. from E. coli serve as repellent receptors for phenol.
Kawagishi. 2002. Dual recognition of the bacte- Trg is also an attractant receptor, whereas Tap is a
rial chemoreceptor by chemotaxis-specific domains repellent receptor for weak organic acids.
of the CheR methyltransferase. J. Biol. Chem.
277:42325–42333. 17. Lukat, G. S., and J. B. Stock. 1993. Response
regulation in bacterial chemotaxis. J. Cell. Biochem.
13. Iino, T., Y. Komeda, K. Kutsukake, R. M. Macnab, 51:41–46.
P. Matsumura, J. S. Parkinson, M. I. Simon, and S.
Yamaguchi. 1988. New unified nomenclature for 18. Parkinson, J. S. 2003. Bacterial chemotaxis: a
the flagellar genes of Escherichia coli and Salmonella new player in response regulator dephosphorylation.
typhimurium. Microbiol. Rev. 52:533–535. J. Bacteriol. 185:1492–1494.
14. Three proteins called FliG, FliM, and FliN, 19. Silversmith, R. E., G. P. Guanga, L. Betts,
coded for by fliG, fliM, and fliN, are the switch pro- C. Chu, R. Zhao, and R. B. Bourret. 2003. CheZ-
teins. These proteins are not in the basal body and mediated dephosphorylation of the Escherichia coli
seem to be peripheral (rather than integral) cell chemotaxis response regulator CheY: role for CheY
membrane proteins closely associated with the basal glutamate 89. J. Bacteriol. 185:1495–1502.
body. A description of the flagellar motor is given in 20. Welch, M., K. Oosawa, S.-I. Aizawa, and M.
Section 1.2.1. Mutants that completely lack these Eisenbach. 1993. Phosphorylation-dependent bind-
www.ebook777.com
ing of a signal molecule to the flagellar switch of bac- stimulates CheA kinase activity is that membranes
teria. Proc. Natl. Acad. Sci. USA 90:8787–8791. prepared from cells in which glutamine is substituted
for glutamate in Tar, or in which the glutamate resi-
21. Ordal, G. W., L. Màrquez-Magana, and M. J.
dues are methylated because the cells were incubated
Chamberlin. 1993. Motility and chemotaxis, pp.
with aspartate, are more active in stimulating the
765–784. In: B. subtilis and Other Gram-Positive
phosphorylation of CheY than membranes contain-
Bacteria: Biochemistry, Physiology, and Molecular
ing unmodified chemoreceptor.
Genetics. A. L. Sonenshein, J. A. Hoch, and R. Losick
(Eds.). ASM Press, Washington, DC. 29. Presumably the binding of chemoattractant
induces a conformational change in the receptor–
22. Parkinson, J. S. 2004. Signal amplification in
transducer protein, exposing additional methylat-
bacterial chemotaxis through receptor teamwork.
ing sites on the receptor–transducer protein, thus
ASM News 70:575–582.
increasing the level of methylation.
23. Ordal, G. W., L. Màrquez-Magana, and M. J.
Chamberlin. 1993. Motility and chemotaxis, pp. 30. Park, C., D. P. Dutton, and G. L. Hazelbauer.
765–784. In: B. subtilis and Other Gram-Positive 1990. Effects of glutamines and glutamates at sites
Bacteria: Biochemistry, Physiology, and Molecular of covalent modification of a methyl-accepting trans-
Genetics. A. L. Sonenshein, J. A. Hoch, and R. Losick ducer. J. Bacteriol. 172:7179–7187.
(Eds.). ASM Press, Washington, DC. 31. Dailey, F. E., and H. C. Berg. 1993. Change
24. Dailey, F. E., and H. C. Berg. 1993. Change in direction of flagellar rotation in Escherichia coli
in direction of flagellar rotation in Escherichia coli mediated by acetate kinase. J. Bacteriol. 175:3236–
mediated by acetate kinase. J. Bacteriol. 175:3236– 3239.
3239. 32. Dunten, P., and D. E. Koshland Jr. 1991. Tuning
25. Barak, R., W. N. Abouhamad, and M. Eisenbach. the responsiveness of a sensory receptor via covalent
1998. Both acetate kinase and acetyl coenzyme A modification. J. Biol. Chem. 266:1491–1496.
synthetase are involved in acetate-stimulated change 33. Borkovich, K. A., L. A. Alex, and M. I. Simon.
in the direction of flagellar rotation in Escherichia 1992. Attenuation of sensory receptor signaling by
coli. J. Bacteriol. 180:985–988. covalent modification. Proc. Natl. Acad. Sci. USA
26. Dunten, P., and D. E. Koshland Jr. 1991. Tuning 89:6756–6760.
the responsiveness of a sensory receptor via covalent 34. Stock, J. B., and M. G. Surette. 1996.
modification. J. Biol. Chem. 266:1491–1496. Chemotaxis, pp. 1103–1129. In: Escherichia coli
27. Borkovich, K. A., L. A. Alex, and M. I. Simon. and Salmonella: Cellular and Molecular Biology.
1992. Attenuation of sensory receptor signaling by Vol. I. F. C. Neidhardt et al. (Eds.). ASM Press,
covalent modification. Proc. Natl. Acad. Sci. USA Washington, DC.
89:6756–6760. 35. Park, C., D. P. Dutton, and G. L. Hazelbauer.
28. It is possible to construct mutants in the aspar- 1990. Effects of glutamines and glutamates at sites
tate chemoreceptor protein (Tar) so that glutamine of covalent modification of a methyl-accepting trans-
(the amidated form of glutamate) is substituted for ducer. J. Bacteriol. 172:7179–7187.
the glutamate residues that are normally methylated. 36. Stock, J., A. Borczuk, F. Chiou, and J. E. B.
The chemotaxis signaling system cannot tell the dif- Burchenal. 1985. Compensatory mutations in recep-
ference between a chemoreceptor that has glutamine tor function: a reevaluation of the role of methylation
and one that has methylated glutamate. Such mutants in bacterial chemotaxis. Proc. Natl. Acad. Sci. USA
show that substituting glutamine for glutamate in a 82:8364–8368.
CheR– and CheB– mutant produces a tumbling behav-
ior. The tumbling behavior can be reversed by adding 37. Stock, J., G. Kersulis, and D. E. Koshland
the attractant aspartate to the medium, reflecting the Jr. 1985. Neither methylating nor demethylating
inhibition of the kinase activity of CheA by the bind- enzymes are required for bacterial chemotaxis. Cell
ing of attractant to the chemoreceptor. Furthermore, 42:683–690.
CheR– mutants swim smoothly and CheB– mutants 38. Stock, J., and A. Stock. 1987. What is the role
tumble more frequently than the wild type, again of receptor methylation in bacterial chemotaxis?
indicating that methylation stimulates tumbling. Trends Biochem. Sci. 12:371–375.
Because the affinity of the methylated chemoreceptor
for the attractant reported is not much different from 39. There are several ways to measure chemotaxis.
that for the nonmethylated chemoreceptor, the effect One way is to use swarm plates. If a chemotactic
of methylation on CheA kinase activity appears to bacterium is inoculated in the center of semisolid
be the dominant reason for adaptation. (However, agar fortified by attractants it can metabolize;
the extent to which methylation changes the affinity then, as the cells grow, the colony swarms outward
of the chemoreceptor to the attractant is a subject from the center. This is because the growing cells
of controversy. See: Blair, D. F. 1995. How bacte- create a gradient of attractant as they utilize it. In
ria sense and swim. Annu. Rev. Microbiol. 49:489– another assay, a capillary is filled with a solution
522.) Additional evidence that methylation of Tar of an attractant and placed in a suspension of cells
without attractant. The cells swim into the capil- 51. Ward, M. J., D. M. Harrison, M. J. Ebner, and J.
lary following the gradient. One then plates out the P. Armitage. 1995. Identification of a methyl-accept-
cells in the capillary to quantitate the chemotaxis. ing chemotaxis protein in Rhodobacter sphaeroides.
A third method is to observe tethered cells with a Mol. Microbiol. 18:115–121.
microscope. The cells can be tethered to a coverslip
52. The R. sphaeroides MCP protein was discovered
coated with anti-flagellin antibody. The coverslip is
by sequencing a fragment of DNA upstream of the
then placed over a chamber through which various
chemotaxis operon. The deduced amino acid sequence
solutions can be added and the clockwise or coun-
of the protein product shows significant similarity to
terclockwise rotation of the bacteria can be moni-
an enteric MCP (MCP Tsr). The protein cross-reacts
tored. For E. coli, CCW rotation is equivalent to
with antisera to the McpA protein of Caulobacter
smooth swimming and CW rotation is equivalent
crescentus and was located primarily in the cytoplas-
to tumbling. One might measure the percentage of
mic fraction, as determined by Western blotting. For
cells that rotated CCW over a specific time period.
Western blotting, the proteins of the membrane an
For example, suppose the measurement was for 15
cytoplasmic cell fractions are first separated by gel
s. Upon addition of attractant, the percentage of
electrophoresis (SDS-PAGE). Then the protein bands
cells that rotate CCW without reversing over a 15 s
are electrophoretically transferred to nitrocellulose.
period might be 100%, then falling to the prestimu-
The proteins on the nitrocellulose are incubated with
lus level, which might be 80%, in 1 or 2 min as adap-
antisera (e.g., to anti-McpA antibody), and the bands
tation occurred.
that bind the antibody are stained by means of an
40. Eisenbach, M., C. Constantinou, H. Aloni, and anti-rabbit secondary antibody (e.g., goat anti-rabbit
M. Shinitzky. 1990. Repellents for Escherichia coli antibody) conjugated to horseradish peroxidase. An
operate neither by changing membrane fluidity nor assay for the peroxidase stains the bands.
by being sensed by periplasmic receptors during
53. Alexandre, G., and I. B. Zhulin. 2001. More than one
chemotaxis. J. Bacteriol. 172:5218–5224.
way to sense chemicals. J. Bacteriol. 183:4681–4686.
41. Reviewed in: Titgemeyer, F. 1993. Signal
54. Packer, H. L., and J. P. Armitage. 1994.
transduction in chemotaxis mediated by the bacte-
The chemokinetic and chemotactic behavior
rial phosphotransferase system. J. Cell. Biochem.
of Rhodobacter sphaeroides: two independent
51:69–74.
responses. J. Bacteriol. 176:206–212.
42. Eisenbach, M. 1996. Control of bacterial
55. Schnitzer, M. J., S. M. Block, H. C. Berg, and E.
chemotaxis. Mol. Microbiol. 20:903–910.
M. Purcell. 1990. Strategies for chemotaxis. Symp.
43. Alexandre, G., and I. B. Zhulin. 2001. More than Soc. Gen. Microbiol. 46:15–34.
one way to sense chemicals. J. Bacteriol. 183:4681–
56. Gotz, R., and R. Schmitt. 1987. Rhizobium
4686.
meliloti swims by unidirectional, intermittent rota-
44. Blair, D. F. 1995. How bacteria sense and swim. tion of right-handed flagellar helices. J. Bacteriol.
Annu. Rev. Microbiol. 49:489–522. 169:3146–3150.
45. Eisenbach, M. 1996. Control of bacterial 57. Schmitt, R. 2002. Sinorhizobial chemotaxis:
chemotaxis. Mol. Microbiol. 20:903–910. a departure from the enterobacterial paradigm.
Microbiology 148:627–631.
46. Porter, S. L., G. H. Wadhams, and J. P. Armitage.
2008. Rhodobacter sphaeroides: complexity in 58. Greck, M., J. Platzer, V. Sourjik, and R. Schmitt.
chemotactic signaling. Trends Microbiol. 16: 251–260. 1955. Analysis of a chemotaxis operon in Rhizobium
meliloti. Mol. Microbiol. 15:989–1000.
47. Armitage, J. P., and R. Schmitt. 1997. Bacterial
chemotaxis: Rhodobacter sphaeroides and 59. Sockett, R. E., J. P Armitage, and M. C. W.
Sinorhizobium meliloti—variations on a theme? Evans. 1987. Methylation-independent and methy-
Microbiology 143:3671–3682. lation-dependent chemotaxis in Rhodobacter spha-
eroides and Rhodospirillum rubrum. J. Bacteriol.
48. Ward, M. J., A. W. Bell, P. A. Hamblin, H. L. 169: 5808–5814.
Packer, and J. P. Armitage. 1995. Identification
of a chemotaxis operon with two cheY genes in 60. Zhulin, I. G., and J. P. Armitage. 1993. Motility,
Rhodobacter sphaeroides. Mol. Microbiol. 17:357– chemokinesis, and methylation-independent
366. chemotaxis in Azospirillum brasilense. J. Bacteriol.
175:952–958.
49. Hamblin, P. A., Maguire, B. A., Grishanin,
R. N., and J. P. Armitage. 1997. Evidence for two 61. Hellingwerf, K. J., W. D. Hoff, and W. Crielaard.
chemosensory pathways in Rhodobacter sphaeroi- 1996. Photobiology of microorganisms: how photo-
des. Mol. Microbiol. 26:1083–1096. sensors catch a photon to initialize signalling. Mol.
50. The chemotaxis operon from R. sphaeroides was
detected by using a cloned cheA homologue from 62. Alexandre, G., S. Greer-Phillips, and I. B. Zhulin.
Rhizobium meliloti. The chemotaxis operon was 2004. Ecological role of energy taxis in microorgan-
sequenced, revealing the presence of the che genes. isms. FEMS Microbiol. Rev. 28:113–126.
www.ebook777.com
63. Alam, M., and D. Oeserhelt. 1984. Morphology, conformations: SR587 or S373, where the subscripts
function and isolation of halobacterial flagella. J. denote maximum absorption peaks, in nanometers.
Mol. Biol. 176:459–475. The two conformations are interconvertible. When
SR587 absorbs light (orange/red) it is converted to
64. Spudich, J. L. 1993. Color sensing in the archaea:
S373. When S373 absorbs light (UV/blue) it is converted
a eukaryotic-like receptor coupled to a prokaryotic
back to SR587. There are approximately 5,000 mol-
transducer. J. Bacteriol. 175:7755–7761.
ecules per cell of SR I, and in the presence of steady
65. Bickel-Sandkötter, S. 1996. Conversion of background light, an equilibrium exists between the
energy in halobacteria: ATP synthesis and photo- two forms. SR587 is the receptor for attractant light.
taxis. Arch. Microbiol. 166:1–11. The absorption of light by SR587 suppresses flagellar
reversals, causing the cells to swim toward higher
66. Armitage, J. P. 1997. Behavioral responses
intensities of orange/red light. S373 is the receptor
of bacteria to light and oxygen. Arch. Microbiol.
for repellent light. The absorption of light by S373
168:249–261.
increases flagellar reversals, causing the cells to swim
67. By way of introduction, the student should away from high intensities of UV/blue light. This is
know that all photosensing involves the absorption an interesting case of the same photoreceptor serving
of a photon of light by a chromophore bound to a as either the attractant or the repellent light receptor,
protein. When the photon is absorbed, one or more depending upon its conformation. It has been pre-
of the following events are induced: (1) excitation sumed that Htr I undergoes a conformational change
transfer, (2) electron transfer, (3) H+ transfer, (4) upon the absorption of light by SR I and transmits
photoisomerization. As we shall see, photo-isomer- the signal to the flagellar motor. It has been demon-
ization and H+ transfer occur when the rhodopsins strated that SR I is capable of light-induced proton
absorb light energy. Two photosensory rhodopsin pumping similar to bacteriorhodopsin proton pump-
pigments, called SR I and SR II, are responsible for ing, but in the presence of tightly bound Htr I the
the photoresponses. SR II is made constitutively at protons circulate, rather than being released into the
relatively low levels, but SR I is induced to high lev- bulk phase. Either the protons circulate through Htr
els under growth conditions that also induce bacte- I residues during SR I photocycle (a direct involve-
riorhodopsin (i.e., low oxygen tensions and intense ment of Htr I in proton circulation) or proton circula-
illumination). It is clear that the rhodopsins that tion is dependent upon conformational interactions
function in photoresponses are similar to bacterior- between SR I and Htr I (an indirect role for Htr I in
hodopsin and halorhodopsin. (See Sections 4.8.4 and proton circulation).
4.9 for a discussion of bacteriorhodopsin and halo-
68. Grishanin, R. N., D. E. Gauden, and J. P.
rhodopsin, respectively.) These pigment proteins are
Armitage. Photoresponses in Rhodobacter spha-
related structurally and presumably have a common
eroides; role of photosynthetic electron transport. J.
origin. Their primary photochemistry is the same: a
Bacteriol. 179:24–30.
light-induced photoisomerization of the retinal from
the all-trans isomer to the 13-cis form. When the rho- 69. Ragatz, L., Z.-Y. Jiang, C. Bauer, and H. Gest.
dopsins absorb light, a signal is transmitted to the 1995. Macroscopic phototactic behavior of the pur-
flagellum motor, probably via a membrane-bound ple photosynthetic bacterium Rhodospirillum cente-
signal-transducing protein, Htr I (halobacterial num. Arch. Microbiol. 163:1–6.
transducer for sensory rhodopsin I) to which SR I is
70. Jiang, Z.-Y., B. G. Rushing, Y. Bail, H. Gest, and
tightly bound. (A second protein, Htr II, is complexed
C. E. Baner, 1998. Isolation of Rhodospirillum cen-
to SR II.) The Htr proteins are similar to the MCP
tenum mutants defective in phototactic colony motil-
proteins in that they have two membrane-spanning
ity by transposon mutagenesis. J. Bacteriol. 180:
domains and a cytoplasmic domain that is homolo-
1248–1255.
gous to the MCPs. Presumably a conformational
change takes place in Htr when light is absorbed by 71. Bibikov, S. I., A. C. Miller, K. K. Gosink, and J.
the chromophore, and a signal is thus generated to S. Parkinson. 2004. Methylation-independent aero-
the Che proteins that signal the flagellum motor. The taxis mediated by the Escherichia coli Aer protein. J.
model proposes that SR I exists in either one of two Bacteriol. 186:3730–3737.
21
Microbial Biofilms—Structured
Multicellular Assemblies
Bacteria rarely exist as the fully dispersed cul- are localized within biofilms, with much greater
tures that most students are familiar with from heterogeneity. In addition, multicellular struc-
the microbiology laboratory. In more natural tures have emergent physical, chemical, and
settings bacteria aggregate with themselves biological properties that could not have nec-
and with other bacteria and other organisms to essarily been predicted by the identity of their
form multicellular structures. These structures individual bacterial constituents.
are often called biofilms, although the term is This chapter provides an introduction to the
applied very broadly to a variety of structures. structure, formation, properties, and relevance
In many ways a bacterial colony on a petri dish of bacterial biofilms and related multicellular
can be considered to be a highly artificial bio- structures.
film. Multicellular biofilms form in virtually
all environments and are in many cases the
dominant mode of existence. Bacteria residing 21.1 Bacterial Multicellular
within biofilms often demonstrate remarkable Structures
resilience in the face of chemical, physical, and For over a century microbiologists have studied
biological assault. For those microbes that cause bacteria in homogeneous, dispersed laboratory
human disease, biofilms have immediate practi- culture and have measured the average prop-
cal importance because they provide increased erties of these relatively uniform populations.
tolerance toward antibiotic treatments, mak- Populations undergoing exponential growth in
ing it more difficult to treat infections in which a dispersed laboratory culture are predictable
biofilms have formed. Conversely, many bio- and balanced, and are relatively easy to evalu-
films can be beneficial, including the floc-form ate by using measurements that average across
microbes that process wastewater in treatment the population. In the natural environment,
plants and the biofilms that form on the exte- however, bacteria seldom exist in this uniform
rior of ripening cheese. In the past ten years, state; rather they exhibit unbalanced patterns
an explosion of research studying microbial of growth such as those observed in the station-
biofilms has changed our understanding of ary phase of batch cultures. Microorganisms
bacteria outside of laboratory cultures. Much often associate with surfaces and with each
of what was learned about bacterial physiology other in multicellular assemblages (Fig. 21.1).
prior to the study of biofilms, as covered com- These multicellular structures can vary from
prehensively in earlier sections of this book, those that form at air–liquid interfaces (called
remains entirely relevant to bacteria residing pellicles or microbial mats) to those that form
within these structures. These physiological in suspension (aggregates) to those that form
phenomena are however more complicated and on surfaces (solid–liquid interfaces) (biofilms).
551
www.ebook777.com
These structures share the basic attribute that 21.2 Prevalence and Importance
they consist of bacterial cells that adhere to one of Biofilms
another and are often connected through self-
Bacterial biofilms and related multicellular
produced exopolymeric material. In general,
structures form in diverse environments and
microorganisms are physiological contortion-
can adopt a range of different conformations,
ists and can rapidly adapt to alterations in their
which can change depending on prevailing
local environment. Cells within a multicellular
environmental conditions. It has been argued
(metazoan) organism are enslaved by, and an
that biofilms are the dominant mode of growth
intrinsic component of, the genetic program
for bacteria. Although there are certainly
of the whole organism. In contrast, microbes
examples of dispersed, suspended growth for
within multicellular structures have tremen-
bacteria, these are relatively rare. Hence, bio-
dous plasticity. These cells have entered into a
films are of central importance to virtually all
pact, and in all but the most developmentally
aspects of microbiology. In addition, biofilms
complex examples, they are able to break free
represent specific opportunities and challenges
of the trappings of multicellularity, to revert
that are inherent to the biofilm mode of growth.
to the single cell state. Even within the struc-
There are many different examples of biofilms,
tured confines of a multicellular assembly, indi-
ranging from communities on rocks in streams
vidual cells can exert their independence and
to infections of artificial joints to the bacteria
enter into a free-living existence. In much of
that colonize vegetables stored in refrigera-
the scientific literature, all these different struc-
tors, which are capable of causing food-borne
tures may be referred as “biofilm,” although it
infections. Any comprehensive discussion of
must be emphasized that the properties of any
the variety of environments, compositions, or
given multicellular structure can vary consider-
structures of biofilms would require a dedicated
ably depending upon the site of formation, the
book, of which there are now several.1,2 For the
microbial composition of the biofilm, and the
purposes of this textbook we will simply discuss
conditions under which the multicellular struc-
several illustrative examples.
ture assembles. For the sake of our discussion in
this chapter, we will refer frequently to biofilms;
keep in mind, however, that this term is used in 21.2.1 Corrosion and clogging of pipes
the broadest sense, to encompass most if not all In a wide variety of industrial settings, fluid is
multicellular microbial structures. transported into and away from processing or
C stratified
complex biofilm
B stratified biofilm
mushroom
A monolayer
channel
Fig. 21.1 Three models of biofilm structure: (A) a monolayer comprising single cells of one species, (B) a
stratified biofilm with multiple species, and (C) a three-dimensionally complex biofilm with channels and
mushroom structures. Biofilm matrix material in mottled gray. (Cells are not drawn to scale.)
microbial biofilms—structured multicellular assemblies 553
collection sites via pipes. These applications food, a frequent subject in the popular press.6
range from water distribution to petroleum Many potentially harmful microbes are capable
refining to gold leaching to maple syrup proof colonizing the surfaces of foodstuffs such as
duction. Depending on the fluid transported, vegetables, fruits, meats, and dairy products
the interior surface of the pipe, the flow rate, during collection, processing, transportation,
and other features of the environment, bio- and preparation for consumption. Biofilms
films will form.3 These biofilms can also act as formed by these pathogens, such as Salmonella
reservoirs for different microbes and in water enterica, can be very difficult to remove by stan-
systems, for example, may serve as a haven for dard food-handling approaches. In the fields,
disease agents. On metal pipes, biofilms can act fruits and vegetables may be contaminated by
as nucleation sites for metal corrosion that can irrigation and animal waste, allowing extensive
eventually compromise the structural integrity and recalcitrant biofilms to grow. Endogenous
of the pipe. Even for pipes that are resistant to microbial biofilms also play a role in plant
corrosion, biofilms can build up and restrict the health. Indeed, the balance between healthy
flow rate, and even clog the pipe. There has been microbiota and those that place the food supply
a great deal of effort to develop materials and at risk is a challenge for agriculture and food
strategies to limit the cost and damage to distri- science into the future.
bution systems due to biofilms. However, this
remains a problem in many different industrial 21.2.4 Dental biofilms
applications. Oral microbiota form extensive biofilms and
are intensively studied because they have a pro-
21.2.2 Wastewater treatment found impact on animal oral health.7 Bacteria
Many biofilms are benign or even beneficial in form structured layers on tooth enamel sur-
terms of human impact. Several examples of faces, as well as below the gum line. Dental
highly beneficial biofilms occur in water treat- plaque is a highly specialized and stratified
ment plants.4 Such a treatment plant can be biofilm (see Fig. 21.1 for an example of a strati-
considered to be a series of beneficial biofilms fied biofilm). Many of the oral microbiota are
that function in the water purification pro- relatively benign, although certain members
cess. Much of the activity that removes waste can be detrimental, such as Streptococcus
materials from wastewater streams, prior to mutans, which causes dental caries (i.e., tooth
chemical treatment and redistribution, is due decay). Colonization of tooth and gum surfaces
to microorganisms. Almost all these microbes is hypothesized to proceed in discrete succes-
are actively breaking down and assimilating the sion, due to specific interactions between differ-
organic material such as sewage that has been ent oral bacteria. Certain late-stage colonists,
introduced into the wastewater. Many if not all such as Fusobacterium species, are commonly
of these wastewater-active microbes are orga- associated with poor dental health. Likewise,
nized as consortia within biofilms (various spe- periodontitis is due to accumulation of distinct
cies are analogous to different workers on an anaerobic biofilms that form beneath the gum
assembly line), and the structure of these bio- lines. Much of dental hygiene practice is aimed
films can have a great impact on the efficacy of at limiting or blocking the growth of these oral
water treatment. For example, in sludge treat- biofilms, and new strategies that target biofilm-
ment, much of the purification is due to bacte- specific processes are working their way into the
ria within flocs (large aggregates) which are in area of dental health.
fact specialized biofilms that degrade the sludge
granules.5 If the biofilm-colonized flocs are lost 21.2.5 Indwelling medical devices
from this step in the process, the sludge process- A major problem in human medicine that is
ing is severely compromised. directly associated with biofilms is the coloniza-
tion of indwelling, man-made medical devices
21.2.3 Food-borne illness such as artificial joints, catheters, and heart
The safety of the human food supply is a continu- valves.8 These foreign surfaces typically do not
ing problem, with incidents of serious illnesses have the antimicrobial capacities or activities
such as salmonellosis due to contaminated exhibited by many animal tissues and therefore
www.ebook777.com
are prime sites for biofilm formation. In addi- associated with other cells? The advent of
tion, installation of the medical devices is often sophisticated fluorescence microscopy, elec-
fairly traumatic to the surrounding tissues, tron microscopy (EM), and scanning laser con-
upsetting the normal microbial balance and focal microscopy (SLCM) radically changed
providing dangerous opportunities for infec- notions about the multicellular structures that
tions. Microbial infections of these devices microbes can form. Biofilms were revealed not
that progress to forming significant biofilms to be monolayers of cells passively adhered to
become very difficult to treat, because they are surfaces, but rather are often thick, structurally
physically recalcitrant and can resist standard complex assemblies which can change dynami-
antibiotic therapy (see later: Section 21.3.2). cally in response to the environment.11 This
These infections can be significant sites of has been a major focus in the field of biofilm
inflammation and can lead to serious complica- research for several decades. The spatial struc-
tions that can become life threatening and may ture that results as a consequence of physical
require additional surgeries, including removal association between cells generates significant
of the device. A great deal of effort is ongoing alterations in the physiology of these bacteria.
to develop materials and procedures that will This is primarily due to the chemical and physi-
limit biofilm formation on these devices, and to cal gradients that occur when cells cannot mix
formulate approaches for removing biofilms in homogeneously.12 Gradients of nutrients and
situ, once they have formed. waste products exert substantial influence on
the growth rate of the microbes. The physiolog-
21.2.6 Lung tissue in cystic fibrosis ical status of cells within the biofilm has been
patients studied using a variety of approaches, including
The genetic disorder called cystic fibrosis (CF) fluorescent reporter genes, transcriptional pro-
results in a salt imbalance in human lungs and filing, and proteomics. Some aspects of biofilm
accumulation of thick sputum.9 CF is the most growth are notably similar to the slow growth
common, potentially lethal, inherited disorder exhibited by bacteria in stationary phase in lab-
among Caucasians, and is especially prevalent oratory culture.
among individuals of northern European ances-
try. The CF mutation is recessive, so individu- 21.3.1 Structural forms and
als must receive one allele from each parent to hydrodynamics
manifest the disease. Cystic fibrosis patients Biofilms can adopt a variety of conformations
are treated their entire lives to limit accumu- that range from relatively sparsely populated
lation of the thick sputum from the lungs, but surfaces to confluent monolayers to highly strat-
inevitably this material forms an ideal growth ified assemblies to three-dimensionally com-
substrate for colonization by opportunistic plex structures with large colonies and channels
bacterial pathogens, most problematically
(Fig. 21.1). These more complex biofilms can
by Pseudomonas aeruginosa and species of
vary structurally for many different attributes
Burkholderia. Biofilm formation within the CF including height, cell density, porosity, stabil-
lung by P. aeruginosa has become the predomi- ity, and the ratio of surface area to volume.
nant model for medically important biofilm Biofilms have been described as viscoelastic sol-
research, and a substantial amount has been ids; they resist deformation, but beyond a cer-
learned about this process.10 Even so, no fully tain point will be stretched, and will not snap
preventive treatments have yet been developed, back after the tension has been released.13 They
and P. aeruginosa infection remains one of the can often be many cell lengths tall (20 μm to
primary causes of mortality in CF patients. several hundred micrometers), but rather than
Biofilm research in this area continues to be a simply being passively assembled piles of cells,
high priority in many laboratories. these thick biofilms can be filled with channels
that act to connect cells within the biofilm to
21.3 Properties of Biofilms the fluid phase. These channels have sometimes
In what fundamental ways does a single bacterial been described as a primitive circulatory sys-
cell differ from a genetically identical cell that, tem for complex biofilms. However, even with
rather than existing independently, is physically these channels, different areas of the biofilm can
experience dramatically different conditions. However, reports suggest that in some cases
For example, oxygen levels are generally high at extracellular DNA (eDNA), released into the
the interface between the biofilm and the fluid biofilm matrix by passive or active processes,
phase but decrease with depth into the biofilm.14 fulfills a structural role.18 More recent work
The growth rate of cells within the biofilm will suggests that certain secreted proteins may
greatly influence the eventual structures that the also play a structural role within the biofilm.19
biofilm adopts. The matrix not only provides biofilms with a
Certain microbial species will favor particular great deal of their structure, it can also serve to
structural forms. Furthermore, ambient envi- chemically protect the microorganisms in resi-
ronmental conditions may also influence the dence from chemical and biological stress. As
type of structure formed. For example, aquatic discussed next, biofilms exhibit elevated toler-
environments with turbulent flow may favor ance for several antibiotics, and this is at least
elongated biofilms with extended streamers, in part due to titration of these compounds
whereas the same microbe may form strikingly into the matrix. Finally, to disperse the bio-
compact structures under static conditions.15 film, many bacteria utilize enzymes, primarily
The presence of other microorganisms within polysaccharases or proteases, that degrade the
the biofilm can also profoundly affect the bio- matrix.
film structure. Mixtures of different microbes
may remain cloistered into clonal colonies or can
21.3.3 Antimicrobial tolerance
be fully dispersed within the biofilm. Metabolic
The physiology of microbes is significantly
collaboration between microbes, called syn-
altered by their presence within a biofilm.
trophism, is often fostered within multicellular
Reflecting this change in physiology is a well-
structures, where nutrients can be effectively
documented increase in tolerance toward anti-
shuttled directly between microbes with little
microbial chemotherapy.20 Bacteria within
opportunity for loss of the nutrients to other
competitors (see Box 21.1 for an example called biofilms can manifest resistance that is orders
ANME). The physical proximity of microbes of magnitude greater than that of their plank-
within a biofilm allows metabolic interactions tonic counterparts. This is a very practical
that cannot be achieved in mixed cultures.16 problem in health care. Many people now have
indwelling medical devices such as artificial
joints, long-term catheters, heart valves, and
21.3.2 The biofilm matrix stents, introduced to counteract specific medi-
Biofilms are usually held together by a com- cal issues.8 These indwelling devices are prone
plex extracellular material described as the to microbial colonization and biofilm forma-
matrix.17 The matrix is generally gelatinous in tion (see Section 21.2), and once the bacteria
composition, but with some degree of poros- have formed a biofilm, they become signifi-
ity that allows the fluid phase to access regions cantly more resistant to antimicrobial therapy
of the biofilm. Different bacteria produce dif- than the same bacteria growing in a dispersed
ferent amounts of matrix material with spe- format such as lab culture. This resistance can
cific compositions. Although the matrix may take the form of tolerance to antibiotics such
be composed of many different compounds, it as tobramycin and tetracycline, but also may
is often dominated by polysaccharides. These extend to antimicrobial agents including hydro-
polysaccharides can be complex mixtures of gen peroxide and heavy metals. The tolerance
diffuse or concentrated material. Extracellular toward these compounds is very complex and is
polysaccharides are released by the bacte- the subject of many different studies. This toler-
ria, and the exact chemistry of the polymeric ance is a combined effect of physical protection
material produced and deposited in the matrix (e.g., exclusion, titration) but also physiological
depends upon the nutritional composition changes resulting in slow growth and formation
of the environment, and upon the growth of of resistance subpopulations called persisters.21
the bacteria. In addition to polysaccharides, Determining how to overcome the enhanced
bacteria release proteins, nucleic acids, and tolerance of biofilm microbes is a major goal
lipids and other common cellular material. of biofilm research and holds tremendous
Some of these may be debris from lysed cells. promise.
www.ebook777.com
BOX 21.1 HISTORICAL PERSPECTIVE: BIOFILMS IN THE OCEAN

