Immunopotentiators in Modern Vaccines

Immunopotentiators in Modern Vaccines

Second Edition

Editors

Virgil E.J.C. Schijns

Wageningen University, Wageningen, The Netherlands

Derek T. O'Hagan

GSK Vaccines, Rockville, MD, United States

Table of Contents

Cover image

Title page

Copyright

List of Contributors

Preface

Chapter 1. Vaccine Adjuvants' Mode of Action: Unraveling ‘‘the Immunologist's Dirty Little Secret"

Introduction

Adjuvants Provide Start Signals for Immune Reactivity and Guide the Response to an Acceptable Magnitude

Regulation of Immune Responses by Antigen Delivery (Signal 1)

Facilitation of Signal 1

Regulation of Signal 2

Facilitation of Signal 2

Release of Immune Brakes

Signal 3: Regulating the Quality of Induced Immune Pathways and Immunity

Facilitation of Signal 3

Outlook

Chapter 2. The Role of Inflammasomes in Adjuvant-Driven Humoral and Cellular Immune Responses

Introduction

Chapter 3. Dendritic Cells as Targets of Vaccines and Adjuvants

Introduction

General Characteristics of Dendritic Cells

Dendritic Cells and T-Cell Activation

The Role of Dendritic Cells in Prophylactic Vaccination

The Role of Dendritic Cells in Therapeutic Vaccination

Conclusions

Chapter 4. Host-Derived Cytokines and Chemokines as Vaccine Adjuvants

Introduction

Interleukin-1

Interleukin-2

Interleukin-6

Interleukin-12

Granulocyte-Macrophage Colony-Stimulating Factor

Chemokines

Interferon

Concluding Remarks

Chapter 5. Discovery of Immune Potentiators as Vaccine Adjuvants

Introduction

Mode of Action of Empirically Derived Adjuvants: Aluminum Salts and Oil-In-Water Emulsions

Pattern Recognition Receptor Agonists as Vaccine Immune Potentiators

Toll-Like Receptors

NOD-Like Receptors

Combination of Immune Potentiators and Delivery Systems

Small Molecule Immune Potentiators as New Adjuvants

Design of Adjuvants Targeting Toll-Like Receptors

Concluding Remarks

Chapter 6. Current Status of Toll-Like Receptor 4 Ligand Vaccine Adjuvants

Introduction

History

Development Status of Clinical Toll-Like Receptor 4 Ligands

Development Status of Selected Preclinical Toll-Like Receptor 4 Ligands

Physicochemical Characterization

Future Outlook

Chapter 7. Flagellins as Adjuvants of Vaccines

List of Abbreviations

Flagellins Are the Main Component of Bacterial Flagella

Flagellin Is Recognized by Innate Receptors

TLR5 Signaling Triggers the Transient Production of Immune Mediators by Myeloid and Epithelial Cells

Direct Activation of Dendritic Cells by Flagellin

Flagellin Also Activates Dendritic Cells by Indirect Signaling Through Structural Cells

Flagellin Is a Potent Adjuvant for Vaccination

Flagellin Potentiates Innate Antiinfectious and Tissue Repair Immune Effectors

Concluding Remarks

Chapter 8. Toll-Like Receptor 7 and 8 Agonists for Vaccine Adjuvant Use

Introduction

Cellular Expression of Toll-Like Receptors 7 and 8 and Cross-Species Expression

Toll-Like Receptor 7/8 Ligands

Adjuvant Action of Toll-Like Receptor 7/8 Agonists

Topical Application of Toll-Like Receptor 7/8 Agonists for Use as Vaccine Adjuvants

Approaches to Optimizing Toll-Like Receptor 7/8 Adjuvant Effects

Use of Toll-Like Receptor 7/8 Agonists in Combination With Other Adjuvants

Conclusions

Chapter 9. CpG Oligodeoxynucleotides as Adjuvants for Clinical Use

Toll-Like Receptor Agonists as Vaccine Adjuvants

CpG Oligodeoxynucleotides

A Brief History of CpG DNA and Toll-Like Receptor 9

Recognition of CpG Motifs

The Signaling Pathway Utilized by Toll-Like Receptor 9

TLR9 Expression by B Cells and Plasmacytoid Dendritic Cells

Classes of CpG Oligodeoxynucleotides

Next-Generation CpG Oligodeoxynucleotides

CpG Oligodeoxynucleotides as Adjuvants for Vaccines Targeting Infectious Pathogens

CpG Oligodeoxynucleotide–Adjuvanted Hepatitis B Virus Vaccine, HEPLISAV

CpG Oligodeoxynucleotide–Adjuvanted Anthrax Vaccine, NuThrax

CpG Oligodeoxynucleotide–Adjuvanted Malaria Vaccines

CpG Oligodeoxynucleotide–Adjuvanted Influenza Vaccines

CpG Oligodeoxynucleotides as Stand-Alone Therapy for Infection

CpG Oligodeoxynucleotides as Adjuvants for Vaccines Targeting Allergy

CpG Oligodeoxynucleotides Coupled to Ragweed Allergen

QbG10 Coadministration With Allergen

QbG10 Without Allergen

CpG Oligodeoxynucleotides as Adjuvants for Vaccines Targeting Cancer

CpG Oligodeoxynucleotides in Cancer Immunotherapy

CpG Oligodeoxynucleotides in Cancer Monotherapy

CpG Oligodeoxynucleotides and Chemotherapy for Cancer

CpG Oligodeoxynucleotides and Radiation Therapy for Cancer

CpG Oligodeoxynucleotides and Therapeutic Monoclonal Antibodies for Cancer

Adverse Events in Clinical Trials

Limitations to the Interpretation of Clinical Data Involving CpG Oligodeoxynucleotides

Conclusion and Perspectives

Chapter 10. Advax Adjuvant: A Potent and Safe Immunopotentiator Composed of Delta Inulin

Delta Inulin Background

Manufacture of Delta Inulin

Advax-Adjuvanted Hepatitis B Vaccine

Advax-Adjuvanted Influenza Vaccines

Other Advax-Adjuvanted Vaccines

Mechanism of Action

Preclinical Safety and Toxicology

Conclusions

Chapter 11. Natural Vaccine Adjuvants and Immunopotentiators Derived From Plants, Fungi, Marine Organisms, and Insects

Introduction

Plant-Derived Immunopotentiators

Potential Adjuvants Derived From Fungi

Marine Organism–Derived Immunopotentiators

Insect-Derived Immunopotentiators

Challenges Faced With Natural Product Research

Methods Used for Screening Vaccine Adjuvants From Natural Products

Conclusions

Chapter 12. Polymeric Particles as Vaccine Delivery Systems

Introduction

Use of PLG for Drug Delivery

Use of PLG for Vaccine Delivery

Conclusions

Chapter 13. MF59: A Safe and Potent Adjuvant for Human Use

Introduction

Initial Development of MF59 Adjuvant

Composition of MF59

Manufacturing of MF59

The Mechanism of Action of MF59

MF59 Safety Profile

MF59-Adjuvanted Influenza Vaccine

MF59 With Other Noninfluenza Vaccines

Future Perspectives on the Use of MF59

Chapter 14. The Development of the Adjuvant System AS01: A Combination of Two Immunostimulants MPL and QS-21 in Liposomes

Introduction

AS01 Formulation Development

Rational for AS01 Selection

Clinical Experience

AS01 Mode of Action

Regulatory Considerations

Conclusions

Chapter 15. Development and Evaluation of AS04, a Novel and Improved Adjuvant System Containing 3-O-Desacyl-4′- Monophosphoryl Lipid A and Aluminum Salt

Introduction

AS04 Adjuvant System

AS04 Formulation Process

3-O-Desacyl-4′- Monophosphoryl Lipid A Dose Selection

Safety Aspects of Adjuvants for Use in Vaccines

Examples of AS04-Based Vaccines

Summary

Chapter 16. ISCOMATRIX Adjuvant in the Development of Prophylactic and Therapeutic Vaccines

