02 Insect Biochemistry Molecular

Insect Biochemistry and Molecular Biology
Volume 35, Issue 10, Pages 1073-1208 (October 2005)

1.

Identification and recombinant expression of a novel chymotrypsin from
Spodoptera exigua ARTICLE
Pages 1073-1082
Salvador Herrero, Eliette Combes, Monique M. Van Oers, J ust M. Vlak, Ruud A.
de Maagd and J ules Beekwilder

2.

Acquisition, transformation and maintenance of plant pyrrolizidine alkaloids
by the polyphagous arctiid Grammia geneura ARTICLE
Pages 1083-1099
T. Hartmann, C. Theuring, T. Beuerle, E.A. Bernays and M.S. Singer

3.

Molecular characterization and evolution of pheromone binding protein genes
in Agrotis moths ARTICLE
Pages 1100-1111
David Abraham, Christer Lfstedt and J ean-Franois Picimbon

4.

The BmChi-h gene, a bacterial-type chitinase gene of Bombyx mori, encodes a
functional exochitinase that plays a role in the chitin degradation during the
molting process ARTICLE
Pages 1112-1123
Takaaki Daimon, Susumu Katsuma, Masashi Iwanaga, WonKyung Kang and Toru
Shimada

5.

Effect of chloroquine on the expression of genes involved in the mosquito
immune response to Plasmodium infection ARTICLE
Pages 1124-1132
P. Abrantes, L.F. Lopes, V.E. do Rosrio and H. Silveira

6.

Accumulation of 23 kDa lipocalin during brain development and injury in
Hyphantria cunea ARTICLE
Pages 1133-1141
Hong J a Kim, Hyun J eong J e, Hyang Mi Cheon, Sun Young Kong, J ikHyun Han,
Chi Young Yun, Yeon Su Han, In Hee Lee, Young J in Kang and Sook J ae Seo

7.

The transcriptome of the salivary glands of the female western black-legged
tick Ixodes pacificus (Acari: Ixodidae) ARTICLE
Pages 1142-1161
Ivo M.B. Francischetti, Van My Pham, Ben J . Mans, J ohn F. Andersen, Thomas N.
Mather, Robert S. Lane and J os M.C. Ribeiro

8.

Development and characterization of a double subgenomic chikungunya virus
infectious clone to express heterologous genes in Aedes aegypti mosqutioes
ARTICLE
Pages 1162-1170
Dana L. Vanlandingham, Konstantin Tsetsarkin, Chao Hong, Kimberly Klingler,
Kate L. McElroy, Michael J . Lehane and Stephen Higgs

9.

Molecular cloning and analysis of a novel teratocyte-specific carboxylesterase
from the parasitic wasp, Dinocampus coccinellae ARTICLE
Pages 1171-1180
Ravikumar Gopalapillai, Keiko Kadono-Okuda and Takashi Okuda

10.

The extensible alloscutal cuticle of the tick, Ixodes ricinus ARTICLE
Pages 1181-1188
Svend Olav Andersen and Peter Roepstorff

11.

Uptake and turn-over of glucosinolates sequestered in the sawfly Athalia
rosae ARTICLE
Pages 1189-1198
Caroline Mller and Ute Wittstock

12.