The process of biofouling, by which activities are extremely common in marine
marine and aquatic organisms colonize environments.2
the hulls of vessels, has been a problem Neuston is the term used by oceanog-
for the shipping industry throughout its raphers to refer to marine life at the ocean
history. Macroscopically, fouling can be surface, and microneuston is the hetero-
seen as barnacles and algae on the under- geneous microbial film that forms there.
sides of vessels; but even microscopic Many marine microbes below the surface
fouling can have a profound impact on are associated with particulate matter
the hydrodynamic properties of the hull, such as so-called marine snow, and these
and fouled hulls make for slower, more contribute to the total particulate organic
fuel-consumptive vessels. The process of matter. Marine plants and animals har-
surface colonization and biofilm forma- bor on their surface and in their internal
tion in marine and aquatic environments tissues large numbers of microorganisms
therefore garnered significant attention that may be symbionts or pathogens. Some
well before other environments. In 1936 marine microbes form complex multi-
pioneer marine microbiologist Claude cellular structures in order to perform
Zobell described a phenomenon called the certain types of metabolism. For exam-
bottle effect, whereby the ratio of surface ple, anaerobic methanotrophic archaea
area to volume in a culture vessel can have (ANME) consume most of the methane
a strong impact on the number of cultivat- produced in the world’s oceans and thus
able bacteria isolated from these environ- help maintain the global methane bal-
ments.1 Zobell and Anderson observed ance. ANME often form small structured,
that in smaller volume flasks (with larger mixed-species aggregates, most commonly
ratios of surface areas to volume) the bac- with specific sulfate-reducing bacteria.
terial densities and activities were much Energetically unfavorable anaerobic oxi-
greater, sometimes by orders of magnitude, dation of methane is thought to be fostered
than in larger culture vessels. Zobell pro- through the collaborative metabolism
posed that this might be due to increased allowed by the close physical association
nutrients on the surface, and therefore an of these different microbial types.3 Other
increased number of bacteria harbored by specific multicellular assemblies have also
the resting place of the surface, relative to been characterized in marine and aquatic
the bulk fluid. It was noted that the pres- environments.
ence of sand suspended in seawater had an
even more dramatic effect. A great deal of REFERENCES
work on what are now known as biofilms
has been predicated on the concept that 1. Zobell, C. E. and D. Q. Anderson. 1936.
the amount of surface area could impact Observations on the multiplication of bacteria
in different volumes of stored sea water and the
the productivity of a community. Zobell influence of oxygen tension and solid surfaces.
was one of the first researchers to intro- Biol. Bull. 71:324–342.
duce sterile slides into the environment, 2. Munn, C. B. 2004. Marine Microbiology:
and then to observe the microscopic life Ecology and Applications, Garland Science,
that accumulated. He and his colleagues New York.
were thus among the first researchers to 3. Knittel, K. and A. Boetius. 2009. Anaerobic
focus on biofilms. In fact, multicellular oxidation of methane: progress with an unknown
structures of all shapes, dimensions and process. Annu. Rev. Microbiol. 63:311–334.
21.4 Progression of Biofilm Formation toward a more permanent attachment.23 For

and Dissolution flagellated bacteria, motility provides the abil-
ity to sample the substratum and the propul-
Biofilm formation is often viewed as a cyclical
sion to overcome surface barriers. In addition,
process (Figure 21.2).22 Free-living individual
flagella may associate with surfaces themselves
cells transition toward the sessile, multicellu-
and act as an initial tether for the cell to the
lar state by initially adhering to surfaces and
surface. Nonmotile bacteria must rely on pas-
to each other, via a process called attachment.
sive forces to mediate initial contact. Once
Subsequent to attachment, growth of the biofilm
many motile and nonmotile bacteria are close
can occur by a number of different mechanisms,
to the surface, extracellular appendages and
including additional recruitment of unattached
elongated surface proteins can act to cross the
cells, clonal growth of attached populations,
boundary and contact the surface. Flagella,
and physical spreading via surface-active motil-
pili, and large cell surface proteins, such as
ity mechanisms. As cells accumulate, micro-
the MSCRAMM (microbial surface compo-
colonies can often form, and eventually, these
nents recognizing adhesive matrix molecules)
microcolonies can coalesce into a full bacterial
proteins of Staphyloccus aureus, fulfill this
biofilm. The distinction between a large micro-
function.24,25 Shortly after initial contact, cells
colony and a small biofilm is not always clear.
may be readily removed from the surface, hence
Biofilm formation is concomitant with produc-
the term reversible attachment. The transition
tion of the matrix material to varying degrees
to irreversible attachment involves the intimate
and with varying compositions, depending
contact of the cells to the surface, via surface
upon the specific microbe and conditions. At
structures and polymers that provide tenacious
virtually any point during biofilm formation,
binding. Often, secreted polysaccharides such as
single cells can release themselves, either to
cellulose provide the glue to maintain irrevers-
reassociate with the biofilm or to move to a dif-
ible binding. Several Pseudomonas species, rod-
ferent location. Under some conditions, large
shaped bacteria, transition from initial surface
sections of the biofilm may dissociate, a process
interactions at their cell poles during reversible
that may be genetically programmed.
attachment to a sidelong interaction in irrevers-
ible attachment. This transition requires the
21.4.1 Attachment function of the extremely large proteins such
Individual bacteria in suspension initiate attach- as LapA (large adhesion protein A), associated
ment to surfaces. This typically involves an ini- with the external face of the cell wall.26 Other
tial reversible attachment phase progressing rod-shaped bacteria remain associated by their
planktonic
cell
mature
biofilm
attachment
growing matrix
microcolony elaboration aggregation
Fig. 21.2 Biofilm formation cycle by which a freely swimming planktonic cell colonizes a surface and pro-
gresses through biofilm formation; capsule and biofilm matrix material depicted in light gray around cells
and adherent biomass. Cells may eventually be released back into the aqueous environment to recolonize
additional sites. (Cells are not drawn to scale.)
www.ebook777.com
poles, even after irreversible attachment.27 of motility under these different conditions. In
Strong attachment of these cells positioned on the citrate-grown medium, P. aeruginosa forms
their poles may require production of polysac- microcolonies on the surface, followed by rapid
charide-based polar adhesives, such as the hold- surface expansion mediated through twitching
fast from Caulobacter crescentus.28 and swarming motility, that results in a confluent
For actively swimming bacteria, acclimation and flat biofilm.33 In contrast, in glucose medium
to the surface involves decelerating or stopping the microaggregates transition into formation of
flagellar rotation, and in many cases flagellar large towers or mushrooms of many cells (see
ejection. For the gram-positive rod Bacillus sub- Fig. 21.1C) interspersed with less densely colo-
tilis, flagellar rotation is disengaged by a flagel- nized surface. Formation of these mushroom-
lar clutch.29 It is not clear whether other bacteria type structures also relies on twitching motility.
use molecular clutches or brakes in the transi- Synthesis of matrix polysaccharides, and in
tion to sessile growth. The regulatory mecha- some cases, extrusion of DNA, also occurs dur-
nisms that control the transition of bacteria to ing maturation of biofilms. For S. epidermidis
surfaces are also poorly understood, but many and S. aureus, the exopolysaccharide poly-
gram-negative bacteria appear to involve the β(1,6)-N-acetylglucosamine (PNAG) functions
intracellular second messenger cyclic diguano- to cohere cells in the growing biofilm to each
sine monophosphate (c-di-GMP).30 In E. coli, other.34 In P. aeruginosa, extrusion of DNA
for example the production of flagella and the into the matrix is under control of the quorum-
synthesis of adhesive appendages such as curli sensing regulatory regulation, presumably link-
and cellulose is controlled a highly complex reg- ing this process to the increasing cell density
ulatory network in which many nodes produce, within the biofilm.35
degrade, or otherwise respond to c-di-GMP.
21.4.3 Maintenance of biofilm structure
Noncontinuous biofilm structures, such as
21.4.2 Biofilm maturation open areas and channels, must be actively
Adherent microbial populations expand across maintained.5 Continued growth of the adher-
surfaces via a combination of clonal growth, sur- ent cells and colonization by additional plank-
face motility, and subsequent colonization from tonic phase cells would effectively occlude these
free-swimming cells.31 Multiplication of cells can channels and make it more difficult for bacteria
lead to additional biofilm cells or active dispersal deep within the biofilm to access the bulk fluid
into the planktonic phase. A subset of microbes phase. Following the formation of mushroom
exhibit surface motilities such as swarming, or tower structures, mutants of P. aeruginosa
twitching, and gliding, and these microbes can that do not synthesize rhamnolipid surfactants
expand laterally but retain association with the (an rhlA null mutant) develop into continuous
surface. Nonmotile microorganisms, on the biofilms with no open channels.36 Furthermore,
other hand, will accumulate at the site of cell the rhamnolipids are required to prevent sub-
division unless they are swept away by turbu- sequent colonization by cells in the planktonic
lence in the fluid phase. The form and amount phase. It is not yet clear whether surfactants
of adherent biomass that emerges is strongly have a similar role for other microbes that form
influenced by the microorganisms present in architecturally complex biofilms.
the local environment, the surface itself, and
the nutritional conditions. Some bacteria form
strikingly different surface communities depend- 21.4.4 Emigration from biofilms
ing on nutritional conditions. For example, P. and dispersal
aeruginosa forms thin, flat films when grown Bacteria attached to surfaces may reenter the
with citrate as a carbon source, whereas when planktonic phase via a process called dissocia-
it is grown with glucose it forms thick, three-dition.22 Detachment of cells from most surfaces is
mensionally complex biofilms.32 Although this likely to be a continual process. Newly released
variety was originally thought to be due to pro- cells may remain planktonic, or they may colo-
duction of different polysaccharides, it appears nize additional surfaces. There is growing evi-
to be primarily influenced by the dominant form dence that dissociation from a biofilm can be
both a passive process (sloughing) that results It has been suggested that biofilm matura-
from physical shear or stress applied to the bio- tion is a developmental process for many bacte-
film and an active process (dissolution) involv- ria, analogous to sporulation in B. subtilis and
ing induction of flagellar-mediated motility and fruiting body formation in the myxobacteria.42
enzymatic degradation of the matrix material.37 Whether biofilm formation is truly a form of
Many varieties of biofilm dissociation have been bacterial development depends upon the crite-
observed, ranging from the release of the prog- ria used to define development and varies across
eny cells derived from the division of attached different bacteria. Irrespectively, biofilm for-
individual cells, to the shearing of large portions mation is mediated by key signal transduction
of a biofilm in high flow. There are soluble factors pathways that drive the process forward and
produced by bacteria that can stimulate the dis- coordinate the intercellular interactions that
solution process, including a recently identified result in higher-order biofilm structure. The Crc
broadly active factor identified as cis-2-decenoic global carbon metabolism regulatory protein of
acid that promotes dissolution of biofilms from P. aeruginosa is required for the formation of
diverse microbes including eukaryotic yeasts mature biofilms containing microcolonies.43
and is produced by P. aeruginosa.38 Induction This deficiency was correlated with decreases
of a lysogenic bacteriophage in mature biofilms in the expression of type IV pilus genes and in
resulting in limited areas of cell death is also twitching motility. It has been postulated that
thought to aid dissolution for P. aeruginosa.39 In Crc facilitates the integration of nutritional sta-
Caulobacter crescentus, its own DNA derived tus with biofilm development in P. aeruginosa.
from dead cells acts to inhibit reattachment of In B. subtilis, a regulator called SinR is thought
progeny cells to surfaces and is hypothesized to to be the master switch for biofilm formation,
promote dispersal away from environments in integrating the truly developmental processes,
which high numbers of Caulobacter cells are such as sporulation, with surface-associated
lysing.40 This inhibitory activity for DNA in growth.44
Caulobacter, which stands in contrast to the pos- The biofilm matrix is typically composed of
itive structural role for DNA reported for other secreted polysaccharides and other extracellular
bacteria, may not be a general phenomenon. polymers, such as proteins and nucleic acids.
Therefore, it is not surprising that regula-
tors of exopolysaccharide biosynthesis would
21.5 Regulation of Biofilm Formation affect biofilm development. Overproduction of
The processes of surface colonization, biofilm alginate in P. aeruginosa due to a mutation in
maturation, and dispersal are extremely com- the MucA anti-sigma factor results in a strain
plex and are influenced by multiple regulatory that overproduces mucoid alginate and is sub-
pathways that vary among different bacteria. stantially altered in its biofilm architecture,
Adaptation to the physicochemical conditions forming larger biofilms that are less dispersed
presented by surfaces results in significant over the surface and remain clustered in large
changes in the physiology of the microbe, and microcolonies.45,46 Synthesis of a different
these changes are mediated through environ- exopolysaccharide known as Pel, required for
mentally responsive regulation.23,41 For motile late stages of biofilm formation in wild-type
bacteria, the transition from the actively swim- P. aeruginosa, is regulated by the c-di-GMP
ming phase to a sessile lifestyle is an important responsive transcription factor called FleQ.47
initial step in the process. Pathways regulated Mutation in the VpsR repressor of Vibrio chol-
through the internal second messenger c-di- erae exopolysaccharide (vps) genes results in
GMP are important in many different bacteria overproduction of VPS and hyperadherent
for mediating the motile-to-sessile transition.30 derivatives that form substantially thicker bio-
Some of these pathways result in changes in films than the wild type.48 However, control of
gene expression patterns, whereas others act the timing for EPS production during surface
to control cellular activity directly. For non- colonization and the subsequent stages of bio-
motile bacteria, acclimation to a surface also film maturation remains poorly understood.
represents a major transition and is accordingly An early study provided evidence for activa-
under extensive control. tion of the P. aeruginosa alginate biosynthetic
www.ebook777.com
gene algC following surface colonization, One approach that has significant potential is
although this observation has not been further to utilize natural products that have evolved to
substantiated.49 A model in which elevation target microorganisms within biofilms. Certain
of EPS biosynthesis is coincident with higher- marine algae produce and secrete compounds
order structure development has been widely known as halogenated furanones, which have
espoused, although there is as yet limited evi- been shown to inhibit the quorum-sensing pro-
dence to support this model. cess in several different bacteria.54 Specific syn-
thetic derivatives of the halogenated furanones,
Quorum-sensing control of biofilm found to be potent inhibitors of P. aeruginosa
formation biofilm formation, were even effective at limit-
Biofilm formation in certain bacteria, such as ing lung infections in a mouse model.55 These
P. aeruginosa and species of Burkholderia, is compounds are, however, relatively toxic, and
controlled at least in part by a form of intercel- thus their use in therapeutic applications is
lular communication called quorum sensing probably limited. However, applications such
(see Chapter 22: Fig. 22.4).50 In gram-negative as incorporation into marine paints that resist
bacteria, diffusible signal molecules known as microbial fouling have significant potential.
acylated homoserine lactones (AHLs) control The recently described dispersal factor cis-2-de-
a variety of complex microbial behaviors. For cenoic acid is also an excellent candidate as an
certain bacteria, biofilms of mutants unable to antibiofilm agent because it dissociates a wide
synthesize specific AHLs are reported to lack diversity of biofilms.38 In this case, the bacteria
higher-order structure and are readily dispersed are hypothesized to actively disperse in response
by mild antimicrobial treatment.51,52 For P. to the signal, rather than succumbing to toxifi-
aeruginosa, quorum-sensing regulation of bio- cation. Biofilms of the gram-positive pathogen
film formation was shown to function through Staphylococcus aureus are effectively inhibited
control of swarming motility.32 Whether quo- and dissociated by enzymatic activity, specifi-
rum sensing plays a direct role in establish- cally a serine protease, presumably reflecting
ing biofilms appears to vary from microbe to cleavage of an important matrix protein.56 For
microbe, although it is very likely that biofilms other systems, addition of the enzymes such
provide the appropriately high populations that as cellulase and DNase can effectively dissoci-
result in the induction of many quorum-sensing ate biofilms.18 These enzymatic approaches are
regulated genes. attractive for certain applications in which their
continued activity is desirable but are not yet
21.6 Inhibition of Biofilm Formation practical for medically relevant biofilms. The
The overall importance of microbial biofilms in use of bacteriophage has been proposed as an
many different areas, including human health, effective way to target specific bacteria within
agriculture, and industry, has generated great biofilms, although phage resistance may limit
interest in the development of strategies to con- the utility of this approach.57
trol or inhibit biofilm growth.53 Ideally, benefi- Another consideration in the design of
cial and benign biofilms can be fostered, whereas approaches to inhibiting biofilms is the release
problematic biofilms can be targeted for removal of bacteria into the planktonic phase, either
or treatment. Thus far, most efforts have been as single cells or as aggregates. These bacteria
focused on the inhibition of detrimental biofilms. may be primed for subsequent colonization
However, owing to the complexity of biofilms, elsewhere and therefore are highly likely to cre-
particularly in environments outside the labora- ate a new problem at one or more additional
tory, it is unlikely that a single strategy will be sites. Therefore, combination approaches that
effective for controlling all biofilms. Given the actively block biofilm formation, or dissoci-
intrinsic resistance of bacteria within biofilms, ate biofilms in the presence of an antimicrobial
standard antibiotic approaches have limited agent that can subsequently incapacitate the
efficacy: Many surface compositions have been planktonic bacteria, may be most desirable.
tested for the prevention of biofilm formation Although biofilm-targeted approaches have
but also have met with limited success—bacterial great potential, these complex structures have
biofilms continue to establish and thrive. thus far defied a single, simple solution.
21.7 Evolutionary Processes in occurs in biofilms and can actually be facili-

Biofilms tated in certain cases.58 Conjugation is often
the mechanism by which genes for antibiotic
Biofilms are densely populated, with a range
resistance are transmitted, harbored on plas-
of microenvironments. The selective pressures
mids or integrative and conjugative elements
experienced by bacteria residing within biofilms
(ICE elements, conjugative transposons). Some
may vary tremendously over very small spatial
plasmids and ICEs are quite large and encode
scales. These attributes lead to conditions that
a broad host range of conjugal functions.
are conducive to genetic exchange, and also are
Transfer of these elements would potentially
highly selective for subpopulations of bacteria,
represent a potent mechanism for acquisition
thus promoting microbial diversification.
of new capabilities by diverse recipient bacte-
ria. Enhanced conjugation is thought to result
21.7.1 Horizontal gene transfer not only from high cell density, but also because
Very rarely do planktonic microorganisms in the bacteria are fixed spatially in close proxim-
the natural environment achieve the same cell ity to one another. One study demonstrated
densities found in biofilms. In terms of micro- that E. coli-bearing conjugative plasmids were
bial evolution, this could have significant con- more efficient at surface colonization and bio-
sequences. One such consequence is increased film formation than the use of an isogenic strain
horizontal gene transfer via the well-established lacking the plasmid.59 It was hypothesized that
routes of transformation, conjugation, and the conjugative plasmids encode functions that
transduction—an important issue in light of direct the host cell to form a biofilm—a lifestyle
the spread of antimicrobial resistance among that would favor further transmission of the
microbial pathogens (Fig. 21.3). plasmid.
It has long been known that the extracellular Another potential mechanism for horizontal
matrix of biofilms contains significant amounts gene transfer is phage transduction. This usu-
of nucleic acids, and more recent work sug- ally results from faulty excision of integrated
gests that DNA can play a structural role in prophage DNA from the host cell chromosome.
biofilms.18 One can comfortably argue that the The phage would then infect a new cell and insert
high concentration of bacteria in a biofilm will small pieces of DNA from the previous host into
result in an unusually high local concentration the chromosome of its new host. There are large
of extracellular DNA. This provides a fertile numbers of bacteriophages in aquatic systems,
ground for horizontal gene transfer through and some have distinct mechanisms for dealing
normal transformation mechanisms, such as with biofilm bacteria such as depolymerases,
competence. Along these lines, several labora- which degrade the extracellular matrix. An
tories have also demonstrated that conjugation interesting examination by Whiteley et al.,
biofilm
matrix
recipient cell co
transduction nju
ga
tio
n
transformation
Fig. 21.3 Illustration of horizontal gene transfer mechanisms fostered within a biofilm structure. Biofilm
matrix material is depicted in light gray. (Cells and bacteriophage are not drawn to scale.)
www.ebook777.com
using a microarray analysis of P. aeruginosa aeruginosa also give rise to phenotypic variants,
biofilms, demonstrated that the most strongly most readily identified with different colony
induced genes in biofilms relative to planktonic morphotypes, but also manifesting other vari-
growth are those encoding a temperate phage ant attributes such as motility and exopolysac-
homologous to the filamentous bacteriophage charide production.64,65 The determination of
Pf1.60 This is also the phage implicated in bio- whether these variants have enhanced fitness
film dissolution in other studies, and so it is properties within the biofilm environment, and
clear that this phage is active in biofilms.39 The whether they manifest emergent properties such
extent to which transduction and the other gene as increased virulence, remains an area of great
transfer mechanisms contribute to horizontal interest.
gene movement in biofilm communities is prob-
ably community specific; however, it is likely 21.8 Summary
that each mechanism is potentially enhanced in
Biofilms represent a highly prevalent and impor-
the context of a biofilm.
tant form of microbial growth in natural envi-
ronments. Most of what we know regarding
21.7.2 Genetic diversification within microbial physiology has however been deter-
biofilms mined by studying dispersed cultures in the labo-
Biofilms appear to be very active in the selec- ratory, despite the realization that a significant
tion of new genetic variants, perhaps owing fraction of microorganisms reside in biofilm-
to the wide range of microenvironments and type communities. There thus remains a great
their associated selective pressures that occur deal to be learned. Biofilms represent challeng-
as a result of the chemical and physical gradi- ing yet fascinating systems to study: they are
ents in a biofilm. Besides the obvious gradients heterogeneous and complex, both physically
of nutrients, gradients of a wide range of com- and chemically. A flurry of work over recent
pounds produced by the microorganisms them- years has led to an increasing understanding
selves (e.g. secondary metabolites or signaling of the basic mechanisms of biofilm formation,
molecules) would exist.12 These selective pres- although there is tremendous diversification
sures could foster significant genotypic diver- among different microbes. Biofilm structures
sity within biofilm populations. The end result can vary among microorganisms and in the same
would be the association of specific phenotypes microbe as a result of changes in environmental
with particular microniches within a biofilm. conditions. Most natural biofilms are comprised
This concept was explored by Remold et. al., of multiple, often times unrelated bacterial spe-
who generated a number of isogenic, random, cies. Microbiologists are still learning the rules
single-transposon insertion mutants in an E. coli for how, why, and when biofilms develop for
REL458 parent that was specifically adapted to specific microbes and in specific environments.
static liquid culture conditions.61 These random The regulation of biofilm formation and disso-
mutants exhibited wide ranges in their capabil- lution are under active study, motivated by the
ity to adapt to shifts in carbon source, demon- realization that identification of these pathways
strating that even single mutations can have a will lead to strategies for controlling biofilm
profound impact on an organism’s ability to growth. There are a variety of current strategies
adapt to a specific niche. Standing liquid cul- for inhibiting biofilm formation, although these
tures of Pseudomonas fluorescens SBW25 were have not realized their potential, and few potent
shown to produce distinct colony-morphology- antibiofilm approaches are in use. Biofilms repre-
phenotypic variants associated with specific sent complex communities that contain diverse
spatial regions or niches of the culture.62 These microorganisms and are a fertile environment for
phenotypic variants were hypothesized to be evolution of new genetic variants via mutation
specifically adapted to successfully compete in and horizontal gene transfer. The importance of
the niche from which they were isolated. One of biofilms as reservoirs for genetic diversification
these variants, termed the “wrinkly spreader,” and perhaps emergent disease remains unclear.
demonstrated elevated biofilm formation Overall, the study of biofilms, which has caused
and was shown to involve production of the microbiologists to rethink or at least modify their
exopolysaccharide cellulose.63 Biofilms of P. perspective on microbial systems, has developed
into a tremendously dynamic topic of study, ripe 2. Kjelleberg, S. and M. Givskov. 2007. The Biofilm
for innovative approaches. Mode of Life: Mechanisms and Adaptations. Horizon
Biosciences, Norfolk, U.K.
3. Beech, I. B., and J. Sunner. 2004. Biocorrosion:
Study Questions towards understanding interactions between biofilms
and metals. Curr. Opin. Biotechnol. 15:181–186.
1. How does growth on a surface differ from 4. Wyndam, R. C., and K. J. Kennedy. 1995.
growth in suspended laboratory culture? Microbial consortia in industrial wastewater treat-
ment, pp. 183–195 In: Microbial Biofilms, H. M.
2. Why is the study of biofilms more challeng- Lappin-Scott and J. W. Costerton (Eds.). University
ing than the study of laboratory culture? Press, Cambridge, U.K.
5. Davey, M. E., and G. A. O’Toole. 2000. Microbial
3. Identify several benefits of biofilm forma- biofilms: from ecology to molecular genetics.
tion for microbes. Are there detrimental Microbiol. Mol. Biol. Rev. 64: 847–867.
costs to microbes that form biofilms?
6. Entis, P. 2007. Food Safety: Old Habits, New
4. Are there microbial biofilms that have a Perspectives. ASM Press, Washington, DC.
direct impact on human health? 7. Kolenbrander, P. E. 2000. Oral microbial com-
munities: biofilms, interactions, and genetic systems.
5. Are there biofilms that human society could Annu. Rev. Microbiol. 54: 413–437.
consider to be beneficial?
8. Waldvogel, F. A., and A. L. Bisno. 2000. Infections
6. How does residence within a biofilm enhance Associated with Indwelling Medical Devices, 3rd ed.
the tolerance of microbes to antibiotics? ASM Press, Washington, DC.
9. George, A. M., P. M. Jones, and P. G. Middleton.
7. What are some of the mechanisms by which 2009. Cystic fibrosis infections: treatment strate-
microbes initially attach to surfaces? gies and prospects. FEMS Microbiol. Lett. 300:
153–164.
8. For motile bacteria, active locomotion
promotes biofilm formation. To be sessile, 10. Hassett, D. J., T. R. Korfhagen, R. T. Irvin, M.
however, motility must be shut down after J. Schurr, K. Sauer, G. W. Lau, M. D. Sutton, H. Yu,
and N. Hoiby. 2010. Pseudomonas aeruginosa bio-
surface colonization. Explain how this film infections in cystic fibrosis: insights into patho-
might be achieved. genic processes and treatment strategies. Expert
Opin. Ther. Targets 14:117–130.
9. How would one define maturation of a bio-
film? What cellular and extracellular com- 11. Lawrence, J. R., D. R. Korber, B. D. Hoyle, J.
ponents are involved in biofilm formation? W. Costerton, and D. E. Caldwell. 1991. Optical
sectioning of microbial biofilms. J. Bacteriol. 173:
10. How does biofilm dissolution occur? Are 6558–6567.
there active mechanisms that promote 12. Stewart, P. S. 2003. Diffusion in biofilms. J.
dispersal? Bacteriol. 185:1485–1491.
11. What processes might regulate biofilm for- 13. Stoodley, P., Z. Lewandowski, J. D. Boyle, and
H. M. Lappin-Scott. 1999. Structural deformation of
mation? Is quorum sensing involved and, if
bacterial biofilms caused by short-term fluctuations
so, how common is it? in fluid shear: an in situ investigation of biofilm rheol-
ogy. Biotechnol. Bioeng. 65:83–92.
12. Why is regulation of exopolysaccharide
production often connected to biofilm 14. de Beer, D., P. Stoodley, F. Roe, and Z.
formation? Lewandowski. 1994. Effects of biofilm structure on
oxygen distribution and mass transport. Biotechnol.
13. Why would biofilms promote the forma- Bioeng. 43:1131–1148.
tion of genetic variants? What are the 15. Stoodley, P., Z. Lewandowski, J. D. Boyle, and
mechanisms by which this might occur? H. M. Lappin-Scott. 1998. Oscillation characteris-
tics of biofilm streamers in turbulent flowing water
as related to drag and pressure drop. Biotechnol.
Bioeng. 57:536–544.
16. Hansen, S. K., P. B. Rainey, J. A. Haagensen, and
1. Ghannoum, M., and G. A. O’Toole. 2004. S. Molin. 2007. Evolution of species interactions in a
Microbial Biofilms. ASM Press, Washington, DC. biofilm community. Nature 445:533–536.
www.ebook777.com
17. Branda, S. S., S. Vik, L. Friedman, and R. dispersal in sessile microbial communities: examples
Kolter. 2005. Biofilms: the matrix revisited. Trends from Pseudomonas aeruginosa and Pseudomonas
Microbiol. 13:20–26. putida model biofilms. FEMS Microbiol. Lett.
261:1–11.
18. Whitchurch, C. B., T. Tolker-Nielsen, P. C.
Ragas, and J. S. Mattick. 2002. Extracellular DNA 32. Shrout, J. D., D. L. Chopp, C. L. Just, M.
required for bacterial biofilm formation. Science Hentzer, M. Givskov, and M. R. Parsek. 2006. The
295:1487. impact of quorum sensing and swarming motility
on Pseudomonas aeruginosa biofilm formation is
19. Borlee, B. R., A. D. Goldman, K. Murakami, R.
nutritionally conditional. Mol. Microbiol. 62:1264–
Samudrala, D. J. Wozniak, and M. R. Parsek. 2010.
1277.
Pseudomonas aeruginosa uses a cyclic-di-GMP-reg-
ulated adhesin to reinforce the biofilm extracellular 33. Klausen, M., A. Heydorn, P. Ragas, L.
matrix. Mol. Microbiol. 75:827–842. Lambertsen, A. Aaes-Jorgensen, S. Molin, and
T. Tolker-Nielsen. 2003. Biofilm formation by
20. Stewart, P. S. 2002. Mechanisms of antibiotic
Pseudomonas aeruginosa wild type, flagella and type
resistance in bacterial biofilms. Int. J. Med. Microbiol.
IV pili mutants. Mol. Microbiol. 48:1511–1524.
292:107–113.
34. Jefferson, K. K. 2009. Poly-N-acetyl-
21. Anderson, G. G., and G. A. O’Toole. 2008.
glucosamine as a mediator of bacterial biofilm for-
Innate and induced resistance mechanisms of bac-
mation, pp. 175–186 In: Bacterial Polysaccharides:
terial biofilms. Curr. Top. Microbiol. Immunol.
Current Innovations and Future Trends. M. Ullrich
322:85–105.
(Ed.). Caister Academic Press, Bremen, Germany.
22. O’Toole, G., H. B. Kaplan, and R. Kolter. 2000.
35. Barken, K. B., S. J. Pamp, L. Yang, et al. 2008.
Biofilm formation as microbial development. Annu.
Roles of type IV pili, flagellum-mediated motility
Rev. Microbiol. 54:49–79.
and extracellular DNA in the formation of mature
23. Sauer, K., A. K. Camper, G. D. Ehrlich, J. W. multicellular structures in Pseudomonas aeruginosa
Costerton, and D. G. Davies. 2002. Pseudomonas biofilms. Environ. Microbiol. 10:2331–2343.
aeruginosa displays multiple phenotypes during devel-
36. Davey, M. E., N. C. Caiazza, and G. A. O’Toole.
opment as a biofilm. J. Bacteriol. 184:1140–1154.
2003. Rhamnolipid surfactant production affects
24. O’Toole, G. A., and R. Kolter. 1998. Flagellar biofilm architecture in Pseudomonas aeruginosa
and twitching motility are necessary for Pseudomonas PAO1. J. Bacteriol. 185:1027–1036.
aeruginosa biofilm development. Mol. Microbiol.
37. Stoodley, P., S. Wilson, L. Hall-Stoodley, J. D.
30:295–304.
Boyle, H. M. Lappin-Scott, and J. W. Costerton.
25. Patti, J. M., B. L. Allen, M. J. McGavin, and 2001. Growth and detachment of cell clusters from
M. Hook. 1994. MSCRAMM-mediated adher- mature mixed-species biofilms. Appl. Environ.
ence of microorganisms to host tissues. Annu. Rev. Microbiol. 67:5608–5613.
38. Davies, D. G., and C. N. Marques. 2009. A fatty
26. Hinsa, S. M., M. Espinosa-Urgel, J. L. Ramos, acid messenger is responsible for inducing dispersion
and G. A. O’Toole. 2003. Transition from reversible in microbial biofilms. J. Bacteriol. 191:1393–1403.
to irreversible attachment during biofilm formation
39. Webb, J. S., L. S. Thompson, S. James, T.
by Pseudomonas fluorescens WCS365 requires an
Charlton, T. Tolker-Nielsen, B. Koch, M. Givskov,
ABC transporter and a large secreted protein. Mol.
and S. Kjelleberg. 2003. Cell death in Pseudomonas
aeruginosa biofilm development. J. Bacteriol.
27. Tomlinson, A. D., and C. Fuqua. 2009. 185:4585–4592.
Mechanisms and regulation of polar surface attach-
40. Berne, C., D. T. Kysela, and Y. V. Brun. 2010.
ment in Agrobacterium tumefaciens. Curr. Opin.
A bacterial extracellular DNA inhibits settling of
Microbiol. 12;708–714.
motile progeny cells within a biofilm. Mol. Microbiol.
28. Tsang, P. H., Li, G., Brun, Y. V., Freund, L. B. J. 77:815–829.
X. Tang. 2006. Adhesion of single bacterial cells in
41. Kuchma, S. L., and G. A. O’Toole. 2000. Surface-
the micronewton range. Proc. Natl. Acad. Sci. USA
induced and biofilm-induced changes in gene expres-
103:5764–5768.
sion. Curr. Opin. Biotechnol. 11:429–433.
29. Blair, K. M., L. Turner, J. T. Winkelman, H. C.
42. Monds, R. D., and G. A. O’Toole. 2009. The
Berg, and D. B. Kearns. 2008. A molecular clutch dis-
developmental model of microbial biofilms: ten
ables flagella in the Bacillus subtilis biofilm. Science
years of a paradigm up for review. Trends Microbiol.
320:1636–1638.
17:73–87.
30. Hengge, R. 2009. Principles of c-di-GMP signal-
43. O’Toole, G. A., K. A. Gibbs, P. W. Hager, P. V.
ling in bacteria. Nat. Rev. Microbiol. 7:263–273.
Phibbs, Jr., and R. Kolter. 2000. The global carbon
31. Klausen, M., M. Gjermansen, J. U. Kreft, and T. metabolism regulator Crc is a component of a signal
Tolker-Nielsen. 2006. Dynamics of development and transduction pathway required for biofilm devel-
opment by Pseudomonas aeruginosa. J. Bacteriol. homoserine lactone-mediated prokaryotic signal-

182:425–431. ling. J. Bacteriol. 178:6618–6622.
44. Kearns, D. B., F. Chu, S. S. Branda, R. Kolter, 55. Hentzer, M., H. Wu, J. B. Andersen, et al. 2003.
and R. Losick. 2005. A master regulator for bio- Attenuation of Pseudomonas aeruginosa virulence by
film formation by Bacillus subtilis. Mol. Microbiol. quorum sensing inhibitors. EMBO J. 22:3803–3815.
55:739–749.
56. Iwase, T., Y. Uehara, H. Shinji, A. Tajima,
45. Hentzer, M., G. M. Teitzel, G. J. Balzer, A. H. Seo, K. Takada, T. Agata, and Y. Mizunoe.
Heydorn, S. Molin, M. Givskov, and M. R. Parsek. 2010. Staphylococcus epidermidis Esp inhibits
2001. Alginate overproduction affects Pseudomonas Staphylococcus aureus biofilm formation and nasal
aeruginosa biofilm structure and function. J. colonization. Nature 465:346–349.
Bacteriol. 183:5395–5401.
57. Fu, W., T. Forster, O. Mayer, J. J. Curtin, S. M.
46. Nivens, D. E., D. E. Ohman, J. Williams, and M. J. Lehman, and R. M. Donlan. 2009. Bacteriophage
Franklin. 2001. Role of alginate and its O-acetylation cocktail for the prevention of biofilm formation by
in formation of Pseudomonas aeruginosa microcolo- Pseudomonas aeruginosa on catheters in an in vitro
nies and biofilms. J. Bacteriol. 183:1047–1057. model system. Antimicrob. Agents Chemother.
54:397–404.
47. Hickman, J. W., and C. S. Harwood. 2008.
Identification of FleQ from Pseudomonas aerugi- 58. Hausner, M., and S. Wuertz. 1999. High rates
nosa as a c-di-GMP-responsive transcription factor. of conjugation in bacterial biofilms as determined
Mol. Microbiol. 69:376–389. by quantitative in situ analysis. Appl. Environ.
Microbiol. 65:3710–3713.
48. Yildiz, F. H., N. A. Dolganov, and G. K.
Schoolnik. 2001. VpsR, a member of the response 59. Ghigo, J. M. 2001. Natural conjugative plas-
regulators of the two-component regulatory systems, mids induce bacterial biofilm development. Nature
is required for expression of vps biosynthesis genes 412:442–445.
and EPS(ETr)-associated phenotypes in Vibrio chol-
60. Whiteley, M., M. G. Bangera, R. E. Bumgarner,
erae O1 El Tor. J. Bacteriol. 183:1716–1726.
M. R. Parsek, G. M. Teitzel, S. Lory, and E. P.
49. Davies, D. G., A. M. Chakrabarty, and G. G. Greenberg. 2001. Gene expression in Pseudomonas
Geesey. 1993. Exopolysaccharide production in bio- aeruginosa biofilms. Nature 413:860–864.
films: substratum activation of alginate gene expres-
61. Remold, S. K., and R. E. Lenski. 2001.
sion by Pseudomonas aeruginosa. Appl. Environ.
Contribution of individual random mutations to
Microbiol. 59:1181–1186.
genotype-by-environment interactions in Escherichia
50. Fuqua, C., and E. P. Greenberg. 2002. Listening coli. Proc. Natl. Acad. Sci. USA 98:11388–11393.
in on bacteria: acylhomoserine lactone signalling.
62. Rainey, P. B., and M. Travisano. 1998. Adaptive
Nat. Rev. Mol. Cell Bio. 3:685–695.
radiation in a heterogeneous environment. Nature
51. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. 394:69–72.
Iglewski, J. W. Costerton, and E. P. Greenberg. 1998.
63. Spiers, A. J., J. Bohannon, S. M. Gehrig, and
The involvement of cell-to-cell signals in the develop-
P. B. Rainey. 2003. Biofilm formation at the air–
ment of a bacterial biofilm. Science 280:295–298.
liquid interface by the Pseudomonas fluorescens
52. Huber, B., K. Riedel, M. Hentzer, A. Heydorn, A. SBW25 wrinkly spreader requires an acetylated form
Gotschlich, M. Givskov, S. Molin, and L. Eberl. 2001. of cellulose. Mol. Microbiol. 50:15–27.
The cep quorum-sensing system of Burkholderia
64. Boles, B. R., M. Thoendel, and P. K. Singh. 2004.
cepacia H111 controls biofilm formation and swarm-
Self-generated diversity produces “insurance effects”
ing motility. Microbiology 147:2517–2528.
in biofilm communities. Proc. Natl. Acad. Sci. USA
53. Costerton, J. W. and P. S. Stewart. 2001. Battling 101:16630–16635.
biofilms. Sci. Am. 285:74–81.
65. Kirisits, M. J., L. Prost, M. Starkey, and M. R.
54. Givskov, M., R. de Nys, M. Manefield, L. Gram, Parsek. 2005. Characterization of colony morphol-
R. Maximilien, L. Eberl, S. Molin, P. D. Steinberg, ogy variants isolated from Pseudomonas aeruginosa
and S. Kjelleberg. 1996. Eukaryotic interference with biofilms. Appl. Environ. Microbiol. 71:4809–4821.
www.ebook777.com
22
Cell–Cell Communication Mechanisms
What many students will learn from this AHLs, and amino acids, and the processes that
chapter is that prokaryotes signal each other are regulated by the signaling. It is very impor-
and that the signaling is important for their tant to realize that for signal transmission to
survival. This chapter describes these signal- occur, each signaling molecule must have a spe-
ing systems and how they work. As we shall cific receptor on the responding cell. Thus, there
learn, cell signaling can involve diffusible sig- is also a wide diversity of receptor molecules.
nal molecules that travel in the environment Also, very importantly, as a consequence of sig-
between cells, as well as cell-associated signal reception, there occurs a signaling cascade
nals that require intimate cell–cell contact. For within the cell that alters gene expression and/
reviews, see refs. 1 through 8. For a summary of or cellular behavior. All of this will be described
early discoveries of prokaryote cell–cell com- in this chapter.
munication, see Box 22.1. The soluble signals
transferred between cells are often referred to
as pheromones. What is a pheromone? By defi- 22.2 Specific Signaling Systems
nition, a pheromone is any chemical secreted This section summarizes the many signaling
into the environment that elicits a behavioral or systems that prokaryotes use to communicate
developmental response from other organisms. with each other. We will begin with oligopep-
Pheromones usually affect cells of the same spe- tide pheromones.
cies and aid in the coordination of activities. As
we shall see, pheromones include amino acids, 22.2.1 Oligopeptide pheromones
lipids, and oligopeptides.
Oligopeptide pheromones can serve as signals.
Commonly found in the gram-positive bacteria,
22.1 Diversity of Diffusible Signal they are small peptides ranging in size from 5 to
Molecules Produced by Bacteria 40 amino acids. The oligopeptide pheromones
The signaling molecules produced by bacteria are commonly processed from larger, riboso-
encompass a wide range of chemical structures mally synthesized peptides prior to or during
that regulate functions including DNA exchange export (Table 22.2).
between bacteria, animal pathogenesis, antibi-
otic production, bioluminescence, and bacte- Location of peptide pheromone receptors
rial development, as well as other activities To have an effect, the peptide pheromone must
necessary for the survival of bacterial popula- bind to a receptor on the recipient cell. In some
tions. By examining Table 22.1, the student will cases the receptor is on the cell surface, and
notice the diversity of the signaling molecules, the pheromone does not enter the cell. Binding
which include oligopeptides, g-butyrolactones, of the pheromone to the receptor generates a
566
cell–cell communication mechanisms 567
BOX 22.1. EARLY DISCOVERIES OF PROKARYOTIC

CELL–CELL COMMUNICATION
The idea of intercellular signaling among observations have proven to be characteristic
prokaryotes is not new. As early as 1905, the of widespread chemical communication sys-
plant pathologist Erwin Smith hypothesized tems shared by other microbes, and relevant
that microbes might interact to become more to a wide range of processes including antibi-
than the sum of their parts, through coordi- otic production, infectious disease, and sym-
nated action. Experimental evidence for such biosis. These early studies have profoundly
mechanisms however, required many years influenced the more recent explosion of work
of work. In 1965 studies on the development on microbial chemical communication.
of genetic competence in Streptococcus pneu-
moniae revealed the activity of a microbial REFERENCES
signal (now known to be an oligopeptide) 1. Tomasz, A. 1965. Control of the competent
important for the timing of competence.1 state in Pneumococcus by a hormone-like cell
Likewise, in 1967, a soluble signaling mole- product: an example of a new type of regulatory
cule called A-factor was identified as essential mechanism in bacteria. Nature 208, 155-159.
for the production of the antibiotic streptomy- 2. Khoklov, A.S., I. I. Tovarova, L.N. Borisova,
cin by the soil microbe Streptomyces griseus.2 S. A. Pliner, L. A. Schevchenko, E.Y. Kornitskaya,
In 1970 Nealson and Hastings reported the N. S. Ivkina, and I.A Rapoport. 1967. A factor
responsible for the biosynthesis of streptomycin
phenomenon of autoinduction for control of by a mutant strain of Actinomyces streptomy-
bioluminescence in Vibrio fischeri, a symbi- cini. Dokl. Akad. Nauk SSSR 177:232–322.
ont of marine animals, whereby the bacteria
3. Nealson, K.H., T. Platt, and J. W. Hastings.
produce a soluble factor (now known to be 1970. Cellular conrol of the synthesis and
an acylated homoserine lactone, or AHL) activity of the bacterial luminescent system.
that activates light production.3 All these J. Bacteriol. 104:313–322.
Table 22.1 Signal molecules and regulated processes
Bacteria Signal molecule Process regulated
Gram-positive
Bacillus subtilis Oligopeptides DNA uptake (competence),
sporulation
Enterococcus faecalis Oligopeptides DNA exchange (conjugation)
Lactic acid bacteria Oligopeptides Bacteriocin production
Staphylococcus aureus Oligopeptides Mammalian pathogenesis
Streptococcus pneumoniae Oligopeptides DNA uptake, pathogenesis
Streptomyces griseus g-Butyrolactones Antibiotic production,
secondary metabolism
Gram-negative
Vibrio fischeri AHLs Bioluminescence
Vibrio harveyi AHL, AI-2 type signal Bioluminescence
Vibrio cholerae AI-2 type, AHK (CAI)- type Virulence and biofilms
Agrobacterium tumefaciens AHL DNA exchange (conjugation)
Pseudomonas aeruginosa AHLs, quinolones Animal pathogenesis
Myxococcus xanthus Amino acids Fruiting body development
Xanthomonas campestris Diffusible signal factor (DSF, Plant pathogenesis
pv. campestris α,β-unsaturated fatty acid)
Salmonella typhimurium AI-2 Animal pathogenesis
Proteus mirabilis Putrescine Swarming motility
Cyanobacteria
Anabaena spp. PCC7120 Oligopeptide Cell differentiation
Source: Adapted from Dunny, G. M., and S. C. Winans. 1999. Cell–Cell Signaling in Bacteria. ASM Press, Washington, DC.
www.ebook777.com
Table 22.2 Examples of oligopeptide pheromones