Introduction

Induction of T-Cell Responses by ISCOMATRIX Vaccines

Induction of Antibody Responses by ISCOMATRIX Vaccines

ISCOMATRIX Adjuvant Promotes Antigen Dose Sparing

Clinical Studies Involving ISCOMATRIX Vaccines

Use of ISCOMATRIX Adjuvant in Infectious Disease Vaccine Development

Evaluation for Use in Hepatitis B Vaccine

Evaluation for Use in Human Immunodeficiency Virus Vaccines

Evaluation for Use in Rabies Vaccines

Evaluation for Use in Dengue Vaccines

Evaluation for Use in Respiratory Syncitial Virus Vaccines

Evaluation for Use in Pandemic Influenza Vaccines

Evaluation for Use in Herpes Simplex Virus 2 Vaccines

Evaluation for Bacterial and Parasitic Vaccines

Combining ISCOMATRIX Adjuvant With Other Immune Stimulators for Cancer Therapy

Conclusion

Chapter 17. Development and Evaluation of CAF01

Introduction

CAF01 Adjuvant Formulation

Preclinical Immunogenicity Assessment

Safety Evaluation

Clinical Immunogenicity Assessment

Adjuvant Mechanism

Combination With Other Adjuvants

Future Perspectives

Chapter 18. Mineral Adjuvants

Introduction

Preparation and Crystalline Structure of Mineral Adjuvants

Application of Mineral Adjuvants

Vaccine Stability and Metallic Ions

Dosing Mineral Adjuvants

Mechanisms of Adjuvant Activity

In Vivo Clearing of Aluminum and Calcium Adjuvants

Side Effect Profile of Mineral Adjuvants

Concluding Remarks

Chapter 19. Toxin-Based Mucosal Adjuvants

Preface

Cholera Toxin and Attempts to Separate Toxicity From Adjuvant Functions

Efforts to Develop Safe and Potent LT-Based Mucosal Adjuvants for Human Use

Pertussis Toxin–Derived Adjuvants

Rational Design of Novel CT-Based Adjuvants With Maintained ADP-Ribosylating Property

Concluding Remarks

Chapter 20. Adjuvants for Skin Vaccination

Introduction

Skin as a Vaccination Site

Routes of Skin Vaccination

Adjuvants for Skin Immunization

Skin Adjuvants in Clinical Trials

Conclusions

Chapter 21. Vaccination to Treat Noninfectious Diseases: Surveying the Opportunities

Introduction

The Immune System

Induction of an Immune Response by Vaccination: General Considerations

Adjuvants in Immunotherapy Based on Induction of Antibodies

Potential and Actual Problems With Active Vaccination

Adjuvants to Stimulate T-Cell Immunity

Adjuvants and Vaccines to Enhance Innate Immunity

Adjuvants and Vaccines Based on Humoral Immunity

Adjuvants and Vaccine Strategies to Enhance the Immunogenicity of Tumor Antigens

Vaccination Against Allergies

Summary and Conclusions

Chapter 22. A Framework for Evaluating Nonclinical Safety of Novel Adjuvants and Adjuvanted Preventive Vaccines

Introduction

Current Status of Relevant, Global Regulatory Guidance and Initiatives

Foundation of a Good Preclinical Safety Package

Adjuvants, Adjuvanted Vaccines, and the Potential for Immune-Mediated Disease

Conclusions

Conclusions—Getting Better

Index

Copyright

Academic Press is an imprint of Elsevier

125 London Wall, London EC2Y 5AS, United Kingdom

525 B Street, Suite 1800, San Diego, CA 92101-4495, United States

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom

Copyright © 2017, 2006 Elsevier Ltd. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-12-804019-5

For information on all Academic Press publications visit our website at https://www.elsevier.com/

Publisher: Sara Tenney

Acquisition Editor: Linda Versteeg-Buschman

Editorial Project Manager: Halima Williams

Production Project Manager: Lucía Pérez

Designer: Matthew Limbert

Typeset by TNQ Books and Journals

List of Contributors

R. Acevedo,     Finlay Institute, Havana, Cuba

N.A. Altwaijry,     University of Strathclyde, Glasgow, Scotland

M.F. Bachmann

University Hospital Bern, Bern, Switzerland

University of Oxford, Oxford, United Kingdom

A.M. Beckmann,     IDRI, Seattle, WA, United States

A. Berger,     GSK Vaccines, Rixensart, Belgium

P.E. Boucher,     PRA Health Sciences, Fort Washington, PA, United States

N. Brock,     Emory University School of Medicine, Atlanta, GA, United States

C. Buonsanti,     GSK Vaccines, Siena, Italy

C. Carnoy

Institut Pasteur de Lille, Lille, France

Inserm, U1019, Lille, France

Univ. Lille, U1019, UMR 8204, CIIL, Center for Infection and Immunity of Lille, Lille, France

CNRS, UMR 8204, Lille, France

CHU Lille, Lille, France

D. Carter

IDRI, Seattle, WA, United States

University of Washington, Seattle, WA, United States

D. Christensen,     Statens Serum Institut, Copenhagen, Denmark

R.W. Compans,     Emory University School of Medicine, Atlanta, GA, United States

U. D'Oro,     GSK Vaccines, Siena, Italy

W.G.J. Degen,     vaxxinova International B.V., Transistorweg, Nijmegen, The Netherlands

A.M. Didierlaurent,     GSK Vaccines, Rixensart, Belgium

N.T. Dobrovolskiene,     National Cancer Institute, Vilnius, Lithuania

L. Duroux,     Brenntag Biosector, Frederikssund, Denmark

V.A. Ferro,     University of Strathclyde, Glasgow, Scotland

C.B. Fox

IDRI, Seattle, WA, United States

University of Washington, Seattle, WA, United States

N. Garçon,     BIOASTER, Lyon, France

A.I. Gray,     University of Strathclyde, Glasgow, Scotland

A.M. Harandi,     University of Gothenburg, Göteborg, Sweden

T.C. Heineman,     Genocea Biosciences, Cambridge, MA, United States

V. Henderickx,     GSK Vaccines, Rixensart, Belgium

J. Igoli

University of Strathclyde, Glasgow, Scotland

University of Agriculture, Makurdi, Benue State, Nigeria

D.M. Klinman,     National Cancer Institute, Frederick, MD, United States

S. Kommareddy,     Research & Development, Seqirus, Cambridge, MA, United States

R.M. Kramer,     IDRI, Seattle, WA, United States

E.C. Lavelle,     Trinity College Dublin, Ireland

N. Lelutiu,     Emory University School of Medicine, Atlanta, GA, United States

E.B. Lindblad,     Brenntag Biosector, Frederikssund, Denmark

N. Lycke,     University of Gothenburg, Göteborg, Sweden

P. Malyala,     GSK Vaccines, Rockville, MD, United States

E. Maraskovsky,     CSL Limited, Bio21 Institute, Parkville, VIC, Australia

S. McCluskey,     Trinity College Dublin, Ireland

A.B. Morelli,     CSL Limited, Bio21 Institute, Parkville, VIC, Australia

N. Muñoz-Wolf,     Trinity College Dublin, Ireland

K. Niwasabutra,     University of Strathclyde, Glasgow, Scotland

D.T. O'Hagan,     GSK Vaccines, Rockville, MD, United States

N. Petroski

Vaxine Pty Ltd, Flinders Medical Centre, Adelaide, SA, Australia

Flinders University, Adelaide, SA, Australia

S.G. Reed

IDRI, Seattle, WA, United States

University of Washington, Seattle, WA, United States

M. Rumbo,     National Universtity of La Plata, La Plata, Argentina

V.E.J.C. Schijns

Wageningen University, The Elst, Wageningen, The Netherlands

Epitopoietic Research Corporation (ERC), Schaijk, The Netherlands and Belgium

H. Shirota,     Tohoku University Hospital, Sendai, Japan

M. Singh,     Takeda Vaccines, Cambridge, MA, United States

J.C. Sirard

Institut Pasteur de Lille, Lille, France

Inserm, U1019, Lille, France

Univ. Lille, U1019, UMR 8204, CIIL, Center for Infection and Immunity of Lille, Lille, France