Bacteria Peptide name Amino acids Modification
mature (precursor)
Lactococcus lactis Nisin A 34 (57) Lanthionine and
methyl–lanthionine
rings, dihydroalanine and
dihydrobutyrate residues
Streptococcus pneumoniae CSP (ComC) 17 (41) Unmodified
Bacillus subtilis ComX 10 (55) Lipid—likely farnesylation of
CSF 5 (40) Trp residue
Unmodified
Staphylococcus aureus AgrD (AIP) 8 (45) Cyclization—internal
thioester (Cys4–Met8)
Enterococcus faecalis CF10 7 (23) Unmodified
iCF10 7 (21) Unmodified
Anabaena sp. PCC 7120 PatS > 5 (17) Unmodified
Source: Adapted from Dunny, G. M., and B. A. B. Leonard. 1997. Cell–cell communication in gram-positive bacteria.
transmembrane signal that initiates a signal cas- Peptide pheromones that initiate two-
cade within the cell. Other peptide pheromones component signaling cascades
must enter the cell to have an effect. The phero- Examples of two-component systems that
mones that enter the cells do so via membrane- respond to peptide pheromones include spor-
associated transporters, usually oligopeptide ulation in Bacillus subtilis, the development of
permeases, and subsequently bind to a cytoplas- competence (DNA uptake) in B. subtilis and
mic receptor, thereby initiating the signal cascade. Streptococcus pneumoniae, and the regulation of
The cytoplasmic receptor might be an enzyme, pathogenicity by Staphylococcus aureus. Figure
or a transcription factor that regulates the tran- 19.1 and Sections 19.1 and 19.1.1 explain how
scription of specific genes, and may be activated an extracellular oligopeptide might trigger a two-
or even inactivated by pheromone binding. component system to stimulate the transcription
of specific genes. In this case it is the oligopep-
Intracellular targets of signaling by tide that interacts with the histidine kinase (HK),
oligopeptide pheromones either directly or indirectly. Alternatively, phero-
As with many other pheromones, the target of mones that enter the cell may inhibit the activity
oligopeptide signaling is the control of gene of phosphatases, and in this way also contribute
expression, and ultimately microbial behavior. to increased transcription of specific genes.
The end result of the signal cascade initiated by
most of the well-studied peptide pheromones 22.2.2 Competence and endosporulation
is the enhancement of transcription of specific in B. subtilis
genes, and in some cases the stimulation of Bacillus subtilis is a gram-positive bacterium
translation of specific mRNAs. Pheromones commonly found in soil. It is a well-studied
such as these oligopeptides alter gene expres- developmental model system for competence
sion by affecting the activity of transcription and endospore formation. Natural competence
factors that regulate these genes. The effect of is defined as the ability to bind and transport
oligopeptide pheromones on the transcription exogenous DNA into a recipient cell, and to
factors is rarely direct; usually it is the outcome recombine the DNA into its genome. (Artificial
of a signaling pathway initiated by perception of competence can be forced in some bacteria with
the pheromone. However, as will be described certain chemical and physical treatments in the
later, some oligopeptide pheromones that enter laboratory.) Natural competence is one of the
the cell do interact directly with transcription ways that genes are transferred between closely
factors. Review the discussion of two-com- related bacteria in the natural environment.
ponent phosphorelay systems in Chapter 19, Endospore formation is a process by which a cell
Section 19.1, for a discussion of how extracel- divides asymmetrically, giving rise to two com-
lular signals can impact gene activity. partments, one of which ultimately becomes
a growth-arrested spore, whereas the other the general secretory pathway (GSP), nor does
one lyses. Competence and endospore forma- it possess the double glycine motif required for
tion can be viewed as alternative developmen- processing and export in the ABC transporter
tal fates. They both tend to occur at the end of pathway. (See Chapter. 17 for a discussion
exponential growth when cell densities are high. of export.) Processing and export of ComX
Whether one or the other occurs is regulated in are presumed to require specialized proteins.
part by oligopeptide pheromones secreted by Interestingly, one of the amino acids in ComX, a
the cells into the extracellular medium. tryptophan residue, is modified by lipid attach-
ment likely to be a C15 polyisoprenoid moiety
Competence (farnesylation).10 Similar modification of fungal
There are two competence-inducing peptide pheromones is well documented, except that
pheromones that accumulate in the extracellular usually a cysteine residue is alkylated.11
media at high cell densities.9 These pheromones The response to ComX requires a two-com-
are ComX and CSF (competence and sporula- ponent system consisting of a histidine kinase
tion factor). ComX is a 10 amino acid oligopep- called ComP with six transmembrane seg-
tide processed from the C-terminal end of a 55 ments and a cytoplasmic response regulator
amino acid precursor peptide (Table 22.2). A called ComA. According to this model ComX
50% maximal response is induced at concen- binds to ComP on the outer surface of the inner
trations between 5 and 10 nM of extracellular membrane and thereby initiates a phosphore-
ComX. The precursor protein does not have a lay whereby ComP autophosphorylates and
leader sequence required for export through this phosphoryl group is transferred to ComA.
Fig. 22.1 Model for regulation of competence by oligopeptide pheromones in B. subtilis. Stimulation of com-
petence by ComX and CSF. Com X binds to the ComX sensor/kinase ComP and stimulates it to autophospho-
rylate; as a result, the phosphoryl group is transferred to the response regulator ComA. Autophosporylation
and phosphotransfer is depicted as ATP → ADP. ComA-P activates the transcription of genes in the compe-
tence pathway. CSF enters through an oligopeptide permease and inactivates the RapC phosphatase, which
dephosphorylates ComA-P. Source: White, D., and C. Fuqua. 2004. Prokaryotic intercellular signaling mech-
anistic diversity and unified themes, in Cell Signaling in Prokaryotes and Lower Metazoa, Fairweather, ed.
Kluwer Academic Publishers, Dordrecht, Netherlands, with kind permission from Springer Science+Business
Media B.V.
www.ebook777.com
ComA-P is a transcriptional activator that stim- SpoOA, the primary positive transcriptional
ulates transcription of genes in the competence regulator for sporulation genes (Fig. 22.2).12
pathway. The undefined sporulation signals apparently
CSF is a 5 amino acid oligopeptide that is also stimulate phosphorylation of SpoOA
processed from the C-terminal end of a secreted through activation of the HKs that initiate the
40 amino acid peptide and accumulates in the phosphorelay.
medium as the cells grow to a high cell density.12
(See Table 22.2.) It has been suggested that a Oligopeptide pheromones and Rap
peptide having between 11 and 25 amino acids phosphatases
is processed from the 40 amino acid precursor It is now recognized that B. subtilis can secrete
and is exported via the GSP pathway. Processing a variety of peptide pheromones (Phr peptides)
of the CSF precursor to the 5 amino acid peptide that are related to the oligopeptide signals just
requires an extracellular peptidase bound to the described.13 Phr peptides are also processed
outer cell surface, but the cleavage is still not from precursor proteins secreted across the cyto-
well understood. plasmic membrane, to mature forms with five
Low concentrations (1–5 nM) of CSF stimu- or more amino acids. The peptides function to
late the competence pathway. CSF enters the inhibit Rap phosphatases, enzymes that can have
cell via an oligopeptide permease and inhibits a profound influence on phosphorelay regulatory
phosphatase (RapC phosphatase) that dephos- pathways such as those that control sporula-
phorylates ComA-P. Thus, both ComX and CSF tion. The Phr peptides are genetically encoded in
increase the amounts of ComA-P. However, as the same operons as the Rap phosphatases, and
explained next, as the extracellular concentra- the B. subtilis genome contains seven distinct
tions of CSF surpass a minimum threshold (> 20 Rap/Phr cassettes. These Phr–Rap systems can
nM), then competence is inhibited and sporula- regulate diverse functions. For example, it was
tion is stimulated. discovered that one of these regulates the hori-
zontal gene transfer of a large, integrated genetic
Sporulation element known as ICEBsuI through release and
The extracellular peptide pheromone that stim- subsequent perception of the Phr signal.14 This
ulates the sporulation pathway is CSF, which regulation enables preferential transfer of the
as previously described also stimulates the ICEBsuI element to B. subtilis strains that lack
competence pathway (Fig. 22.1). In addition, the element.
there are as yet undefined sporulation factors
that are also integrated with the effect of CSF. 22.2.3 Competence in Streptococcus
Whether CSF stimulates competence or sporu- pneumoniae
lation depends upon its extracellular concentra- S. pneumoniae, a gram-positive bacterium that
tion. If the extracellular concentration of CSF is found in the upper respiratory tract of many
is sufficiently high (>20 nM), then sporulation, healthy people, is a common cause of middle ear
rather than competence, is stimulated. It has infections, bacterial meningitis in children, and
been suggested that high intracellular levels of pneumonia. As with B. subtilis, competence in
CSF inhibit the histidine kinase ComP, resulting S. pneumoniae is stimulated by an oligopeptide
in lower levels of the phosphorylated ComA-P pheromone that activates a two-component
(Fig. 22.2). The result would be decreased regulatory system.15 The signal transduction
expression of ComA-P dependent genes in the pathway is similar to the generalized pattern
competence pathway. drawn in Fig. 19.1. (See Section 19.1 for a gen-
If this occurs and conditions are appropri- eral description of two-component systems.)
ate for sporulation, then the cells will sporu- The oligopeptide pheromone that stimulates the
late rather than become competent. CSF also competence pathway is a competence-inducing
stimulates sporulation by enhancing phospho- factor called CSP, a 17 amino acid peptide made
transfer (Fig. 22.2). Evidence suggests that CSF from a larger precursor protein and likely pro-
inhibits a phosphatase (RapB phosphatase) cessed during export via an ABC-type permease,
that dephosphorylates SpoOF-P, which is part ComAB. (For a review of this permease and other
of a phosphorelay cascade that phosphorylates protein secretion systems, see Chapter. 18).
Fig. 22.2 CSF sporulation control in Bacillus subtilis. There are at least two histidine kinases: cytoplasmic KinA
and membrane-associated KinB. KinB autophosphorylates, and the phosphoryl group is transferred to a response
regulator protein SpoOF. Autophosporylation and phosphotransfer is depicted as ATP→ADP. The phosphoryl
group is passed from SpoOF to SpoOB, and in turn is transferred SpoOA. SpoOA-P is a transcriptional activa-
tor. In addition to the histidine kinases, three phosphatases regulate the levels of SpoOA-P. RapA and RapB
dephosphorylate SpoOF-P, and SpoOE dephosphorylates SpoOA-P. The oligopeptide pheromone CSF enters
the cell through the oligopeptide permease and inactivates RapB. RapA is regulated in an independent manner.
Adapted from White, D. 2000. The Physiology and Biochemistry of Prokaryotes, 2nd ed. Oxford University
Press, New York. Source: White, D., and C. Fuqua. 2004. Prokaryotic intercellular signaling mechanistic diver-
sity and unified themes, in Cell Signaling in Prokaryotes and Lower Metazoa, Fairweather, ed. Kluwer Academic
Publishers, Dordrecht, Netherlands, with kind permission from Springer Science+Business Media B.V.
CSP accumulates in the extracellular media in the transcription of genes in the competence
proportion to cell density and binds to a recep- pathway.
tor on the surface of target cells. The receptor,
called ComD, is the histidine kinase portion of 22.2.4 Virulence genes in Staphylococcus
a two-component system. The binding of CSP aureus
to ComD stimulates autophosphorylation of S. aureus is a gram-positive pathogenic bacte-
ComD and the subsequent phosphorylation of rium that is responsible for a variety of diseases
the response regulator protein, called ComE. ranging from skin diseases such as impetigo and
The phosphorylated form of ComE stimulates infection of hair follicles (boils) to more serious
www.ebook777.com
infections of the lungs, heart, and blood. The suggested to activate the transcription of genes
bacteria produce a variety of virulence factors that result in the biosynthesis of nisin (see Ref. 16
that contribute to pathogenesis. The expression for a review). Thus, in this case the bacteria use
of many of the virulence genes is dependent upon the bacteriocin to signal the rest of the popula-
the Agr (accessory gene regulation) regulatory tion to make more nisin. There are, however,
system, which responds to an oligopeptide-type variations on this theme. Lactic acid bacteria
quorum-sensing signal.4 Two of the gene prod- that synthesize class II bacteriocins (unmodified
ucts of the Agr system, AgrC and AgrA, comprise peptide antibiotics) secrete peptide pheromones
a two-component regulatory system, AgrC being that activate bacteriocin production but are not
a membrane-bound histidine kinase and AgrA a themselves bacteriocins.
cytoplasmic response regulator. AgrC is activated
by an extracellular oligopeptide pheromone 22.2.6 Enterococcus signaling
of 8 amino acids called AIP for Agr autoinduc- Not all peptide pheromones bind to histidine
ing peptide, which functions in quorum sensing kinase/receptors on the cell surface and stimu-
(Table 22.2). AIPs are called autoinducing pep- late two-component signaling systems, nor do
tides because the first ones ones discovered were they all enter the cell like CSF to prevent the
found to stimulate the expression of genes that dephosphorylation of a response regulator.
encoded them. AIP is processed from the larger One peptide signaling system that does not
AgrD gene product and is internally cyclized, appear to involve a two-component pathway is
containing a thioester bond between the cysteine responsible for mating and plasmid transfer in
residue at position 4 and the carboxy terminus of species of Enterococcus.17
the peptide. The maturation of AIP requires the E. faecalis is a gram-positive bacterium that
AgrB gene product. In the current model, exter- is part of the normal flora in the human intes-
nal AIP binds to AgrC on the cell surface and tine. It can infect the urinary tract and other
stimulates autophosphorylation. The phospho- parts of the body, and is a frequent cause of
ryl group is passed to the cytoplasmic RR AgrA. hospital-acquired infections (nosocomial
AgrA-P does not appear to bind to DNA, but it infections). A strain of E. faecalis harboring a
may stimulate expression by interacting with a particular plasmid can transfer a copy of the
DNA-binding transcription factor. plasmid to a strain that lacks the plasmid via
a process called conjugation or mating. Such
22.2.5 Bacteriocin production plasmid transfer is stimulated by oligopeptide
Many bacteria that live in animals secrete bac- pheromones 7 or 8 amino acids long produced
teriocins, which are antibiotic proteins or pep- by the recipient strain (Table 22.2). In contrast
tides that inhibit or kill other bacteria that may to the pheromone systems described earlier,
be competitors for food or space. For example, stimulation by the enterococcal pheromones
Lactococcus lactis secretes a 34 amino acid bac- does not appear to involve two-component
teriocin called nisin that inhibits the growth of signal relay systems. The enterococci can carry
many different gram-positive bacteria (Table several plasmids belonging to different fami-
22.2). Nisin is considered to be a class I bacte- lies, the transfer of each one of which is respon-
riocin, synthesized as a larger gene product that sive to a specific cognate peptide pheromone.
is processed and post-translationally modified As a consequence of a cell reacting to a peptide
with internal lanthionine bonds (intramolecular pheromone, several genes required for mating
thioether bridges) and dehydroalanine as well and the transfer of the cognate plasmid to a
as dehydrobutyrine residues.16 The production recipient cell are transcribed. Once a cell has
of nisin by L. lactis is stimulated by extracellu- received a plasmid, the production of phero-
lar nisin, which in this context acts like an oli- mone for that particular plasmid is reduced,
gopeptide pheromone. When nisin accumulates although it can continue to secrete pheromones
in the extracellular media, it binds to a mem- for other families of plasmids. This specific inhi-
brane-bound histidine kinase (NisK), activating bition functions through the inhibiton of pher-
it. The activated NisK autophosphorylates and omones by conjugal plasmids. The inhibitors
transfers the phosphoryl group to the response are also oligopeptides, released in excess to the
regulator called NisR. Phosphorylated NisR is endogenously generated inducing pheromone
and thereby preventing self-activation. Mutant which are responsible for the growth and the
donor cells that cannot synthesize the inhibitor production of new filaments. In Anabaena sp.
peptides are activated by their own pheromone PCC 7120, conditions of nitrogen depriva-
signal and are not responsive to recipients. tion cause proheterocyst cells to synthesize a
A model has been proposed for the func- 17 amino acid oligopeptide called PatS, which
tion of the peptide pheromone cAD1 that is trimmed, retaining the C-terminal end (at
stimulates the transfer of plasmid pAD1.18 least the terminal 5 amino acid residues) that
The pheromone is transported into the cell by comprises the oligopeptide signaling molecule
a membrane-associated permease. Once inside called PatS (Table 22.2).20 PatS diffuses into
the cell, the oligopeptide binds to a negative adjacent cells and prevents them from devel-
transcription regulator called TraA, inactivat- oping into heterocysts. Mutant strains unable
ing it and thus derepressing the transcription to produce PatS make numerous heterocysts
of genes that promote mating and plasmid unevenly spaced even when grown in the pres-
transfer. One of the genes whose transcription ence of nitrogen.
is enhanced encodes TraE1, a positive regula- To account for the distinct pattern (one in
tor of the transcription of genes required for roughly every 10 cells) of heterocysts found
pAD1 conjugation. Thus the pheromone con- in wild-type filaments, a model has been pro-
trols conjugation by stimulating the synthesis posed postulating that PatS diffuses through
of a positive transcription factor for the expres- the filament between the plasma membrane
sion of genes required for mating and plasmid and the outer portion of the cell wall (the
transfer. It must be stated, however, that the periplasmic space). This creates a gradient
molecular control mechanisms are not the originating at the proheterocyst and dissipat-
same for all of the pheromones and their cog- ing laterally along the filament. This type of
nate plasmids. For example, in the case of the signaling differs from those discussed previ-
plasmid pCF10 and its cognate pheromone, ously in that the signal acts specifically on
regulation includes control at the translational cells within the same filament, and not on cells
level.17 It is not clear why different E. faecalis outside the filament. In this respect, it is more
plasmids have evolved such different mecha- like a hormone produced by a multicellular
nisms to transduce oligopeptide signals, rather animal. The signal transduction system is not
than utilize a two-component phosphorelay as known; based upon other known oligopeptide
in other gram-positive bacteria. One unique signaling systems, however, a phosphorelay
requirement in the E. faecalis system is the has been suggested.
necessity to inhibit pheromone release, hence
self-activation in donors as well as recipients 22.2.8 Intercellular signaling in
that receive the plasmid. If one considers the myxobacterial development
donors and recipients to be discrete popula- Myxobacteria are unique among the bacteria
tions, mating pheromone signaling is in fact an in that they form multicellular fruiting bodies
excellent example of alloinduction. A complete (reviews of the biology of myxobacteria can
discussion of pheromones in enterococci can be be found in refs. 21 and 22. A review of signal-
found in ref. 19. ing in myxobacteria can be found in ref. 23).
Chapter 23 (Bacterial Development) provides
22.2.7 Heterocyst formation in a a complete review of the myxobacteria life
cyanobacterium cycle. This section will concentrate on signaling
Species of the genus Anabaena are filamen- between Myxococcus xanthus, using a diffus-
tous photosynthetic cyanobacteria (blue-green ible signal called the A-signal, which initiates
algae) that form nitrogen-fixing cells called het- development. M. xanthus also employs cell–cell
erocysts. The heterocysts, which form in wild- contact signaling, and this will be discussed later
type organisms only when the medium is devoid (Section 22.3). A different genus of myxobac-
of fixed nitrogen, are nondividing cells located teria, the Stigmatella, does not use A-signaling
approximately 10 cell intervals apart in the fila- but does use an excreted lipoidal pheromone for
ment. They supply fixed nitrogen to the rest of early stages of aggregation and fruiting body
the cells (called vegetative cells) in the filament, formation.24
www.ebook777.com
A-Signal reasonable understanding of the bioluminescence

As we will discuss in Chapter 23, myxobacte- control in these microbes (see refs. 28 and 29). V.
ria aggregate and form fruiting bodies when fischeri synthesizes the AHL 3-oxo-hexanoyl-ho-
nutrients are depleted and when the cell den- moserine lactone (3-oxo-C6-HSL) via an enzyme
sity is sufficiently high. A diffusible signal that called LuxI.30,31 This signal molecule is freely
activates developmental genes when nutrients diffusible across the bacterial envelope. At low
are depleted from a dense population of cells population densities, as found when V. fischeri is
is produced by M. xanthus, and it is called the free in seawater, the AHL effectively exits the cell
A-signal.25 The A-signal is a subset of extracellu- down a steep concentration gradient, resulting in
lar amino acids that accumulates in the medium a low intracellular concentration that is insuffi-
about 1 to 2 hours poststarvation, accumulat- cient to induce luminescence. Colonization of the
ing in the medium to reach a threshold level animal host light organ by V. fischeri results in
leading to activation of developmental genes. clonal growth of the population and eventually a
M. xanthus grows on amino acids and pep- high cell density that is paralleled by a concomi-
tides and, if the cells are not starving, then the tant increase in the 3-oxo-C6-HSL concentra-
A-signal amino acids in the medium are above tion. As the diffusion gradient across the cellular
the threshold level for activating developmental envelope is reduced, less AHL is released from
genes and the cells continue to grow. the cell and the intracellular concentration of
How does this all begin? Upon starvation, a pheromone increases, fostering interaction with
signaling pathway within the cells is activated, a cytoplasmic AHL receptor called LuxR. LuxR
and one of the results is the activation of genes binds to 3-oxo-C6-HSL, in turn associates with
responsible for the secretion of proteases. The a DNA regulatory element upstream of the bio-
proteases degrade cell surface proteins and, as a luminescence (lux) genes, activates expression of
result, the subset of amino acids that constitute the downstream operon, and turns on the lights.
the A-signal is released into the medium. When For this microbe, the high population density
the combined concentration of the amino acids provided by the light organ environment leads
of the A-signal exceeds 10 μM, a membrane- to accumulation of the pheromone, and subse-
bound histidine kinase is activated and a signal quent activation of lux genes. The dramatic dif-
cascade within the cells is initiated that results in ference in the V. fischeri populations in the light
the activation of developmental genes required organ compared to those cells in the free-living
for aggregation and subsequently fruiting body state results in a symbiosis-specific regulation
formation. See Chapter 23 (Fig. 23.4). of light production. However, artificially high
population densities, such as those provided in
22.2.9 Acylated homoserine lactones (AHLs) the laboratory, also effectively activate the AHL
AHLs were first identified as signaling mol- regulatory system. Surprisingly, light regulation
ecules in several bioluminescent marine Vibrio in the closely related V. harveyi, while sharing a
species; V. harveyi colonizes the intestines of population density sensing component and uti-
certain fishes and is prevalent in seawater, and lizing an AHL, functions via an entirely distinct
V. fischeri is a symbiont that inhabits the “light mechanism (see later: Section 22.2.11).
organs” of specific fishes and squids and pro-
duces light for its animal host (thought to allow AHLs are a common intercellular signal in
the animal either to identify prey in the dark or, gram-negative bacteria
conversely, to enable the creature itself to avoid Although the original observations of lumines-
predation).26,27 These Vibrio species produce cence regulation in the marine vibrios were made
light in standard laboratory culture, but lumines- in the early 1970s, it was not until 1992 that other
cence is significantly delayed until the late stages bacteria were found to also produce AHLs.32 Since
of culture growth. Light production can be aber- that time there has been a tremendous explosion
rantly induced in fresh cultures by the addition of information on AHLs and on regulators of the
of conditioned medium from dense, biolumines- LuxI–LuxR type (accelerated by the rate of micro-
cent cultures, implicating the presence of one or bial genome sequencing) from many different
more bacterially derived or modified factors that genera of Proteobacteria, and one report of AHL
signal the younger cells to begin producing light. activity from cyanobacteria.33,34 The AHLs syn-
A great deal of work over many years has led to a thesized by these bacteria share the same general
Table 22.3 Examples of AHL quorum-sensing systems

Bacteria Regulatorsa Chain length, B–R groupc Abbreviation
nb
Vibrio fischeri LuxR–LuxI 6 (1) 5O 3-oxo-C6-HSL
AinR–AinSd 8 (2) –H C8-HSL
Vibrio harveyi LuxN–LuxMd 4 (0) –OH 3-OH-C4-HSL
Pseudomonas aeruginosa LasR–LasI 12 (8) 5O 3-oxo-C12-HSL
RhlR–RhlI 4 (0) –H C4-HSL
Agrobacterium tumefaciens TraR–TraI 8 (4) 5O 3-oxo-C8-HSL
Rhodobacter sphaeroides CerR–CerI 14 (10) –H D7-C14-HSLe
Vibrio anguillarum VanR–VanI 10 (6) 5O 3-oxo-C10-HSL
Yersenia enterocolitica YenR–YenI 6 (2) –H, =O C6-HSL
3-oxo-C6-HSL
Sources: For details and references, see Fuqua, C., and A. Eberhard. 1999. Signal generation in autoinduction systems:
synthesis of acylated homoserine lactones by LuxI-type proteins, pp. 211–230. In S. C. Winans and G. M. Dunny (Eds.).
Cell–Cell Signaling in Bacteria. ASM Press, Washington, DC, and Fuqua, C., and E. P. Greenberg. 2002. Listening in on
bacteria: acylhomoserine lactone signalling. Nat. Rev. Mol. Cell Biol. 3:685–695.
a
Quorum-sensing regulatory proteins. Except where indicated, proteins are members of the LuxR–LuxI family; R, recep-
tor/regulator; I/S/M, AHL synthases.
b
Corresponds to n in diagram.
c
Chemical moiety at b position, R in diagram.
d
AinR and LuxN are homologous to two-component sensor kinases, AinS and LuxM are AHL synthases, but not homol-
ogous to LuxI.
e
An unsaturated bond exists between positions 7 and 8 on the acyl chain.
structure as the V. fischeri pheromone, but may signaling for the maturation and maintenance
vary in three respects: (1) length (with an even of these multicellular structures. It is arguable
number of alkyl groups ranging from 4 to 18), that biofilms are the structures for which the
(2) modification at the third (b) position (fully AHL quorum sensors, as well as other quorum-
reduced, hydroxylated, or carrying a carbonyl sensing systems, evolved to function. Indeed,
group), and (3) the presence or absence of unsatu- AHLs have been detected in natural biofilms
rated bonds in the acyl chain (Table 22.3). In most from a wide variety of sources. For several bac-
cases, the AHL is synthesized by an enzyme that teria, quorum sensing plays a role in formation
shares sequence homology with LuxI. Perception of the biofilm. For example, in the well-studied
of the AHL and subsequent transcriptional regu- biofilm former P. aeruginosa, mutants that can-
lation of target genes is also usually mediated by not synthesize 3-oxo-C12-HSL, one of its two
a LuxR homologue. The target functions con- major AHLs, form biofilms that lack any of the
trolled by AHLs and LuxI–LuxR regulatory cir- complex structures of wild-type biofilms and are
cuits are strikingly diverse. More often than not, easily removed from surfaces.35 The ubiquity of
the AHL-regulated circuits are either directly or AHLs as biofilm signal molecules is currently
indirectly involved in interactions between the an area of active study, but it is implausible
microbe and host organisms. to suggest that AHLs play this role in all bio-
films (e.g., those dominated by gram-positive
AHLs in bacterial biofilms microbes). Other intercellular signaling systems
Bacteria grow in structured surface communities probably perform this function along with, or
called biofilms on many if not all nonsterile sur- in place of, AHLs in biofilms of different bacte-
faces (see Chapter 21). Some biofilms uniformly ria. Understanding the role of AHLs and other
cover large surface areas; others comprise dis- signaling molecules in these complex microbial
continuous smaller aggregates commonly called communities is an exciting area of future work.
microcolonies. Individual cells in biofilms are
commonly embedded in exopolymeric matrix A second AHL synthase family
material, where they exist at high population There is at least one additional mechanism for
density. Chapter 21 is devoted to biofilms. Here AHL synthesis which involves enzymes that
we will describe the importance of cell–cell are not directly related to LuxI-type proteins.36
www.ebook777.com
Fig. 22.3 General model for AHL signaling. LuxR-type protein and LuxI-type protein are represented as
spheres labeled R and I, respectively; AHL is a solid circle. Solid arrows indicate noncovalent association
and catalysis; dashed arrow implies that activation of I gene expression is variable among different bacteria.
Squiggles indicate transcription and translation of the I gene. On the generalized AHL structure, R can be
H, OH, or O, and n equals the number of additional two-carbon groups beyond the minimal carbon length
of 4, depicted for the acyl chain, 0–5. Abbrevations: ACP, acyl carrier protein; 5´-MTA; 5´- methylthioad-
enosine; SAM, S-adenosyl–L-methionine. Source: White, D., and C. Fuqua. 2004. Prokaryotic intercellular
signaling mechanistic diversity and unified themes, in Cell Signaling in Prokaryotes and Lower Metazoa,
Fairweather, ed. Kluwer Academic Publishers, Dordrecht, Netherlands, with kind permission from Springer
Science+Business Media B.V.
These were also identified in Vibrio species and example, diketopiperazines, or cyclic dipep-
are called AinS and LuxM, in V. fisheri and V. tides, are compounds produced by a number
harveyi, respectively. In V. harveyi LuxM is one of different bacteria and found in rich culture
component of a dual-channel quorum-sens- media that, at high concentrations, can effec-
ing system, synthesizing 3-OH-C4-HSL, and tively activate several LuxR-type proteins.37 In
thereby influencing bioluminescence as well as combination with AHLs, the diketopiperazines
a variety of other target functions. In V. fischeri can also act as competitive inhibitors. The phys-
AinS is responsible for synthesizing C8-HSL and iological relevance of this observation is not
also influences the LuxR regulatory network of yet clear, but it does have implications for the
this microbe. Neither AHL synthase bears any biochemical specificity of LuxR-type proteins
significant sequence resemblance to LuxI-type for AHLs. Several compounds synthesized by
proteins, but AinS has been shown to synthe- eukaryotes can also impact Lux-type systems,
size C8-HSL in vitro. It is not yet clear how including halogenated furanones produced
widespread these alternate AHLs synthases are by a red alga that inhibit quorum sensing, and
among other bacteria, although general surveys unidentified compounds synthesized by several
of bacterial genome sequences suggest that they plants that can either activate or inhibit LuxR-
are less common than LuxI-type synthases. type proteins.38,39 It is unclear whether these
compounds are synthesized for this purpose or
22.2.10 Non-AHL molecules that affect whether their impact on bacterial quorum sens-
the activity of Lux-type systems ing is fortuitous. They have, however, made
There are reports of non-AHL molecules that attractive targets for developing quorum-sens-
affect the activity of Lux-type systems. For ing inhibitors (QSIs).
22.2.11 AHL signaling without LuxR and

LuxI-type proteins
Bioluminescence in V. harveyi is regulated
by population density via three diffusible sig-
nals: the 3-OH-C4-HSL signal just described,
a second signal called autoinducer-2 (AI-2,
see Section 22.2.12), and possibly by a third
signal that represents a new class of com-
pounds called a-hydroxyketones (AHKs, see
Section 22.2.13).40–42 Perception of 3-OH-C4-
HSL involves the activity of a two-component-
type sensor kinase called LuxN that engages a
complex phosphorelay system (Fig. 22.4). The
phosphoryl group is transferred from LuxN
to a protein called LuxU, which then catalyzes
phosphotransfer to the LuxO response regu-
lator. Under noninducing conditions (i.e., at
low quorum-sensing signal concentration), the
phosphotransfer cascade is active and phospho-
rylated LuxO activates expression of a series
of small regulatory RNAs (quorum-regulated
RNAs, or Qrrs).43 The Qrrs inhibit transla-
tion of a transcription factor historically called
LuxR (confusingly, a protein with no homol-
ogy to AHL-responsive LuxR proteins such as
that from V. fischeri) through interaction with
its mRNA transcript. The 3-OH-C4-HSL sig-
nal inactivates the phosphorylation cascade,
Fig. 22.4 Quorum-sensing signal transduction in Vibrio
preventing LuxO-stimulated expression of the harveyi and Vibrio cholerae. See text for details. 3-OH-
Qrrs, relieving inhibition of LuxR, and allow- C4-HSL plus AI-2 are functional in V. harveyi, and
ing it to activate bioluminescence and the other AI-2 plus CAI-1 are functional in V. cholerae. The
target functions. The differences between the downstream components of the regulatory pathway
V. harveyi LuxN/LuxU/LuxO/Qrr/LuxR AHL are conserved. AI-2 interacts with a periplasmic bind-
phosphotransfer cascade are striking compared ing protein that is LuxP in V. harveyi. The pathway is
with the direct AHL interaction with the V. depicted in the “off” state. Signal binding by the recep-
fischeri-type LuxR transcription factors, and tors prevents phosphotransfer and derepresses LuxR
exemplify the mechanistic diversity of quorum expression, resulting in elevated target gene expression.
Source: White, D., and C. Fuqua. 2004. Prokaryotic
sensing. The complex AHL-responsive phos-
intercellular signaling mechanistic diversity and uni-
photransfer cascade is linked to control of small
fied themes, in Cell Signaling in Prokaryotes and
regulatory RNAs, which are also employed by Lower Metazoa, Fairweather, ed. Kluwer Academic
other vibrios as well as by more diverse pro- Publishers, Dordrecht, Netherlands, with kind permis-
teobacteria, although they may provide response sion from Springer Science+Business Media B.V.
to quorum-sensing signals other than AHLs.44
22.2.12 Signaling via AI-2 type factors more of the furanosyl borate diester diffusible sig-
As described earlier, a second quorum-sensing cue nal molecules called autoinducer 2 (AI-2).45 AI-2
was discovered in V. harveyi and has proven to be is perceived via the LuxQ two-component sensor
a common signal among highly diverse bacteria, kinase and engages the same LuxU/LuxO phos-
including gram-positive bacteria and spirochaetes. phorelay system that 3-OH-C4-HSL activates
V. harveyi mutants that cannot synthesize the (Fig. 22.4). Therefore, in mutants that cannot
AHL still exhibit cell-density-dependent regula- synthesize the AHLs, AI-2 provides a compensa-
tion of bioluminescence. This is due to this second tory signal. Likewise, strains that carry mutations
quorum-sensing system, which generates one or in LuxQ, hence cannot perceive AI-2, maintain
www.ebook777.com
their population density responsiveness to the Although V. cholerae does produce AI-2, it does
AHL signal. Double mutants defective for synthe- not synthesize 3-OH-C4-HSL. Instead, it syn-
sis of both signals cannot activate lux genes. Using thesizes a different signal originally described
a mutant V. harveyi strain that responds only to as cholerae autoinducer 1 (CAI-1) and also
AI-2, Bassler et al. found that a range of differ- produced by V. harveyi. CAI-1 was found to be
ent bacteria produce factors capable of activating 3-hydroxytridecan-4-one,49 the founding mem-
the AI-2-dependent system—the first evidence ber of a new class of signal molecules collectively
that this signal molecule may provide interspecies called the a-hydroxyketones (AHKs). Synthesis
information to V. harveyi and other bacteria.46 of CAI-1 requires the synthase enzyme CqsA,
and perception of CAI-1 signal requires the
AI-2-type signals are produced by diverse two-component sensor kinase CqsS, which also
bacteria relies on a LuxO/LuxU/Qrr pathway, as do the
AI-2-like factors are in fact synthesized by V. harveyi signals described earlier in Sections
a broad range of bacteria across a striking 22.2.11 and 22.2.12. An AHK quorum-sensing
range of diversity.47 Synthesis of AI-2 in V. system with almost all the components of that
harveyi and other bacteria requires the luxS in V. cholerae has also been characterized in the
gene, which is now known to function as a distantly related human pathogen Legionella
S-ribosylhomocysteinase in the activated pneumophila (the cause of Legionnaire’s dis-
methyl cycle involved in recovering the amino ease) and related bacteria, and homologous
acid methionine from other metabolites. LuxS genes for this pathway are also present in other
produces AI-2 as a side reaction to its primary proteobacteria.44
role in this cycle. V. harveyi has evolved the
capacity to detect the AI-2 signal(s) as a means 22.2.14 g-Butyrolactones: Regulators
of regulating bioluminescence, most likely using of antibiotic synthesis and secondary
this to gauge its own population density, as well metabolism in Streptomyces
as that of other bacteria in its environment. It One of the first hints at intercellular signal-
is clear that many bacteria synthesize AI-2, but ing in prokaryotes came from the pioneering
it remains controversial whether this functions work of Kholkov on the filamentous, fungus-
as a signal for all these microbes. Even so, there like microbes of the genus Streptomyces.50 The
is growing acceptance that at least a subset of streptomycetes are gram-positive soil bacteria
bacteria do utilize AI-2 signaling in quorum- that grow as a branching, filamentous, non-
sensing-type processes. For example, AI-2-type septate mycelium, superficially resembling
signaling apparently plays a role in regulat- fungi. Older parts of the colony send up aerial
ing the transcription of genes encoding a type filaments called hyphae or sporophores. Septa
III secretion system in enteropathogenic and form at regular intervals in the aerial hyphae,
enterohemorrhagic E. coli strains.48 The type and each septated compartment develops into a
III secretion system is involved in attachment of spore called a conidium (pl. conidia). Members
these pathogenic E. coli strains to intestinal cells of the genus Streptomyces are ubiquitous soil
during infection, and the injection of virulence microbes and are well recognized as the primary
proteins. (For a review of type III secretion sys- source of pharmaceutically important antibiot-
tems, refer to Section 18.5.2 in Chapter 18.) ics. It is easy to imagine that Streptomyces has
evolved antibiotic production as a mechanism
22.2.13 a-Hydroxyketone signaling in to compete with other soil microbes for limit-
gram-negative bacteria ing nutrients. Typically synthesized in the late
Surprisingly, analysis of quorum sensing in stages of growth in laboratory culture, antibiot-
Vibrio species has led to the discovery of yet ics represent a product of so-called secondary
another type of signal molecule. The human metabolism (metabolism occurring after the
pathogen V. cholerae is a nonbioluminescent primary stage of growth in laboratory culture
relative of V. harveyi and V. fischeri that controls which results in the secretion of products not
its virulence by means a quorum-sensing mech- required for growth). We will now return to the
anism and the same type of phosphorelay sys- g-butyrolactone signal called A-factor, intro-
tem described earlier for V. harveyi (Fig. 22.4). duced in Box 22.1
Mechanism of A-factor signaling butyrolactones rarely, if ever, occurs between