CNRS, UMR 8204, Lille, France

CHU Lille, Lille, France

I. Skountzou,     Emory University School of Medicine, Atlanta, GA, United States

M.M. Strioga,     National Cancer Institute, Vilnius, Lithuania

F. Tavares Da Silva,     GSK Vaccines, Rixensart, Belgium

M.A. Tomai,     3M Co., Drug Delivery Systems, St. Paul, MN, United States

J. Tusiimire,     University of Strathclyde, Glasgow, Scotland

J.P. Vasilakos,     3M Co., Drug Delivery Systems, St. Paul, MN, United States

J. Vekemans,     GSK Vaccines, Rixensart, Belgium

M. Vogel,     University Hospital Bern, Bern, Switzerland

G. Voss,     SciPeo Sprl, Grez-Doiceau, Belgium

D.G. Watson,     University of Strathclyde, Glasgow, Scotland

N. Woods,     University of Strathclyde, Glasgow, Scotland

Preface

Adjuvants are essential components of vaccines to enhance and direct immune responses, particularly for those vaccines that are composed of highly purified antigens, which are poorly immunogenic. The selection of an appropriate adjuvant is essential to vaccine efficacy and requires a deep understanding of the mode of action of the adjuvant. This is especially relevant for vaccines that require cell-mediated and/or mucosal immunity.

The first edition of this book was very well received by an audience of students, vaccine experts in academia and companies, regulatory officers, and journalists. The intent was to give an overview of novel and existing vaccine adjuvant approaches that were available at the time, including those in licensed vaccines, those under review by regulatory authorities, and those that were still in early stages of preclinical research. The material presented focused on the mechanisms governing the activity of the adjuvants. However, the mechanisms were largely unknown until recently and much was still to be learnt.

Now more than 10  years later, we have significantly updated the book and highlighted the important progress that has been made in the field of vaccine adjuvants. This new book better defines the biological effects of different types of adjuvants and provides the latest knowledge on critical immunological pathways, which is often depicted in comprehensive illustrations. In addition to more detailed knowledge of the structure and chemical composition of specific adjuvants, and their formulations, novel insights into immunological pathways, the role of key immune cells, and novel receptors will also be presented.

We believe that the information in this new edition will enable more rational vaccine design. It will further enhance knowledge on diverse vaccine adjuvant strategies, and will provide the reader with a cutting-edge overview of the vaccine adjuvant field.

We thank all our coauthors for valuable information presented in this book.

Virgil E.J.C. Schijns, and Derek T. O'Hagan

Chapter 1

Vaccine Adjuvants' Mode of Action

Unraveling ‘‘the Immunologist's Dirty Little Secret"

V.E.J.C. Schijns     Wageningen University, The Elst, Wageningen, The Netherlands     Epitopoietic Research Corporation (ERC), Schaijk, The Netherlands and Belgium

Abstract

Most modern vaccines, based only on well-defined purified antigens, are usually insufficient for proper immune induction in the absence of co-administrated immunostimulatory components called adjuvants (adjuvare (Latin) to help). Vaccine adjuvants come in many forms, which are not unified by a common structure. The choice of a suitable adjuvant for a certain vaccine antigen formulation is often difficult due to the fact that little is known about the mechanisms underlying adjuvant activity in general. As a result, the local and systemic reactions of the antigen–adjuvant combination have to be determined by trial and error. Because of this uncertainty adjuvants have been called the immunologist's dirty little secret (Janeway, 1989). Ideally, future vaccines contain adjuvants with predictable activity. During the last decades we have learned in more detail how adaptive immune responses in a normal individual evolve. A number of immunological theories may explain the critical mechanisms underlying adjuvanticity, each of them championing either distinct pathways or distinct key steps within immunological pathways. Here, the most important concepts are discussed in relation to the most recent knowledge in vaccine adjuvant activity.

Keywords

Antibodies; Antigen-presenting cells; Adjuvant; Dendritic cells; Immune system; Vaccines

Introduction

The immune system has evolved to free the host from potentially noxious pathogens and to identify and attack abnormal, potentially tumorigenic cells. Upon first exposure to a pathogen, the immune system reacts with the natural innate and subsequent primary adaptive immune responses. Primary adaptive immune responses need at least a few days to develop before they become effective immune effector responses. This delay in the primary natural immune response is the reason for the variable success of the naive host to attack the invading microorganism in the case of a rapidly replicating invader, which is not stopped by innate immune functions. Vaccination induces an artificial immune response against microorganisms ideally facilitating the formation of long-lived T and B memory cells to conserved antigens, which become rapidly activated after secondary infection. In addition, vaccination may generate readily available immune effector elements, such as circulating antibodies with various functional capacities.

Classic vaccines come in two form, either as attenuated, less virulent, replicating microorganisms, which, however, may carry the risk of reversion to virulence and adverse reactions in immunocompromised individuals, or as nonreplicating inactivated microbes or their components. The latter category is most safe and therefore preferred. Unfortunately, immunization with purified antigen alone is usually insufficient for proper immune induction. Initiation, amplification, and guidance of an appropriate adaptive immune response of sufficient magnitude and duration therefore requires the coadministration of immunostimulatory components called adjuvants [adjuvare (Latin) meaning to help]. Many different types of adjuvants have been described, which are not unified by a common structure. The proper choice of adjuvant for a certain vaccine formulation is often difficult. The onset, magnitude, and type of immune pathway and the duration of immune response of the antigen–adjuvant combination are often unpredictable. This is largely due to the fact that little is known about the mechanisms underlying adjuvant activity in general. In addition, depending on the antigen–adjuvant combination, the local and systemic reactions may vary. Therefore, it is difficult to predict what type of immune reaction and immune effector response will be elicited to the vaccine antigen by the chosen adjuvant. Moreover, the side effects often cannot be foreseen. Therefore, adjuvants have been called the immunologist's dirty little secret.⁵⁸ Although vaccines are the most successful medical invention of the past century, it is obvious that future vaccines require adjuvants with predictable activity.

Today we know that adaptive immune responses in a normal individual initially involve the activation of antigen-specific T-helper cells, which amplify and regulate—via either soluble cytokines or membrane-bound costimulatory molecules—the activities of antimicrobial effector cells, microbiocidal macrophages, cytolytic T cells, and/or B cells. Activation of naive antigen-specific T-helper cells occurs in lymphoid organs, like the lymph nodes, by dendritic cells (DCs), the so-called professional antigen-presenting cells (APCs). Upon delivery or expression of antigen in peripheral tissues, DCs take up the antigen via pinocytosis, phagocytosis, or following infection by the microorganism. During virulent infection, DCs receive stimuli from (structures of) the pathogen, leading to maturation and activation. In the absence of these stimuli, DCs are presumed to tolarize antigen-specific T helper cells, resulting in insufficient priming for an effective T helper cell–dependent immune response. Currently, a number of immunological theories may explain the critical mechanisms underlying adjuvanticity, each of them championing either distinct pathways or different key steps in immunological pathways.¹⁰⁷,¹⁰⁸ Here, the most important concepts are discussed in relation to the most recent knowledge in vaccine adjuvant activity.

Adjuvants Provide Start Signals for Immune Reactivity and Guide the Response to an Acceptable Magnitude

As mentioned earlier, vaccination aims to generate memory immune effector responses of adaptive T and/or B cells specific for a, preferentially conserved, epitope of the pathogen, tumor, or allergen of interest. T helper cells are critical amplifiers and guiders of antigen-specific immune effector cell reactions, such as those of B cells and cytolytic T cells. In addition, they amplify the microbiocidal activity of macrophages. Hence, the priming and clonal expansion of antigen-specific T helper cells is initially critical for adequate adaptive immunity. According to the two-signal model of immune reactivity, activation of T helper cells, apart from the delivery of antigen signals to T cell receptors (signal 1), critically depends on costimulatory signals (signal 2) in the form of soluble cytokines or membrane-bound surface molecules at the time of antigen recognition. According to the classic two-signal model, antigen presentation in the absence of costimulation results in T-cell anergy, tolerance, or deletion.¹³,¹⁶,⁷³

Following on from these thoughts, efficient vaccines should be able to amplify and direct adaptive immune responses, most of which are under regulatory control of antigen-specific major histocompatibility complex (MHC) class II–restricted T helper cells. Although the two-signal theory is well accepted and sustained by numerous publications, it does not explain all immunological events and is at variance with other experimental data.