A-factor is a g-butyrolactone, specifically 2-iso- streptomycetes.
caprolyl-3R-hydroxymethyl-g-butyrolactone,
which is effective at a concentration of 1 nM. 22.2.15 Diffusible signaling factor (DSF)
Sporulation and synthesis of the aminoglyco- and relatives: Novel α,β-unsaturated fatty
side streptomycin by Streptomyces griseus is acid signals
delayed until late stages of culture growth, and Xanthomonos campestris pv. campestris causes
its synthesis depends on a diffusible signal mol- black rot disease on cruciferous plants (a large
ecule called A-factor (for a detailed review see family of plants with four-petaled flowers;
ref. 51.). The AfsA protein is a probable A-factor includes mustards, cabbages, broccoli, turnips,
synthase, utilizing fatty acid biosynthetic pre- and their many relatives). Production of an
cursors and glycerol derivatives as substrates. EPS, in this case called xanthan gum, is the pri-
A-factor shares structural similarity with the mary cause of disease symptoms. Production of
AHLs and as such is also freely diffusible across extracellular enzymes, such as endoglucanases
the bacterial envelope. A-factor is thought to and polygalacturonate lyases, probably adds to
coordinate antibiotic synthesis in adjacent cells the damage of host plant tissue. Xanthan gum
along a filament, as well as between different synthesis and extracellular enzyme activity are
filaments in the same population. Resistance regulated in a complex manner that includes the
genes are also under the control of A-factor, and effect of a self-produced diffusible extracellular
it seems reasonable that coordination of anti- factor called DSF.52 DSF is a small molecule that,
biotic synthesis and resistance gene expression like the AHLs, can be extracted into the solvent
might ensure that members of a population of S. ethyl acetate. DSF synthesis is dependent on
griseus will not suffer growth inhibition due to the rpfB and rpfF genes, the products of which
bursts of antibiotic production in adjacent parts share homology with proteins involved in fatty
of the population. acid metabolism. Xanthan gum and extracellu-
A-factor binds to the cytoplasmic ArpA pro- lar enzyme production initiates in late stage cul-
tein, a repressor that inhibits several genes in tures, parallel with accumulation of DSF. DSF
the absence of A-factor (Fig. 22.5). The ArpA synthesized by rpfB was found to be 11-methyl-
amino acid sequence is not similar to other cis-2-dodecenoic acid. DSF is perceived through
quorum-sensing system regulators. A-factor a two-component system comprising the RpfC
binding, presumably to a site in the carboxy- sensor kinase and the RpfG response regulator.
terminal portion of ArpA, causes derepression In response to DSF accumulation outside the
in which the protein releases its DNA-binding cell, phosphorylated RpfG activates a complex
site, resulting in target gene induction that intracellular signal transduction pathway that
affects streptomycin biosynthetic and resis- influences a large number of target processes in
tance genes, and other genes involved in micro- addition to EPS synthesis.
bial development.
DSF-type signals in other bacteria and fungi
g-Butyrolactones in other Streptomyces spp. DSF signaling systems have been identified
Similar g-butyrolactone molecules, identified and characterized for members of the genus
in several other species of Streptomyces, have Xylella, species of which are very important
been shown to regulate antibiotic synthesis and plant pathogens that are transmitted to plants
cell differentiation. Virginiamycin synthesis in by sap-feeding insects.53 The Xylella DSF sig-
S. virginiae is regulated by a group of several naling factor appears to govern expression of
g-butyrolactones called the virginiae butano- insect-specific adhesion factors, which promote
lides (VBs) in much the same way as A-factor colonization of the insect host and subsequent
functions in S. griseus.7 Despite a VB receptor transmission to the plant during sap feeding. In
protein that shares homology with ArpA, a high addition to Xylella, species of the Burkholderia
degree of specificity apparently prevents cross- and Stenotrophomonas, some of which are
recognition between Streptomyces g-butyro- human pathogens, have also been shown to
lactones from different species. Therefore it is employ or to be influenced by DSF-type signal-
likely that interspecies quorum sensing via the ings systems involving a,b-unsaturated fatty
www.ebook777.com
Fig. 22.5 Streptomyces griseus signaling between mycelia. Exchange of A-factor between two adjacent S.
griseus filaments; solid circle indicates A-factor. Intrafilament A-factor signaling is not depicted but is thought
to occur in the same fashion. See text for details. Source: White, D., and C. Fuqua. 2004. Prokaryotic intercel-
lular signaling mechanistic diversity and unified themes, in Cell Signaling in Prokaryotes and Lower Metazoa,
Fairweather, ed. Kluwer Academic Publishers, Dordrecht, Netherlands, with kind permission from Springer
Science+Business Media B.V.
acids.54 Finally, farnesoic acid, which is known approximately 2 μm) with several thin flagella,
to function as a quorum-sensing signal in the into an elongated (up to 80 μm long), multinu-
yeast Candida albicans, the cause of yeast infec- cleoid form called a swarmer cell that is covered
tions in humans, is structurally related to the with thousands of flagella.55 In the laboratory,
DSF-type signals, and in fact, DSF itself from swarmer cells form when Proteus is cultivated on
X. campestris can function in place of farnesoic agar surfaces of the appropriate growth medium.
acid. Owing to its activity in both bacteria and Swarming growth moves from the point of inoc-
fungi, DSF is considered to be a cross-kingdom ulum outward in concentric waves that reflect
signal molecule.52 the differentiation of P. mirabilis swimmer cells
into swarmers, and subsequent concerted move-
22.2.16 Putrescine control of swarming ment away from the point of inoculation. After
motility in Proteus several hours of colony expansion, the swarmers
Another example of signaling concerns swarm- stop moving and differentiate back into swim-
ing motility in Proteus. Swarming is group mer cells, and the growth ring consolidates by
motility by which the entire colony spreads via vegetative multiplication. At a specific time fol-
movement of cells across a surface. Swarming lowing consolidation, the swimmers begin dif-
motility, which has been quite extensively stud- ferentiating into swarmer cells, and a new wave
ied in P. mirabilis, entails differentiation from a of swarming occurs. This cycle repeats itself until
standard, rod-shaped cell (called a swimmer cell, the entire plate is covered with growth. Swarming
motility in the host animal probably allows the damaging toxins and enzymes from the bacteria
bacteria to persist in the high flow rate environ- directly into the cytoplasm of eukaryotic cells,
ment of the urinary tract. radically altering their regulatory balance.
Although there are many factors that influ-
ence swarming motility, intercellular signaling 22.3.1 Fruiting body formation and
plays an important role. In P. mirabilis, the gliding in Myxococcus xanthus
polyamine putrescine has been identified as a For myxobacteria to aggregate and form fruit-
potent and specific signal for swarming motili- ing bodies, both diffusible and cell–cell contact
ty.56 Mutants that cannot synthesize putrescine signaling must take place. Diffusible signaling
are delayed for the differentiation into swarmer via A-signals was discussed in Section 22.2.8.
cells, and these mutants are rescued by addition Contact-dependent signaling is discussed here.
of putrescine. The specific signaling pathways Chapter 23 provides a fuller discussion of fruit-
through which the putrescine response is trans- ing body formation in myxobacteria.
duced are not yet defined. Although in P. mira-
bilis there is no evidence for AHL-dependent
C-Signaling and development
control of swarming, in Serratia liquefaciens,
another swarming bacterium, a LuxI–LuxR One signaling pathway requiring cell-to-cell
type of regulatory system is involved in regulat- contact relies on the C-signal.59 Mutants defec-
ing the swarming process.57 Clearly, intercel- tive in C-signaling have altered motility pat-
terns and delayed development; they construct
lular signaling plays an important role in the
abnormal aggregates instead of fruiting bod-
coordinated behavior of swarming bacteria,
ies. Furthermore, they sporulate at a reduced
and several distinct mechanisms of cell-to-cell
efficiency, and reduce or abolish expression of
communication have evolved.
genes that are expressed late during wild-type
development. M. xanthus cells are rod shaped,
22.3 Cell–Cell Signaling That Requires and cells that are aligned end-to-end stimulate
Contact each other via the C-signal to complete the devel-
In addition to signaling via diffusible chemicals, opmental program. A model for C-signaling is
bacteria can signal each other, as well as host shown in Fig. 22.6. It has been suggested that
tissues, via contact-dependent communication. the C-signal is a protein on the surface of the
For instance, the process of bacterial conjuga- ends of the cells that binds to a receptor at the
tion is a form of contact-dependent signaling. end of the second cell, resulting in activation
In most forms of conjugation, the donor and of C-signal-dependent pathways. Cells unable
recipient cell first fuse membranes to form a to signal other cells have a mutation in a gene
mating pair. The donor then transfers genetic called csgA; the gene is not strongly expressed
information in the form of plasmid or chro- during vegetative growth but increases mark-
mosomal DNA to the recipient. It is clear that edly during development. C-signaling results in
recipients are not passive in this process, and the activation of the transcription factor FruA.
therefore conjugation represents a stabilized, The FruA protein either directly or indirectly
physical interaction between bacteria. Another activates transcription of the dev operon, which
recent example is that of the syntrophic metab- contains genes required for sporulation and
olism in a symbiosis between a bacterium and aggregation. Examination of the gliding motil-
an archaeon. In this case, physical contact ity of single csgA mutant cells to which C-signal
between the two partner cells, via proteins in has been added has revealed that the C-signal
the bacterial flagellum, regulates the activation (1) stimulates cells to glide faster, (2) increases
of their collaborative metabolism.58 There are the duration of the gliding interval, and (3)
also numerous examples of contact-dependent decreases the number of stops per minute.60 As
signal exchange between bacteria and eukary- a consequence, the C-signal causes the cells to
otic cells. One important example is that of glide further per unit time. It was suggested that
the type III secretion systems of pathogenic this encourages the cells to move into aggrega-
bacteria (described earlier: see Section 18.5.2). tion centers, which is an initial step in fruiting
These secretion systems have evolved to act as a body formation. None of these effects were seen
contact-dependent syringe for injection of cell- in cells that had mutations in the Frz motility
www.ebook777.com
system, which controls the frequency of rever- communication stimulates gliding motility in
sals in the direction of gliding (see discussed both systems. The signaling systems that stimu-
next), suggesting that the Frz proteins are late single-cell and group motility are not well
required to ensure response to the C-signal. The understood, but both forms appear to involve
addition of C-signal causes the methylation of cell–cell contact in both the A- and S-motility
one of the Frz proteins (FrzCD), and the signal- systems.63 The signaling molecules have not
ing pathway resulting in methylation requires been identified, but presumably they are bound
FruA. (See Fig. 22.6 and ref. 61 for a review.). to the cell surface or to fibrils or pili that extend
from the cell surface. Genetic analysis and extra-
Signaling required for gliding motility cellular complementation studies of mutants
Individual cells in a myxobacterial popula- have suggested that there are five signaling mol-
tion must coordinate their gliding relative to ecules for the A-motility system. Very little is
one another. The coordination is required to known about the signal transduction pathways
keep the swarm intact as a communally feed- for the A-motility signals.
ing entity, as well as to provide the necessary Gliding using the S-motility system requires
contact-mediated signals for aggregation and pili that are present at the cell poles. Pili are short
fruiting body development. Gliding motility protein fibrils that protrude from the surfaces
in myxobacteria is coordinated by several cell- of most gram-negative bacteria. In general, they
contact-mediated signaling systems. mediate attachment to other bacterial, animal,
Gliding motility in M. xanthus is controlled plant, or fungal cells and are very important
by two sets of genes; adventurous (A) and social for colonization of surfaces. The M. xanthus
(S).62 Mutants in system A genes (A−S+) can glide polar pili that are required for S motility belong
only as groups, indicating that system A genes to the class called type IV pili.64 It is believed
are required for single-cell motility. Mutants in that a signal required for S motility is transmit-
system S genes (A+S−) show a motility pattern in ted from cell to cell via the polar pili. Mutants
which they glide as single cells or as small clus- with lesions in the gene tglA are defective in the
ters, suggesting that system S genes are required assembly of pilin (the major protein compo-
for group motility. nent of pili) into pili and consequently cannot
Both single-cell and group motility are sig- glide via the S motility system. However, these
nificantly stimulated when the cell density is mutants can be induced to transiently produce
increased, reflecting the fact that cell-to-cell pili and glide if they make end-to-end contact
early signals
1
FruA
2 dev TRS sporulation
Active FruA
3 csgA expression
CsgA on cell surface C-factor sensor
Fig. 22.6 C-signaling in M. xanthus. Positive control is indicated by arrows, negative control by lines ending
in a bar. Upon receiving the C-signal, a cell activates a transcription factor called FruA. FruA activates the
transcription of the operon devTRS whose products are required for sporulation. FruA also stimulates the
transcription of csgA which encodes the C-factor, providing positive feedback. There is also negative feedback
on devTRS expression. Adapted from Kaiser, D. 1999. Cell fate and organogenesis in bacteria. Trends Genet.
15:273–277. Source: White, D., and C. Fuqua. 2004. Prokaryotic intercellular signaling mechanistic diversity
and unified themes, in Cell Signaling in Prokaryotes and Lower Metazoa, Fairweather, ed. Kluwer Academic
Publishers, Dordrecht, Netherlands, with kind permission from Springer Science+Business Media B.V.
with wild-type cells. It has been suggested that cell surfaces of interacting cells including spe-
the TglA protein is localized at the poles of the cific receptors, and (3) the relative alignment
cells where the type IV pili are located and that between the signal-donating partner and the
the protein stimulates pilus assembly at the responding cell.
poles of recipient cells that are properly aligned
with the poles of the stimulatory cells.59 It is of
interest that type IV pili have been shown to be Study Questions
involved in several different processes in a wide
variety of other gram-negative bacteria, includ- 1. Diffusible signals are often called quorum
ing the adherence of pathogenic bacteria to host sensors. Why is that?
cells and twitching motility.65 2. What is meant by the term pheromone?
Give some examples of pheromones and
The role of fibrils in signaling via cell contact their effect on cells. What is the difference
Protruding from the cell surface of M. xanthus between a pheromone and a hormone?
are thin appendages called fibrils (distinct from
polar pili) that are composed of protein attached 3. For a diffusible signal to cause a change in
to a polysaccharide backbone.66 The absence the behavior of a recipient cell, must it enter
of fibrils results in loss of cohesion, a defect in the cell? Explain.
S motility, and inability to form fruiting bod- 4. Propose a model for how an intercellular
ies. The addition of isolated fibrils to fibril-less signal can activate or repress a gene in a
mutants restores the wild-type phenotype. The recipient cell.
fibrils are formed under conditions of high cell
density, perhaps reflecting the fact that their 5. What is meant by “a two-component sig-
production is stimulated by cell-cell contact. naling cascade”?
The cells use the fibrils to attach to one another, 6. What are bacteriocins, and why are they
and it has been suggested that they are involved important for bacteria that produce them?
in contact-mediated cell–cell signaling.
7. Can you think of six different bacterial pro-
cesses that require intercellular signaling?
22.4 Summary What are they? Are there more?
In this chapter we have introduced the diversity
of intercellular signaling mechanisms employed 8. Give two examples of cell–cell contact
by bacteria. Signal molecules of many different dependent signaling.
types play important roles in microbial behav- 9. What are acylated homoserine lactones?
ior. The primary considerations for diffusible Give an overview of their use as signals,
signals are (1) how the signal is released from including specific examples.
the source cell(s), (2) what dictates accumu-
lation of the signal, and (3) perception of the
signal by the responding cells. The lipid-based REFERENCES AND NOTES
signals often can cross the cellular envelope by
passive diffusion as they exit and enter cells, 1. Fuqua, C., and D. White. 2004. Prokaryotic inter-
although their receptors can be cytoplasmic cellular signaling: mechanistic diversity and unified
themes, pp. 27–71 In: Cell Signalling in Prokaryotes
or the external portion of transmembrane and Lower Metazoa, I. Fairweather (Ed.). Kluwer
proteins. The oligopeptide pheromones must Academic Publishers, Dordrecht, Netherlands.
be actively secreted from cells or generated
2. Blango, M. G., and M. A. Mulvey. 2009. Bacterial
from external substrates. Perception of the landlines: contact-dependent signaling in bacterial
oligopeptide depends on transmembrane sen- populations. Curr. Opin. Microbiol 12:177–181.
sors or transport of the oligopeptide into the 3. Ng, W. L., and B. L. Bassler. 2009. Bacterial
cytoplasm of responding cells. Cell-associated quorum-sensing network architectures. Annu. Rev.
signaling is less well understood, but clearly Genet. 43:197–222.
important aspects of signal exchange are (1) 4. Novick, R. P., and E. Geisinger. 2008. Quorum
the anchoring and presentation of the signal sensing in staphylococci. Annu. Rev. Genet. 42:541–
by the donating cell, (2) interactions between 564.
www.ebook777.com
5. Parker, C. T., and V. Sperandio. 2009. Cell-to- 18. Clewell, D. 1999. Sex pheromone systems in
cell signalling during pathogenesis. Cell. Microbiol. enterococci, pp. 47–65. In: Cell–Cell Signaling in
11:363–369. Bacteria, G. M. Dunny and S. C. Winans (Eds.). ASM
6. Shank, E. A., and R. Kolter. 2009. New devel- Press, Washington, DC.
opments in microbial interspecies signaling. Curr. 19. Wirth, R. 2000. Sex pheromones and gene transfer
Opin. Microbiol. 12:205–214. in Enterococcus faecalis. Res. Microbiol. 151:493–496.
7. Takano, E. 2006. Gamma-butyrolactones: 20. Aldea, M. R., K. Kumar, and J. W. Golden. 2008.
Streptomyces signalling molecules regulating anti- Heterocyst development and pattern formation,
biotic production and differentiation. Curr. Opin. pp. 75–90. In: Chemical Communication among
Microbiol. 9:287–294. Bacteria, S. C. Winans and B. L. Bassler (Eds.). ASM
8. Winans, S. C., and B. L. Bassler. 2008. Chemical Press, Washington, DC.
Communication among Bacteria. ASM Press, 21. Brun, Y. V., and L. J. Shimkets. 2000. Prokaryotic
Washington, DC. Development. ASM Press, Washington DC.
9. Lazazzera, B. A., T. Plamer, J. Quisel, and A. D. 22. Whitworth, D. E. 2008. Myxobacteria:
Grossman. 1999. Cell density control of gene expres- Multicellularity and Differentiation. ASM Press,
sion and development in Bacillus subtilis, pp. 27–46 Washington, DC.
In: Cell–Cell Signaling in Bacteria, G. M. Dunny and
S. C. Winans (Eds.). ASM Press, Washington, DC. 23. Kaiser, D. 2004. Signaling in myxobacteria.
10. Magnuson, R., J. Solomon, and A. D. Grossman.
1994. Biochemical and genetic characterization 24. White, D. 2000. The Physiology and Biochemistry
of a competence pheromone from B. subtilis. Cell of Prokaryotes, 2nd ed. Oxford University Press,
77:207–216. New York.
11. Caldwell, G. A., F. Naider, and J. M. Becker. 25. Kaplan, H. B. 2003. Multicellular development
1995. Fungal lipopeptide mating pheromones: a and gliding motility in Myxococcus xanthus. Curr.
model system for the study of prenylation. Microbiol. Opin. Microbiol. 6:572–577.
Rev. 59:406–422. 26. Nealson, K. H., T. Platt, and J. W. Hastings.
12. Solomon, J. M., B. A. Lazazzera, and A. D. 1970. Cellular control of the synthesis and activ-
Grossman. 1996. Purification and characterization ity of the bacterial luminescent system. J. Bacteriol.
of an extracellular peptide factor that affects two dif- 104:313–322.
ferent developmental pathways in B. subtilis. Genes
27. Nealson, K. H. 1977. Autoinduction of bacterial
Dev. 10:2014–2024.
luciferase: occurrence, mechanism, and significance.
13. Pottathil, M., and B. A. Lazazzera. 2003. The Arch. Microbiol. 112:73–79.
extracellular Phr peptide–Rap phosphatase signaling
circuit of Bacillus subtilis. Front. Biosci. 8:d32–45. 28. Defoirdt, T., N. Boon, P. Sorgeloos, W.
Verstraete, and P. Bossier. 2008. Quorum sensing
14. Auchtung, J. M., C. A. Lee, R. E. Monson, A. P. and quorum quenching in Vibrio harveyi: lessons
Lehman, and A. D. Grossman. 2005. Regulation of learned from in vivo work. ISME J. 2:19–26.
a Bacillus subtilis mobile genetic element by intercel-
lular signaling and the global DNA damage response. 29. Stabb, E. V., A. L. Schaefer, J. L. Bose, and E.
Proc. Natl. Acad. Sci. USA 102:12554–12559. G. Ruby. 2008. Quorum signaling and symbiosis
in the marine luminous bacterium Vibrio fischeri,
15. Oggioni, M. R., and D. A. Morrison. 2008. pp. 233–250. In: Chemical Communication among
Cooperative regulation of competence development Bacteria, S. C. Winans and B. L. Bassler (Eds.). ASM
in Streptococcus pneumoniae: cell-to-cell signaling Press, Washington, DC.
via a peptide pheromone and an alternative sigma
factor, pp. 345–362. In: Chemical Communication 30. Eberhard, A., A. L. Burlingame, C. Eberhard, G.
among Bacteria, S. C. Winans and B. L. Bassler L. Kenyon, K. H. Nealson, and N. J. Oppenheimer.
(Eds.). ASM Press, Washington, DC. 1981. Structural identification of autoinducer of
Photobacterium fischeri luciferase. Biochemistry
16. Kleerebezem, M., W. M. de Vos, and O. P. 20:2444–2449.
Kuipers. 1999. The lantibiotics nisin and subtilin act
as extracellular regulators of their own biosynthe- 31. Churchill, M. E., and J. P. Herman. 2008.
sis, pp. 159–174 In: Cell–Cell Signaling in Bacteria, Acylhomoserine lactone biosynthesis: structure and
G. M. Dunny and S. C. Winans (Eds.). ASM Press, mechanism, pp. 275–290. In: Chemical Communi-
Washington, DC. cation among Bacteria, S. C. Winans and B. L. Bassler
(Eds.). ASM Press, Washington, DC.
17. Dunny, G. M. 2007. The peptide pheromone–
inducible conjugation system of Enterococcus faeca- 32. Bainton, N. J., B. W. Bycroft, S. R. Chhabra,
lis plasmid pCF10: cell–cell signalling, gene transfer, et al. 1992. A general role for the lux autoinducer in
complexity and evolution. Philos. Trans. R. Soc. bacterial cell signalling: control of antibiotic biosyn-
Lond. B. Biol. Sci. 362:1185–1193. thesis in Erwinia. Gene 116:87–91.
33. Fuqua, C., and E. P. Greenberg. 2002. Listening in the quorum-sensing bacterium Vibrio harveyi. J.
on bacteria: acylhomoserine lactone signalling. Nat. Bacteriol. 179:4043–4045.
Revi. Mol. Cell Biol. 3:685–695.
47. Schauder, S., K. Shokat, M. G. Surette, and B. L.
34. Lerat, E. and N. A. Moran. 2004. Evolutionary Bassler. 2001. The LuxS family of bacterial autoin-
history of quorum-sensing systems in bacteria. Mol. ducers: biosynthesis of a novel quorum-sensing sig-
Biol. Evol. 21:903–913. nal molecule. Mol. Microbiol. 41:463–476.
35. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. 48. Sperandio, V., J. L. Mellies, W. Nguyen, S. Shin,
Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. and J. B. Kaper. 1999. Quorum sensing controls
The involvement of cell-to-cell signals in the develop- expression of the type III secretion gene transcrip-
ment of a bacterial biofilm. Science 280:295–298. tion and protein secretion in enterohemorrhagic and
enteropathogenic Escherichia coli. Proc. Natl. Acad.
36. Hanzelka, B. L., M. R. Parsek, D. L. Val, P. V. Sci. USA 96:15196–15201.
Dunlap, J. E. J. Cronan, and E. P. Greenberg. 1999.
Acylhomoserine lactone synthase activity of the 49. Higgins, D. A., M. E. Pomianek, C. M. Kraml, R.
Vibrio fischeri AinS protein. J. Bacteriol. 181:5766– K. Taylor, M. F. Semmelhack, and B. Bassler. 2007.
5770. The major Vibrio cholerae autoinducer and its role in
virulence factor production. Nature 450:883–886.
37. Holden, M. T., S. R. Chhabra, R. de Nys, et al.
1999. Quorum-sensing crosstalk: isolation and 50. Khoklov, A. S., I. I. Tovarova, L. N. Borisova, S.
chemical characterization of cyclic dipeptides from A. Pliner, L. A. Schevchenko, E. Y. Kornitskaya, N.
Pseudomonas aeruginosa and other gram-negative S. Ivkina, and I. A. Rapoport. 1967. A factor respon-
bacteria. Mol. Microbiol. 33:1254–1266. sible for the biosynthesis of streptomycin by a mutant
strain of Actinomyces streptomycini. Dokl. Akad.
38. Givskov, M., R. de Nys, M. Manefield, L. Gram, Nauk SSSR 177:232–235.
R. Maximilien, L. Eberl, S. Molin, P. D. Steinberg,
and S. Kjelleberg. 1996. Eukaryotic interference with 51. Horinouchi, S. 2008. The A factor regulatory
homoserine lactone-mediated prokaryotic signal- cascade that triggers secondary metabolism and
ling. J. Bacteriol. 178:6618–6622. morphological differentiation in Streptomyces, pp.
363–378. In: Chemical Communication among
39. Teplitski, M., J. B. Robinson, and W. D. Bauer. Bacteria, S. C. Winans and B. L. Bassler (Eds.). ASM
2000. Plants secrete substances that mimic bacte- Press, Washington, DC.
rial N-acyl homoserine lactone signal activities and
affect population density-dependent behaviors in 52. Wang, L. H., Y. He, Y. Gao, and et al. 2004. A
associated bacteria. Mol. Plant–Microbe Interact. bacterial cell–cell communication signal with cross-
13:637–648. kingdom structural analogues. Mol. Microbiol.
51:903–912.
40. Cao, J. G., and E. A. Meighen. 1989. Purification
and structural identification of an autoinducer for 53. Newman, K. L., R. P. P. Almeida, A. H. Purcell,
the luminescence system of V. harveyi. J. Biol. Chem. and S. E. Lindow. 2004. Cell–cell signaling controls
264:21670–21676. Xylella fastidiosa interactions with both insects
and plants. Proc. Natl. Acad. Sci. USA 101:1737–
41. Surette, M. G., and B. L. Bassler. 1998. Quorum 1742.
sensing in Escherichia coli and Salmonella typhimu-
rium. Proc. Natl. Acad. Sci. USA 95:7046–7050. 54. Boon, C., Y. Deng, L. H. Wang, Y. He, J. L.
Xu, Y. Fan, S. Q. Pan, and L. H. Zhang. 2008. A
42. Henke, J. M., and B. L. Bassler. 2004. Three par- novel DSF-like signal from Burkholderia cenocepa-
allel quorum-sensing systems regulate gene expres- cia interferes with Candida albicans morphological
sion in Vibrio harveyi. J. Bacteriol. 186:6902–6914. transition. ISME J. 2:27–36.
43. Lenz, D. H., K. C. Mok, B. N. Lilley, R. V. 55. Belas, R. 1997. Proteus mirabilis and other
Kulkarni, N. S. Wingreen, and B. L. Bassler. 2004. swarming bacteria, pp. 183–219. In: Bacteria as
The small RNA chaperone Hfq and multiple small Multicellular Organisms, J. A. Shapiro and M.
RNAs control quorum sensing in Vibrio harveyi and Dworkin (Eds.). Oxford University Press, New
Vibrio cholerae. Cell 118:69–82.. York.
44. Tiaden, A., T. Spirig, and H. Hilbi. 2010. 56. Rather, P. N. 2005. Swarmer cell differentiation
Bacterial gene regulation by alpha-hydroxyketone in Proteus mirabilis. Environ. Microbiol. 7:1065–
signaling. Trends Microbiol. 18:288–297. 1073.
45. Chen, X., S. Schauder, N. Potier, A. Van 57. Eberl, L., M. K. Winson, C. Sternberg, et al.
Dorsselaer, I. Pelczer, B. L. Bassler, and F. M. 1996. Involvement of N-acyl-L-homoserine lactone
Hughson. 2002. Structural identification of a bacte- autoinducers in control of multicellular behavior of
rial quorum-sensing signal containing boron. Nature Serratia liquefaciens. Mol. Microbiol. 20:127–136.
415:545–549.
58. Shimoyama, T., S. Kato, S. Ishii, and K.
46. Bassler, B. L., E. P. Greenberg, and A. M. Stevens. Watanabe. 2009. Flagellum mediates symbiosis.
1997. Cross-species induction of luminescence in Science 323:1574.
www.ebook777.com
59. Kaiser, D. 2008. C-signal control of aggre- 63. Hartzell, P., W. Shi, and P. Youderian. 2008.
gation and sporulation, pp. 51–64. In: Chemical Gliding motility in Myxococcus xanthus, pp.
Communication among Bacteria, S. C. Winans and 103–122. In: Mycobacteria: Multicellularity and
B. L. Bassler (Eds.). ASM Press, Washington, DC. Differentiation, D. E. Whitworth (Ed.). ASM Press,
Washington, DC.
60. Jelsbak, L., and L. Sogaard-Andersen. 1999. The
cell surface–associated intercellular C-signal induces 64. Wu, S. S., and D. Kaiser. 1995. Genetic and
behavioral changes in individual Myxococcus xan- functional evidence that type IV pili are required for
thus cells during fruiting body morphogenesis. Proc. social gliding motility in Myxococcus xanthus. Mol.
Natl. Acad. Sci. USA 96:5031–5036. Microbiol. 18:547–558.
61. Ward, M. J. and D. Zusman. 2000. Developmental 65. Mattick, J. S. 2002. Type IV pili and twitch-
aggregation and fruiting body formation in the gliding motility. Annu. Rev. Microbiol. 56:289–
ing bacterium Myxococcus xanthus, pp. 243–262. 314.
In: Prokaryotic Development, Y. V. Brun and L. J.
66. Dworkin, M. 1999. Fibrils as extracellular
Shimkets (Eds.). ASM Press, Washington, DC.
appendages of bacteria: their role in contact-medi-
62. Shimkets, L. J. 1999. Intercellular signaling dur- ated cell–cell interactions in Myxococcus xanthus.
ing fruiting-body development of Myxococcus xan- BioEssays 21:590–595.
thus. Annu. Rev. Microbiol. 53:525–549.
23
Bacterial Development
Bacterial development in its broadest sense can University recognized through his extensive
refer to (1) cellular differentiation in which studies of fungi that these organisms are actu-
a cell acquires physiological properties and ally bacteria, not fungi, as reported by some in
changes in morphology that clearly differen- the earlier literature. Professor Thaxter pub-
tiate it from a precursor cell, (2) cellular dif- lished papers on myxobacteria in 1892, 1897,
ferentiation in which a cell divides to produce and 1904.
two daughter cells that can be distinguished
morphologically and/or physiologically, or (3) 23.1.1 Life cycle
multicellular development in which members
Rod-shaped, gram-negative gliding bacteria
of a population of cells interact to form special-
belonging to the δ-proteobacteria group, the
ized structures, such as fruiting bodies, or inter-
myxobacteria can be isolated from soil, dung
active communities within biofilms. (Biofilms
pellets, and decaying vegetation on the for-
are discussed in Chapter 21.) Differential gene
est floor. (For reviews, see refs.1–3.) The cells
expression occurs during these developmen-
glide on solid surfaces such as soil particles
tal processes and is often due to cell-to-cell
and vegetation in thin, spreading populations.
signaling (e.g., quorum sensing, described in
These populations of cells are referred to in the
Chapter 22), as well as signaling within cells,
myxobacteria literature as swarms, although
for example, via two-component (or multiple-
the mode of motility is not the same as the
component) phosphorelay systems (discussed
swarming motility that occurs with flagellated
in Chapter 19) that can be triggered by envi-
bacteria, discussed in Section 1.2.1. Gliding
ronmental cues. This chapter describes how
motility by myxobacteria does not require fla-
the myxobacteria, Caulobacter, and Bacillus
gella. Rather, it is a smooth, nonrotating for-
develop. What this chapter explains is how
ward or backward movement on a solid surface
intercellular and intracellular signaling regulate
in the direction of the long axis of the cell. The
the developmental progression in each of these
mechanism is discussed in Section 23.1.2. It
bacteria. It is important to realize that similar
should be pointed out that it is common for bac-
mechanisms apply to many other prokaryotes,
teria, not simply myxobacteria, to live in nature
and Sections 21.1 through 21.3 simply discuss
as populations of interacting bacteria on a solid
examples of prokaryotic development.
surface. Such populations on a solid surface are
called biofilms. (See Chapter 21 for a discussion
23.1 Myxobacteria of biofilms.)
The myxobacteria are unique among the bac- Myxobacteria are unique among the known
teria in forming multicellular fruiting bodies. prokaryotes in that the cells aggregate and con-
Over a century ago, Roland Thaxter of Harvard struct multicellular fruiting bodies, which can
587
www.ebook777.com
be induced in the laboratory by subjecting the or proteins such as casein. This is because the
cells to starvation on agar. Thus in myxobac- digested products of the lytic enzymes secreted
teria, as in many microbial developmental sys- by the myxobacteria are shared. Thus, the
tems, starvation is a signal for development. dispersal of fruiting bodies can be viewed as a
Because the shapes of the fruiting bodies are means of ensuring that when the myxospores
due to cell movements rather than cell growth, germinate, a feeding population will be pro-
the movements are referred to as morphoge- duced, consisting of at least 100,000 cells.
netic movements.
During the course of multicellular develop- Cell-to-cell signaling
ment, individual cells convert to resting cells To move as coherent groups of cells in the veg-
called myxospores, which are housed in the etative swarms while seeking nutrients, and
fruiting bodies. The shape of the fruiting body is also to move cooperatively to construct mul-
species specific and ranges from simple mounds ticellular fruiting bodies, the myxobacteria
of cells containing resting cells called myxo- have evolved systems of intercellular signaling.
spores, as in Myxococcus xanthus, to elaborate Intercellular signaling and gliding motility are
fruiting bodies with stalks supporting several discussed in Section 23.1.2, and the relationship
compartments, or sporangioles, that house of intercellular signaling to the construction of
the myxospores, as in Stigmatella aurantiaca. fruiting bodies and myxospores is discussed in
Figure 23.1A is a photomicrograph of an S. Section 23.1.3.
aurantiaca fruiting body. Most research is done
with M. xanthus, however, and this section will 23.1.2 Intercellular signaling and
focus on the results of that research. gliding motility
Two motility systems
Two stages in the population life cycle There are two types of gliding motility in M.
The myxobacteria have two stages in the xanthus, controlled by two sets of genes: those
population life cycle, a vegetative stage and a of system A (for adventurous motility) and
developmental stage (Fig. 23.1B). The vegeta- those of system S (for social motility). Mutants
tive stage is the time of feeding and growing. in system A (A–S+) can glide only as groups; that
During this period the cells in the swarms is, system-A genes are required for single-cell
move in groups and feed on the source of nutri- motility. Mutants in system-S genes (A+S–) show
ent, which in their natural habitat is usually a motility pattern indicating that they glide as
other bacteria that they lyse by secreting lytic single cells (well-separated cells, with some cell
enzymes. In the laboratory they can be grown clusters); that is, system S genes are required
on mixtures of peptides or amino acids. During for group motility. Mutants in the S-motility
the developmental stage, which is triggered by system show some defects in aggregation and
nutrient depletion, the myxobacteria cells move fruiting body formation, whereas those in the
into numerous aggregation centers consist- A-motility system usually show more minor
ing of approximately 100,000 cells each. As a developmental defects. Although, as we shall
result of cell movements, the aggregation cen- see next, the mechanisms of motility via the
ters develop into multicellular fruiting bodies. two systems differ, there does exist at least one
Within the fruiting bodies, the cells differenti- gene, nla24, that is required for both A- and
ate into myxospores. Eventually, the fruiting S-motility, and it has been suggested that Nla24
bodies are dispersed by nonspecific means, such might be a transcriptional activator of certain
as sticking to the bristles of roaming insects, to genes whose products are required for either
new locations, where they germinate and pro- the A- or S-motility systems.4 (See note 5 for a
duce a new population of cells that enters the further explanation.) Another gene required for
vegetative stage. both the A- and S-motility systems is the mglA
gene that codes for a cytoplasmic GDP/GTP
Cooperative feeding binding protein involved in the regulation of
It is important to realize that myxobacteria feed motility. (See Section 23.1.7 for a further expla-
cooperatively; that is, when they are together in nation.) As will be explained shortly, in con-
a population, they grow best on other bacteria nection with S-motility, when both the A- and
bacterial development 589
Fig. 23.1 Fruiting bodies and life cycles. (A) Fruiting

body of Stigmatella aurantiaca. The fruiting body
consists of a cellular stalk supporting several spo-
rangioles. Each sporangiole houses myxospores. (B)
Life cycle of Myxococcus xanthus. (1) Vegetative
growth: cells grow on solid surfaces in dense popu-
lations called swarms. (2) Aggregation: when nutri-
ents are depleted, cells glide into aggregation centers,
each one consisting of many thousands of cells. (3)
Mound formation: each aggregation center becomes
a mound of cells as bacteria continue to accumulate.
(4) The mound develops into a fruiting body when the
cells differentiate into resting cells called myxospores.
Each myxospore is surrounded by a coat (capsule).
(5) When nutrient becomes available once more, the
myxospores germinate into vegetative cells, return-
ing the population to a new growth phase. Once the
cells have depleted the supply of nutrients, they can
aggregate on agar and form myxospore-filled fruiting
bodies within 72 to 96 h.
S-motility systems are functioning in the same A-motility system works at the rear pole, push-
cell, they work coordinately at opposite poles ing the cell forward. When the cell reverses
of the cell. The S-motility system works at the its direction of motility, the A- and S-motors
forward pole, pulling the cell forward, and the switch poles.
www.ebook777.com
1. Social motility (S-motility) regulates developmental gene expression, and