Regulation of Immune Responses by Antigen Delivery (Signal 1)

Naive T cells continuously survey and recirculate between lymph nodes and the spleen. They are unable to access the nonlymphoid areas of the body. Only memory and effector cells can do this. To be recognized by antigen-specific T and B cells of the adaptive immune system, antigen administered as a vaccine to peripheral tissues, like skin and muscle tissue, must first reach the peripheral lymphoid organs. Mice lacking secondary lymph nodes, due to a genetic mutation or as a result of surgical ablation, are strongly compromised in cellular and humoral responses.⁶⁴ Also, interruption of afferent lymphatic vessels prevents immune responses.¹¹,³⁵ Antigen present within peripheral tissues drains to regional lymph nodes, either in free form or after uptake by local immature DCs. Especially DCs can prime naive T cells and are, therefore, called professional APCs and also nature's adjuvant.¹⁰,¹²²

On arrival in the lymph node, DCs have processed the antigen in peptide fragments, which are then exposed in MHC molecules on their cell surface (signal 1). According to the geographical concept of immune induction¹³⁵ and the classical depot theory ³³,³⁴,⁵¹ this facilitation of signal 1 expression in secondary lymphoid organs is most critical for immune reactivity and a durable immune response, respectively [see Fig. 1.1 (upper panel) and Fig. 1.2].

A number of observations are in accordance with this view. Kündig et al. ⁷¹ noted that repeated immunization with peptide leads to a functional T-cell response even in CD28-deficient mice, whereas a single immunization resulted in T-cell anergy.

Immunization of syngeneic mice with fibroblasts transfected with glycoprotein G of lymphocytic choriomeningitis virus (LCMV) evoked cytotoxic T lymphocyte (CTL)-mediated immunity against lethal LCMV challenge and resistance to challenge infection with recombinant vaccinia virus expressing LCMV-G protein. Intraperitoneal injection of the transfected cells required only 1% of the cells if injected subcutaneously to induce CTL, whereas as few as 500  cells were sufficient following direct injection into the spleen.⁷⁰

The fibrosarcoma cell line (MC57), capable of CTL induction, was in fact a prototypic nonprofessional APC, expressing only MHC–peptide complexes without any other known costimulatory molecule.⁷⁰ This study shows that immune responsiveness can occur in the absence of signal 2 if the antigen is able to reach the secondary lymphoid organs in sufficient amounts.

Figure 1.1  Schematic of the key events during immune induction. (A) Antigen localization in the lymph node determines immune induction. (B) Staggered release of antigen from the inoculation site facilitates immune induction. (C) Non-self-discrimination by innate immune cells starts an immune response by the upregulation of costimulatory signals. (D) Necrotic or stressed cells evoke increased costimulation of antigen-presenting cells. Each step may explain the mechanism underlying a particular type of adjuvant: 0, signal 0; 1, signal 1; and 2, signal 2. After Schijns VEJC. Induction and direction of immune responses by vaccine adjuvants. Crit Rev Immunol 2001;21:75–85. With permission.

Also, direct injection of vesicular stomatitis virus particles into the mesenteric lymph nodes of splenectomized and anti-CD4–treated mice, primed for IgM responses largely without T-cell help,⁹² whereas injection of comparable amounts of antigen subcutaneously does not reach the spleen or lymph nodes in sufficient amounts and fails to activate B cells without T-cell help. From these studies, it was concluded that the presence of a sufficient amount of antigen in secondary lymphoid tissues over a given time period (i.e., immunoavailability) is critical for immune reactivity. Hence, localization, dose, and time of antigen within secondary lymphoid organs determine immune reactivity.⁹²,¹³⁴

Figure 1.2  Adjuvants recruit, target, or activate antigen-presenting dendritic cells (DCs). Hence, DC/antigen-presenting cell (APC) migration (signal 1 facilitation) or maturation (signal 2 facilitation) by microbial or nonmicrobial immunopotentiators is key to primary immune induction.

Facilitation of Signal 1

We observed that, upon intramuscular injection of antigen in a saline solution stained with the colorant methylene blue, the majority of the inoculum is detected in the liver and urine within minutes (Schijns et al., unpublished observations). Arguably, higher doses of injected antigen are less rapidly cleared by phagocytes or degraded by enzymatic reactions and have an increased chance to be sampled by resident immature DCs or B cells. The half-life of labeled gD2 antigen after intramuscular injection is only 2.5  h. Only trace amounts of antigen reach the lymph nodes.²⁸

Certain cytokines, such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and fetal liver tyrosine kinase 3 (Flt3) ligand, influence the level of influx and type of DCs migrating toward the antigen injection site and their migration capacity toward the draining lymph node. Thereby, they regulate the amount of antigen (signal 1) presented as processed peptide–MHC complexes on the APC surface that can be recognized.⁹⁰,¹⁰⁰ As early as 1966, Herbert demonstrated that repeated, daily immunizations of minute amounts of antigen (1  mg/dose) in saline solution for a period of 50  days evoked antibody formation with kinetics similar to those of responses evoked by a single immunization of a high dose of antigen (2000  mg) formulated in a water-in-oil depot.⁵¹ From this experiment, Herbert concluded that a slow release of antigen could fully explain the great effectiveness of depot-forming adjuvant.⁵¹ Remarkably, the minute amounts of antigen are apparently able to increase antibody titers in the face of circulating antigen-specific antibodies. It is also remarkable that, under in vitro conditions, antigen included in a water-in-oil emulsion adjuvant is not released from this depot for a period of more than 3  weeks, whereas, given in vivo, such a formulation evokes significant antibody responses within the same 3-week period after inoculation.⁷,⁶¹ By contrast, antigen dissolved in saline or antigen formulated with an oil-in-water adjuvant emulsion is released quickly in vitro, but evokes generally less antibody responses despite immediate availability of antigen.

According to the geographical concept of immune induction and the depot theory, the facilitation of signal 1 expression is most critical and sufficient for immune reactivity. Hence, immune induction critically depends on the facilitation of antigen delivery and presentation in the lymph node. This site is visited by circulating, potentially reactive, resting T or B cells and, importantly, contains sufficient amounts of signal 2 molecules, which do not necessarily need to be upregulated.¹³⁴ Indeed, an antigen unable to reach the lymph node does not elicit a response.⁶⁴

Apart from antigen localization, dose, and timing, the antigenic structure also influences immune reactivity. In particular, highly ordered, repetitive structures mimic pathogen signatures and, therefore, are able to induce immune responses without adjuvants, when given at high concentrations.⁹² This may explain why purified viruslike particles, or even virosomes devoid of genetic material, lacking presumed danger signals such as RNA or DNA, are able to evoke an immune response in the apparent absence of any danger signal.⁸⁹ Instead, they possess a surface structure in a highly repetitive configuration, ordered such that they may cross-link B-cell receptors and thereby provide a direct activation signal for proliferation and antibody synthesis.⁹,³¹,³² In addition, they can target surface receptors on APCs.¹³⁶ This may explain why viral antigens, when decorated in a highly ordered pattern on a particulate carrier, are more immunogenic than solubilized antigens.⁶²,⁶⁵ In 1968, Dresser noted that repeated immunization with aggregate-free soluble bovine g globulin (BGG) caused immunological tolerance, unless particle aggregates of BGG were removed.²⁶ This observation is in apparent conflict with the data described by Herbert.⁵¹ Indeed, empty microparticles per se, with no antigen incorporated or adsorbed, may exert adjuvant activity for antigen that is just admixed⁷ (Schijns et al., unpublished data).

Although it is nowadays accepted that a minimal amount of antigen (concentration) has to reach the secondary lymphoid organs, the optimal kinetics of antigen delivery are unknown. Studies using osmotic pumps, or solid implants continuously delivering antigen over a prolonged period of weeks or months, readily evoked adequate immune reactions⁶⁷,¹²⁹; for a review, see Ref. 78. These observations contrast with older reports suggesting that continuous antigen delivery leads to tolerance or so-called immunological paralysis.²⁴,²⁷,⁸⁸ However, in those studies, exceptionally high or low antigen doses were used.

Although an immune response can be evoked by soluble antigen when it is able to reach the lymph nodes, in general, the strengths and the duration of the reaction strongly benefit from the codelivery of an adjuvant, i.e., an immunopotentiator or a delivery system. This is particularly true for the subcutaneous or intramuscular immunization routes, with reduced chances of antigen uptake by the lymphoid organs before antigen degradation.⁹²

The activity of a number of well-known facilitating adjuvants fits precisely within this concept.¹⁰⁷ All adjuvants that prolong the presence of antigen after inoculation, not necessarily leading to upregulation of signal 2 molecules on DCs, can be categorized accordingly. In this respect, an interesting study by Sun et al.¹²³ showed that alum, nonionic surfactant vesicles, and, to limited levels, also poly(lactide-co-glycolide) failed to upregulate the important classical signal 2 molecules (such as CD80, CD86, and CD40) on DCs, in contrast to lipopolysaccharide (LPS), but proved to stimulate antigen-specific T cells in vitro.¹²³

Although not formally proven, signal 1–facilitating adjuvants likely include the aforementioned depot-type water-in-oil adjuvants⁵¹ and, for a limited time and amount of antigen, also oil-in-water microemulsions, and double-oil emulsions,⁷ aluminum³⁹ or calcium salts,⁴⁶,⁵⁷ liposome-based delivery systems,⁶⁸ and the diverse group of antigen delivery systems based on polymers. They may briefly prolong antigen retention and/or half-life after administration. In addition, all adjuvants that facilitate recruitment of increased numbers of APCs, or which facilitate the targeting, loading, processing, and presentation of antigen on APC, fit within this concept. Such adjuvants may include certain natural or recombinant host-derived chemokines, Flt3 ligand and GM-CSF, and interferon (IFN)-γ or inducers of such molecules.⁹⁰,¹⁰⁰,¹⁰⁶,¹³² Due to their ability to stimulate proliferation of hematopoietic progenitor cells of both lymphoid and myeloid origins, they all lead to increased numbers of antigen sampling DCs at the inoculation site.⁹³ In addition, host opsonins, such as complement or (natural) antibodies, naturally targeting receptors on APCs, can be engaged to improve receptor-mediated delivery of antigen to APCs. Dempsey et al.²² showed that fusion of two to three copies of complement Cd3 to a recombinant model antigen, hen egg lysozyme (HEL), made the antigen 100- to 1000-fold more immunogenic, as evidenced by facilitated antibody formation, when compared with HEL alone. Similarly, targeting of poorly immunogenic antigen to DCs, for example, through selective engagement of receptors for Fc portions of immunoglobulin (Fc-g-R), has been shown to overcome nonresponsiveness and induce immunity to model and tumor antigens.²,¹⁰² The recognition and concentration of pathogen-derived proteins by nonspecific natural antibodies can act as a trigger for complement-dependent pathways, leading to improved immune responses, including interleukin (IL)-4 production and CD8+ T-cell priming.¹²⁰

Antigens present on or in particles are better taken up by APC relative to soluble antigens and hence more immunogenic. They may activate endogenous pro-inflammatory cytokines which trigger antigen-specific immune activation pathways.¹¹⁴

Regulation of Signal 2

Signal 0 (Recognition of Non-Self, Stranger) Concept

The immune system has evolved to protect the host from potentially noxious pathogens. In particular, innate immune cells, including immature DCs in peripheral tissues, are able to recognize molecular patterns on the surface of microorganisms that provide a microorganism-specific signal and an initiation stimulus for the immune system. These patterns represent a conserved signature of certain classes of pathogens and are not found on host cells. The recognition of these so-called pathogen-associated molecular patterns (PAMP) occurs by pathogen recognition receptors (PRRs), such as the toll-like receptors (TLRs), mannose receptors (MRs), and complement receptors.¹²,⁵⁸,⁵⁹,¹³³ These receptors are germline encoded, evolutionary conserved, and found in evolutionary distinct species, from Drosophila to mammals. Activation of PRRs, defined as signal 0, leads to transcriptional activation of proinflammatory cytokine and chemokine genes. In addition, PRR engagement leads to the upregulation of costimulatory molecules, such as members of the B7 family of molecules expressed on the surface of APCs, like DCs and accessory innate immune cells, e.g., macrophages. The upregulated soluble and membrane-bound costimulatory signals on DCs are designated as signal 2 and considered essential for activation of resting T cells and the productive prolongation of the immune response. Indeed, MyD88-deficient mice, which cannot signal through most known TLRs, showed reduced inflammatory responses and selectively impaired immune responses following exposure to microbial stimuli, such as Mycobacterium tuberculosis in complete Freund adjuvant (CFA)¹¹³ [see Fig. 1.1 (lower panel) and Fig. 1.3].

Signaling via different PRRs, expressed by different subsets of innate immune cells, may evoke qualitatively different types of immune reactions that are necessary for proper elimination of the pathogens.⁶³ Also, expression of TLR is regulated by infection-associated signals. Hence, IFN-α/β and IFN-γ activate a number of different TLRs on macrophages.⁸⁶ Also, TLR expression is inducible by LPS.³ Analysis of gene expression profiles of human DCs exposed to different pathogens or their components revealed both common and pathogen-specific gene activation programs. Interestingly, microbial components of these pathogens (like LPS from the Escherichia coli cell wall, yeast cell wall–derived mannan, and double-stranded RNA evoked by influenza infection, which act as ligands for known PRRs) induced only a subset of genes induced by the complete infectious pathogen. Moreover, the levels of gene expression upon recognition of the microbial ligand were reduced when compared with the infection of the DCs by the live pathogen.⁵⁵ Both in humans and mice, specialized DCs have been identified, which sense the presence of viruses, even without the need to become infected, and respond with rapid production of type I IFN.⁶,¹¹⁸

In accordance with the two-signal model, agonistic engagement of pathogen structures with PRR leads to the upregulation of signal 2 molecules on APCs, as a result of direct maturation of DCs; indirectly, this may be supported by cytokines produced by the APCs themselves, or by bystander macrophages or other accessory cells (Figs. 1.1 and 1.2).