Social motility (S-motility) requires type IV pili the Che4 system regulates S-motility reversal.
(also referred to as type 4 pili or TFP), which Type IV pili are also found in other bacteria
are made of PilA subunits; they are located at (e.g., Pseudomonas aeruginosa, Neisseria gonor-
any one time at either one, but not both, of the rhoeae) in which the pili are responsible for pull-
cell poles. ing groups of bacteria across wet surfaces via a
The type IV pili constitute the motility appa- form of gliding motility called twitching, which is
ratus for social motility, and mutants defective important for the colonization of these surfaces.
in the production and function of type IV pili Twitching motility can be important for the mat-
cannot move via the S system. According to the uration of biofilms formed by bacteria possessing
model for type IV pili function in S-motility, the type IV pili. Twitching and social gliding motil-
pili at the leading pole adhere to the polysaccha- ity in myxobacteria are identical with respect to
ride portion of extracellular fibrils (consisting of mechanism and requirement for type IV pili.
50% protein and 50% polysaccharide), extend-
ing from a cell in front, and then retract, pulling 2. Adventurous motility (A-motility)
the rear cell forward (Fig. 23.2). When retrac- The mechanism of A-motility is different from
tion is completed, the cell in front is released. that of S-motility in that the former does not
Mutant studies have shown that S-motility also require type IV pili and has been reported to be
requires the lipopolysaccharide (LPS) O-antigen correlated with the secretion of slime (extracel-
in the outer membrane. It has been suggested lular polyelectrolyte gel whose chemical nature
that the LPS O-antigen mutants may not be able is not well characterized) from nozzlelike struc-
to retract the type IV pili that normally pass tures at the rear pole of the moving cell. The noz-
through the LPS O-antigen layer during retrac- zlelike structures can be seen at both cell poles
tion. Motility via type IV pili requires cells to be by using electron microscopy, but the slime is
close enough to adhere to one another, and the secreted only from the rear pole of the moving
cells move in groups, rather than as single cells. cell. (See Fig. 23.2.) (It is possible that the slime
The conclusion that extracellular fibrils play is also secreted from the front and sides of the
a role in S-motility is based upon experimen- cell as the cells move over it.)
tal results showing that S-motility and fruiting According to one model, the slime is intro-
body formation are defective in dif (defective duced into the nozzlelike structures in a dehy-
in fruiting) mutants, even if type IV pili are drated form and then becomes hydrated, causing
present. The dif genes encode proteins that are it to swell and be extruded. The model further
homologous to proteins in the E. coli chemot- proposes that the extruded slime adheres to the
axis signaling system (Chapter 20). These are substrate; and as a consequence, force is gener-
the methyl-accepting chemotaxis proteins ated, pushing the cell forward. A similar mecha-
(MCPs), CheA, CheY, and CheW. Mutations in nism has been proposed for gliding filamentous
some of the dif genes have been demonstrated to cyanobacteria. (See Section 1.2.1.) The model
lead to defects in the production of cell surface suggests that the motility motors at the poles
fibrils. It has been suggested that the Dif homo- function coordinately in that the pili at the front
logues of the chemotaxis-like signal proteins “pull” the cell forward and the secretion of
constitute a phosphotransfer signaling system slime at the rear “pushes” the cell forward. The
in Myxococcus that regulates the biogenesis of model also predicts that the A- and S-motility
cell surface peritrichous fibrils, and that these systems must switch poles when the cell reverses
fibrils play a necessary role in social motility. For its direction of motility.
example, as noted earlier, polar type IV pili are There is a more recent model that does not
thought to adhere to fibrils on the cell in front, postulate motility via slime secretion, but rather
then retract, and then pull the cell forward. proposes that focal adhesion clusters, con-
M. xanthus has three other systems (Frz, nected to the A-engines, are attached to the agar
Che3, and Che4) that encode homologues of substrate. When the A-engines, which are pos-
chemotaxis-like signal proteins. One of the sys- tulated to be localized along the length of the
tems, Frz, regulates the frequency of reversal cell, exert force against the adhesion complexes
of the direction of motility. The Che3 system attached to the agar substrate, the force moves
S-engine Direction of cell gliding A-engine
Pili pull by Slime

disassembly ribbons
push by
secretion
Fig. 23.2 Gliding motility via type IV pili and slime secretion. The A- and S-motors function in a cooperative
manner in A+S+ cells. This is deduced from the observation that the maximum swarming rate in A+S+ cells
is greater than in A+S– or A–S+ cells. One hypothesis is that the A-motor secretes slime at one pole and this
pushes the cell forward. Support for this suggestion, which has yet to appear, would include more information
regarding the chemical and physical nature of the slime, and how it is attached to the substrate. Experimental
support exists, however, for a model according to which the S-motor extends type IV pili at the opposite end
and, when the pili attach to the cell in front (or to the substrate) and retract, the cell is pulled forward. When
the cell reverses its direction of gliding, the activities of pili extension/retraction and slime secretion switch
poles. Source: Reproduced with permission from Nature Reviews in Microbiology. Kaiser, D. 2003. Coupling
cell movements to multicellular development in myxobacteria. Nat. Rev. Microbiol. 1:45–54. Macmillan
Publishers Ltd.
the cell forward. The A-engines and the adhe-

sion complexes eventually become located at
the rear part of the forward-moving cell, where
they are disassembled and reassembled at the
leading edge of the forward-moving cell.6
Intercellular signaling stimulates swarm

expansion
The rate of swarm expansion (also called colony
expansion) via the gliding of cells away from the
outer edge of colonies by means of either the A-
or S-motility system is stimulated when the cell
density is increased (Fig. 23.3). This suggests Fig. 23.3 Rate of swarm expansion as a func-
that both motility systems are stimulated by tion of initial cell density A–S+ and A+S– mutants.
cell–cell interactions, although as we shall see Microdroplets of cells at various density units (100
later, system A drives single-cell motility and density units = 4 × 108 cells/mL) were placed on
its operation does not require that cells be close nutrient agar and the rate of swarm expansion was
together, whereas system S drives group motil- measured. The swarms expanded linearly with time
ity and requires cell–cell contact for movement as the cells glided away from the center of the drop.
to occur. Furthermore, other experiments have Approximately 90% of swarm expansion was due
to movement and 10% to growth, as determined
shown that the maximum colony expansion
by comparing swarm expansion via wild-type cells
rate in A+S+ cells is greater than in A+S– or A–S+
with swarm expansion via nonmotile mutants.
cells, indicating that the two motor types for Source: Data from Kaiser, D., and C. Crosby. 1983.
motility act cooperatively. Cell movement and its coordination in swarms of
Myxococcus xanthus. Cell Motil. 3:227–245.
23.1.3 Intercellular signaling and
multicellular development7
Early evidence for cell-to-cell signaling involved measure of the ability of the cells to aggregate
in the formation of fruiting bodies and myxo- and form fruiting bodies.) These cells had muta-
spores came from the isolation of Myxococcus tions that could be separated into five classes
mutants that were unable to form myxospores. (asg, bsg, csg, dsg, and esg), and were able to
(The formation of myxospores is a convenient sporulate when mixed with wild-type cells.
www.ebook777.com
They would also sporulate when mixed with Overview of A- and C-signaling
each other in complementary pairs (e.g., bsg+/ For a more detailed description of the A- and
csg– and bsg–/csg+). C-signaling pathways, see Sections 23.1.3 and
The existence of five classes of mutants that 23.1.7. The signaling pathway steps listed
are capable of extracellular complementation next correspond to the steps numbered in
indicates that five extracellular signals (A, B, Fig. 23.4.
C, D, E) exist for development. As we shall
describe later, two signals, the A- and C-signals, 1. A-signal
have been isolated, and these are the ones best The A-signal has been referred to as a cell den-
characterized. Signaling via the A-, B-, D-, and sity signal. It is a subset of extracellular amino
E-signals is necessary for the initiation of expres- acids that accumulates in the medium about 1
sion of developmental genes during the first 5 h to 2 h poststarvation and functions until about
of development, whereas C-signal is essential 4 h poststarvation. When the combined extra-
for the initiation of expression of developmental cellular concentration of the amino acids that
genes after 6 h, the time at which the cells begin comprise A-signal reaches a threshold value of
to aggregate into small aggregation centers that about 10 μM, the expression of many devel-
after 24 h become fruiting bodies housing the opmental genes whose products are made
myxospores. early in development is stimulated. Mutants
A-signal (a subset of
amino acids)
nutrient limitation 5
proteases
aggregation
4
AsgB
AsgA expression of genes
2 3
starvation signal ? ? required for secretion
+ AsgC
pathway (ppGpp) AsgA-P of proteases that
1 AsgB * generate A-signal
(+) fruA SasS

7 SasR-P
6 A-
10 14
SasS signal
FruA SasR
11 actABCD HPK 8 P
csgA 9 early
HPK
13 FruA-P (low levels) developmental
15 genes
16
12
CsgA Frz aggregation
FruA-P 17
(high levels) dev operon sporulation
Fig. 23.4 Model of A-signaling and C-signaling in Myxococcus xanthus. The A-signaling pathway (steps 1–9),
induced by nutrient limitation and high cell density, is responsible for the transcription of early developmental
genes. AsgA is a cytoplasmic histidine protein kinase (HPK), and AsgB is a DNA-binding protein but not a
response regulator. Studies with mutants that are deficient in the production of A-signal indicate that three
other proteins are involved in the A-signal generation signal pathway. The mutations are in the following
genes: asgC, an allele of sigA, which encodes the major sigma factor; asgD, which encodes a histidine protein
kinase similar to AsgA; and asgE, which encodes a protein whose amino acid sequence suggests that it spans
the membrane. The C-signaling pathway (steps 10–17) is induced by cell–cell contact and is responsible for
gene transcription after 6 h. Note that the early expressed genes (step 9) are A-signal dependent/C-signal
independent, whereas the later expressed genes (steps 16 and 17) are A-signal and C-signal dependent owing
to step 14. Also, steps 11 and 12 constitute an amplifying feedback loop for the expression of csgA, the gene
encoding the C-signal. Symbols: solid circles, C-signals; solid rectangle, precursor to C-signal: half-circle,
C-signal receptor.
that do not make A-signal (asg mutants) are Step 5. The proteases degrade cell surface pro-
arrested in development at around 2 h. Thus, teins to generate A-signal which, as described
A-signal can be thought of as a cell density later, is a subset of amino acids.
signal that signals starving cells when the cell Step 6. A-signal accumulates in the extracellular
density is sufficient for aggregation and fruit- medium. When the combined concentration
ing body formation. From this point of view, of the amino acids of the A-signal exceeds
the A-signal is analogous to other quorum sig- 10 μM, a membrane-bound histidine kinase,
nals, including acylated homoserine lactones SasS, is activated by autophosphoryla-
produced by a wide range of gram-negative tion and in turn phosphorylates its cognate
bacteria and peptide signals used by various response regulator, SasR. (In these enzyme
gram-positive bacteria. If the concentration names, Sas stands for suppressor of asg.)
of A-signal amino acids is above 10 mM, the Step 7. SasR-P stimulates the transcription of
cells continue to grow, rather than entering the gene fruA. This occurs early in develop-
the developmental pathway, reflecting the fact ment, that is, at approximately 6 h.
that starvation as well as cell density stimu- Step 8. It has been suggested that the gene prod-
lates development. uct FruA is phosphorylated by a putative
histidine protein kinase, HPK, to produce a
Step 1. A-signaling begins when Myxococcus
low level of FruA-P.
xanthus responds to nutrient limitation
Step 9. Low levels of FruA-P have been pro-
by activating a starvation signal pathway.
posed to stimulate the transcription of devel-
Starvation for amino acids, carbon, energy,
opmental genes that are expressed prior to
or phosphorus, but not for purine or pyrimi-
aggregation.
dine, triggers development. The earliest stages
Steps 10 through 17 are listed in the next sub-
of the signaling pathway involve the synthesis
section, on C-signaling.
of an intracellular signal called (p)ppGpp,
which refers to guanosine 3′-diphosphate, 2. C-signaling
5′-diphosphate (ppGpp), and guanosine C-signaling begins to function approximately 6
3′-diphosphate 5′-triphosphate (pppGpp). h poststarvation, after A-signaling. The C-signal
[For more discussion of the role of (p) is a cell surface protein. At the present time, the
ppGpp in regulating cell metabolism, read identity of its receptor is unknown. The C-signal
the description of the stringent response in molecule is a 17 kDa protein that is processed
Section 2.2.2.] There exists strong evidence from a 25 kDa protein by a cell surface pro-
that (p)ppGpp is part of the signaling path- tease. The 25 kDa protein is encoded by the
way in M. xanthus that turns on the expres- csgA gene. C-signaling begins at around 6 h and
sion of developmental genes when the cells becomes primarily responsible either directly or
are faced with nutrient limitation. The (p) indirectly for the transcription of developmen-
ppGpp signal is a global regulator, wide- tal genes expressed during aggregation, fruiting
spread in bacteria. body formation, and myxospore formation.
Step 2. The starvation signal pathway in step Consequently, mutants in C-signaling fail to
1, including the guanosine derivatives, acti- aggregate and to sporulate.
vates a cytoplasmic phosphotransfer system
that includes a protein histidine kinase called Step 10. C-signal activates a cytoplasmic sig-
AsgA, as well as a putative DNA-binding naling pathway when cells are in contact
protein called AsgB. AsgB is not a response with each other, end to end. C-signaling is
regulatory protein. enhanced during aggregation and fruiting
Step 3. AsgB activates transcription of genes body formation because the cells make end-
responsible for secretion of proteases, which to-end contact as they move in chains and in
as we shall see generate the A-signal. AsgB parallel arrays called streams into aggrega-
has a helix–turn–helix motif similar to many tion centers and within the developing fruit-
DNA-binding proteins. The sigma factor ing bodies. The location of the C-signal has
responsible for the AsgB-activated tran- been suggested to be at the cell poles, but at
scription is AsgC. the present time this has not been demon-
Step 4. Proteases are made and secreted. strated.
www.ebook777.com
Step 11. As a consequence of more end-to- has been shown to involve a specific set of
end contact, there is increased C-signaling amino acids. The six most active amino acids are
between the ends of the cells, stimulating their tyrosine, proline, phenylalanine, tryptophan,
cognate receptors. This sets up an amplifica- leucine, and isoleucine. It has been demonstrated
tion loop resulting in increased transcription that the amino acids restore development when
of the act operon, which contains the gene added to asg mutants. When the concentration
for C-signal, csgA. Transcription of the csgA of amino acids in the A-signal exceeds 10 μM,
gene is controlled by four genes in the act which corresponds to what is produced at a
operon. The four proteins encoded by these cell density greater than 3 × 108/mL, the signal
genes are ActA and ActB, whose functions informs the population of cells that the cell den-
are to stimulate the transcription of csgA, sity has reached the requisite number for initiat-
and ActC and ActD, both of which are ing the developmental program.
important for the timing of csgA transcrip-
tion. Sequence analysis indicates that actA Sensing A-signal
encodes a response regulator and that actB As discussed earlier, the Asg proteins appear to
encodes a σ54 activator protein. be involved in a starvation-induced signaling
Step 12. The product of the csgA gene is CsgA, cascade that activates genes required to produce
a 25 kDa protein, which is inserted into the A-signal. Once A-signal has been produced and
outer membranes. is accumulating in the external medium, how is
Step 13. The 25 kDa protein is processed by a it sensed by the cells? At least one pathway for
cell surface protease to a 17 kDa protein that sensing A-signal involves the product of the sasS
is the actual C-signal. gene.8 The nucleotide sequence predicts that the
Step 14. Increased C-signaling resulting in the product SasS will be a membrane-bound sen-
activation of a cognate FruA histidine pro- sor histidine kinase. According to the model,
tein kinase (HPK). SasS responds to A-signal by phosphorylating a
Step 15. The HPK increases the levels of FruA-P. cognate response regulator protein (SasR) that
(Recall from steps 6 and 7 that the fruA gene regulates the transcription of genes expressed
itself is transcribed as a result of prior activa- early in development. To find sasS, it was neces-
tion by the A-signal.) sary to use a developmental reporter gene (i.e.,
Step 16. The C-signal transduction pathway a gene that is expressed only during develop-
branches at FruA-P. One branch leads to the ment). The reporter gene that was used, gene
frz operon and aggregation. The frz operon 4521, is fused to Tn5 lac such that transcription
contains genes that regulate the frequency of from the 4521 promoter results in the synthe-
reversal of the direction of gliding motility. sis of β-galactosidase, which can be measured.
Step 17. The second branch from FruA-P leads The expression of the reporter gene was shown
to the dev operon (devTRS) and sporulation. to increase early in development and to require
The dev operon consists of five genes that are A-signal. (The addition of A-signal rescues the
expressed in the fruiting body. Mutants in expression of the 4521 gene in asg mutants.)
these genes do not form myxospores. As the However, expression of 4521 requires both
cells aggregate, more cell-to-cell end contacts starvation and the A-signal; that is, neither one
occur, resulting in increased C-signaling and alone is sufficient. Second site mutations in the
higher amounts of FruA-P. The aggregation mutant asg strains that expressed 4521 during
pathway (step 16) is activated at a lower level growth as well as starvation were isolated, and
of FruA-P than is the pathway of this final these mapped to the sasB locus. Thus, muta-
step, myxospore formation. This aspect of tions in the sasB locus expressed 4521 despite
the C-signaling pathway accounts for why the absence of the A-signal and regardless of
aggregation precedes myxospore formation. whether the cells were starved. These variants
are called suppressor mutants because they sup-
23.1.4 A-signal identity and generation press the mutant asg phenotype.
A-signal activity that has been isolated from One of the mutant genes in the sasB
buffer in which wild-type cells have been sub- locus, is sasS. SasS responds to the A-signal
jected to starvation (shaking the cells in buffer) and to starvation by phosphorylating itself
(autophosphorylation) and then transferring collide. As development proceeds, end-to-end

the phosphoryl group to SasR, which then acti- contact becomes more frequent, and the pro-
vates transcription of 4521. (It appears that duction of C-signal increases as a result of the
both starvation and cell density signals feed self-amplifying loop (Fig. 23.4, steps 11 and
into a signaling pathway that results in the 12). At around 6 h the amount of C-signal per
autophosphorylation of SasS during starva- cell reaches a threshold level, and at that time
tion, but not during growth, when the set of C-signaling decreases motility reversal. This
amino acids comprising the A signal exceeds a promotes the gliding of cells within streams into
threshold value.) The suppressor mutation in aggregation centers. At still higher concentra-
sasS presumably locks it into the autophospho- tions of C-signal, sporulation genes are induced
rylating conformation so that it does not need in the fruiting bodies.
the A-signal or the starvation signal. Null sasS
mutants do not express 4521 even in the pres- C-signaling requires end-to-end cell contact
ence of A-signal, in agreement with the hypoth- In contrast to A-signaling, C-signaling requires
esis that SasS is part of the signal cascade that close cell–cell contact, in particular end-to-end
couples A-signal to the expression of 4521. In contact. Nonmotile cells do not express C-signal-
addition, the null sasS mutants are not able to dependent genes, not do they sporulate, despite
form fruiting bodies and sporulate. having the wild-type csg gene, because they
cannot align themselves properly for C-signal
23.1.5 C-signal identity and transmission between the cells. However,
signaling pathway when the nonmotile cells were placed on an
Mutants defective in making C-signal (csgA agar surface and allowed to settle into grooves
mutants) do not form fruiting bodies, nor do created by lightly scratching the agar with
they form myxospores. They construct abnor- emery paper, the cells in the grooves became
mal aggregates, and only after a significant delay. aligned end-to-end parallel to each other and
C-signal is thus critical for the morphogenetic were capable of stimulating the expression of
movements of cells during aggregate and fruit- C-signal-dependent lacZ fusion reporter genes
ing body construction as well as the expression as well as sporulation.9 Cells outside the grooves
of developmental genes. Studies with develop- did not express the fusion. When exogenous
mentally expressed lacZ fusion reporter genes C-signal was added, the nonmotile mutants
indicate that C-signal acts at a developmental expressed C-signal-dependent genes and sporu-
stage after A-signal. Indeed, most of the genes lated, although they did not aggregate.
that are developmentally expressed after 6 h As discussed later in the context of the frz
are induced directly or indirectly by C-signal, genes, a hypothesis has been developed to
whereas the other extracellular signals act ear- explain how C-signal stimulates aggregation. It
lier. The signaling cascade leading to develop- has been proposed that the exchange of C-signal
mental gene expression initiated by C-signal is between cells that are moving into aggregates
being characterized. CsgA is a membrane-asso- decreases the frequency of motility reversals
ciated protein, and C-signal itself is a processed and that this is important for cells to continue
form of the protein that is located on the surface moving into aggregates.
of cell poles. The C-signal binds to a receptor at
the cell pole of the recipient cell. 23.1.6 Preventing developmental gene
expression during growth: SasN
C-signaling regulates reversal frequency The amino acids that comprise A-signal are
of gliding motility as well as myxospore also present in the growth medium. Therefore,
formation we would expect to find mechanisms for pre-
Prior to aggregate formation, when the amounts venting A-signal-dependent genes from being
of C-signal molecules on the cell surface are expressed during growth. In fact, this is the
relatively low, cells that run into each other case. The expression of the developmental gene
and make end-to-end contact both transmit 4521 (Ω4521 Tn5 lac insertion) requires both
C-signal and reverse their direction of motil- starvation and A-signal. One of the suppressors
ity. This occurs when traveling waves of cells of mutant asg is a gene called sasN that encodes
www.ebook777.com
the SasN protein, a negative regulator of devel- takes part in partially disassembling the A- and
opmental gene expression during growth. In the S-motility apparatus at their old poles so that at
absence of SasN, owing to a null mutation in any one time, one pole (the rear pole) has only
sasN, the gene 4521 is highly expressed during the A-motor and the other pole (the leading
growth and development. It has been suggested pole) has only the S-motor. Thus, in the absence
that SasN, a membrane-associated protein, of MglA both motility motors remain active at
interacts with SasS, blocking its autophospho- both poles, with the result that the cell attempts
rylation during growth. Upon starvation, SasN to move simultaneously in both directions.
is inactivated, allowing SasS to respond to the (Reassembly does not require MglA because
A-signal. mglA deletion mutants are able to assemble the
A- and S-systems.) As described next, a model
23.1.7 Switching of the A- and S-motility has been proposed suggesting that the Frz pro-
systems from one cell pole to the other teins inactivate MglA, and in this way cause an
increase in reversal frequency.
Since the A-system “pushes” the cell forward
at the lagging pole, and the S-system “pulls”
the cell forward at the leading pole, these two The Frz proteins stimulate pole switching and
systems must not only operate at opposite cell the frequency of reversals
poles but, when the cell reverses its direction The Frz proteins are part of a signal transduc-
of motility, the A- and S-motility motors must tion pathway that increases the frequency of
switch poles. This occurs on the average of reversals. (For the following discussion, see
every 10 min during growth, every 8 min during Fig. 23.5.) The frz genes are so named because
early development, and much less frequently the mutations originally isolated in these genes
later in development, when the cells are gliding resulted in the formation during development of
into aggregation centers during fruiting body abnormal aggregates that have a frizzy appear-
formation. For a review, see ref. 10. How is all ance rather than resembling discrete mounds.
this regulated? The frizzy aggregates have a tangled, swirling
Although the process is not well understood, pattern and do not develop into fruiting bodies.
several different proteins that are involved The model shown in Fig. 23.5 indicates that the
have been identified. These include the MglA Frz proteins stimulate reversals by inhibiting
protein, which regulates the switching of the MglA, the protein that decreases the frequency
A- and S-motors from pole to pole; the Frz of motility reversals.
proteins, which regulate the activity of MglA;
and C-signal, which regulates the activity of Increased C-signaling inhibits pole switching
the Frz proteins. Importantly, late in develop- and the reversal of gliding motility by
ment when the cells are traveling end-to-end in inhibiting the Frz system
streams toward and within developing fruiting Increased C-signaling that occurs later in devel-
bodies, the increased C-signaling inhibits rever- opment inhibits motility reversal. The model
sals. How this all occurs will now be described, is shown in Fig. 23.5. C-signaling does this by
starting with the MglA protein. increasing the methylation of FrzCD, a cyto-
plasmic homologue of the methyl-accepting
The MglA protein inhibits pole switching and proteins (MCPs) in the enteric bacteria; and
the reversal of gliding methyl–FrzCD inhibits Frz signaling. Indeed,
MglA decreases the frequency of motor reversals it has been demonstrated that the addition of
so that net movement can occur. In fact, mutants purified C-signal to csgA mutants results in full
defective in mglA reverse motility so rapidly methylation of FrzCD and that increased meth-
that there is no net movement in either direc- ylation of FrzCD is correlated with a reduced
tion. As a consequence, spreading of the colo- frequency of reversals. The inhibition of motor
nies does not take place. The mglA gene encodes reversals by increased C-signaling is important
a cytoplasmic GTPase. The active form of MglA for aggregation and fruiting body formation
is MglA-GTP, and without GTPase activity, it because it results in chains and streams of cells
could not act to decrease the frequency of motil- (parallel chains of cells) moving into aggregates
ity reversals. One suggestion is that MglA-GTP and remaining in the aggregates as they move
FruA
C-signal
FrzCD No C-signaling
Frz
FruA-P
methyl-FrzCD Frz MglA
decreases
motility reversals
Fig. 23.5 The Frz signaling pathway for motor reversal includes the core components FrzCD and FrzE
(not shown) that signal the reversal switch. There are two signaling pathways. In one pathway increased
C-signaling, which occurs later in development, sends a signal that decreases the frequency of reversals;
in the other pathway (during growth) C-signal is not involved, and the cells reverse more frequently. Both
the C-signal-dependent pathway and the C-signal-independent pathway converge at MglA, a protein that
decreases the frequency of reversals by ensuring that one pole retains only a functioning A-motor, whereas the
opposite pole retains only a functioning S-motor. The C-signal-dependent model results in the phosphoryla-
tion of FruA. FruA-P then stimulates the methylation of FrzCD. Increased methylation of FrzCD is correlated
with a decrease in reversals. The model proposes that methyl–FrzCD inhibits reversal frequency by inhibit-
ing the Frz system, which inhibits MglA. As a consequence, MglA remains active, and motility reversal is
decreased. In the absence of C-signaling, the Frz system inhibits MglA and, as a consequence, motility reversal
is increased. Some of the evidence supporting this model is that null mutations of frzCD and frzE reverse much
less frequently than wild type. The model is summarized from Søgaard-Anderson, L. 2004. Cell polarity,
intercellular signalling and morphogenetic cell movements in Myxococcus xanthus. Curr. Opin. Microbiol.
7:587–593.
in concentric circles. As shown in steps 14 and homologues of the chemotaxis two-component

15 in Fig. 23.4, and in Fig. 23.5, the signaling phosphorelay signaling proteins, and these sys-
pathway from C-signal requires the C-signal- tems contribute to the developmental program.
dependent phosphorylation of the response (Other bacteria, with the exception of E. coli,
regulator FruA. Thus the signaling pathway is a are known to have multiple homologues of Che
“two-component” phosphorelay pathway. proteins, reflecting different roles for these pro-
teins in phosphorelay signaling pathways.13)
23.1.8 Che3 and Che4 are also
important for development 23.1.9 Myxobacteria chemotaxis
In addition to the Dif and Frz systems that M. xanthus moves up phosphoethanolamine
utilize homologues of the E. coli chemotaxis (PE) gradients consisting of certain fatty acids
proteins to signal extracellular matrix (ECM) and displays adaptation to PE, two activities
formation and motility reversal, respectively, that are consistent with chemotaxis. Adaptation
the Myxococcus xanthus genome encodes at is correlated with increased methylation of
least two other systems that utilize chemotaxis- FrzCD, which is part of the Frz system discussed
like proteins for signaling. These two systems earlier. (Reviewed in ref. 14.) The cells glide up
are the Che3 and Che4 systems.11,12 The Che3 the gradient because PE decreases the frequency
system regulates developmental gene expres- of reversal of gliding motility. It has been pos-
sion, and the Che4 system regulates S-motility tulated that binding of PE to a receptor on the
reversal. Mutants in the Che3 system aggregate extracellular matrix that covers the cell surface
early when starved, and they show increased stimulates a signal transduction pathway that
and premature expression of developmental regulates motor reversal. Mutants that lack the
genes. Deletion of the che4 operon prevents ECM do not respond to PE.
aggregation and myxospore formation in A–S+ In addition to the ECM, the excitatory
cells (but not A+S+ cells). Thus, M. xanthus has response to PE requires a protein called DifA,
at least four sets of signaling systems utilizing which is a homologue to the MCPs and is
www.ebook777.com
believed to be a membrane protein. It has been of the swarmer cell. (The chemosensory proteins
proposed that PE binds to a surface receptor are placed at the pole opposite the stalked pole
and that a signal that decreases reversals is sent in the dividing cell and are used for chemotaxis
to the motility motors via DifA. Since the A- and by the swarmer cell. They are degraded when
S-motility motors must coordinate with each the swarmer cell becomes a stalked cell.) What
other for net movement to occur, PE must affect factors are involved in the regulation of the tim-
the reversal frequency of both motors. ing and the spatial location of these events? One
important regulatory molecule is CtrA, and this
23.2 Caulobacter Development: is discussed next.
Control of DNA Replication and
23.2.2 CtrA, a global regulator of
Cell Cycle Genes gene expression
Caulobacter crescentus has been a model sys- CtrA is a response regulator protein (cell cycle
tem for the study of bacterial cell differentiation transcription regulator A) that is activated by
because at each cell division, two different cell phosphorylation due to a phosphorelay path-
types are produced: a stalked cell and a swarmer way activated by histidine kinases. CtrA-P can be
cell. In the past few years much has been learned viewed as a global regulator of gene expression
about how the predivisional cell uses intracel- during the cell cycle. We will first examine the role
lular signals to regulate cell cycle events. For a of CtrA in regulating gene expression in both the
review, see ref. 15. stalked cell and the swarmer cell. Then we will
see how its levels and distribution between the
23.2.1 The Caulobacter life cycle stalked and swarmer cells are regulated. Finally,
Caulobacter crescentus belongs to the α sub- we will describe the pathway of phosphorylation
division of the Proteobacteria. It is an aquatic, of CtrA that results in its activation.
crescent-shaped, gram-negative bacterium that
undergoes an asymmetric cell division to produce The role of CtrA-P in regulating DNA
two very different cell types: a motile swarmer replication
cell and a sessile stalked cell (Fig. 23.6). We will begin with the control of DNA replica-
The swarmer cell differs morphologically tion by CtrA-P. Since the swarmer cell indeed
from the stalked cell not only in lacking a stalk contains CtrA-P, which is a repressor of the
but in having a flagellum and several pili at one initiation of DNA replication, this process does
pole. The two cells are also very different physi- not take place in the swarmer cell. One might
ologically. Whereas the stalked cell synthesizes ask how phosphorylated CtrA represses the
DNA and divides, the swarmer cell swims away initiation of DNA replication. For DNA repli-
and does not initiate DNA synthesis and cell cation to be initiated, transcription must occur
division until it has shed its flagellum, retracted from a promoter within the origin of replica-
its pili, and grown a stalk at the same pole that tion. CtrA-P binds to the origin of replication in
previously harbored the flagellum and pili. the swarmer cell and in so doing represses this
Thus, the swarmer cell serves to disperse C. transcription. Hence it prevents the initiation of
crescentus, whereas the stalked cell attaches to DNA replication. As shown in Fig. 23.7, CtrA
surfaces via an N-acetylglucosamine adhesin levels are high in the swarmer cell, drop in the
(called a holdfast) at the tip of the stalk, grows, young stalked cell, allowing DNA replication to
and reproduces swarmer cells. Organic nutri- be initiated, and then rise in the stalked cell, pre-
ent is often concentrated at surfaces, and this is venting the reinitiation of DNA replication.
probably related to why Caulobacter attaches CtrA-P regulates DNA replication in another
to surfaces. way. When DNA is synthesized in the stalked
Clearly the Caulobacter cell cycle is complex. cell, the new strand is not methylated and there-
Processes such as the replication of DNA, the fore the replicated DNA is hemimethylated.
synthesis of flagella and pili, the production of Unless DNA becomes fully methylated, replica-
the chemosensory proteins, and cell septation tion cannot be initiated in the stalked cell in the
take place at specific times and in specific cellular next cell cycle. However, near the end of DNA
locations in the stalked cell prior to production replication in the late predivisional cell, CtrA-P
Fig. 23.6 Life cycle of Caulobacter crescentus. The swarmer cell is a swimming, nongrowing cell specialized
for dispersal. It has a single flagellum, several pili, and chemosensory receptor proteins at the same cell pole.
After a period of swimming it settles down to become a sessile stalked cell that synthesizes DNA, grows, and
divides. During the transition it degrades the chemosensory receptor proteins, sheds its flagellum, retracts its
pili, and produces a stalk at the same cell pole. The stalk is a cylindrical extension of the cell and has a poly-
saccharide adhesive cap, called a holdfast, that anchors the cell to the substratum. As the stalked cell grows,
it produces a flagellum, pili, and chemosensory proteins at the pole opposite the stalk. As a consequence, cell
division produces two cell types: a motile swarmer cell and a stalked cell that remains sessile. The stalked
cell continues to grow and divide. Source: Adapted from Dworkin, M. 1985. Developmental Biology of the
Bacteria. Benjamin/Cummings, San Francisco.
activates transcription of ccrM, the gene for accounting, in part, for why there is no FtsZ in
CcrM DNA methylase. As a consequence, the swarmer cells. Recall that FtsZ forms the sep-
DNA becomes fully methylated, and thus rep- tal ring required for cytokinesis (Section 2.6.2).
lication can be initiated in the early stalked cell Since swarmer cells do not divide, they do not
when the CtrA levels drop. need FtsZ.
The repression of the ftsZ gene occurs only
The role of CtrA-P in regulating after sufficient FtsZ has accumulated for sep-
transcription of ftsZ tation to occur in the stalked cell. As septation
CtrA-P represses the transcription of ftsZ progresses, the concentration of FtsZ declines
in swarmer cells and predivisional cells, owing to proteolysis. At the time of cell division,
www.ebook777.com
Fig. 23.7 Relative levels of CtrA (shaded areas) during the Caulobacter cell cycle are controlled by spatially
regulated proteolysis. CtrA prevents the initiation of DNA replication and the transcription of ftsZ. It is pres-
ent in swarmer cells and is degraded during the transition phase when swarmer cells become stalked cells. The
ctrA gene is transcribed in the stalked cell; but proteolysis in the half of the predivisional cell destined to be the
progeny stalked cell ensures that only the swarmer cell is born with CtrA. The ellipse within the cell represents
DNA. Source: Adapted from Quon, K. C., B. Yang, I. J. Domian, L. Shapiro, and G. T. Marczynski. 1998.
Negative control of bacterial DNA replication by a cell cycle regulatory protein that binds at the chromosome
origin. Proc. Natl. Acad. Sci. USA 95:120–125.
essentially all of the FtsZ has been degraded. the early stalked cell does not. This is because
The stalked cell resynthesizes FtsZ for the next the ClpXP protease complex destroys CtrA in
division, but the swarmer cell does not start syn- the early stalked cell derived from a swarmer
thesis again until it has begun to differentiate cell, allowing DNA replication to begin in the
into a stalked cell. early stalked cell. Then CtrA is made and is dis-
tributed throughout the predivisional cell prior
CtrA-P activates the transcription of many to flagellum synthesis. It is then destroyed in the
genes in the predivisional cell stalked cell half of the predivisional cell so that
In addition to binding to the origin of replica- once cell division has taken place, DNA replica-
tion and preventing transcription, and thus the tion can begin in the progeny stalked cell, but
initiation of DNA replication in the swarmer not in the progeny swarmer cell.
cell, and also repressing ftsZ so that FtsZ is not The destruction of CtrA is due to spatially
made in the swarmer cell but only during the regulated proteolysis in the stalked cell half of
swarmer-to-stalked-cell transition period and the predivisional cell, as well as during the tran-
in the young stalked cell, CtrA-P activates the sition from swarmer to stalked cell. Temporal
transcription of many genes. These include cell and spatially regulated proteolysis implies that
division genes (ftsI, ftsQ, ftsA), genes required there are cell cycle signals that activate prote-
for the synthesis of flagellar components, genes olysis of CtrA at certain times (temporal con-
required for chemotaxis, and pilA, the gene trol) and that mechanisms exist to ensure that
for the synthesis of the pilus subunit, as well as proteolysis takes place only in certain regions
ccrM, the gene for a DNA methyltransferase of the cell (spatial control). One key player is
that is required to fully methylate newly repli- the response regulator protein DivK, which is
cated DNA so that DNA replication can be ini- phosphorylated throughout the cell cycle (Fig.
tiated in the early stalked cell. 23.8). DivK activates the proteolysis of CtrA
via the ClpXP protease, presumably by phos-
How the levels and distribution of CtrA are phorylating components of a phosphorelay sys-
regulated tem that activates the protease. Cold-sensitive
Examine Fig. 23.7, in which the shaded areas (cs) mutants of divK make a mutant protein
denote CtrA. The swarmer cell has CtrA, but (DivK-cs) at the restrictive temperature and do
not degrade CtrA (although the mutant form of cell is inherited from the predivisional cell.
DivK is phosphorylated). This results in the pre- However, once the swarmer cell has become
vention of the initiation of DNA replication and a stalked cell, the ctrA gene is transcribed and
cell cycle progression beyond the stalked stage CtrA accumulates.
when swarmer cells are incubated at the restric-
tive temperature (20 °C). The phenotype is fila- 23.3 Sporulation in Bacillus subtilis
mentous cells with abnormally long stalks.
For reviews of B. subtilis sporulation, see refs.
Figure 23.8 suggests that DivK-P receives
16 and 17. Bacillus subtilis is a gram-positive
a phosphoryl group and transfers it to CtrA.
bacterium that lives in the upper portions of
However, the amount of CtrA-P in DivK-cs
soil, and like other bacteria has evolved ways
cells incubated at the restrictive temperature
to adjust to stressful changes in the natural
is similar to that found in wild-type cells incu-
environment. One of these ways is to sporulate
bated under the same conditions. Because of
when faced with limiting supplies of a carbon or
this, it has been suggested that DivK-P transfers
nitrogen source. In the laboratory, this occurs
its phosphoryl group to a protein other than
when the population enters stationary phase.
CtrA, and that this is part of a signaling path-
During a relatively brief period of the cell cycle,
way that activates the ClpXP protease.
B. subtilis decides whether to form a midcell
The levels of CtrA are also regulated at the
septum and continue cell division, or to form
transcriptional level. The ctrA gene is tran-
a polar septum and sporulate. The decision to
scribed in the stalked cell but not in the swarmer
sporulate is regulated by a phosphorelay signal
cell; hence the CtrA protein in the swarmer
transduction system to be described later.
B. subtilis experiences several physiological
changes during adaptation to nutrient depri-
Signal CckA Signal Chromosome
Signal
replication vation, and sporulation is only one of them.
x DNA Adaptation includes synthesizing a complex
methylation motility and chemotaxis system, which in the
PleC DivK CtrA Cell division natural habitat would increase the chances of
Flagellum
finding nutrient, and producing antibiotics that
DivJ Pili
inhibit the growth of other organisms that com-
Chemotaxis
Signal DivL pete for the limiting nutrient. In addition, the
Signal
motile cells secrete proteases and other degra-
dative enzymes that might generate nutrients
Fig. 23.8 Model of the signal transduction net- from carbon sources in the natural environ-
work controlling the activity of the master response ment. If starvation continues, the cells develop
regulator, CtrA. Response regulators and the pro-
competence for the uptake of exogenous DNA
tein kinases are shown as circles and boxes, respec-
tively. CckA, PleC, and DivJ are histidine kinases.
similar to their own DNA (and therefore in the
DivL is a tyrosine kinase. DivJ phosphorylates natural environment from other B. subtilis cells
DivK, and PleC plays a role in dephosphorylating in the population), and later the population spo-
DivK-P or prevents the phosphorylation of DivK. rulates. Sporulation is a last resort that occurs
CtrA is phosphorylated and stable at the restrictive when the population of cells has increased to a
temperature in cells that have a temperature-sensi- high cell density and growth is no longer possi-
tive mutation in the gene encoding DivK. Because ble because there is not sufficient nutrient for the
of this, it has been suggested that DivK-P partici- dense population of cells to continue to grow.
pates in a phosphorelay pathway that activates The spore is a dormant stage in the life cycle
the ClpXP protease that degrades CtrA, rather and is resistant to environmental stresses such as
than phosphorylating CtrA, as implied in the fig-
heat, ultraviolet radiation, and toxic chemicals.
ure. Source: Ausmees, N., and C. Jacobs-Wagner.
2003. Spatial and temporal control of differen-
Spores can remain dormant for hundreds of
tiation and cell cycle progression in Caulobacter years but will germinate into growing cells (veg-
crescentus. Annu. Rev. Microbiol. 57:225–247. etative cells) when nutrient becomes available.
Reprinted with permission, from the Annual It is well known that B. subtilis will sporulate
Review of Microbiology, Volume 57, © 2003, by more efficiently at high cell densities. This is cor-
Annual Reviews, www.annualreviews.org. related with the production and accumulation
www.ebook777.com
in the media of extracellular pentapeptides, some bacteria at the tips of cells by a process
which serve as cell density signals. Peptides are similar to budding.
often used by gram-positive bacteria as inter- The stages of endospore formation are shown
cellular signals. When the cell densities are in Fig. 23.9. The growing vegetative cells are said
low, Spo0F-P, which is part of the phosphore- to be in stage 0. As the cells get ready for spo-
lay system that activates sporulation genes, is rulation, the two copies of the chromosomes,
dephosphorylated by Rap phosphatases. As the which may be fully or partially replicated, con-
cell density increases, pentapeptide signaling dense and elongate into an axial filament. This
molecules produced by the bacteria accumulate has been referred to as stage I. Then a septum is
in the medium, enter the cells, and inhibit the laid down asymmetrically near one pole of the
Rap phosphatases so that sporulation occurs. cell.
Once the septum has been completed, the cell
23.3.1 Stages in sporulation (now called a sporangium) is divided into two
Morphological description compartments: a mother cell compartment and
The spore forms inside the cytoplasm of the pro- a smaller forespore compartment. (Sometimes
genitor cell as a consequence of an asymmetric the forespore compartment is referred to as
septation and is therefore called an endospore, a prespore compartment.) At this early stage
as opposed to exospores, which are formed by (stage II), about one-third of the spore DNA is
I II
V–VI IV III
coat
cortex
VII
Fig. 23.9 Stages in sporulation of Bacillus. Vegetative cells are in stage 0. The two chromosomes from the
completed round of replication become aligned along the long axis of the cell in stage I, with their origins
attached to opposite poles of the cell. At stage II a septum has formed near one pole, dividing the cell into
a mother cell compartment and a forespore, trapping approximately one-third of the forespore chromo-
some in the forespore. DNA translocation across the forespore septum takes place, causing the forespore
chromosome to be moved entirely into the forespore. Once the septum has formed, the cell is referred to as a
sporangium. The mother cell engulfs the forespore, bringing the sporangium to stage III, where the forespore
is set free in the cytoplasm of the mother cell as a protoplast. There is a space between the inner and outer
membranes of the spore protoplast, and in stage IV a cortex consisting of peptidoglycan is synthesized in
that space. The cortex covers a peptidoglycan cell wall (the germ cell wall) that is laid down on the surface of
the inner forespore membrane. A polypeptide multilayered coat is synthesized by the mother cell around the
developing spore in stage V. A proteinaceous exosporium is also made. The exosporium covers the spore coat
as a layer that is either loosely fitting or more closely fitting, depending upon the species of Bacillus. The spore
(also called an endospore) matures during stage VI, developing resistance properties, and is released from the
lysed sporangium in stage VII. The entire process, including the release of the spore, takes between 8 and 10
h. Each flattened circle (wavy line) represents a nucleoid.
trapped in the forespore compartment. After division, it is reasonable to conclude that sporu-
this has occurred, the remaining two-thirds of lation is an example of cell division that leads to
the spore DNA is quickly translocated into the two cells with identical genes but very different
forespore compartment by the DNA translo- developmental fates owing to differential gene
case, SpoIIIE. During stage III the mother cell expression.
septum grows around the forespore, a process
called engulfment, and in so doing pinches off Sporulation genes
the forespore as a protoplast floating in the cyto- B. subtilis sporulates when it expresses spo
plasm within the mother cell compartment. genes, which were discovered by examining
When engulfment is complete, the forespore mutants that failed to complete sporulation.
in stage III has two sets of membranes, an inner The genes are named after the stage of block-
and an outer membrane. age (0, I, II, etc.) and are distinguished from
In stage IV, a cortex, consisting of peptido- one another by a letter. For example, mutants
glycan, is synthesized in the space between the in spo0A fail to initiate sporulation and do not
inner and outer forespore membranes. The cor- proceed to stage I, and spoIIA mutants com-
tex covers another layer of peptidoglycan called plete stage II (septation), but fail to proceed to
the germ cell wall, which is made on the surface stage III (engulfment).
of the inner forespore membrane. During stage
V a spore coat is synthesized surrounding the Factors that regulate the expression of
outer membrane. The coat is made of protein sporulation genes
and is produced by the mother cell. At the end 1. Sigma factors
of stage V, the prespore is dehydrated and phase The vegetative sigma factor is σA, which is
bright. The resistance properties of the spore required for the expression of genes during veg-
develop during stage VI, and the mature spore etative growth as well as certain genes required
is released as a result of lysis of the mother cell for sporulation. In addition to σA there exist
during stage VII. The entire process, including several sporulation-specific sigma factors that
the release of the spore, takes around 8 to 10 h are made in a sequential order during sporula-
and requires the activities of at least 113 sporu- tion. These sigma factors, σH, σF, σE, σG, and σK,
lation-specific genes. are located in either the mother cell compart-
ment or the forespore compartment, where they
Sporulation is an example of cell division transcribe genes. For a summary of the location
ending in two different developmental fates and roles that these sigma factors play in their
for the daughter cells respective compartments, as well as a discus-
sion of how their activities are regulated, see
The polar septum that forms during sporulation
Box 23.2. In addition to the sigma factors, a
is similar to the septum that forms at midcell
master transcription factor called Spo0A, which
during cell division when the cells are growing
is activated via a two-component phosphorelay
vegetatively. Both require the same genes for
system when the cells receive signals to sporu-
synthesis, and both consist of two septal mem-
late, is also critical. This is described next.
branes with a layer of peptidoglycan between
them. However, there are some differences. 2. Spo0A
For example, the sporulation septum is much Spo0A is activated by phosphorylation to
thinner than the midcell septum, and the pepti- Spo0A-P. Spo0A-P is a transcription factor
doglycan in the sporulation septum is autolyzed whose amounts increase at the beginning of
with complete loss of wall material. This does sporulation. Spo0A activates transcription of
not occur during vegetative division, and the genes required for sporulation and represses
wall material in the septum remains at the poles the transcription of other genes expressed dur-
of the new cells after vegetative cell division. ing postexponential growth. It has been aptly
Although at the time of polar septation the fore- referred to as the master regulator of sporula-
spore has only one-third of its chromosome, the tion genes. Spo0A-P is responsible for axial fila-
remaining two-thirds is rapidly moved into the ment formation, polar septation leading to the
forespore via the DNA translocase, SpoIIIE. forespore, and compartmentalized gene expres-
Despite these differences with ordinary cell sion in the mother cell and forespore. A brief
www.ebook777.com
BOX 23.1 PROTEINS INVOLVED IN FORMATION OF THE SPORE