Danger Concept (Recognition of Hidden Self) Concept

A related, although distinct, concept for signal 2 induction is the "danger theory coined by Matzinger.⁸³ The danger theory is largely in accordance with the two-signal model, but stresses that phenotypic and functional maturation of APCs results not only from the presence of stranger molecules (microbial conserved molecules like PAMPs) but also from the presence of the so-called danger signals. This model emphasizes that rather than distinguishing between microbial nonself (stranger) and non–microbial self, the immune system discriminates between harmless, healthy signals and signals resulting from tissue destruction (danger). The stress signals may be released endogenously and are not necessarily provided by conserved structures of microbial origin.³⁷ They can be generated from organelles of cells undergoing pathological necrotic death in stressed, damaged tissue, normally not found during a healthy situation (hidden self), but are not very clearly defined at the molecular level. However, it was proved that mitochondrial and nuclear fractions of necrotic cells,⁷⁵ as well as heat shock proteins (HSP), are responsible for DC maturation and subsequent immune activation. Interestingly, TLR4-deficient C3H/HeJ mice failed to respond to HSP60-induced macrophage activation,⁹⁴ suggesting that HSP60 is a putative endogenous ligand for the TLR4 complex. By contrast, apoptotic cells, which maintain intact cell membranes, and do not release cell content before being cleared by phagocytosis, are unable to cause phenotypic and function maturation of DCs.⁷⁵ Besides TLR4, HSP are also claimed to activate TLR2, although there has been speculation about the possibility of endotoxin contamination.¹³⁰ Steroidal ginseng saponins, known to cause substantial muscle necrosis at the injection site, proved to cause maturation of DCs in vitro and to drive Th1 polarization.¹²⁵ Transplantation of skin grafts evokes immune-activating danger signals, leading to transplant rejection, even a month after adoptive transfer of repopulating fetal liver stem cells.⁴ Also, mammalian double-stranded DNA proved to activate APCs.⁵⁶ At conflict with this hypothesis is the observation that both necrotic and apoptotic cells failed to induce maturation of human monocyte-derived immature DCs, except in cases when the cells were contaminated by Mycoplasma, an exogenous microbial stimulus. When Mycoplasma was eliminated by cyprorin treatment, the maturation-inducing effect was lost.¹⁰³ Shi et al.¹¹⁵ identified a molecule purified by chromatography in fractionated cytoplasm of necrotic cells that acts as an alarming endogenous danger signal. They identified uric acid as an activator of DCs and adjuvant for particulate HIVgp120 antigen. It has been demonstrated that adjuvants based on alum promote local necrosis in muscle tissue⁴⁰ and in the mouse peritoneum.⁸¹ However, the nature of alum-induced cell death is not fully characterized.⁹⁵ A role for endogenous danger signals, such as uric acid⁶⁹ and host DNA⁸¹ released during alum-induced cell death in driving immune responses has been demonstrated.

Facilitation of Signal 2

Ligands for Pattern Recognition Receptors

A number of well-known experimental adjuvants act via pattern recognition receptors, which recognize stranger signals from microbial origin or danger signal molecules. Ligands for these receptors are discussed in more detail for their adjuvant function in several distinct chapters in this book. This is the case for various bacterial lipoproteins that have been shown to engage TLR2. Examples include M. tuberculosis in CFA, or mycobacterial derivatives, such as purified protein derivative (PPD), muramyl dipeptide (MDP), and threonyl MDP, coadministered in free form or included as a component in well-known commercially available adjuvants. Also, synthetic poly I:C, mimicking double-stranded (ds) RNA, provides a conserved pattern characteristic of a natural viral infection. This pattern is recognized by TLR3 leading to IFN-α/β production.³ IFN-α/β upregulates expression of TLR-1, 2, 3, and 7⁸⁶; activates DCs¹⁵; and is required for Th1-dependent responses following vaccination with naked DNA plasmids in mice.¹²⁷ In addition, LPS and monophosphoryl lipid A are recognized by TLR4, whereas a highly conserved structure of bacterial flagellin is seen by TLR5.¹,⁸⁴ Both nonmethylated CpG motifs, found abundantly on bacterial DNA, including naked DNA of vaccine plasmids (but to limited amounts in self, eukaryotic DNA), and synthetic CpG-rich oligodeoxynucleotides are recognized as a molecular pattern by TLR9.⁴⁸ Terminal α-D-mannopyranosyl residues are common glycoprotein structures of parasites, bacteria, yeasts, and enveloped virus. MRs found on macrophages and immature DCs have been described to recognize such carbohydrate patterns on microorganisms.¹⁰⁴,¹²¹ These receptors also recognize chitin structures in chitosan or mannans in yeast cell walls, both common vaccine adjuvants.¹¹⁶,¹¹⁷ MRs may act as signal transducing receptors and have been shown to trigger cytokine secretion and DC activation.¹⁰⁴,¹¹⁶,¹³³ In the past few years, the list of identified receptors on innate immune cells has expanded dramatically. For a number of microbial structures, the innate immune cell pattern recognition receptors have yet to be determined. Among viruses, the viral surface G protein of respiratory syncytial virus has been identified as a potential ligand for TLR4.⁴⁵,⁷² However, purified inactivated, nonreplicating viral particles of distinct virus families are able to evoke immune responses in the absence of adjuvants. Repetitive highly organized surface antigens on various virus types, such as the prototypic rhabdoviruses, are likely to cross-link B-cell receptors in the presumed absence of signal 2,⁶² as mentioned earlier. However, even randomly or poorly organized antigens on other virus families, e.g., herpes or corona viridae, are able to evoke T helper and IgG responses without the help of adjuvant.²⁰ These viruses contain dsDNA or dsRNA sequences, which, upon cell entry, possibly activate MHC gene expression and genes involved in antigen presentation¹²⁴ and cytokine production.⁴² Indeed, TLR8 recognizes single-stranded RNA,²³,⁵⁰ whereas TLR2 and TLR9 recognize herpes virus elements.¹⁷,⁵³ Also, the nonmicrobial, low-molecular-weight adenine or guanosine analogs, such as imidazoquinoline compounds and loxoribine, are able to activate inflammatory responses of immune cells via TLR7.⁴⁹

Terminal α-D-mannopyranosyl residues are glycoprotein PAMPs, common on bacteria, yeasts, parasites, and viruses, and less abundantly expressed on eukaryotic cells. They are recognized with relatively high specificity by MRs, C-type lectins, on macrophages and DCs.³⁰,¹⁰⁴ Antigens naturally exposing mannosylated glycoproteins or engineered to express mannose residues are efficiently internalized by DCs and concentrated in the endocytotic pathway. This leads to more efficient presentation to specific T cells when compared with internalization of fluid-phase antigens.¹⁸,³⁰ In addition, IFN-α is induced following recognition of enveloped viruses.⁸⁷ Similarly, artificial synthetic particles of nonmicrobial origin, decorated with terminal sugar residues, including mannose, fucose, or N-acetyl-D-glucosamine, are recognized by mannose or β-glucan receptors on macrophages, resulting in IL-12, tumor necrosis factor (TNF), and IFN-γ production.¹¹⁶,¹¹⁷ The increased immunogenicity of the latter particulate structures is probably explained by the activation of innate immune cells upon recognition of mimics of microbes in phagocytosable form (510  nm). Larger particles of 50–100  nm fail to activate such events. Therefore, mannosylation of antigens is assumed to facilitate immunogenicity.

When using such microbe component–based adjuvants, it is important to realize that receptors for microbial patterns can be differentially expressed among species, and on innate immune cells or their subtypes residing in different tissues, including vaccine inoculation sites. The qualitative and quantitative variation in microbial recognition receptor expression may explain differences in the quality of immunological responses.¹⁰⁸

Endogenous Immune Enhancers

In mice studies, it has been shown that endogenous type I IFN proved essential for Th1-type humoral immune responses induced by typical adjuvants, including LPS, Poly I:C,⁵⁴ incomplete Freund adjuvant (IFA), CFA, and CpG motifs.⁹⁹ Endogenous soluble factors such as IL-12, but not IL-6, proved essential for mediating immune stimulating complex (ISCOM) adjuvanticity,¹¹⁹ whereas IL-4 and IL-13 function¹⁴ and MyD88¹¹³ proved redundant for adjuvant activity of aluminum hydroxide.

After identification as critical elements induced by known adjuvants, host-derived signal 2 molecules themselves have been demonstrated to act as molecular adjuvants. Prototypic molecules include soluble cytokines, facilitating either APC subtype recruitment and activation, or acting more downstream at the level of T helper–CTL or T helper–B-cell interaction. Examples include various cytokines such as the earlier-mentioned hematopoietic growth factors, GM-CSF, and Flt3 ligand, which may selectively recruit response-modulating APCs, as well as the proinflammatory IFN, IL-1, IL-6, IL-12, IL-18, or T cell-derived IL-2, IL-4, etc.⁴⁴,⁹⁹,¹⁰¹

Many recombinant cytokines have been proved to be successful in murine model systems⁹⁹,¹¹⁰–¹¹² or in veterinary species⁴⁴; some are in clinical vaccine trials, whereas others are already components of registered medicinal products (IFN, IL-2, GM-CSF). However, the degree and duration of efficacy boosting may depend on the nature and dose of the antigen and may vary between distinct cytokines and the species of interest (Schijns, unpublished data). The relatively high costs, the short half-life, and the required cytokine concentration may require fine-tuning to make a successful product of the vaccine type of interest.

In addition to secreted cytokines, membrane-bound costimulatory molecules exert adjuvant activity, e.g., TNF ligand superfamily members, such as CD40 ligand,⁸⁰,⁸⁵,⁹¹ OX40-ligand,⁷⁶ and B-cell activating factor,⁷⁹ or the lymphocyte activation gene-3.⁵ Under physiological conditions, these molecules act as natural amplifiers of antigen presentation by APCs, as products of activated T cells. Activation by these ligands of their natural receptors on APCs induces further expression of immune costimulatory molecules, thereby further enhancing APC activity.

When administered as recombinant molecules, expressed in vivo either by plasmids or recombinant vectors, or delivered as recombinant (antigen fusion) proteins, immunopotentiation may result in optimized vaccine immunity.²⁹,⁴¹,⁴⁷,⁹⁸,¹³¹

Release of Immune Brakes

Within a normally functioning immune system, a variety of inhibitory pathways are responsible for immune system self-control. The so-called immune checkpoints represent a group of regulatory steps, aimed at mitigating or preventing exaggerated immunologic responses.⁹⁶ Unfortunately, tumors and many chronic infections are able to facilitate these otherwise indispensable immune regulatory homeostatic mechanisms and exploit them to prevent and escape immune-mediated destruction.

Hence, novel—often immunotherapeutic—vaccination strategies aim to restore the immune balance during cancer or chronic infections. Targeting of immune checkpoint mechanisms represents one of the modern tumor vaccine immunotherapy approaches⁹⁷ and facilitates immune signal 2 by releasing immune brakes. Currently, more and more immune checkpoints are being discovered and targeted to evaluate their potential benefit in therapeutic vaccines.

Signal 3: Regulating the Quality of Induced Immune Pathways and Immunity

The differential outgrowth of certain subsets of distinct antigen-specific Th cells, including Th1, Th2, Th17, Th9, Th22, and also regulatory T cell (Treg) populations, together shape and represent signal 3. This signal results from events that take place during antigen recognition at the level of the APC, its surrounding cells, and microenvironment, and the antigen-recognizing adaptive T and/or B lymphocyte.

DCs come in various forms related to their localization, phenotype, and function.⁷⁷ They may express different sets of PRR or danger receptors depending on their origin and surrounding tissue signals.⁶³ Professional APCs judge the conditions during antigen uptake and, depending on the cytokine or endogenous danger molecule constellation, they may be stably conditioned to differentiate into the so-called DC1-, DC2-, or DC3-type APC,¹⁹,⁷⁴,⁷⁷ associated with imprinting of Th1-, Th2-, and Th3-type immune reactions, respectively.²⁵,⁵² A cell type lacking classical DC phenotype markers with plasmacytoid morphology has been identified as the major type I IFN–producing cell, sensing the presence of viruses.⁶,³⁸ However, when addressing the capacity and response skewing ability of distinct DC subsets following interaction with viral structures, we came to the conclusion that the nature of the antigen may dictate the type of immune response rather than the DC subset.²¹

MyD88 is an adaptor protein mediating signal transduction by TLR. Interestingly, MyD88 knockout mice, which have an impaired ability to signal through many TLR, show no defects in immune responses against ovalbumin when coadministered with alum,¹¹³ although they lacked an inflammatory response to bacterial components.⁶⁶ This indicates that certain adjuvants, especially those of a nonmicrobial nature, do not necessarily trigger the TLR/nuclear factor-κB pathway in DCs, but possibly other PRR or PRR-independent pathways.

As outlined before, T helper cells amplify antigen-specific adaptive immune effector response pathways. Under extreme conditions, they can polarize these antigen-specific responses into Th1-, Th2-, Th9-, Th17-, or Treg-type responses, as evidenced by the pattern of cytokines they produce and the constellation of membrane-bound costimulatory factors they express during antigen presentation and antigen recognition. The instructive signals for this Th subset polarization may originate most upstream in the activation cascades during APC activation (Fig. 1.3). Collectively, the local cytokine pattern, and the expression of costimulatory molecules and local chemokines attracting other cytokine-producing cells, will determine the differential outgrowth of certain subsets of distinct antigen–specific Th cells, including Th1, Th2, Th17, Th9, Th22, and also Treg populations, together shaping and representing signal 3.¹²⁸

According to the geographical concept, immune induction and potentiation benefits from delivery of antigen in secondary lymphoid tissues. For this concept, the antigen dose, place, and residence time regulates the onset, strength, and type of immune responses. Consequently, in the absence of upregulated signal 2, the quality of immune response may be an intrinsic feature of the antigen or its structure. Depending on the route of inoculation or the type of adjuvant, functionally different APCs will sample the antigen, which may potentially influence the nature of the immune response and also the targeting of effector cells (see also the paragraph on signal 4 below).

Figure 1.3  Immunopotentiators influence the quality of immune reactions. Recognition of adjuvants by innate immune cells generates soluble, secreted, or membrane-bound signals, which shape the direction of downstream Th phenotypes upon priming by activated and conditioned antigen-presenting dendritic cells. APC , antigen-presenting cells; CFA , complete Freund adjuvant; DDA, dimethyldioctadecylammonium ; LPS , lipopolysaccharide; IFA , incomplete Freund adjuvant; IFN , interferon; IL , interleukin; LT/CT, ​Escherichia coli enterotoxin/ Cholera toxin ; MC/EO, ​mast cell/eosinophil ; MDP , muramyl dipeptide; MPL , monophosphoryl lipid A; NK , natural killer; PMN, polymorphonuclear leucocyte ; W/O/W , ​water-in oil-in water emulsion.

Facilitation of Signal 3

As described earlier, apart from antigen (signal 1), the quality of immune responses is under the influence of signal 2 molecules (Figs. 1.1–1.3). This may occur by endogenous danger signals or by PRR activation by exogenous microbial structures, leading to predominant Th1 differentiation as a result of IL-12 and IL-18 production, or IFN-γ production by natural killer cells. Conversely, commitment for the Th2 pathway is proposed to develop in the absence of such microbial stimuli, resulting in the alternative default Th2 pathway.⁶⁰ For example, immunomodulatory adjuvants like alum and chitosan are not only antigen delivery systems affecting immune signal 1 but also modulate DC activity and enhance vaccine-specific Th2 responses and their associated cytokine patterns including synthesis of IL-4, IL-5, and IL-13, which profoundly stimulate B cells to produce specific antibodies and thereby influence the quality of the adaptive immune response (signal 3). These adjuvants have been extensively tested and found to be safe for human use. Also, (biodegradable) micro- and nanoparticles, and particulate forms of antigen in general, have strong effects on humoral and cellular vaccine-specific immunity by modulating the activity and antigen-presenting capacity of APCs.⁸²,¹¹⁴ Hence signal 3 eventually depends on the constellation of membrane-bound costimulatory molecules like CD28 or (cytotoxic T lymphocyte associated protein 4 (CTLA-4).⁶⁰

Interestingly, signal 0 activation may not always be associated with Th1 responses. For example, LPS from Porphyromonas gingivalis, which activates via a TLR4-independent pathway, stimulates IL-4 production, whereas hyphae from Candida albicans and a filarial nematode product stimulate IL-4 production by DC and subsequent Th2 responses.²⁵,⁵² Also, cholera toxin polarizes the maturation of DCs into Th2-priming APCs.³⁶ In addition, protein extracts from the helminth, Schistosoma mansoni, were able to induce DC-promoting Th2 cell (designated DC2 cells) development in vitro.¹⁹ Notably, typical PRR conditioning for a Th2-type response has not been identified yet.