SEPTUM AND CHROMOSOME PARTITIONING
FtsZ transfer proteins (Tra proteins) of conju-

gative plasmids of Streptomyces, and that
The student should read Chapter 1 for a dis- mutations in this region block chromosome
cussion of the role of FtsZ as well as other transfer into the prespore compartment.1
cell division genes in septum formation and
cell division in E. coli. Prior to the forma-
tion of the asymmetric septum in stage Spo0J
II, the cells form an FtsZ ring at both cell
poles. The establishment of the FtsZ ring at Spo0J is one of three proteins that are
the cell poles rather than in the cell center believed to recruit the oriC
C region to the cell
is an early event that distinguishes cell divi- poles. The other two proteins are DivIVA
sion from sporulation. and RacA.
Spo0J localizes to the cell poles along
with the chromosomal origins of repli-
SpoIIIE cation (oriC). It has been suggested that
Spo0J aids in positioning the origin of
The translocation of the chromosome replication of the chromosome to the cell
into the forespore compartment after sep- poles, perhaps by binding to sites near oriC
tum formation requires the product of the and to proteins at the pole. Indeed, it has
spoIIIE gene, which codes for an ATP- been demonstrated that SpOJ binds to sites
dependent DNA translocase located at the near oriCC of the B. subtilis chromosome.2
leading edge of the growing septum that SpoJ also functions during chromosome
partitions the forespore compartment from partitioning during vegetative growth.
the mother cell compartment. Consistent with this, null mutants of spo0J
Septation actually bisects the prespore produce a significant increase in cells with-
chromosome, trapping inside the forespore out DNA during vegetative growth.
compartment, perhaps attached to the cell
pole, approximately one-third of the pre-
spore chromosome, which includes the ori- RacA
gin of replication. The product of spoIIIE is
required to move the remaining two-thirds RacA (remodeling and anchoring of chro-
through the septum into the forespore. mosome A) is synthesized during sporula-
The SpoIIIE protein forms a pore through tion but not during vegetative growth.3
which the remainder of the prespore DNA RacA binds nonspecifically to the DNA but
travels. preferentially near oriC. It may be involved,
The mechanism of transfer of the remain- along with Spo0J, in the binding of the
ing two-thirds of the chromosome into the chromosome to the cell pole during sporu-
prespore compartment through the spore lation. Mutations in racA frequently result
septum is believed to be similar to the in forespores without DNA. RacA interacts
transfer of plasmid DNA during conjuga- at the cell pole with DivIVA.
tion between cells. (The prespore compart-
ment and the mother cell compartment
are analogous to two separate cells.) This DivIVA
conclusion is supported by the finding that
the carboxy-terminal domain of SpoIIIE DivIVA is an anchor protein at the cell poles.
has significant sequence similarity to DNA It probably binds directly or indirectly to
RacA and to Spo0J, which themselves are partitioning during sporulation in subtilis.
bound to the chromosome in the area of the Genes Dev. 9:1316–1326.
oriC
C region. 2. Lin, D. C.-H., and A. D. Grossman. 1998.
Interestingly, the binding of RacA to Identification and characterization of a bac-
DivIVA may promote polar septation. It terial chromosome partitioning site. Cell
92:675–685.
has been suggested that the binding of RacA
to DivIVA displaces the division inhibitor, 3. Wu, L. J., and J. Errington. 2003. RacA and
MinCD, at the pole, thus allowing septum the Soj–Spo0J system combine to effect polar
chromosome segregation in sporulating Bacillus
formation at the pole.4 The suggestion is
subtilis. Mol. Microbiol. 49:1463.
based upon an earlier publication showing
that DivIVA sequesters MinCD at the poles 4. Ben-Yehuda, S., D. Z. Rudner, and R.
Losick. 2003. RacA, a bacterial protein that
so that cell division occurs at midcell during
anchors chromosomes to the cell poles. Science
vegetative growth.5 299:532–536.
REFERENCES 5. Marston, A. L., H. B. Thomaides, D. H.

Edwards, M. E. Sharpe, and J. Errington. 1998.
1. Wu, L. J., P. J. Lewis, R. Allmansberger, P. Polar localization of the MinD protein of Bacillus
M. Hauser, and J. Ellington. 1995. A conjuga- subtilis and its role in selection of the mid-cell
tion-like mechanism for prespore chromosome division site. Genes Dev. 12:3419–3430.
summary of how Spo0A-P accomplishes all of the forespore. These proteins are FtsZ, which
this follows. forms a ring that will develop into the polar
Spo0A-P directly regulates the expression of septum, SpoIIIE, which moves the DNA from
genes in the Spo0A regulon. An estimate puts the mother cell into the forespore compart-
the number of genes in the Spo0A regulon at ment, and Spo0J, RacA, and DivIVA, which
121, of which 40 are positively regulated and bring the oriC regions to the cell poles during
81 are negatively regulated. Twenty-five of the chromosome partitioning. These proteins were
genes in the Spo0A regulon are themselves tran- described in Box 23.1.
scription factors, setting the number of genes
indirectly regulated by Spo0A at around 400. 23.3.3 Phosphorelay system for initiation
Mutations in spoA result in cells blocked at of sporulation
stage 0 of sporulation. The genes activated by The decision to sporulate rests upon the inte-
Spo0A-P include genes responsible for the for- gration of a variety of environmental and physi-
mation of the axial filament (including racA), ological signals that are transduced via various
as well as a gene (spoIIE) that encodes a pro- signaling pathways that result in the phosphor-
tein that facilitates the formation of polar FtsZ ylation of Spo0A, which is the master response
rings. Spo0A-P also directly activates transcrip- regulator protein for activating the transcrip-
tion of spoIIA and spoIIG, which encode the tion of the sporulation genes. The phosphore-
first prespore-specific sigma factor, σF, and lay system has intermediate phosphoryl donors
the first mother-cell-specific sigma factor, σE, (phosphorelay proteins) between the histidine
respectively. kinase and Spo0A.
23.3.2 Proteins involved in formation The phosphorelay model for the initiation of
of the spore septum and chromosome sporulation
partitioning The model shown in Fig. 23.10 stipulates that
There are several proteins that are impor- a histidine kinase receives a signal for sporu-
tant for the formation of the spore septum lation and as a consequence autophosphory-
and for the partitioning of the chromosome lates. There are actually five histidine kinases
between the mother cell compartment and involved in sporulation. However, two are
www.ebook777.com
BOX 23.2 SPORULATION-SPECIFIC SIGMA FACTORS
σH the chromosome near the origin region at

the cell poles during sporulation (but not
The earliest acting sporulation-specific during vegetative growth). The spoIIA
sigma factor is σH, which also controls the operon and racA are also under the con-
transcription of some stationary phase trol of Spo0A.
genes. The main vegetative sigma factor,
σA, transcribes the σH gene (spo0H) weakly
during exponential growth but greatly σE
increases the transcription of the gene at
the initiation of sporulation. The transcrip- Sigma factor σE is important for the tran-
tion of the gene for σH is stimulated by the scription of genes in the mother cell com-
active form of the response regulator pro- partment after the forespore septum has
tein Spo0A. (Spo0A does this indirectly. formed. The gene for σE is transcribed prior
It represses transcription of the gene for to the formation of the forespore septum
AbrB, which is a negative regulator of the at about the same time as the gene for σF,
gene for σH.) In the predivision cell, σH is a forespore-specific sigma factor, is tran-
required for the transcription of genes scribed. A model for how σE is restricted to
important for axial filament and polar sep- the mother cell compartment despite being
tum formation. The genes transcribed by made prior to septation can be found in ref.
σH-RNA polymerase include genes encod- 1. The protein is synthesized in an inactive
ing proteins in the phosphorelay pathway, form, called pro-σE, which is presumed to
including spo0A, kinA, kinE, and spo0F, be degraded in the forespore compartment,
as well as the operon that contains the gene accounting for the restriction of σE to the
for σF. The subsequent formation of the mother cell compartment. (An alterna-
septum at one of the two FtsZ rings also tive suggestion is that pro-σE exists in both
requires σH, which activates the transcrip- compartments and the σE is degraded in the
tion of the ftsAZ operon, whose products forespore compartment.)
(FtsA and FtsZ) are required for formation Activation of pro-σE to σE requires
of the septum. A cell septum must also form proteolytic cleavage of an N-terminal
during vegetative growth. However, the sequence of 27 amino acids. The putative
promoter that initiates transcription of the protease, SpoIIGA, which cleaves pro-σE
ftsAZ operon during sporulation is differ- to active σE, is made prior to septation, is
ent from the promoter that is used during initially inactive in the septal membrane,
vegetative growth and is dependent upon and is activated by a protein made in the
σH. In the absence of σH, a polar septum forespore as a result of transcription of by
does not form. σF-polymerase.
Other genes regulated by σH include the The protein that activates the protease is
spoIIA operon, which includes the gene called SpoIIR, and it migrates to the fore-
encoding σF, whose product is necessary spore side of the septum to activate the
to transcribe genes in the forespore, as well protease. This is another example of com-
as racA, required for axial filament for- munication between the mother cell com-
mation. As described in Box 23.1, RacA partment and the forespore.
binds preferentially near oriC, and less The functions of σE include triggering
specifically throughout the DNA. RacA is engulfment of the forespore and the initia-
required for the extension of the nucleoid tion of the assembly of the spore coat, as
into an axial filament during sporulation, well as stimulating the synthesis of σK in the
and it is also required for the anchoring of mother cell compartment.
σF by σF, which explains why σG is localized

to the forespore. Initially, σG is inactive
The initiation of forespore-specific gene because it binds to an inhibitor protein.
transcription requires σF. The gene for σF The activation of σG in the forespore is due
is transcribed, along with the gene for σE, to a complex of membrane proteins made
before the forespore septum is made, and in the mother cell, as directed by σE. The
it is kept in an inactive form in both the activating proteins are associated with the
mother cell compartment and the forespore. mother cell membrane before engulfment
Transcription requires the positive tran- and become part of the outer membrane of
scription regulator, Spo0A, which is part the forespore. This is one of the ways that
of the phosphorelay signaling system, and the mother cell ensures the proper develop-
σH. It is kept inactive by an anti-sigma fac- ment of the forespore.
tor (SpoIIAB) that binds to it and prevents it
from binding to the RNA polymerase. The
inhibition is reversed by an anti-anti-sigma σK
factor (SpoIIAA) in the forespore, but not
in the mother cell compartment. The later stages of sporulation depend upon
In its dephosphorylated form, SpoIIAA the activity of σK in the mother cell com-
interacts with SpoIIAB, resulting in the release partment. Transcription of the gene for σK
of σF. This happens only in the forespore is directed by σE, which explains why σK is
compartment because there exists a septum- found only in the mother cell compartment.
bound phosphatase, called SpoIIE, produced The protein product, σK, which is inactive,
in the predivisional cell, that dephosphory- is activated by proteolytic removal of a 20
lates SpoIIAA-P in the forespore. amino acid sequence from the N-terminal
The functions of σF include transcrib- end. The activation of pro-σK in the mother
ing spoIIIG, which encodes σG, which in cell compartment depends upon σG in the
turn activates late gene transcription in forespore. It has been proposed that σG
the forespore. As mentioned previously, σF directs the transcription of a gene encoding
participates in the expression of mother cell an inner membrane forespore protein that
compartment genes by directing the expres- interacts with outer membrane forespore
sion of a forespore gene that encodes a pro- proteins synthesized in mother cell compart-
tein that migrates to the forespore side of ment (as a result of σE-polymerase), and the
the septum and stimulates a septum-associ- result is cleavage of pro-σK in the mother cell
ated protease that converts pro-σE to active compartment to active σK. Active σK stimu-
σE in the mother cell compartment. lates the transcription of genes important
for cortex and coat synthesis, maturation of
the spore, and release of the spore.
σG
REFERENCE
After engulfment, σ replaces σ as the pri-
G F
1. Hilbert, D. W., and P. J. Piggott. 2004.
mary sigma factor for the transcription of Compartmentalization of gene expression dur-
sporulation genes in the forespore. The ing Bacillus subtilis spore formation. Microbiol.
transcription of the gene for σG is directed Mol. Biol. Rev. 68:234–262.
most important for sporulation in laboratory residue in a phosphotransferase protein called

media. One is KinA, which is cytoplasmic, and Spo0B. Spo0B-P then transfers the phospho-
the other is KinB, which is membrane bound. ryl group to an aspartate residue in a response
Spo0F-P functions as a phosphorelay protein regulator protein called Spo0A. As discussed
and transfers the phosphoryl group to a histidine earlier, Spo0A-P is the transcription factor that
www.ebook777.com
proteins, Spo0F and Spo0A, via a phospho-

transferase Spo0B. The phosphorelay system
offers more sites for fine tuning or control than
some of the more simple two-component sys-
tems. This may be necessary to ensure that B.
subtilis does not stop multiplying and enter the
sporulation pathway prematurely.
Regulatory role of phosphatases

In many two-component systems the level of
phosphorylation of the response regulatory
protein is determined by the kinase and phos-
phatase activities of the histidine kinase. In other
words, many histidine kinases are bifunctional
enzymes and have both kinase and phosphatase
activities. However, in the phosphorelay that
controls sporulation, the levels of the phos-
phorylated response regulator proteins are
determined by separate phosphatases. In fact,
there exist several phosphatases that take part
in adjusting the levels of the phosphorylated
response proteins.
The existence of multiple phosphatases
allows multiple inputs into the regulation of
Fig. 23.10 Phosphorelay system during initiation of the phosphorelay pathway. One of the phos-
sporulation in B. subtilis. A histidine kinase auto-
phatases, Spo0E, dephosphorylates the mas-
phosphorylates in response to a signal and then
transfers the phosphoryl group to a phosphorelay
ter transcription regulator, Spo0A-P. It is
protein, Spo0F. There are two histidine kinases not known how Spo0E is regulated. Two of
shown, KinA (cytoplasmic) and KinB (membrane the phosphatases, RapA (response regulator
bound). Three other histidine kinases, Kin C, Kin D, aspartyl phosphatase A) and RapB, dephos-
and KinE, also exist and are discussed in the text. phorylate the phosphorelay protein, Spo0F-P.
Spo0F-P then phosphorylates the phosphorelay pro- A third phosphatase, RapE, which dephos-
tein, Spo0B, which in turn transfers the phospho- phorylates Spo0F-P and whose activity is con-
ryl group to a response regulator protein, Spo0A. trolled in a similar fashion to that of RapA, has
Spo0A-P is a transcriptional activator for sporula- been reported. However, it is less significant
tion genes. (Spo0A-P also represses transcription of than RapA in regulating the onset of sporula-
the negative regulator gene, abrB, which turns off
tion, and it appears to play an accessory role.
transcription of genes normally expressed during the
transition phase between exponential growth and
Regulation of the level of phosphorylated
sporulation.) At least three phosphatases are impor- response regulators by separate phosphatases
tant for regulating the level of Spo0A-P. They are occurs in other phosphorelay systems.
RapA (Spo0L) and RapB (Spo0P), which dephos-
phorylate Spo0F-P, and Spo0E, which dephospho- How the RapA phosphatase is controlled
rylates Spo0A-P. It is clearly important to understand how B.
subtilis regulates the amounts and activities
is the primary regulator of the expression of the of the phosphatases that control the phospho-
sporulation genes. relay pathway. A model for how the RapA
phosphatase is regulated was published in
Advantage to the phosphorelay 1996, and it involves a two-component sig-
The phosphorelay system described for B. sub- naling system. (See Fig. 23.11.) There exists a
tilis sporulation differs from standard two- two-component signaling system consisting
component systems in that the phosphate is of a membrane-bound histidine kinase called
transferred between two response regulator ComP and a response regulator protein called
ComA. Upon receiving a signal, ComP phos- However, there must be additional steps in
phorylates ComA, which then activates the the regulatory pathway. This is because the
transcription of the rapA operon. There are rapA operon encodes the RapA phosphatase,
two genes in the rapA operon. One is rapA, in addition to the phosphatase inhibitor. It
which encodes the RapA phosphatase protein, may be that control of initiation of sporula-
and the other is phrA, (phosphatase regulator tion involves, in part, regulation of the secre-
A), which encodes the protein PhrA, which tion and processing of PhrA and/or its uptake
when processed inhibits the activity of RapA into the cell.
phosphatase.
Since the ComP/CompA two-component How the RapB phosphatase is controlled
system produces an inhibitor of the RapA There is a third phosphatase that dephospho-
phosphatase, it stimulates sporulation. To rylates Spo0F-P, thereby preventing sporula-
inhibit the RapA phosphates, PhrA must be tion. It is called RapB. Cells cannot sporulate
processed. PhrA is secreted from the cell and if the levels of RapB are too high. However,
cleaved extracellularly to a pentapeptide unlike rapA (and rapE), the ComA/ComP
(PhrA peptide) that is imported back into the two-component system is not involved in the
cell through an oligopeptide transport system transcription of rapB, and a Phr peptide is not
called Opp. involved in inhibiting the activity of RapB.
Once inside the cell, PhrA peptide acts as an Instead, rapB is constitutively transcribed
inhibitor of RapA phosphatase activity and, throughout the growth cycle, and the RapB
as a consequence, sporulation is not inhib- phosphatase is inhibited by the CSF peptide, a
ited by the RapA phosphatase. Genetic evi- quorum sensor that increases as the cell density
dence for this is as follows: if the phrA gene increases during growth, reaching active levels
is deleted, then RapA phosphatase activity is at the end of exponential growth and the onset
very high and sporulation is inhibited. It has of stationary phase. Thus, the CSF peptide
also been demonstrated that the PhrA peptide should stimulate sporulation when the levels
inhibits RapA phosphatase activity in vitro. rise appropriately.
Fig. 23.11 Regulation of RapA phosphatase by PhrA. An unknown signal for competence stimulates the
ComP histidine kinase and causes phosphorylation of the ComA response regulator. The phosphorylated
ComA then stimulates the transcription of the rapA operon, which encodes RapA phosphatase and PhrA.
The RapA phosphatase dephosphorylates Spo0F-P so that sporulation does not occur while the cells are in
the competence process. The phrA gene product is secreted to the cell surface, where it is processed to a pen-
tapeptide (PEP5) that reenters the cell via the oligopeptide permease. PEP5 inhibits RapA so that sporulation
can take place. Certain steps in the secretion and/or processing or uptake of PhrA are probably regulated to
prevent the inhibition of the RapA phosphatase while the cell is in the competence pathway. Source: Perego,
M. 1997. A peptide export–import control circuit modulating bacterial development regulates protein phos-
phatases of the phosphorelay. Proc. Natl. Acad. Sci. USA 94:8612–8617.
www.ebook777.com
There are some similarities between CSF a replicating cell. The stalked cell (“mother
and the Phr peptide. Each is synthesized as a 40 cell”) is a sessile, feeding cell that under-
amino acid protein precursor that is exported goes cell division to produce a non-replicat-
by the general secretory pathway out of the cell, ing, nonfeeding, swimming cell called the
where it is processed to a 5 amino acid peptide swarmer cell (“daughter cell”). Eventually,
that is brought back into the cell via the same the swarmer cell settles down and develops
oligopeptide permease (Opp), and ultimately into a stalked cell, and cell division resumes.
inhibits a phosphatase (RapB) that dephospho- A global regulator of gene expression is a
rylates Spo0F-P. response regulator protein call CtrA. CtrA is
activated by phosphorylation due to a phos-
phorelay pathway involving histidine kinases.
23.4 Summary CtrA-P regulates DNA replication as well as
Bacteria often sense signals that alter develop- the transcription of many genes that are acti-
mental patterns. The signals transfer informa- vated during cellular differentiation.
tion to the genome, usually via two-component Bacillus subtilis undergoes several changes
phosphorelay systems, and the consequence as part of its adaptation to nutrient depriva-
is a change in the expression of specific genes tion. One of these changes is sporulation, which
that results in changes in metabolism and/or occurs more efficiently at high cell densities.
developmental patterns. This chapter discusses Extracellular pentapeptide signals produced
these events as they take place in myxobacteria, by B. subtilis serve as cell density signals. The
Caulobacter, and Bacillus. pentapeptides enter the cells and inhibit the
Myxococcus xanthus produces five extracel- Rap phosphatases that prevent sporulation by
lular signals (A, B, C, D, E) to regulate gene dephosphorylating Spo-OF-P, which in turn is
expression and cellular movement that results part of the phosphorelay system that activates
in fruiting body formation and myxospore for- sporulation genes.
mation when cells face starvation conditions.
One of these signals, the A-signal, is postulated
Study Questions
to be sensed by a membrane-bound histidine
kinase (SasA) and this begins a phosphorelay
1. Rationalize the formation of fruiting bod-
signal transduction system that regulates the
ies in myxobacteria in terms of their feeding
expression of developmental genes that are
habits and dispersal.
important for early stages in development.
Another one of the five signals is C-signal, 2. Describe the mechanisms proposed for
which acts later than the other signals. C-signal motility in myxobacteria. Give a possible
is a membrane protein located on the surface of explanation of why myxobacteria have two
the cell poles. The signaling cascade initiated motility systems.
by C-signal transfer between cells stimulates
3. What is the model for how extracellu-
the movement of cells into aggregation centers
lar signals influence gene transcription in
and stimulates the expression of genes required
myxobacteria? Do extracellular signals
for myxospore formation. The myxospores are
affect Bacillus sporulation? Explain.
housed in mature fruiting bodies. In addition,
Myxococcus xanthus has at least four sets of 4. How would you describe the overall differ-
signaling systems that use homologues of the ence and similarity in cell division between a
chemotaxis two-component phosphorelay sporulating Bacillus and Caulobacter? How
signaling systems in E. coli, and these homo- does Myxococcus fit into this picture?
logues are part of the developmental program
5. Discuss the role of two-component phos-
in M. xanthus.
phorelay systems, with specific examples,
Caulobacter development does not rely
in bacterial development
on external signals, but rather on a complex
internal signaling process that results in cells 6. Why is CtrA called a global regulator of gene
of two very different types when cell division expression in Caulobacter? Summarize the
occurs. One cell is the stalked cell, which is role of CtrA.
7. List the adaptive changes that Bacillus 8. Yang, C., and H. B. Kaplan. 1997. Myxococcus
subtilis undergoes in response to nutrient xanthus sasS encodes a sensor histidine kinase
required for early developmental gene expression. J.
adaptation.
Bacteriol. 179:7759–7767.
8. What is the role of extracellular pentapep- 9. Kim, S. K., and D. Kaiser. 1990. Cell alignment
tides in sporulation of Bacillus subtilis? required in differentiation of Myxococcus xanthus.
Science 249:926–928.
REFERENCES AND NOTES 10. Kaiser, D., and R. Yu. 2005. Reversing cell polar-
ity: evidence and hypothesis. Curr. Opin. Microbiol.
1. Dworkin, M., and D. Kaiser (Eds.). 1993. 8:216–221.
Myxobacteria II. ASM Press, Washington, DC.
11. Kirby, J. R., and D. R. Zusman. 2003.
2. Kaplan, H. B. 2003. Multicellular development Chemosensory regulation of developmental gene
and gliding motility in Myxococcus xanthus. Curr. expression in Myxococcos xanthus. Proc. Natl.
Opin. Microbiol. 6:572–577. Acad. Sci. USA 100:2008–2013.
3. Kaplan, H., and D. E. Whitworth. 2007. 12. Vlamakis, H C., J. R. Kirby, and D. R. Zusman.
Myxobacteria: Multicellularity and Differentiation. 2004. The Che4 pathway of Myxococcus xanthus
ASM Press, Washington, DC. regulates type IV pilus-mediated motility. Mol.
4. Lancero, H., N B. Caberoy, S. Castaneda, Y. Microbiol. 52:1799–1811.
Li, A. Lu, D. Dutton, X.-Y Duan, H. B. Kaplan, 13. Berleman, J. E., and C. E. Bauer. 2005. A Che-like
W. Shi, and A. G. Garza. 2004. Characterization signal transduction cascade involved in controlling
of a Myxococcus xanthus mutant that is defec- flagella biosynthesis in Rhodospirillum centenum.
tive for adventurous motility and social motility. Mol. Microbiol. 55:1390–1402.
Microbiology 150:4085–4093.
14. Kearns, D. B., and L. J. Shimkets. 2001. Lipid
5. Mutants in nla24 are defective in the production chemotaxis and signal transduction in Myxococcus
of extracellular polymeric substance (EPS), which is xanthus. Trends Microbiol. 9:126–129.
part of the polysaccharide/protein fibril matrix that
plays a role in S-motility, as well as the expression of 15. Curtis, P. D., and Y. Brun. 2010. Getting in the
agllU and calB that are required for A-motility. (The loop: regulation of development in Caulobacter cres-
EPS and associated protein are sometimes referred to centus. Microbiol. Mol. Biol. Rev. 74:13–41.
as the extracellular matrix. 16. Hilbert, D. W., and P. J. Piggot. 2004.
6. Mignot, T., J. W. Shaevitz, P. L. Hartzell, and Compartmentalization of gene expression during
D. R. Zusman. 2007. Evidence that focal adhesion Bacillus subtilis spore formation. Microbiol. Mol.
complexes power bacterial gliding motility. Science Biol. Rev. 68:234–262.
315:853–856.
17. Piggot, P. J., and D. W. Hilbert. 2004. Sporulation
7. Kaiser, D. 2004. Signaling in myxobacteria. of Bacillus subtilis. Curr. Opin. Microbiol. 7:
Annu. Rev. Microbiol. 58:75–98. 579–586.
www.ebook777.com
BOXED MATERIAL
1.1. Phylogeny 4
1.2. Nonflagellar Motility 8
1.3. Historical Perspective: Christian Gram 18
1.4. Tuberculosis 23
1.5. Leprosy 26
1.6. Cannulae and Hami: Cell Surface Structures of Archaea 46
2.1. Transcriptional, Translational, and Post-Translational Regulation of Levels of rpoS 62
3.1. Historical Perspective: The Discovery of DNA and Its Role 78
3.2. Historical Perspective: The Structure of DNA 79
4.1. Historical Perspective: Oxidative Phosphorylation 112
5.1. Historical Perspective: Cellular Respiration 153
8.1. Historical Perspective: Energy Transfer in the Cytosol 208
9.1. Historical Perspective: Cell-Free Yeast Fermentation and the Beginnings of Biochemistry 224
9.2. Vitamins 239
9.3. Historical Perspective: The Citric Acid Cycle 241
12.1. Capsules as Virulence Factors 327
16.1. Historical Perspective: The Morse Code 421
18.1. SRP-Dependent Protein Translocation across the Endoplasmic Reticulum
Membrane in Eukaryotes 460
21.1. Historical Perspective: Biofilms in the Ocean 556
22.1. Early Discoveries of Prokaryotic Cell–Cell Communication 567
23.1. Proteins Involved in Formation of the Spore Septum and Chromosome Partioning 604
23.2. Sporulation-Specific Sigma Factors 606
xiii
CONVERSION FACTORS, EQUATIONS,

AND UNITS OF ENERGY
Electrode potential at pH 7
E′h = E′0 + [RT/nF] ln([ox]/[red]) volts, where n refers to the number of electrons, and [ox] and
[red] refer to the concentrations of oxidized and reduced forms, respectively. When [ox] = [red],
then E′h = E′0 = E′m:
E′h = E′0 + (60/n)log10([ox]/[red]) mV at 30 °C
Electron, charge
1.6023 × 10−19 C
Gibbs energy
For the reaction aA + bB ↔ cC + dD,
ΔG = ΔG0 + RT ln[C]c[D]d/[A]a[B]b
Gibbs energy and equilibrium constant
ΔG′0 = −RT ln K′eq or −2.303RT log10K′eq
Gibbs energy for solute uptake and concentration gradient
ΔG = RT ln[C]in/[C]out at 30 °C
ΔG = 5.8 log10[C]in/[C]out kJ/mol
or
ΔG/F = 60 log10[C]in/[C]out mV
Growth
g(k) = 0.693
x = x02Y
where Y is the number of generations and x is mass or any parameter that changes linearly with
mass: this is the equation for exponential growth
Y = t/g
Light, energy in a quantum
E = νh = hc/λ
where ν = frequency (c/λ), h = Planck’s constant, λ = wavelength
E (kJ) = 1.986 × 10−19/λ
where λ is in nanometers
E (eV) = 1.24 × 103/λ
xix
www.ebook777.com
xx conversion factors, equations, and units of energy
Light, energy in an einstein

E = Nhν = Nhc/λ = 1.197 × 105 kJ/λ
where λ is the wavelength in nanometers, N is Avogadro’s number, and c is the speed of light
E (kJ) = 1.196 × 105/λ
Nernst equation
ΔΨ = −(RT/nF) ln[S]in/[S]out V or ΔΨ = −60/n log [S]in/[S]out mV at 30 °C
where [S] is the concentration of a diffusible cation (Sin > Sout) of valence n, and ΔΨ is the mem-
brane potential, inside negative. The equation states that at equilibrium the chemical driving
force due to the outward diffusion of S along its concentration gradient is equal to the electrical
driving force drawing S into the cell. According to the equation, each 10-fold concentration dif-
ference of a permeant monovalent cation corresponds to a potential difference of 60 mV.
Phosphorylation potential
ΔGp = ΔG′0 + RT ln[ATP]/[ADP][Pi]
Proton potential
Δp = ΔμH+/F = ΔΨ − 60 ΔpH mV at 30 °C
Proton potential and ΔEh at equilibrium
−nΔEh = yΔp
where n is the number of electrons transferred over a redox potential difference of ΔEh volts, and
y is the number of protons translocated over a proton potential difference of Δp volts.
2.303RT
5.8 kJ/mol, or 1.39 kcal/mol, at 30 °C
2.303RT/F
0.06 V, or 60 mV, at 30 °C
DEFINITIONS
Acetogenic bacteria Anaerobic bacteria that of grams per mole are not appropriate (e.g.,
synthesize acetic acid from CO2 and secrete when one is referring to ribosomes).
the acetic acid into the medium. Dehydrogenase Dehydrogenases are enzymes
Acidophile Organism that grows between pH 1 that catalyze oxidation–reduction reactions
and 4 and not at neutral pH. in which hydrogens as well as electrons are
Aerobe Organism that uses oxygen as an elec- transferred. They are named after one of the
tron acceptor during respiration. substrates (e.g., pyruvate dehydrogenase).
Aerotolerant anaerobe Organism that cannot Drug Any chemical that affects the physiology
use oxygen as an electron acceptor during of an organism. This includes drugs such as
respiration but can grow in its presence. alcohol and caffeine, as well as therapeutic
Anaerobe Organism that does not use oxygen drugs to treat disease, and antimicrobial
as an electron acceptor during respiration. drugs to treat infections. Antimicrobial
Alkaliphile (alkalophile) Organism that grows drugs include antibiotics, which are made
at pH above 9, often with an optimum by microorganisms and kill or prevent the
between 10 and 12. growth of other microorganisms, and syn-
Antibiotic Antimicrobial compound produced thetic drugs.
by a microorganism. Einstein One “mole” of light (6.023 × 1023
Antimicrobial agents Antibiotics and chemi- quanta).
cally synthesized antimicrobial compounds, Electrode potential The tendency of a molecule
among others. to accept an electron from another molecule is
Autotroph Organism that uses CO2 as sole or given by its electrode potential, E, also called
major source of carbon. the reduction potential, the redox potential,
Chaperone proteins Proteins that transiently or the oxidation–reduction potential.
bind to other proteins and assist in proper Facultative anaerobes Organisms that can
folding of the target protein and/or transport grow anaerobically in the absence of oxygen
to a correct cellular site. or will grow by respiration if oxygen is avail-
Chaperonins Multisubunit complexes of chap- able.
erone proteins. Facultative autotroph Organisms that can
Chemolithotroph See Lithotroph. grow on CO2 as sole or major source of
Cytoplasm The fluid material enclosed by the carbon or on organic carbon.
cell membrane. Gel electrophoresis Procedure in which macro-
Cytosol The liquid portion of the cytoplasm. molecules such as proteins or nucleic acids
Dalton (Da) All atomic and molecular weights are applied to a polyacrylamide or agarose
refer to the carbon isotope, 12C, which is 12 gel and subjected to an electrical field. The
Da or 1.661 × 10−24 g. Daltons are numeri- macromolecules having a net electrical
cally equal to molecular weights and can be charge migrate in the electrical field and can
used as units when molecular weight units be separated according to their size.
xxi
www.ebook777.com
xxii definitions
Growth yield constant, Y The amount of dry amino acids are joined together, the molecule
weight of cells produced per weight of nutri- is referred to as a polypeptide (MW < 10,000)
ent used. or a protein (MW > 10,000).
Halophile Organism that requires high salt Phosphorylation potential The energy required
concentrations for growth. to phosphorylate one mole of ADP by using
Heat-shock proteins Proteins that transiently physiological concentrations of ADP, Pi, and
increase in amount relative to most cell ATP.
proteins when the temperature is elevated. Photon Quantum; a particle of light.
Several are chaperone proteins. Photosynthesis The use of light as a source of
Heterotroph Organism that uses organic car- energy for growth.
bon as a major source of carbon. Photoautotroph Organism that uses light as a
Holliday junction An intermediate in homolo- source of energy for growth and CO2 as the
gous recombination in which a cruciform source of carbon.
(cross-shaped) structure is formed. Photoheterotroph Organism that uses light as
Hyperthermophile Organism whose growth a source of energy for growth and organic
temperature optimum is 80 °C or greater. carbon as the source of cell carbon.
Isomerase Enzymes that catalyze the transfer Phototroph Organism that uses light as the
of chemical groups (e.g., hydrogen or phos- source of energy for growth.
phate) within molecules to produce isomeric Psychrophile Organism that grows best at tem-
forms of a molecule with the same chemical peratures of 15°C or lower, and does not
formula. For example, glucose (C6H12O6) grow above 20°C.
and fructose (C6H12O6) are isomers of each Quantum A particle of light (photon).
other because they have the same chemical Regulon A set of noncontiguous genes con-
formula, although their chemical structures trolled by the same transcription regulator.
are different. Standard conditions All reactants and prod-
Leader region The region of messenger RNA ucts are in their “standard states.” This
for an operon that is 5′ of the coding region means that solutes are at a concentration
for the first gene. of one molar (1 M = one mole of solute
Lithotroph Organism that oxidizes inor- per liter) and gases are at one atmosphere.
ganic compounds as a source of energy for Biochemists usually take the standard state
growth. of H+ as 10−7 M (pH 7). By convention, if
Mesophile Organism whose growth tempera- water is a reactant or product, its concentra-
ture optimum is between 25 and 40 °C. tion is set at 1.0 M, even though it is 55.5 M
Midpoint potential (Em) The electrode potential in dilute solutions.
when the molecule is 50% oxidized. Strict anaerobes Organisms that will grow only
Mutases A subclass of isomerases; enzymes in the absence of oxygen and are killed by
that transfer a functional group (e.g., a phos- traces of oxygen.
phate group) from one part of a molecule to Tautomerism Refers to an isomerization in
another. which the isomeric forms are easily intercon-
Neutrophile Organism that grows at a pH opti- vertible and an equilibrium exists between
mum near neutrality. the isomers. An example is the enol–keto
Obligate anaerobes Organisms that will grow tautomerization in which enol pyruvic acid
only in the absence of oxygen but are not readily converts to pyruvic acid.
necessarily killed by oxygen Thermophile Organism that can grow at tem-
Oligopeptide Refers to a molecule of a few amino peratures greater than 55°C.
acids joined together by peptide bonds. Also Ym Molar growth yield constant; grams of dry
referred to as a peptide. A pentapeptide is weight of cells produced per mole of nutrient
an example of an oligopeptide. When many used.
SYMBOLS
c speed of light (3.0 × 108 m/s)