In addition, recruitment of DC precursors by pretreatment with GM-CSF, which depends on the concentration, favors polarization toward Th2-type immunity, whereas pretreatment with Flt3 ligand facilitates Th1-associated IgG2a formation for the same ovalbumin antigen.¹⁰¹

Adjuvants May Also Provide Signal 4, Regulating the Homing of Induced Immune Effector Cells

The imprinting of homing signals of vaccine-induced immune effector cells is called signal 4. Remarkably, a wrongly chosen adjuvant, such as IFA for a particular experimental tumor vaccine, may direct tumor antigen–specific effector immune cells to the vaccine injection site, rather than the tumor, whereas the same tumor vaccine antigen injection in combination with a nonrepository adjuvant has been shown to direct effector responses to the tumor.⁴³

This example illustrates that certain vaccine adjuvants, alone, or administered via a specific route, may selectively influence the imprinting of signal 4. The correct imprinting of signal 4 is especially important for vaccines that require immune effector cells to travel to the anatomical site where the effector immune response is needed, including tumors and chronic infections. For example, several reports have demonstrated that anti-tumor-specific CD8+ T cells detected in the blood did not lead to significant intratumoral T-cell infiltration. For mucosal tumors, mucosal imprinting seems important for correct homing of vaccine-induced CD8-positive T lymphocytes to inhibit the growth of mucosal tumors.¹⁰⁵ Immune potentiators intentionally designed for instructing DCs to provide the correct homing signal 4 to T cells remain to be discovered and developed. Vaccines that aim to induce circulating antibodies, which may even reach anatomical sites in the mucosal tissues, are less dependent on the signal 4–imprinting adjuvants.

Outlook

The majority of new-generation vaccines will consist of purified recombinant, synthetic, or natural antigens to avoid the risk of unwanted adverse reactions. Unfortunately, purified antigens are rather passive elements unable to prime or sustain proper immune responses. They require the colocalization of adjuvants to induce, amplify, and guide an innate and subsequent adaptive immune response of sufficient magnitude and quality. Without a proper adjuvant, even the most suitable antigen will not result in an efficacious vaccine when administered in a single or a few shots. Adjuvants for new-generation vaccines will be designed as precision instruments able to trigger appropriate immunological target molecules with minimal local or systemic side reactions. These modern immunopotentiators can nowadays be searched for in a rational approach by considering one or more of the immunological concepts described earlier. Combinations of differentially acting molecules may broaden the choices of available immunopotentiators and will further enhance vaccine efficacy for defined antigens and delivery routes.

The following chapters address in more detail the characteristics of modern and classic immunopotentiators, their presumed or confirmed mode(s) of action, and their efficacy and safety properties in prototype vaccines.

Acknowledgments

The technical expertise of John Jansen, Monique Teeuwen, and Dick Coppus in the preparation of the illustrations is gratefully acknowledged. The author is grateful to Dr. William Enright for the critical reading of the first version of this manuscript. I thank many present and former colleagues and collaborators for discussions that proved helpful to write this chapter.

References

1. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol. 2001;2:675–680.

2. Akiyama K, Ebihara S, Yada A, Matsumura K, Aiba S, Nukiwa T, et al. Targeting apoptotic tumor cells to Fc gamma R provides efficient and versatile vaccination against tumors by dendritic cells. J Immunol. 2003;170:1641–1648.

3. Alexopoulou L, Czopik Holt A, Medzhitov R, Flavell R. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature. 2001;413:732–738.

4. Anderson C.C, Carroll J.M, Gallucci S, Ridge J.P, Cheever A.W, Matzinger P. Testing, time-, ignorance-, and danger-based models of tolerance. J Immunol. 2001;166:3663–3671.

5. Andreae S, Piras F, Burdin N, Triebel F. Maturation and activation of dendritic cells induced by lymphocyte activation gene-3 (CD223). J Immunol. 2002;168:3874–3880.

6. Asselin-Paturel C, Boonstra A, Dalod M, Durand I, Yessaad N, Dezutter-Dambyuant C, et al. Mouse type I IFN-producing cells are immature APCs with plasmacytoid morphology. Nat Immunol. 2001;2:1144–1150.

7. Aucouturier J, Dupuis L, Ganne V. Adjuvants designed for veterinary and human vaccines. Vaccine. 2001;19:2666–2672.

8. Deleted in review.

9. Bachmann M.F, Zinkernagel R.M. Neutralizing antiviral B cell responses. Ann Rev Immunol. 1997;15:235–270.

10. Bancherau J, Briere F, Caux C, Davoust J, Lebecque S, Liu Y.J, et al. Immunobiology of dendritic cells. Ann Rev Immunol. 2000;18:767–811.

11. Barker C.F, Billingham R.E. The role of regional lymphatics in the skin homograft response. Transplantation. 1967;5:962–966.

12. Barrington R, Zhang M, Fisher M, Carroll M.C. The role of complement in inflammation and adaptive immunity. Immunol Rev. 2001;180:5–15.

13. Bretscher P, Cohn M. A theory of self-nonself discrimination. Science. 1970;169:1042–1049.

14. Brewer J.M, Conacher M, Hunter C.A, Mohrs M, Brombacher F, Alexander J. Aluminium hydroxide adjuvant initiates strong antigen-specific Th2 responses in the absence of IL-4-or IL-13-mediated signaling. J Immunol. 1999;163:6448–6454.

15. Cella M, Salio M, Sakakibara Y, Langen H, Julkunen I, Lanzavecchia A. Maturation, activation and protection of dendritic cells by double-stranded RNA. J Exp Med. 1999;189:821–829.

16. Cohn M, Langman R.E. The protection: the unit of humoral immunity selected by evolution. Immunol Rev. 1990;115:11–147.

17. Compton T, Kurt-Jones E, Boehme K.W, Belko J, Latz E, Golenbock D.T, et al. Human cytomegalovirus activates inflammatory cytokine responses via CD14 and toll-like receptor 2. J Virol. 2003;77:4588–4596.

18. Copland M.J, Baird M.A, Rades T, McKenzie J.L, Becker B, Reck F, et al. Liposomal delivery of antigen to human dendritic cells. Vaccine. 2003;21:883–890.

19. De Jong E.C, Vieira P.L, Kalinski P, Schuitemanker J.H, Tanaka Y, Wierenga E.A, et al. Microbial compounds selectively induce Th1 cell-promoting or Th2 cell-promoting dendritic cells in vitro with diverse Th cell-polarizing signals. J Immunol. 2002;168:1704–1709.

20. De Wit M, Horzinek M.C, Haagmans B.L, Schijns V.E.J.C. Host-dependent type-1 cytokine response driven by inactivated viruses may fail to default in the absence of IL-12 and IFN-a/b. J Gen Virol. 2004;85:795–803.

21. De Wit M, Horzinek M.C, Haagmans B.L, Schijns V.E.J.C. Immunisation with virion-loaded plasmacytoid or myeloid dendritic cells induces primary Th-1 immune responses. Vaccine. 2005;23:1343–1350.

22. Dempsey P.W, Allison M.E, Akkaraju S, Goodnow C.C, Fearon D.T. C3d of complement as molecular adjuvant: bridging innate and acquired immunity. Science. 1996;271:348–350.

23. Diebold S.S, Kaisho T, Hemmi H, Akira S, Reis e Sousa C. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science. 2004;303:1529–1531.

24. Dixon F.J, Maurer P.H. Immunologic unresponsiveness induced by protein antigens. J Exp Med. 1955;101:237–245.

25. D'Ostiani C, Del Sero Bacci G.A, Montagnoli C, Spreca A, Mencacci A.P, et al. Dendritic cells discriminate between yeasts and hyphae of the fungus Candida albicans: implications for initiation of T helper cell immunity in vitro and in vivo. J Exp Med. 2000;191:1661–1674.

26. Dresser D.W. Adjuvanticity of vitamin A. Nature. 1968;217:527–529.

27. Dresser D.W, Gowland G. Immunological paralysis induced in adult rabbits by small amounts of protein antigen. Nature. 1964;203:275–292.

28. Dupuis M, McDonald D.M, Ott G. Distribution of adjuvant MF59 and