C coulomb
cal calorie (4.184 J)
E0 standard redox potential (reduction potential) at pH 0
E′0 standard redox potential at pH 7
Em midpoint potential at a specified pH (e.g., Em′,7); for pH 7, also written as E′m, which
is numerically equal to E′0
Eh actual redox potential at a specified pH; for pH 7, E′h or Eh,7
eV electron volt. The work required to raise one electron through a potential differ-
ence of one volt. It is also the work required to raise a monovalent ion (e.g., a
proton) through an electrochemical potential difference of one volt. One electron
volt = 1.6 × 10−19 J.
F Faraday constant (approximately 96,500 C/mol). The charge carried by one mole
of electrons or monovalent ion. It is the product of the charge carried by a single
electron and Avogadro’s constant.
g generation time (time/generation)
ΔG0 standard Gibbs free energy change at pH 0 (J/mol or cal/mol)
ΔG′0 Gibbs free energy at pH 7
ΔGp phosphorylation “potential” (energy required to synthesize ATP via physiological
concentrations of ADP, inorganic phosphate, and ATP)
h Planck’s constant (6.626 × 10−34 J · s)
J joule: one coulomb-volt (C × V). The work required to raise one coulomb through a
potential difference of one volt.
k instantaneous growth rate constant (time−1)
kJ kilojoule
K absolute equilibrium constant
K′ apparent equilibrium constant at pH 7
xvii
www.ebook777.com
xviii symbols
Δμion electrochemical potential difference, expressed in joules, between two solutions of

an ion separated by a membrane
Δμion /F electrochemical potential difference expressed as volts or millivolts
ΔμH+/F the proton motive force; electrochemical potential difference in volts or millivolts of
protons between two solutions separated by a membrane. Also written as Δp.
N Avogadro’s number (6.023 × 1023 particles/mol)
Δp See ΔμH+/F
ΔpH difference in pH between the inside and outside of the cell; usually, pHin − pHout
R the ideal gas constant (8.3144 J K−1 mol−1 or 1.9872 cal K−1 mol−1)
V volt; the potential difference across an electric field
ΔΨ membrane potential; usually Ψin − Ψout
Π osmotic pressure (Pa)
INDEX
Index Terms Links
A. See Absorbance
ABC transporters. See ATP-binding cassette
transporters
Absorbance (A) 55–56 56f
Accessory pigments. See Light-harvesting pigments
Acetate
acetyl-CoA formation during growth on 219
fermentation of 391–92 391f
methanogenesis from 368f 370–71
oxidation of, in acetyl-CoA pathway 364f 372
Acetoacetyl-ACP 279n8
Acetogenesis 391–92 391f
Acetogenic bacteria 380n4
Acetyl-CoA
acetyl phosphate formation from 218–19
cell material incorporation of 363–65
condensation reactions 247 247f 248f
formation of 217–18 217f 218f
381n33
in acetyl-CoA pathway 363 364f
during growth on acetate 219
from pyruvate 394
methanogen incorporation of 371–72 371f
pyruvate oxidation to 217–18 217f 218f
238–41 240f 394
Acetyl-CoA carboxylase 279n6
This page has been reformatted by Knovel to provide easier navigation.
www.ebook777.com
Index Terms Links
Acetyl-CoA pathway 363–65

acetate oxidation 364f 372
acetyl-CoA production in 363 364f
autotrophic growth with 364f 365–66
in Clostridium thermoaceticum 365–66
CODH in 366
in methanogens 364f 366–69 367f
368f 369f
Acetyl-CoA synthetase, ADP-dependent 219
Acetyl-P, formation of, from pyruvate 394
Acetyl phosphate 216–19
ATP formation from 219
flagella biosynthesis and 509
formation of 218–19
phosphorylation of 539
in two-component signaling systems 508–9
Acid-fast bacteria 28
Acidophiles 5
Δp contributions of ΔpH and ΔΨ in 118
pH homeostasis in 403 404t 406
AcrAB–TolC drug efflux system 445–46 446f
Acridine dyes 100–101 110n56
Acrylate pathway, propionate fermentation via 388–89
Acylated homoserine lactones (AHLs) 574–76
in biofilms 575
in gram-negative bacteria 574–75 575t
signaling without LuxR and LuxI-type proteins 577 577f
synthase family 575–76
Acyl derivatives, group transfer potential of 212–13 212f
Adenosine monophosphate (AMP), biosynthesis
from IMP 267 269f
Adenosine triphosphate (ATP)
as donor 211f 213 214f
formation of, from acetyl phosphate 219

Index Terms Links
Adenosine triphosphate (ATP) (Cont.)

group transfer potentials and 209t 210–11
hydrolysis of, in Δp generation 128
peptide bond formation with 214–15 215f
solute transport driven by 437–41
ABC transporters 438–40 439f
ATP-driven K+ influx 440–41
shock-sensitive transport system 438
synthesis of
in coupling sites 158 158f
with lithotrophy 353
oxalate:formate exchange in 131
oxidative phosphorylation in 112
Δp and 138–39
with photosynthesis 186–87
by substrate-level phosphorylation 209f 209t 215–20
216f 216t
Adventurous motility, in myxobacteria 590–91 591f
Aerobic repression, of photosynthetic pigment
synthesis 505
Aerobic respiration 147
in Escherichia coli 163–64 164f
in Paracoccus denitrificans 166–67 166f
Aerobic respiratory control (Arc) system
anaerobiosis response of 490–92 491f
Aerotaxis 546
Agrobacterium tumefaciens, virulence genes in 521–22
AHLs. See Acylated homoserine lactones
AI-2. See Autoinducer 2
Alanine, synthesis of 271 273f
Aldol cleavage reactions, in glycolysis 229–30 229f
Aldonic acids, ED pathway and 236–38
Algae, light-harvesting pigments in 190–91 192f
www.ebook777.com
Index Terms Links
Aliphatic hydrocarbons 273–75

degradative pathways of 273–75
hydroxylation of
at one end 273–75 277f
on penultimate carbon 275 277f
Alkaliphiles
Δp and ATP synthesis in 138–39
pH homeostasis in 403 404t
+ +
Na /H antiporter for 404f 405–6
Allosteric effectors 199
Amino acids 267–73 271t
catabolism of 272–73 275f
synthesis of 269–72
aspartate and alanine 271 273f
glutamate and glutamine 269–71 272f
serine, glycine, and cysteine 270f 271–72 274f
transamination reactions from glutamate 271 273f
tRNA attachment of, in protein synthesis 299 300f
Aminoacyl tRNAs
in protein synthesis 305–6
tRNA and 313n18
Ammonia-oxidizing bacteria 344t 345–46 345t
346f 356n26
AMP. See Adenosine monophosphate
Amphipathic phospholipids 34–35
Anabolic pathway
citric acid cycle 244
glycolysis 227–28 227f 228f
Anaerobic bacteria
energy conservation by 384 385f
oxygen toxicity of 383–84 384t
Anaerobic food chain 386–87 386f

Index Terms Links
Anaerobic induction, of photosynthetic pigment

synthesis 505
Anaerobic respiration 147
in Escherichia coli 164–65 165f
in Paracoccus denitrificans 166f 167
Anaerobiosis, facultative anaerobes response to 488–94
Arc system 490–92 491f
FblA regulon 491
FNR system 489f 490 492–94
493f
formate-hydrogen lyase pathway 489f 494
metabolic changes 488–90 489f
Nar regulatory system 490–91
regulatory systems of gene expression 490–94
Anoxygenic phototrophs 176–78 176t
Antibiotics
protein synthesis inhibition 310 310t
transcription inhibition 296
Antimicrobial drugs
biofilms and 555
resistance to 451n58
Anucleate cells 91
Appendages 6–16
fimbriae, pili, and fibrils 14–16 14f
flagella 6–14 9f
Aquaporins, of bacterial cell membranes 35–36
Arabinogalactan 27–28 28f
Archaeal cell walls 31–32 31f
Archaeal flagella 14
Arc system. See Aerobic respiratory control system
A-signal 574
Aspartate, synthesis of 271 273f
ATP. See Adenosine triphosphate
AT pathway. See Autotransporter pathway
www.ebook777.com
Index Terms Links
ATP-binding cassette (ABC) transporters 438–40 439f 461

in EPS translocation 330–31 331f
ATP-dependent proteases, Hsps 415
ATP synthase
description of 123–24 123f
mechanism of 124 125f
Δp and 123–24 450n53
solute transport and 444 444t
Autoinducer 2 (AI-2), signaling via 577–78 577f
Autotransporter (AT) pathway 469–71 470f
immunoglobulin A protease 470–71
two stages of 469–70
Available hydrogen method, for balancing
fermentations 387 388t
Axial filaments 12
Azospirillum brasilense 543
b6f complex 198n24

Bacillary dysentery 520–21
Bacillus subtilis
chromosome partitioning in 99–100 107n20
factors for 99–100
proteins involved in 605
during sporulation 97f 99
during vegetative growth 97f 99
competence and endosporulation in 568–70 568t 569f
571f
sporulation in 601–10
factors for 603–5
genes for 603
phosphorelay system for initiation of 605–10
RapA phosphatase 608–9 609f

Index Terms Links
Bacillus subtilis (Cont.)

RapB phosphatase 609–10
sigma factors in 606–7
stages in 602–5 602f
Bacterial cell membrane 34–36 35f
Bacterial development 587–611
Caulobacter crescentus 598–601
CtrA in 598–601 600f 601f
life cycle of 598 599f
myxobacteria 587–611
Bacterial electron transport chains 152–56 155f
branching of 155–56 155f
different terminal oxidases 154–55
Bacteriochlorophyll 189–90 197n14
oxidation of 178–80 179f 180f
structure of 191–92 193f
Bacteriocin production 572
Bacteriorhodopsin 5
Δp generation by 135–37 136f 137f
138f
Bactoprenol 319–20 320f
Basal body 7 9 9f
base-excision repair, for deaminated bases 418–19 419f
Bayer’s patches 32
bc1 complex 160–62 161f 198n24
Beer-Lambert law 55
Betaine 428n24
Bifidum pathway 392–94
Biofilms 551–65
AHLs in 575–76
evolutionary processes in 561–62
genetic diversification 562
horizontal gene transfer 561–62 561f
formation and dissolution of 557 557f
www.ebook777.com
Index Terms Links
Biofilms (Cont.)
attachment of 557–58
dispersal and emigration from 558–59
maintenance of 558
maturation of 552f 558
inhibition of 560
multicellular structures 551–52 552f
in ocean 556
prevalence and importance of 552–54
corrosion and clogging of pipes 552–53
dental 552f 553
food-borne illness 553
in-dwelling medical devices 553–54
lung tissue in CF patients 554
wastewater treatment 553
properties of 554–55
antimicrobial tolerance 555
matrix 555
structural forms and hydrodynamics 552f 554–55
regulation of formation of 559–60
1,3-Bisphosphoglycerate 216 216f 217f
Butanediol, fermentation of 394–97 396f
Butanol-acetone, fermentation of 398–400 399f
Butyrate, fermentation of 397–400 398f 399f
γ-Butyrolactones 578 580f
bvg genes, pertussis and 519–20
C1 metabolism 358–82
carbon dioxide fixation systems 358–74
methylotrophs 374–77
Calvin cycle 358–63
carbon balance of 361–62
glycolysis and PPP and 362–63 362f

Index Terms Links
Calvin cycle (Cont.)

reactions of 359
stages of 359–61 359f 360f
cAMP system, Cra system compared with 512
Capping proteins, of flagella 10
CAPs. See Cationic antimicrobial peptides
Capsules 16
E. coli EPS 326–28 327f
as virulence factors 327
Carbohydrate metabolism. See Central metabolic
pathways
Carbohydrate synthesis, with photosynthesis 187
Carbon balance
of Calvin cycle 361–62
for citric acid cycle 244
for glyoxylate cycle 249
for PPP 234–35
Carbon dioxide (CO2), methanogenesis from H2
and 368–70 368f 369f
Carbon dioxide (CO2) fixation systems 358–74
acetyl-CoA incorporation by methanogens 371–72 371f
acetyl-CoA pathway 363–65
acetate oxidation 364f 372
in Clostridium thermoaceticum 365–66
in methanogens 364f 366–69 367f
368f 369f
Calvin cycle 358–63
methanogenesis
from acetate 368f 370–71
from CO2 and H2 368–70 368f 369f
energy conservation during 368f 369–70 369f
reductive citric acid cycle 372–74 373f
Carbon monoxide dehydrogenase (CODH), in
acetyl-CoA pathway 366
www.ebook777.com
Index Terms Links
Carbon sources, response to 510–15

Carboxydobacteria 344t 345 345t
Carboxysomes 39
Carnitine 381n37
Carotenoids 187–88 188t 197n15
structure of 192 192f
Catabolite repression 442
in E. coli 510–13
by glucose 64–65
by other molecules 65
via CcpA system 513–14 513f
Catabolite repressor/activator (Cra) system 510
cAMP system compared with 512
gene expression and repression by 510 511f
nitrite reductase gene repression 512–13 512f
Catalase 384 384t
Cationic antimicrobial peptides (CAPs) 533n142
Caulobacter crescentus 598–601
cell cycle of 100
chromosome partitioning in 100
CtrA in 598–601
DNA replication 598–99 600f
ftsZ transcription 599–600
gene transcription 600
regulation of 600–601 600f 601f
life cycle of 598 599f
CcpA system, catabolite repression via 513–14 513f
Cell-cell communication. See Signaling
Cell cycle, of Caulobacter crescentus 100
Cell division 69–71
DNA replication and 108n23
proteins required for 70 70f
septum formation 69–70 70f

Index Terms Links
Cell division (Cont.)

sequence of events in 70–71
sporulation as 603
Cell-free yeast fermentation 224–25
Cell membrane 34–38
archaeal 36–38
bacterial 34–36 35f
phospholipids and structure of 260 262f
Cell shape 42–44
Cell size, nutrient limitation and 59
Cell structure 6–44
appendages 6–16
archaea surface 46
cell membrane 34–38
cell shape 42–44
cell walls 18–32
cytoplasm 38–42
glycocalyx 16–18 17f
periplasm 32–34 32f
Cell walls 18–32
archaeal 31–32 31f
gram-negative 28–31 29f
gram-positive 19f 20–22 21f
22f 27–28
periplasm in 34
Gram stain 18–20 19f
Central metabolic pathways 222–54 223f
citric acid cycle 241–45
chemistry of reactions in 247–48
reductive cycle modification of 246–47 246f
ED pathway 236–38 236t 237f
glycolysis 224–30 226f
modified pathway 231 231f
glyoxylate cycle 248–49 249f
www.ebook777.com
Index Terms Links
Central metabolic pathways (Cont.)

NADH
fate of 230
structures of 230–31 230f
PEP formation 249–50
PPP 231–36
pyruvate and PEP carboxylases 245–46 245f 251f
pyruvate dehydrogenase reaction 238–41 240f
pyruvate formation from malate 251
relationship summary of 251 251f
CF. See Cystic fibrosis
Chaperone proteins 308–10
DnaK 309
GroE system 308–9
Hsps 414
measuring flux of proteins through 308
for protein transport 454
role of 308
trigger factor 309–10
Chaperone/usher (CU) pathway 470f 471
Chemical energy 113
Chemiosmotic theory 111–12 113f 114f
357n36
Chemoeffectors 534
Chemostat 68 68f
Chemotaxis 534–35
See also Methyl-accepting
chemotaxis protein
adaptation 536
without methylation 540
model for 539–40
different systems of 541–43
Azospirillum brasilense 543
non-tumbling 542

Index Terms Links
Chemotaxis (Cont.)
Rhizobium meliloti 543
Rhodobacter sphaeroides 542–43
without MCPs 541
measurement of 548n39
model for 537–40 540t
acetyl phosphate phosphorylation 539
adaptation 539–40
signal gain 538–39
stimulation 537–38 538f
of myxobacteria 597–98
proteins required for 536–37
cytoplasmic 536–37
membrane 537
repellant action mechanism 541
tumbling 535 535f 536f
Che proteins 536–37
Chloride pump, halorhodopsin as 137–38
Chlorophyll 187 188f
structure of 191–92 193f
Chloroplasts
photosynthesis of 184–86
photosynthetic electron transport in 185–86 186f
two light reactions in 184–85 185f
Chlorosomes 40 187 189
190f
Cholera, ToxR and 515–17
Cholera toxin (CT) 476n40
secretion of 469
structure of 531n128
Chromosome partitioning and separation 91–92
in Bacillus subtilis 97f 99–100 107n20
605
www.ebook777.com
Index Terms Links
Chromosome partitioning and separation (Cont.)

in Caulobacter crescentus 100
cell division and 108n23
decatenation 92
directionality in 96
in eukaryotes 93
mutant phenotypes 91
Par proteins 96–97
in prokaryotes 93–94 95f
proteins involved in 95–96
RacA 96 97f
replisome in 94–95 95f 108n35
SetB and MreB 98–99
in sister chromosomes 94–96
site-specific recombination at dif 92 93f
SMC and Muk proteins 97–98 98f
topoisomerase IV 91–92 92f
Citrate lyase 381n33
Citric acid cycle 222 223f 241–45
as anabolic pathway 244
carbon balance for 244
chemistry of reactions in 247–48
acetyl-CoA condensation reactions 247 247f 248f
decarboxylation reactions 229f 232f 248
248f
distribution of 244–45
historical perspective on 241–42
reactions of 242–44 243f
reductive cycle modification of 246–47 246f 372–74
373f
regulation of 244
relationship with other pathways 251 251f
summing up 244
Clostridium acetobutylicum 398–400 399f

Index Terms Links
Clostridium propionicum, fermentation pathway of 388–89 389f

Clostridium thermoaceticum, acetyl-CoA pathway in 365–66
CO2. See Carbon dioxide
CoASH 279n5
CODH. See Carbon monoxide dehydrogenase
Codons, tRNA and 313n15
Coenzymes 172n4
Cofactors 172n4
Cold-shock proteins 412
Competence
in Bacillus subtilis 568–70 568t
in Streptococcus pneumoniae 570–71
Complex II. See Succinic dehydrogenase
Conjugation, sex pili in 15–16
Continuous growth 68–69
Cooperative feeding, of myxobacteria 588
Cord factor 22 27–28 28f
Corrinoid enzymes 380n6
Coupled translation 307–8 308f
Coupling sites 154f 156–58
ATP to electron ratio in 158 158f
Escherichia coli and electron flow 164
identification of 156–57 157f
of mitochondria 146–47
proton potential created at 157f 159–63
proton pumps 160f 162–63
Q cycle 159–62 161f
Q loops 159 160f
Covalent modification 199 204 205t
C period 75n32 75n33
Cpx system 414
Cra system. See Catabolite repressor/activator
system
Crenarchaeota 3
www.ebook777.com
Index Terms Links
Crescentin 44
Crown gall tumors 521
C-signaling 581–82 582f
CT. See Cholera toxin
CU pathway. See Chaperone/usher pathway
Cyanobacteria 176 176t
heterocyst formation in 568t 573
light-harvesting pigments in 187–88 190 191f
192f
photosynthesis of 184–86
photosynthetic electron transport in 185–86 186f
two light reactions in 184–85 185f
Cyclic photophosphorylation 178–80 180f
Cysteine, synthesis of 270f 271–72 274f
Cystic fibrosis (CF) lung tissue, biofilms in 554
Cytochrome c, in periplasm 33
Cytochromes
classes of 150 151f 152f
dual-beam spectroscopy of 146t 150–52
as electron carriers 147 148t 150–52
reduced minus oxidized spectra 150
Cytokines 479n82
Cytoplasm 38–42
cytosol 42
inclusion bodies, multienzyme aggregates, and
granules 39–42
intracytoplasmic membranes 38–39
Cytoskeleton proteins
cell shape and 43–44
eukaryotic 42 53n131
in Mollicutes 44
prokaryotic 43
Cytosol 42

Index Terms Links
Cytosol bioenergetics 207–21

ATP synthesis 209f 209t 215–20
216f 216t
group transfer potential 207–10 209f 209t
acyl derivatives 212–13 212f
ATP 209t 210–11
central role of 213–15
PEP 209t 211–12
phosphoenolpyruvate 220 220f
succinyl-coenzyme A 219–20 219f
Dam methylase. See Deoxyadenosine methylase

Dark avoidance 544–45
Deaminases 273 276f
Death 58
Decarboxylation reactions, of citric acid cycle 229f 232f 248
248f
Decatenation, at dif 92
Deletion induced filamentation (dif), site-specific
recombination at 92 93f
Delocalized excitons 193–94 194f
Denitrification 337–38
Dental biofilms 552f 553
Deoxyadenosine methylase (Dam methylase) 102 106n13
Deoxyribonucleotide, synthesis of 267 271f
Desulfivibrio, electron path to sulfate in 338–39 339f
Detoxifying agents, in periplasm 33
Developmental gene expression, in myxobacteria 595–96
Dextran synthesis 331–32
Diauxic growth 64 65f
dif. See Deletion induced filamentation
Diffusible signaling factor (DSF) 579–80
Dissimilatory nitrate reduction 337–38
www.ebook777.com
Index Terms Links
Dissimilatory sulfate reduction 338–39

electron path to sulfate in Desulfivibrio 338–39 339f
sulfate reducers 338
DivIVA 604–5
in Bacillus subtilis 100
DNA
damage to, types of 416 416f
discovery of 78–79
repair systems for 415–21
base-excision repair 418–19 419f
editing errors 101 102f
GO system 419–21 419f
methyl-directed mismatch errors 101–3 103f
by recombination 417–18 418f
UV-damage 416–17 416f
via nucleotide excision repair 417 417f
replication of
bidirectional 106n12
in Caulobacter crescentus 598–99 600f
cell division and 108n23
DNA polymerases 85–90
inhibitors of 100–101 101f
initiation of 106n13
repairing errors during 101–3
replication fork 80–82 82f 83f
semiconservative 77 77f 80f
81f
termination 90 90f
topological problem in 80–82 82f 83f
structure of 79
DnaA 106n13
DNA-binding proteins, nucleoids and 41
DNA gyrase 81 84f 105n3
DNA helicase, in sister chromosome separation 95–96

Index Terms Links
DnaK 309 314n30

DNA photolyase 416–17
DNA polymerases 85–90 431n71
lagging strand synthesis 87 87f
replication fork and 88 88f
leading strand synthesis 87 87f
Okazaki fragments 86–87 87f
attaching 88–90 89f
opposite polarity and 84f 85
in sister chromosome separation 95–96
sliding clamps 89–90
subunits of 107n14
Donnan potential 429n37
D period 75n32 75n33
Drug-export systems 445–46 446f
Dry weight 57
DsbB 473 481n101
DSF. See Diffusible signaling factor
Editing repair 101 102f

ED pathway. See Entner–Doudoroff pathway
Electrical energy 113
Electrochemical energy 112–18
generating ΔΨ and ΔpH 116–18
of protons 113–16
chemical component of 115–16
derivation of 116
electrical component of 115
proton motive force 113–14
units for 115
Electrochemical gradient, solute transport and 436–37 436f
Electrogenic flow, membrane potential and 116–17
www.ebook777.com
Index Terms Links
Electron carriers 146–52 148t

bacterial organization of 152–56 155f
cytochromes 147 148t 150–52
151f 152f
flavoproteins 147–48 149f
iron–sulfur proteins 147–50 149f
mitochondrial organization of 152 154f
quinones 146–48 149f
Electronic cell counting 56
Electron sinks, in fermentations 385–86
Electron source, for nitrogen fixation 342
Electron spin resonance spectroscopy 172n6
Electron transfer, in purple photosynthetic bacteria 179f 181–82 181f
Electron transport 146–74 147f
See also Bacterial electron transport chains;
Photosynthetic electron transport
aerobic and anaerobic respiration 147
coupling sites 154f 156–58
proton potential created at 157f 159–63
electron carriers 147–52 148t
in mitochondria 154f
oxidative phosphorylation in 112
patterns in individual bacterial species 163–70
Escherichia coli 163–65
Paracoccus denitrificans 165–68 166f
Rhodobacter sphaeroides 168
Wolinella succinogenes 168–70 169f
reversed 127
Electron volts (eV) 115
Embden–Meyerhof–Parnas pathway (EMP). See
Glycolysis
EMP pathway. See Glycolysis
Endocytosis 478n65

Index Terms Links
Endoplasmic membrane (ER), protein translocation

across 460
Endosporulation. See Sporulation
End-product inhibition. See Feedback inhibition
Enterococcus signaling 568t 572–73
Entner–Doudoroff (ED) pathway 222 223f 236–38
236t 237f
hexose degradation with 238
partly nonphosphorylated 238
physiological role for 236–38
reactions of 236 237f
Environmental stress 60–61
responses to 403–31
EnvZ/OmpR regulatory system 506–7
Enzyme activity, osmolarity effect on 409–10
Epimers 254n8
EPS. See Extracellular polysaccharide
ER. See Endoplasmic membrane
Erythrose phosphate 231
Escherichia coli
catabolite repression in 510–13
electron flow in 163–65
aerobic respiratory chains 163–64 164f
anaerobic respiratory chains 164–65 165f
coupling sites and 164
EPS capsule of 326–28 327f
Hemolysin secretion by 462–63 462f
Hsps of 414–15
ATP-dependent proteases 415
folding newly synthesized proteins 414–15
for protein export 415
refold denatured proteins 415
roles for 414
www.ebook777.com
Index Terms Links
Escherichia coli (Cont.)

K1 capsule
biosynthesis of 329–30
chemical nature and biological role of 329
osmotic homeostasis in 408–9 409f
Eukaryotes, chromosome partitioning in 93
Euryarchaeota 3
eV. See Electron volts
Evolutionary distance method 4
Excitons, delocalized 193–94 194f
Exergonic reactions, Δp generation with 124–28
Exonuclease 102
Exponential growth equation 66–67 66f
Extracellular polysaccharide (EPS) 16 326–31 327f
E. coli capsules 326–28 327f
export of 330–31
based on ABC transporters 330–31 331f
based on undecaprenol diphosphate
intermediates 330 331f
synthesis of 328
in E. coli group 2 K1 antigen 329–30
in gram-negative bacteria 328
in K. aerogenes 328–29
without undecaprenol intermediates 330
via undecaprenol diphosphate intermediates 320f 326f 328–29
Extracellular protein secretion 461–72
chaperone/usher pathway 470f 471
type III pathway 463–67
common properties of 464–66
flagellin secretion apparatus and 462f 464 465f
secretion apparatus 463–64 464f
in Yersinia 466–67
type II pathway 467–69
cholera toxin and protease secretion 469

Index Terms Links
Extracellular protein secretion (Cont.)

Out system 468–69
secretion apparatus for 467–68 468f
specificity of 469
two stages of 462f 467 468f
type I pathway 461–63
hemolysin secretion 462–63 462f
secretion apparatus 461–62
type IV pathway 471–72
Sec-dependent system 472 472f
of VirB system 471–72 472f
type V pathway: autotransporters 469–71 470f
immunoglobulin A protease 470–71
two stages of 469–70
Extrusion-capture mode 94 95f
Factor for inversion stimulation (Fis) 63 75n27 75n30

106n13
Facultative anaerobes, anaerobiosis response of 488–94
Arc system 490–92 491f
FblA regulon 491
FNR system 490
formate-hydrogen lyase pathway 489f 494
metabolic changes 488–90 489f
regulatory systems of gene expression 490–94
FAD. See Flavin adenine dinucleotide
Fatty acids 255–60
comparisons between Archaea, Bacteria, and
Eucarya in 6 7t
conformation of 279n10
in lipid A 322
www.ebook777.com
Index Terms Links
Fatty acids (Cont.)

β-oxidation of 256 256t 257f
258f
role of 255–56
synthesis of 256–60 259f
unsaturated 259–60
types of 255
Fatty acid synthase 42 53n130
FblA regulon, anaerobiosis response of 491
FDH. See Formate dehydrogenases
FdUMP. See 5-Fluorodeoxyuridine-5′-phosphate
Feedback inhibition, in metabolic pathway
regulation 199–201 200f
Fermentation 383–402
of acetate 391–92 391f
anaerobic food chain 386–87 386f
balancing of 387
available hydrogen method 387 388t
O/R method 387
of butanediol 394–97 396f
of butyrate 397–400 398f 399f
electron sinks 385–86
energy conservation in 384 385f
of lactate 392–94
mixed-acid 394–95 395f
NADH and 230
oxygen toxicity 383–84 384t
of pentoses 399–400 399f
of propionate
via acrylate pathway 388–89
via succinate-propionate pathway 389–91 389f 390f
Ferrous sulfate, growth on 351–52 352f
Fibrils. See Pili
Filament, of flagella 10

Index Terms Links
Fimbriae. See Pili

Fis. See Factor for inversion stimulation
Flagella 6–14 9f
archaeal 14
assembly of 10–11
OmpR and 509
bacterial compared with eukaryotic 47n16
basal body 7 9 9f
filament and capping proteins of 10
gene expression for 11
growth of 11 49n34
hook and HAP proteins of 10
motility with 6–7
motor 9–10 9f 48n26
48n28
number of 12–13
signal transfer to 544
site of insertion of 12–13
spirochaete 12 49n41 49n43
structural differences in 11–12
swarming and 13–14
switch of 10
in tactic responses and virulence 13–14
Flagellin secretion apparatus, T3SSs and 462f 464 465f
Flavin 148
Flavin adenine dinucleotide (FAD) 147–48 148t
Flavin mononucleotide (FMN) 147–48 148t
Flavoprotein oxidase 272–73 276f
Flavoproteins (Fp), as electron carriers 147–48 149f
5-Fluorodeoxyuridine-5′-phosphate (FdUMP) 101 101f
FMN. See Flavin mononucleotide
FNR system. See Fumarate nitrate reductase system
Folded proteins, translocation of 459–61
Folding of new proteins 308–10
www.ebook777.com
Index Terms Links
Food-borne illness, biofilms in 553

Formate 381n37
Formate dehydrogenases (FDH) 527n43
Formate-hydrogen lyase pathway, anaerobiosis
response of 489f 494
Fp. See Flavoproteins
Frameshift mutations 110n56
Fructose, Cra removal by 510–12
Fructose-6-phosphate 228 228f
Fruiting body formation, in Myxococcus xanthus 581–83
Frz protein 596–97 597f
FtsZ 43 604
Fumarate nitrate reductase (FNR) system
anaerobiosis response of 489f 490 492–94
493f
Fumarate reductase 390
Fumarate respiration 168–70 169f 173n38
254n2
Gas vesicles 39
Gene expression
by Cra system 510 511f
nutrient limitation and 61
oxygen and light and 504–5
ToxR and 515 517f
Gene fusions, lacZ 60 73n16 526n26
Gene regulation, Nar regulatory system and 495–96 496f
Gene repression, by Cra system 510 511f
Genetic diversification, in biofilms 562
GFP group. See Green fluorescent bacteria group
Gliding motility
in GFP group 8
of myxobacteria 588–91

Index Terms Links
Gliding motility (Cont.)

in Myxococcus xanthus 581–83
signaling required for 582–83
slime extrusion and 8
of type IV pili 8
Global regulator 292
Globules 40–41
Glucose
catabolite repression by 64–65
Cra removal by 510–12
sugar accumulation prevention by 514–15
Glucose-6-phosphate 229 229f
Glutamate
in bacteria 428n17
transamination reactions from 271 273f
Glutamate dehydrogenase 498
Glutamate synthase 498–99
Glutamine, synthesis of 269–71 272f
Glutamine synthase (GS) 336
Glutamine synthetase 498
PII-UMP and PII effect on 502 502f 502t
Glutaredoxins 425
Glycerol phosphate dehydrogenase 261
Glycine, synthesis of 270f 271–72 274f
Glycocalyx 16–18 17f
chemical composition of 16–17
PPP relationship with 235–36
role of 17–18 18f
Glycogen synthesis 332
Glycolipids, of gram-positive cell walls 22 27–28 28f
Glycolysis 222 223f 224–30
226f
as anabolic pathway 227–28 227f 228f
www.ebook777.com
Index Terms Links
Glycolysis (Cont.)
Calvin cycle and 362–63 362f
isomerization and aldol cleavage reactions in 229–30 229f
modified pathway 231 231f
phosphorylation of intermediates 230
regulation of 227f 228–29
stages of 224–27 226f
Glyoxylate cycle 248–49 249f
carbon balance for 249
regulation of 249 249f
GMP. See Guanosine monophosphate
GO. See 8-Oxoguanine
Gram, Christian 18
Gram-negative cell walls 28–31 29f
Gram-positive cell walls 19f 20–22 21f
22f 27–28
periplasm in 34
Gram stain 18–20 19f
peptidoglycan in 19–20 19f 51n80
Granules 40–41
Green fluorescent bacteria (GFP) group, gliding
motility in 8
Green gliding bacteria. See Green nonsulfur
phototrophs
Green nonsulfur phototrophs 176–77 176t 197n9
light-harvesting pigments of 187 189–90 190f
195f
Green sulfur phototrophs 176–77 176t 183–84
light-harvesting pigments of 187 189–90 190f
195f
photosynthetic electron transport of 183–84
GroE system 308–9 314n28 430n57

Index Terms Links
Group transfer potential 207–10 209f 209t

acyl derivatives 212–13 212f
ATP 209t 210–11
PEP 209t 211–12
Growth 55–76
See also Cell division
kinetics of 66–68
exponential growth equation 66–67 66f
growth rate and nutrient concentration 67–68 67f
measurement of 55–57
physiology of 57–65
catabolite repression by glucose 64–65
catabolite repression by other molecules 65
diauxic 64 65f
macromolecular composition and 63–64 63f
nutrient limitation 58–63 63f
population growth phases 57–58 57f
steady state and continuous 68–69
turgor pressure and 36 407–8
yields 65–66
GS. See Glutamine synthase
GSP pathway 467–68 480n87
Guanosine monophosphate (GMP), biosynthesis
from IMP 267 269f
H2. See Hydrogen

Halobacteria
osmotic homeostasis in 408
photoresponse of 544
swimming behavior of 544
Halophilic archaea 5 144n68
respiration of 144n72
www.ebook777.com
Index Terms Links
Halorhodopsin 5
as chloride pump 137–38
HAP proteins, of flagella 10
Heat-shock proteins (Hsps) 412–13
of E. coli 414–15
ATP-dependent proteases 415
folding newly synthesized proteins 414–15
for protein export 415
refold denatured proteins 415
roles for 414
Heat-shock response (HSR) 412–15
Cpx system 414
heat-shock proteins 412–13
σ32 regulon 413
E
σ regulon 413–14
s
σ regulon 414
Heliobacteria 176–78 176t
Helix-turn-helix motif, in transcription regulation 290 290f
Heme. See Cytochromes
Hemolysin secretion 462–63 462f
Heterocyst formation, in cyanobacteria 568t 573
Heterocysts, nitrogen fixation 343–44 343f
Heterofermentative lactate fermentation 392–93 393f
Hexose degradation, with ED pathway 238
High-energy molecules 207
High-osmolarity media, adaptation to 408–10
Histidine kinase (HK) protein 483–85 483f
amino acid sequences and 486–88 486f 487f
chemotaxis 534
transmitter domain of 524n3
Histidine permease system 449n27
Histones, comparisons between archaea, bacteria,
and eucarya in 6 7t
HK protein. See Histidine kinase protein

Index Terms Links
H-NS 63
Homofermentative lactate fermentation 392
Hook, of flagella 10
Horizontal gene transfer, in biofilms 561–62 561f
Hpr, phosphorylation of 514
Hsps. See Heat-shock proteins
HSR. See Heat-shock response
HvrA 529n86
Hydrogen (H2)
interspecies transfer of 386–87 400
methanogenesis from CO2 and 368–70 368f 369f
Hydrogenase, NADH and 230
Hydrogen-oxidizing bacteria 344t 345 345t
Hydrolytic enzymes, in periplasm 33
α-Hydroxyketone signaling 578
Hypo-osmotic stress 36 52n110
IHF. See Integration host factor

IMP. See Inosinic acid
Inducer exclusion 64–65
Inducer expulsion 514–15
Inductive resonance transfer 193 194f
In-dwelling medical devices, biofilms in 553–54
Inflammatory response 479n82
70S Initiation complex 297f 302
Inorganic metabolism 335–57
dissimilatory nitrate reduction 337–38
denitrification 337–38
dissimilatory sulfate reduction 338–39
lithotrophy 344–53
lithotrophs 344–53 344t 345f
345t
nitrate assimilation 335–36 336f
www.ebook777.com
Index Terms Links
Inorganic metabolism (Cont.)

nitrogen fixation 339–44
sulfate assimilation 335–37 337f
Inosinic acid (IMP), AMP and GMP synthesis
from 267 269f
Inside-out vesicles 143n49
Integral proteins 35
Integration host factor (IHF) 106n13
Ntr regulon and 500f 502
ompF and ompC and 506
regulatory role of 497
Intercellular signaling
of myxobacteria
A-signaling 592–95
C-signaling 592f 593–96
multicellular development and 591–94
switching poles for 596–97
in myxobacterial development 573–74 591 591f
Interspecies transfer of hydrogen 386–87 400
Intracytoplasmic membranes 38–39
Ionophores 119–20 120f 121f
uncouplers 119–20
Iron-oxidizing bacteria 344t 345t 351–53
energy from 356n34
ferrous sulfate 351–52 352f
pyrite 352–53 352f
Iron–sulfur proteins 172n5
as electron carriers 147–50 149f
in green sulfur bacteria 183
Isocitrate lyase 248
Isomerization reactions
in glycolysis 229–30 229f
in PPP 233 233f
Isoprenoids 198

Index Terms Links
Isoprenyl alcohol 263–64
Joules (J) 115
K+. See Potassium

K antigens 327–28
KdpABC operon 507–8
2-Keto-3-deoxy-6-P-gluconate (KDPG) 236 237f
α-Ketocarboxylic acids, decarboxylation of 396–97 397f
α-Ketoglutarate, decarboxylation of 248 248f
α-Ketoglutarate dehydrogenase 42
Klebsiella aerogenes, EPS synthesis in 328–29
Klebsiella pneumoniae, oxaloacetate decarboxylase
in 129–30
Korarchaeota 3
Lactate, fermentation of 392–94

Lactate efflux, Δp generation by 133–34 134f
Lactate oxidase 394
Lactic acid bacteria 392–94
acetyl-CoA and acetyl-P synthesis from pyruvate 394
bifidum pathway 392–94
heterofermentative lactate fermentation 392–93 393f
homofermentative lactate fermentation 392
lacZ gene fusions 60 73n16 526n26
Lagging strand
at replication fork 88 88f
synthesis of 87 87f
Lag phase 58
Leader peptide 453–54 454f
www.ebook777.com
Index Terms Links
Leading strand, synthesis of 87 87f

Lecithin 381n37
Leprosy 26–27
Levan synthesis 331–32
Light energy
photosynthetic genes and 504–5
proton pump driven by 136 136f
Light-harvesting pigments 187–91 188f 188t
in algae 190–91 192f
in cyanobacteria 187–88 190 191f
192f
energy transfer from 193–94
of green photosynthetic bacteria 187 189–90 190f
195f
of purple photosynthetic bacteria 187–89 189f
Lineweaver–Burke plot 203 203f
Linoleate 279n9
Lipid A
fatty acids in 322
synthesis of 319f 322–23 324f
Lipid monolayer 38 38f
Lipids. See also Fatty acids; Phospholipids
of archaeal cell membranes 36 37f
of bacterial cell membranes 34–35 35f
Lipoic acid 239–40
Lipopolysaccharide (LPS) 321–26
fatty acids in lipid A 322
function of 28–30 29f 51n89
structure of 28–30 29f 51n89
321–22 322f 323f
assembly of 325–26 326f
core 323 326f

Index Terms Links
Lipopolysaccharide (LPS) (Cont.)

lipid A 319f 322–23 324f
O-antigen region 323–24 325f
Lipoproteins 30 30f 31f
52n92
Lipoteichoic acids (LTA) 21–22 51n85
Lithotrophy 344–53
energetic considerations for 353
ATP synthesis 353
NADH reduction 353
lithotrophs 344–53 344t 345f
345t
ammonia-oxidizers 344t 345–46 345t
346f
carboxydobacteria 344t 345 345t
hydrogen-oxidizers 344t 345 345t
iron-oxidizers 344t 345t 351–53
nitrite-oxidizers 344t 345t 346–48
347f
sulfur-oxidizers 344t 345t 348–51
349f
Low-osmolarity media, adaptation to 410–11
LPS. See Lipopolysaccharide
LTA. See Lipoteichoic acids
Lux-type systems, non-AHL molecules in signaling in 566–81
Magnesium levels, PhoP/PhoQ regulatory system and 517–18 518f

Magnetosomes 40
Malate, pyruvate formation from 251
Malate lyase 248
Mating. See Conjugation
Maximum parsimony method 4
MCP. See Methyl-accepting chemotaxis protein
www.ebook777.com
Index Terms Links
MDR. See Multidrug resistance

Mechanosensitive (MS) channels 36
osmolarity and 410
Membrane bioenergetics 111–45
chemiosmotic theory 111–12 113f 114f
ΔΨ generation 129–37
electrochemical energy 112–18
halorhodopsin as chloride pump 137–38
ionophores 119–20 120f 121f
Δp and ATP synthesis in alkaliphiles 138–39
Δp contributions of ΔpH and ΔΨ 118–19
Δp generation 124–37
Δp measurement 120–22 121f 122f
work done with Δp 114f 122–24 123f
Membrane-bound proteins, translocation of 457–59
cytoplasmic membrane destination 457 458f
outer membrane destination 457–59 458f
Membrane fusion protein (MFP) 445
Membrane potential (ΔΨ) 113
generation of 116–17 129–37 132f
133f 142n14
ionophores and 119
measurement of 120–21 121f 122f
Δp contributions of 118–19
Menaquinones 148 149f
Metabolic activity, nutrient limitation and 59
Metabolic pathway regulation 199–206
by covalent modification 204 205t
nonregulatory enzymes 201–4
patterns of 199–201
feedback inhibition 199–201 200f
positive regulation 200f 201
regulatory enzymes 200f 201

Index Terms Links
Metabolic pathway regulation (Cont.)

regulatory enzymes 201–4
conformational changes in 204 204f
Methane, growth on 374–77 375f 381n42
ribulose-monophosphate cycle 377 378f
serine pathway 375–77 376f
Methanogens 4 381n23
acetyl-CoA incorporation by 371–72 371f
acetyl-CoA pathway in 364f 366–69 367f
368f 369f
Methanogenesis
from acetate 368f 370–71
from CO2 and H2 368–70 368f 369f
energy conservation during 368f 369–70 369f
Methanol 381n37
periplasmic oxidation of 167–68 168f
Methyl-accepting chemotaxis protein (MCP) 534 537 541
547n16 549n52
Methylamines 381n37
Methyl-directed mismatch repair 101–3 103f
Methylotrophs 374–77
methane growth of 374–77 375f
ribulose-monophosphate cycle 377 378f
serine pathway 375–77 376f
Methylreductase system 368–69 368f 369f
MFP. See Membrane fusion protein
MglA protein 596
Michaelis-Menten equation 201–3 202f 203f
Mismatch repair, methyl-directed 101–3 103f
Mitochondria
coupling sites of 146–47
electron carriers of 152 154f
electron transport scheme in 154f
Mitomycin C 100
www.ebook777.com
Index Terms Links
Mixed-acid fermentation 394–95 395f

MoFe protein 340–41 355n18 356n19
Mollicutes 54n141
cytoskeleton components in 44
Monooxygenase 274
Morse code 421
Motility. See also Swarming
flagellar 6–7
nonflagellar 8
types of 8
Motor, of flagella 9–10 9f
M protein 17 50n73 51n85
MreB 98–99
Mre proteins 43–44
MS channels. See Mechanosensitive channels
Muk proteins 96–98 98f 109n45
Multidrug resistance (MDR) transport systems 445
Multienzyme complexes 41–42
Murein lipoproteins 30 30f 31f
Mutagenesis, SOS response and 423
MutM 420 420f
Mut proteins 421
MutT 420–21
MutY 419f 420 420f
Mycolic acid 22 27–28 28f
Mycoplasma 44
Myxobacteria
chemotaxis of 597–98
developmental gene expression in 595–96
development of 587–611
Che3 and Che4 in 597
gliding motility of 588–91
intercellular signaling of 573–74 591 591f
A-signaling 592–95

Index Terms Links
Myxobacteria (Cont.)
C-signaling 592f 593–96
multicellular development and 591–94
switching poles for 596–97
life cycle of 587–88 589f
cell-to-cell signaling 588
cooperative feeding 588
two stages in 588 589f
Myxobacterial development, intercellular signaling
in 573–74
Myxococcus xanthus, fruiting body formation and
gliding in 581–83
NADH. See Nicotinamide adenine dinucleotide

NADH oxidoreductase 400
NADPH
green sulfur bacteria and 183–84 184f
production of 231–32
NADPH synthesis, with photosynthesis 186–87
+ +
Na /H antiporter, for pH homeostasis 404f 405–6
Nalidixic acid 100
Nanoarchaeota 3
NarQ 496–98 496f
anaerobiosis response of 490–91
discovery of 528n54
enzymes in 495
gene regulation and 495–96 496f
IHF regulatory role 497
NarQ 496–98 496f
NarX 496–98 496f 497f
nitrate reduction pathway 495
NarX 496–98 496f 497f
www.ebook777.com
Index Terms Links
Neutrophilic bacteria
pH homeostasis in 403 404t
Nick translation 88
Nicotinamide adenine dinucleotide (NADH)
fate of 230
in fumarate respiration 254n2
reduction of, with lithotrophy 353
structures of 230–31 230f
Nigericin 119
Nitrate assimilation 335–36 336f
Nitrate reductase 167 174n46 355n5
Nitrate reduction
dissimilatory 337–38
Nar regulatory system and 495
Nitrifiers 344t 345–46 345t
346f
Nitrite-oxidizing bacteria 344t 345t 346–48
347f
Nitrite reductase 167 174n46
Cra repression of gene for 512–13 512f
Nitrogenase 340–41
additional 341–42
in heterocysts 343–44 343f
oxygen protection for 342–43
reaction 341 341f
Nitrogen fixation 339–44
electron source for 342
heterocysts 343–44 343f
nitrogenase 340–41
additional 341–42
oxygen protection for 342–43
reaction 341 341f

Index Terms Links
Nitrogen fixation (Cont.)

organisms 340
pathway of 340–44
Nonflagellar motility 8
Nonregulatory enzymes, kinetics of 201–3 202f
Novobiocin 100
Ntr regulon 498–502 528n64
IHF regulatory role in 500f 502
PII-UMP and PII effect 502 502f 502t
regulation of 497f 499–501 500f
of other operons 500f 501
Nucleoid 41
Nucleophilic displacements 208
Nucleotide excision repair, for damaged DNA 417 417f
Nucleotides 264–67
AMP and GMP synthesis from IMP 267 269f
deoxyribonucleotide synthesis 267 271f
nomenclature and structures of 264 265f
synthesis of
purine 265–67 267f
pyrimidine 264–65 265f 266f
THF 267 270f
Nutrient concentration, growth rate and 67–68 67f
Nutrient limitation, adaptive responses to 58–63 63f
O-antigen region, synthesis of 323–24 325f

OD. See Optical density
Okazaki fragments 86–90 87f 88f
89f
Oligopeptide pheromones 566–68 568t
intracellular targets of 568
Rap phosphatases and 570
www.ebook777.com
Index Terms Links
Oligopeptide pheromones (Cont.)

receptors for 566–68
two-component signaling cascades and 568
Oligosaccharides, in periplasm 33
ompC, expression regulation of 506–7
ompF, expression regulation of 506–7
OmpR. See also EnvZ/OmpR regulatory system
flagella biosynthesis and 509
Open complex 82–85 86f
Operons, in transcription regulation 289 292–93
Opines 521–22
Optical density (OD) 55–56 56f
oriC 96 107n13
O/R method, for balancing fermentations 387
Osmolarity. See also High-osmolarity media;
Lowosmolarity media
response to 507–8
transcription and enzyme activity 409–10
Osmolytes 36
Osmosensors 412
Osmosensory transporters 36
Osmotic homeostasis
in E. coli 408–9 409f
in halobacteria 408
in periplasm 411
Osmotic potential 407
Osmotic pressure 406–7 429n37
See also Turgor
pressure
response to 506–7
Out system 468–69
Overwinding 80–81 84f
Oxalate:formate exchange, Δp generation by 130–31 132f

Index Terms Links
Oxaloacetate
PEP formation from 249–50 250f
replenishment of 245–46 245f 251f
Oxaloacetate decarboxylase, in Klebsiella pneumoniae 129–30
Oxalobacter formigenes 130–31 132f
β-Oxidation, of fatty acids 256 256t 257f
258f
Oxidation-decarboxylation reactions, in PPP 232–33
Oxidation-reduction reactions 126–27
Δp generation with 126–27
respiration coupled to sodium ion efflux 127–28
reversed electron transport 127
Oxidative phosphorylation 112
Oxidative stress 423–25
proteins made in response to 424
toxic forms of oxygen 423–24
transcriptional regulation of genes induced by 424–25
Oxidoreductase 355n5
8-Oxoguanine (GO) lesions, protection from 419–21 419f
Oxygen
nitrogenase protection from 342–43
photosynthetic genes and 504–5
production of, with photosynthesis 187
toxic forms of 423–24
toxicity of, fermentation and 383–84 384t
Oxygenic phototrophs 176 176t
OxyR 425
Δp. See Proton motive force

Paracoccus denitrificans, electron transport in 165–68
aerobic pathway 166–67 166f
www.ebook777.com
Index Terms Links
Paracoccus denitrificans, electron transport in (Cont.)

anaerobic pathway 166f 167
periplasmic oxidation of methanol 167–68 168f
Par proteins 96–97
in Bacillus subtilis 99
in plasmids 109n41
visualization of 110n54
Penicillin-binding proteins (PBPs) 53n137 321
Pentose phosphate pathway (PPP) 222 223f 231–36
Calvin cycle and 362–63 362f
carbon balance for 234–35
glycolysis relationship with 235–36
reactions of 232–34 232f 233f
sugar catabolism with 235
summarizing 234 234f
Pentoses, fermentation of 399–400 399f
PEP. See Phosphoenolpyruvate
Peptide bond, formation of 303–4 303f
Peptidoglycan 316–21
chemical composition of 316–17 317f
comparisons between archaea, bacteria, and
eucarya in 6 7t
cross-linking in 317 318f
in Gram stain 19–20 19f 51n80
pseudopeptidoglycan compared with 31–32
structure of 316 316f
penicillin-binding proteins 321
peptide cross-link 321 321f
reactions in membrane 319–21 320f
UDP derivatives 317–19 319f
Peripheral proteins 35

Index Terms Links
Periplasm 32–34 32f

components of 33–34
in gram-positive bacteria 34
osmotic homeostasis in 411
volume of 428n32
Periplasmic oxidation, of methanol 167–68 168f
Periplasmic proteins
disulfide bonds corrections in 473–74
folding of 473–74
intramolecular disulfide bonds for 473
thiol-redox enzymes and 473 473f
Permeability, of bacterial cell membranes 35
Permease induction 515
Pertussis, bvg genes and 519–20
PGALD. See Phosphoglyceraldehyde
ΔpH. See pH gradient
Phenotypes 3–6
pH gradient (ΔpH)
Δp contributions of 118–19
generation of 117–18 117f
ionophores and 119
maintaining 403–6 404t
measurement of 121–22
pH homeostasis
demonstrating 403 404f
mechanisms of 403–6 404f
in acidophiles 403 404f 406
+
K for 404f 405 405f
Na+/H+ antiporter in alkaliphiles for 404f 405–6
proton pumping 403–5 404f
PhoP/PhoQ regulatory system 517–19
Pho regulon 503–4
signal transduction pathway 503–4 504f
www.ebook777.com
Index Terms Links
Phosphatase 484–85
See also RapA phosphatase;
RapB phosphatase
Phosphoenolpyruvate (PEP) 220 220f
carboxylation of 390f 391
formation of 249–50
from oxaloacetate 249–50 250f
from pyruvate 250
group transfer potential of 209t 211–12 212f
Phosphoenolpyruvate (PEP) carboxykinase 402n5
Phosphoenolpyruvate (PEP) carboxylase 245–46 245f 251f
402n5
Phosphoenolpyruvate (PEP) synthetase 250 250f
Phosphoglyceraldehyde (PGALD) 229 229f
Calvin cycle for 358
Phosphoglycerides 260 261f
Phospholipids
cell membrane structure and 260 262f
inhibition of 262–63
Phosphonates 529n74
Phosphorelay system, Bacillus subtilis sporulation 605–10
advantages to 608
Phosphorylation potential 173n20
Phosphoryl group transfer reaction 207–10 209f
to acceptors other than water 210 210f
of ATP 209t 210–11
of PEP 209t 211–12 212f
Phosphotransferase system (PTS) 441–44 442f
catabolite regulation by 441f 442–44
in chemotaxis 541
Photoactive yellow protein (PYP) 543–44

Index Terms Links
Photocycle, proton pump and 136–37 138f

Photoreaction repair, of UV-damaged DNA 416–17 416f
Photoresponses 543–44
dark avoidance 544–45
halobacteria 544
of photosynthetic bacteria 544–46
photosynthetic electron transport role in 545
phototaxis of colonies 545–46
Photosynthesis 175–98
cyanobacteria and chloroplasts 184–86 185f 186f
efficiency of 186–87
energy transfer in 193–94
green sulfur photosynthetic bacteria 183–84
photosynthetic membrane structures 194 195f
photosynthetic pigments 187–92
phototrophic prokaryotes 175–78 176t 196n4
purple photosynthetic bacteria 178–83
Photosynthetic electron transport 545
in cyanobacteria and chloroplasts 185–86 186f
of green sulfur phototrophs 183–84
of purple photosynthetic bacteria 178–80 179f 180f
Photosynthetic gene cluster 504
Photosynthetic membranes, structure of 194 195f
Photosynthetic pigments 187–92
light-harvesting 187–91 188f 188t
structures of 191–92
synthesis of, oxygen and light response of 504–5
Phototaxis of colonies 545–46
Phototrophic prokaryotes 175–78 176t
anoxygenic 176–78 176t
oxygenic 176 176t
Phycobilins 187–88 188t
Phycobiliproteins 190 192f
Phycobilisomes 187–88 190 191f
www.ebook777.com
Index Terms Links
Phylogenetic tree
of all life-forms 3f
construction of 4–5
Phylogeny 2t 3–6 3f
Archaea 3–6
eucarya 6 7t
phylogenetic tree construction 4–5
PII, glutamine synthetase and 502 502f 502t
PII-UMP, glutamine synthetase and 502 502f 502t
Pili 14–16 14f
medical significance of 15
role in signaling requiring contact 583
sex 15–16 16f
type IV 8 50n63
Pipes, biofilm corrosion and clogging of 552–53
Plasmids 16 50n65
Plastoquinones 148 149f
PLR. See Polar localization region
PmrA/PmrB regulatory system 519
Polar localization region (PLR) 96
Polycistronic messages, in protein synthesis 306–7 307f
Polysaccharides, neutral 20–21
Polysomes, in translation 306 306f
Population growth, phases of 57–58 57f
Porins 30–31
synthesis of, regulation of 506–7
Porphyrins 150
+
Potassium (K )
ATP-driven influx of 440–41
in bacteria 428n17
for pH homeostasis 404f 405 405f
response to 507–8

Index Terms Links
ppGpp 61 62–63 73n22

74n25
PPP. See Pentose phosphate pathway
Prepriming complex 82–85 86f
Prokaryotes
chromosome partitioning in 93–94 95f
major subdivisions of 2t
Proofreading 101
Propionate, fermentation of
via acrylate pathway 388–89
via succinate-propionate pathway 389–91 389f 390f
Propionigenium modestum 129
Prosthetic group 147
Protease secretion 469
Proteins
of archaeal cell membranes 38
of bacterial cell membranes 35
Hsps
for export of 415
for folding newly synthesized 414–15
to refold denatured 415
oxidative stress and 424
synthesis of. See Translation
transport and secretion of 452–81
across ER 460
extracellular protein secretion 461–72
folded 459–61
folding of periplasmic proteins 473–74
Sec system 453–57
SRP 459
translocation of membrane-bound 457–59
Protein measurements 57
www.ebook777.com
Index Terms Links
Proteoliposomes 144n54
preparation of 432–33 433f
solute transport study with 433
Proton current 111–12 113f 114f
Proton motive force (Δp) 113–14
ATP synthase and 123–24 450n53
ATP synthesis and 138–39
electron flow and establishment of 169–70
generation of 124–37
ATP hydrolysis 128
bacteriorhodopsin light absorption 135–37
decarboxylation of acids 131 133f
end-product efflux 131–35
oxalate:formate exchange 130–31 132f
oxidation-reduction reactions 126–27
respiration coupled to sodium ion efflux 127–28
reversed electron transport 127
measurement of 120–22 121f 122f
ΔpH and ΔΨ contributions to 118–19
purple photosynthetic bacteria reaction center
and 182
SecA effects compared with 146n26
solute transport and 444–45 444t
work done with 122–24
solute uptake 114f 123
Proton potential, coupling site creation of 157f 159–63
proton pumps 160f 162–63
Q cycle 159–62 161f
Q loops 159 160f
Proton pump
bacteriorhodopsin as 136 137f
coupling site creation of proton potential at 160f 162–63
light energy to drive 136 136f

Index Terms Links
Proton pump (Cont.)

for pH homeostasis 403–5 404f
photocycle and 136–37 138f
Protons
electrochemical energy of 113–16
chemical component of 115–16
derivation of 116
electrical component of 115
proton motive force 113–14
units for 115
membrane potential and 117
Protoplasts 34 52n104
Pseudopeptidoglycan 6
peptidoglycan compared with 31–32
ΔΨ See Membrane potential
Pst system 529n78
PTS. See Phosphotransferase system
PulD proteins 468 480n88
Pullulan 480n86
Purine nucleotides, synthesis of 265–67 267f
enzymatic reactions 265–67 268f
Purple photosynthetic bacteria 176–83 176t
electron sources for growth 179f 182–83 183f
light-harvesting pigments of 187–89 189f
oxygen and light and gene expression in 504–5
photosynthetic electron transport of 178–80 179f 180f
reaction center of 179f 180–82 181f
Putrescine control of swarming motility 580–81
PYP. See Photoactive yellow protein
Pyrimidine nucleotides, synthesis of 264–65 265f 266f
Pyrite, growth on 352–53 352f
Pyrococcus furiosus 231 231f
www.ebook777.com
Index Terms Links
Pyruvate
acetyl-CoA formation from 217–18 217f 218f
238–41 240f 394
acetyl-P synthesis from 394
carboxylation of 390f 391
formation of, from malate 251
PEP formation from 250
Pyruvate carboxylase 245–46 245f 251f
402n5
Pyruvate dehydrogenase 41–42 217–18 218f
Pyruvate dehydrogenase reaction 238–41 240f
physiological control of 240–41 240f
Pyruvate-ferredoxin oxidoreductase 218 218f
Pyruvate-formate lyase 218 219f
Pyruvate oxidase 394
Pyruvate-phosphate dikinase reactions 250 250f
Q cycle 159–62 161f

Q loops 159 160f
Quinones
in bacterial electron transport chains 152 154
as electron carriers 146–48 149f
Quorum-sensing, in biofilm formation 560
RacA 96 97f 604

in Bacillus subtilis 99–100
RapA phosphatase, Bacillus subtilis and 608–9 609f
RapB phosphatase, Bacillus subtilis and 609–10
Rap phosphatases, oligopeptide pheromones and 570
Reaction center
energy transfer to 193–94

Index Terms Links
Reaction center (Cont.)

of purple photosynthetic bacteria 180–82
electron transfer in 179f 181–82 181f
Δp contribution of 182
structure and composition of 180–81
Recombinases 92
Recombination, for DNA repair 417–18 418f
Reductases 152
RegB/RegA system 504–5
Regulatory enzymes
conformational changes in 204 204f
kinetics of 201 203–4 203f
in metabolic pathway regulation 200f 201
subunits of 204
Release factors, in protein synthesis 304 305f
Repair systems, for DNA replication 101–3
editing repair 101 102f
methyl-directed mismatch repair 101–3 103f
Replication fork 80–82 82f 83f
lagging strand synthesis at 88 88f
open complex at 83–84 86f
opposite polarity at 84f 85
prepriming complex at 84
Replisome 82 94 95f
in chromosome separation 94–95 95f 108n35
Repressors, in transcription 291f 292–93
Respiration. See also Aerobic respiration;
Anaerobic respiration
fumarate 168–70 169f
of halobacteria 144n72
historical perspective on 153–54
NADH and 230
uncouplers and 119–20
www.ebook777.com
Index Terms Links
Response regulator (RR) protein 483–85 483f

amino acid sequences and 486–88 486f 487f
Reversed electron transport, Δp generation with 127
Rhizobium meliloti 515 543
Rhodobacter capsulatus 504–5
Rhodobacter sphaeroides 542–43
electron transport in 168
Ribosomal RNA (rRNA) synthesis 61–63 63f 293
295f
Ribosomes 41
eucarya in 6 7t
in protein synthesis 298
key sites within 297f 298
Ribozymes 312n11
Ribulose-1,5-bisphosphate (RuBP)
carboxylation of 359 359f
regeneration of 360f 361
Ribulose-monophosphate (RuMP) cycle 377 378f
Right-side-out vesicles 143n49
RNA holoenzyme structure 282–83 283f
in transcription 283–87 284f 285f
RNA polymerases (RNAP)
eucarya in 6 7t
in transcription 282–83 283f
RNA synthesis. See Transcription
RpoH 413
RpoS 414 430n50
nutrient limitation and 60–61
rRNA. See Ribosomal RNA
16S rRNA sequences 4–5
RR protein. See Response regulator protein
RuBP. See Ribulose-1,5-bisphosphate

Index Terms Links
Ruminococcus albus 400 401f

RuMP cycle. See Ribulose-monophosphate
Salmonella, PhoP/PhoQ regulatory system and 517–18 518f

Scotophobia 544–45
SecA
nucleotide binding sites of 476n24
Δp effects compared with 146n26
recycling of 456–57
study of 475n21
translocase and 454–55
SecB, chaperone proteins and 454
Secretins 468
Sec system 453–57
components of 453–55
chaperone proteins 454
leader peptide 453–54 454f
translocase 454–55
cotranslational translocation 457
model for protein export 455–57 456f
SecA recycling 456–57
translocation independent of 457
SecYEG 455 475n12
Semiconservative replication 77 77f 80f
81f
Sensor kinase. See Histidine kinase protein
Septum formation 69–70 70f
period before 69
proteins required for 70 70f
Serine, synthesis of 270f 271–72 274f
Serine pathway 375–77 376f
SetB 98–99
Sex pili 15–16 16f
www.ebook777.com
Index Terms Links
Shine-Dalgarno sequence 302 302f

Shock-sensitive transport system 438
Siderophores 33 52n97
Sigma 54 polymerases 528n69
Sigma factors, in sporulation of Bacillus subtilis 606–7
32
σ regulon. See RpoH
σE regulon 413–14
σs regulon. See RpoS
Signaling 566–86
early discoveries of 567
of myxobacteria 588
required for gliding motility 582–83
requiring contact 581–83
C-signaling 581–82
fibrils role in 583
fruiting body formation and gliding 581–83
signaling molecules for 566 567t
systems for 566–81
acylated homoserine lactones 574–76
AHL signaling without LuxR and LuxI-type
proteins 577 577f
bacteriocin production 572
γ-butyrolactones 578 580f
competence and endosporulation in Bacillus
subtilis 568–70 568t 569f
571f
competence in Streptococcus pneumoniae 570–71
diffusible signaling factor 579–80
enterococcus 568t 572–73
heterocyst formation in cyanobacteria 568t 573
α-hydroxyketone 578
intercellular signaling in myxobacterial
development 573–74

Index Terms Links
Signaling (Cont.)
non-AHL molecules in Lux-type systems 566–81
oligopeptide pheromones 566–68 568t
putrescine control of swarming motility 580–81
via AI-2 type factors 577–78 577f
virulence genes in Staphylococcus aureus 568t 571–72
Signal recognition particle (SRP) 459
Sister chromosomes separation 94–96
in Bacillus subtilis 99–100
factors for 99–100
during sporulation 97f 99
during vegetative growth 97f 99
directionality in 96
proteins involved in 95–96
replisomes in 94–95 95f
S layers 16
Sliding clamps 89–90 107n16
Slime capsules 16
SMC. See Structural maintenance of chromosomes
Social motility, in myxobacteria 590 591f
Sodium current 111–12 114f
Sodium ion efflux, Δp generation with 127–28
Sodium potential, sodium transport decarboxylase
creation of 129–30
Sodium transport decarboxylase, sodium potential
creation by 129–30
Solfataras 47n9
Solute-binding proteins, in periplasm 33
Solute efflux
Δp generation by 131–35
energetics of 132–33
lactate 133–34 134f
physiological significance of 134–35
www.ebook777.com
Index Terms Links
Solute efflux (Cont.)

succinate 134 135f
symport of protons and sodium ions with 133 134f 135f
Solute transport 432–51
in bacteria 446 447f
drug-export systems 445–46 446f
electrochemical gradient and 436–37 436f
energy-dependent transport 434–44 436f
group translocation 434–35
phosphotransferase system 441–44 442f
primary transport systems 434–35 437–41
secondary transport systems 434–37 436f
solute/proton or solute/sodium symport 437 438f
energy source for 434–45 444t
kinetics of 433–34
in absence of transporter 434 434f
transporter-mediated uptake 433–34 434f
proteoliposomes to study 432–33 433f
Solute uptake, Δp for 114f 123
SOS response 421–23
genes stimulated during 422–23 422t
regulation of 421–23 422f
translesion synthesis and mutagenesis 423
SoxR 424 424f 431n78
SoxS 424 424f
Spirochaete flagella 12 49n41 49n43
Spiroplasma 44
Spo0A, in B. subtilis 603–5
Spo0J 604
SpoIIIE gene 604
Sporulation, in Bacillus subtilis 568–70 569f 571f
601–10
factors for 603–5
genes for 603

Index Terms Links
Sporulation, in Bacillus subtilis (Cont.)

phosphorelay system for initiation of 605–10
RapA phosphatase 608–9 609f
RapB phosphatase 609–10
sigma factors in 606–7
stages in 602–5 602f
SpoT 62–63
SRP. See Signal recognition particle
Staphylococcus aureus, virulence genes in 521 568t 571–72
Start codon, in protein synthesis 302
Stationary phase 58
Steady state growth 68–69
Streptococcus pneumoniae, competence in 570–71
Stringent response 61–62
inhibition of phospholipid synthesis during 262–63
Structural maintenance of chromosomes (SMC) 96–98 98f
Substrate-level phosphorylation, ATP synthesis by 209f 209t 215–20
216f 216t
Succinate efflux, Δp generation by 134 135f
Succinate-propionate pathway, propionate
fermentation via 389–91 389f 390f
Succinic dehydrogenase 172n12 525n24
Succinyl-coenzyme A 219–20 219f
Sugar rearrangement reactions, in PPP 233–34 233f
Sulfate assimilation 335–37 337f
Sulfate reduction, dissimilatory 338–39
electron path to sulfate in Desulfivibrio 338–39 339f
sulfate reducers 338
Sulfur-oxidizing prokaryotes 344t 345t 348–51
349f
Sulfur oxygenase 350–51
Supercoiling 80 82f 105n2
positive v. negative 80–81
www.ebook777.com
Index Terms Links
Superoxide dismutase 383–84 384t

Surface properties, nutrient limitation and 59
Swarming
flagella and 13–14
putrescine control 580–81
Swimming behavior, of halobacteria 544
Switch, of flagella 10
T3SSs. See Type III secretion systems

T4SSs. See Type IV secretion systems
TA reactions. See Transaldolase
Tat system 459–61
T-DNA, of Agrobacterium tumefaciens 521–22
Teichoic acids 20 22f
Teichuronic acids 20
Temperature, response to 506–7
Terminal oxidases, in bacterial electron transport
chains 154–55
Ter sites 90 90f
of Bacillus subtilis 108n20
Tetrahydrofolic acid (THF) 267 270f
Tetrapyrroles 150
TF. See Trigger factor
Thermoacidophiles 5
Thermophilic archaea 5
THF. See Tetrahydrofolic acid
Thiamine pyrophosphate, α-ketocarboxylic acids
decarboxylation by 396–97 397f
Thiol-redox enzymes 473 473f
Thioredoxins 425
Thiosulfate oxidation 348–50 349f 350f
Thylakoids 187–88
TK reactions. See Transketolase reactions

Index Terms Links
TnphoA 532n131
TonB protein, in periplasm 32f 33–34
Topoisomerases 105n3
Topoisomerase II. See DNA gyrase
Topoisomerase IV 91–92 92f
Total cell counts 56
ToxR
cholera and 515–17
gene expression and 515 517f
Transaldolase (TA) reactions, in PPP 233–34 233f
Transcarboxylase reaction 390
Transcription 281–96 282f
chain elongation, steps in 283–87 285f
elongation 284–85
inhibitors of 296
initiation of 283 284f
frequency of 287
steps in 283–87 285f
osmolarity effect on 409–10
regulation of 289–93
of genes induced by oxidative stress 424–25
helix-turn-helix motif 290 290f
operons 289 292–93
transcription factors 289–93 291f
of ribosomal RNA 293 295f
RNA holoenzyme structure 282–83 283f
closed complex 284 285f
open complex 284 285f
sigma subunit 287–89
anti-sigma factors 288–89
termination 285–87
factor-independent 286 286f
Rho-dependent 286–87
www.ebook777.com
Index Terms Links
Transcription (Cont.)
topoisomerases in 287
of transfer RNA 293–94 295f 296f
Transcriptional regulators 510
Transcription factors
as activators or repressors 293
operation of 291–93 291f
negative regulators 291f 292–93
positive regulators 291–92
in transcription regulation 289–90
Transfer RNA (tRNA) 295f 296f 299
aminoacyl tRNAs and 313n18
codons and 313n15
processing of 293–94 295f 296f
in protein synthesis 298–301
amino acid attachment to 299 300f
aminoacyl 305–6
fMet-tRNA 300–301 301f
proofreading 299
A site binding to 302–3 303f
Transition elements 402n3
Transketolase (TK) reactions, in PPP 233–34 233f
Translation 281–82 282f 296–310
chaperone proteins 308–10
DnaK 309
GroE system 308–9
measuring flux of proteins through 308
role of 308
trigger factor 309–10
coupled 307–8 308f
cytosolic eukaryotic protein synthesis
comparison with 297–98
elongation 297 302–4
EF-Tu–GDP recycling 303

Index Terms Links
Translation (Cont.)
formation of peptide bond 303–4 303f
leader sequence 302
rate of 297
start codon 302
translocation 303f 304
tRNA binding to A site 302–3 303f
eukaryotic and bacterial comparison of 313n13
formyl group and methionine removal 297
inhibitors of 310 310t
initiation 297 301–2
70S initiation complex 297f 302
preinitiation complex 301–2 301f
overview of 296–98 297f
polarity in 307
polycistronic messages in 306–7 307f
polysomes in 306 306f
ribosomes in 298
key sites within 297f 298
termination 297 304–5
GTP requirements 304–5
release factors 304 305f
tRNA in 298–301
amino acid attachment to 299 300f
aminoacyl 305–6
fMet-tRNA 300–301 301f
proofreading 299
Translational coupling 74n24
Translesion synthesis 421
SOS response and 423
Translocase 454–55
Translocon 477n64
TraR 533n151
Trigger factor (TF) 309–10
www.ebook777.com
Index Terms Links
Tuberculosis 23–26
Tumbling 535 535f 536f
Turbidity 55–56 56f
Turgor pressure, growth and 36 407–8
Tus 90 90f
Twitching motility, of type IV pili 8
Two-component signaling systems 483–88 483f
acetyl phosphate 508–9
components of 485
EnvZ/OmpR 506–7
KdpABC operon 507–8
Ntr regulon 498–502
oligopeptide pheromones and 568
peptide pheromone and 521
PhoP/PhoQ 517–19
Pho regulon 503–4
PmrA/PmrB 519
RegB/RegA 504–5
signal transduction in 483f 485–86
VirA-VirG 521–22
Type III secretion systems (T3SSs) 463–67 477n53
common properties of 464–66
flagellin secretion apparatus and 462f 464 465f
secretion apparatus 463–64 464f
in Yersinia 466–67
Type IV pili 8 50n63
motility of 8
Type IV secretion systems (T4SSs) 471–72
Sec-dependent system 472 472f
of VirB system 471–72 472f
Uncouplers, respiration and 119–20

Index Terms Links
Undecaprenol diphosphate intermediates

in EPS synthesis 320f 326f 328–29
in EPS translocation 330 331f
Undecaprenyl phosphate 319–20 320f
Uniport 435
UV-damaged DNA, photoreaction repair of 416–17 416f
V. See Volts
Valinomycin 119 121f
Viable cell counts 57
VirA-VirG system 521–22
VirB system, T4SSs of 471–72 472f
Vir regulon, bvg genes and 519–20
Virulence factors
bvg genes and pertussis 519–20
capsules as 327
cholera and ToxR 515–17 517f
PhoP/PhoQ and Salmonella 517–18 518f
Virulence genes
in Agrobacterium tumefaciens 521–22
bacillary dysentery and 520–21
in Staphylococcus aureus 521 568t 571–72
Vitamins 239
Volts (V) 115
Wastewater treatment, biofilms in 553

Wolinella succinogenes, electron transport in 168–70 169f
electron flow and Δp establishment 169–70
topology of components in 168–69
www.ebook777.com
Index Terms Links
Yersinia 479n73
T3SSs in 466–67
Yop proteins 467 480n84

The Phy and Bio of Pro

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

The Phy and Bio of Pro

Încărcat de

Drepturi de autor:

Formate disponibile

Free ebooks ==> www.ebook777.

New York Oxford

Oxford New York

Copyright © 2012, 2007, 2000, 1995 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

Library of Congress Cataloging-in-Publication Data

White, David, 1936-

Printed in the United States of America

Introduction physiology of speciﬁc groups of prokaryotes is

that established the central tenets of micro- Acknowledgments

Boxed Material xiii

Chapter 1. Structure and Function 1

Chapter 2. Growth and Cell Division 55

Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77

Chapter 4. Membrane Bioenergetics: The Proton Potential 111

Chapter 5. Electron Transport 146

Chapter 6. Photosynthesis 175

Chapter 7. The Regulation of Metabolic Pathways 199

Chapter 8. Bioenergetics in the Cytosol 207

Chapter 9. Central Metabolic Pathways 222

Chapter 11. RNA and Protein Synthesis 281

Chapter 12. Cell Wall and Capsule Biosynthesis 316

Chapter 13. Inorganic Metabolism 335

Chapter 14. C1 Metabolism 358

Chapter 15. Fermentations 383

Chapter 16. Responses to Environmental Stress 403

Chapter 17. Solute Transport 432

Chapter 18. Protein Transport and Secretion 452

Chapter 19. Responses to Environmental Cues 482

Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534

Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551

Chapter 22. Cell–Cell Communication Mechanisms 566

Chapter 23. Bacterial Development 587

Boxed Material xiii

Chapter 1. Structure and Function 1

Chapter 2. Growth and Cell Division 55

Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77

Chapter 4. Membrane Bioenergetics: The Proton Potential 111

4.9 Halorhodopsin, a Light-Driven Chloride Pump 137

Chapter 5. Electron Transport 146

Chapter 6. Photosynthesis 175

Chapter 7. The Regulation of Metabolic Pathways 199

Chapter 8. Bioenergetics in the Cytosol 207

Chapter 9. Central Metabolic Pathways 222

9.6 The Entner–Doudoroff Pathway 236

Chapter 11. RNA and Protein Synthesis 281

Chapter 12. Cell Wall and Capsule Biosynthesis 316

Chapter 13. Inorganic Metabolism 335

Chapter 14. C1 Metabolism 358

Study Questions 380

Chapter 15. Fermentations 383

Chapter 16. Responses to Environmental Stress 403

Chapter 17. Solute Transport 432

Chapter 18. Protein Transport and Secretion 452

Chapter 19. Responses to Environmental Cues 482

19.2 Responses by Facultative Anaerobes to Anaerobiosis 488

Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534

Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551

Chapter 22. Cell–Cell Communication Mechanisms 566

22.4 Summary 583

Chapter 23. Bacterial Development 587