Mutant Mos1 mariner transposons are hyperactive in Aedes aegypti
ARTICLE
Pages 1199-1207
David W. Pledger and Craig J . Coates
Copyright 2006 Elsevier Ltd. All rights reserved
Insect
Biochemistry
and
Molecular
Biology
Insect Biochemistry and Molecular Biology 35 (2005) 10731082
Identication and recombinant expression of a novel chymotrypsin
from Spodoptera exigua
Salvador Herrero
a,b,
, Eliette Combes
b
, Monique M. Van Oers
b
, Just M. Vlak
b
,
Ruud A. de Maagd
a
, Jules Beekwilder
a
a
Business Unit Bioscience, Plant Research International B.V., Wageningen University and Research Centre, Wageningen, The Netherlands
b
Laboratory of Virology, Wageningen University, Binnenhaven 11, 6709 PD Wageningen, The Netherlands
Received 2 March 2005; received in revised form 29 April 2005; accepted 2 May 2005
Abstract
A novel chymotrypsin which is expressed in the midgut of the lepidopteran insect Spodoptera exigua is described. This enzyme,
referred to as SeCT34, represents a novel class of chymotrypsins. Its amino-acid sequence shares common features of gut
chymotrpysins, but can be clearly distinguished from other serine proteinases that are expressed in the insect gut. Most notable,
SeCT34 contains a chymotrypsin activation site and the highly conserved motive DSGGP in the catalytic domain around the active-
site serine is changed to DSGSA. Recombinant expression of SeCT34 was achieved in Sf21 insect cells using a special baculovirus
vector, which has been engineered for optimized protein production. This is the rst example of recombinant expression of an active
serine proteinase which functions in the lepidopteran digestive tract. Puried recombinant SeCT34 enzyme was characterized by its
ability to hydrolyze various synthetic substrates and its susceptibility to proteinase inhibitors. It appeared to be highly selective for
substrates carrying a phenylalanine residue at the cleavage site. SeCT34 showed a pH-dependence and sensitivity to inhibitors,
which is characteristic for semi-puried lepidopteran gut proteinases. Expression analysis revealed that SeCT34 was only expressed
in the midgut of larvae at the end of their last instar, just before the onset of pupation. This suggests a possible role of this protein in
the proteolytic remodelling that occurs in the gut during the larval to pupal molt.
r 2005 Elsevier Ltd. All rights reserved.
Keywords: Serine proteinase; Chymotrypsin; Trypsin; Baculovirus; Proteinase inhibitor; Lepidoptera
1. Introduction
Serine proteinases (SP) belong to one of the largest
gene families in the animal kingdom. Within the human
genome, for instance, around 500 proteinase-encoding
genes have been identied, of which around 30% are SP
or SP homologues (SPH) (Southan, 2001). A similar
complexity exists in the Drosophila melanogaster gen-
ome, where around 200 SP- and SPH-encoding genes
have been identied (Ross et al., 2003). SPs are involved
in a wide range of physiological functions, including
digestion of dietary proteins, blood coagulation, im-
mune responses, signal transduction, hormone activa-
tion and development (Barrett et al., 2003). In insects,
the most abundant and best studied group of SPs
contains those expressed in the larval midgut, and these
are supposed to be involved in the digestion of dietary
protein.
Usually, the architecture of such proteinases is
comparatively simple. While most regulatory SPs, for
ARTICLE IN PRESS
www.elsevier.com/locate/ibmb
0965-1748/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ibmb.2005.05.006
Abbreviations: SeCT34, Spodoptera exigua chymotrypsin 34; BAp-
NA, Na-benzoyl-L-arginine p-nitroanilide; SAAPFpNA, N-succinyl-
alanine-alanine-proline-phenylalanine p-nitroanilide; SAAPLpNA, N-
succinyl-alanine-alanine-proline-leucine p-nitroanilide; SAAApNA, N-
succinyl-alanine-alanine-alanine p-nitroanilide; EFLpNA, pyrogluta-
myl-phenylalanine-leucine p-nitroanilide; BBI, BowmanBirk trypsin
inhibitor; PMSF, phenylmethylsulfonyl uoride; TPCK, N-tosyl-L-
phenylalanine chloromethyl ketone; EDTA, ethylenediamine tetra-
acetic acid.
Corresponding author. Present address: Department of Genetics,

University of Valencia, 46100 Burjassot, Spain. Tel.: +34 96 354 30 06;
fax: +34 96 354 30 29.
E-mail address: sherrero@uv.es (S. Herrero).
instance those involved in polyphenol-oxidase activation
(Cerenius and Soderhall, 2004; Ji et al., 2004), or those
involved in dorsoventral patterning (Rose et al., 2003),
have a number of non-proteolytic protein modules
attached to their N-terminus, the SP genes isolated
from lepidopteran midgut do not contain such modules,
and have a relatively small size (i.e. less than 300 amino-
acids) (Bown et al., 1997). Generally, the immature
protein (also called zymogen) contains a signal for its
secretion into the gut lumen and a pro-protein part
which keeps the protein in an inactive form until it is
cleaved off (Barrett et al., 2003).
The study of digestive proteinases in lepidoptera is
generally motivated by the fact that many lepidoptera
are severe agricultural pests and that their digestive
system is a suitable target for crop-protection strategies.
For instance, herbivory of Manduca sexta on tobacco
plants can be reduced by expressing a recombinant
potato proteinase inhibitor in the leaves (Johnson et al.,
1989). Proteinase inhibitors are also employed by the
natural defence of plants against insects (Zavala et al.,
2004). The inhibitors function by blocking the digestive
proteinases in the larval gut, thereby limiting the release
of amino acids from food protein. As a consequence, the
larvae are arrested in development and eventually die.
However, this strategy has not worked in all cases.
Polyphagous insects like Helicoverpa zea and Spodop-
tera exigua have been shown to adapt to the presence of
proteinase inhibitors in their diet, by switching to the
production of proteinases that are resistant to plant
proteinase inhibitors (Jongsma et al., 1995; Mazumdar-
Leighton and Broadway, 2001b). Lepidopteran midgut
SPs have also been studied in relation to their
interaction with the Cry toxins from the entomopatho-
genic bacterium Bacillus thuringiensis (Oppert, 1999).
Cry toxins accumulate in the bacteria in a protoxin form
which, upon ingestion by the insect, is converted into an
active form by action of the insects SP. In addition, SP
are also involved in the inactivation of such toxins by
degradation. Resistance to Cry toxins has been de-
scribed to be mediated both by down-regulation of
proteinase expression thereby decreasing the activation
of the protoxin (Oppert et al., 1997; Herrero et al., 2001)
as well as by up-regulation of SPs increasing toxin
inactivation (Forcada et al, 1996).
Despite their importance, not much is known about
the catalytic properties of individual midgut SPs from
lepidopteran insects. They have been studied following
two different approaches. In a biochemical approach,
the SPs have been puried from the midgut of the
insects, which allowed characterization of their activity
(Volpicella et al., 2003). By this approach, only the most
abundant proteins in the mixture have been identied
and characterized. In a genomic approach, sequences
from different proteinases have been obtained from
cDNA libraries (Bown et al., 1997) or by RT-PCR
techniques using conserved primers (Mazumdar-Leight-
on and Broadway, 2001a). This approach does consider
low abundant proteins, but no information on the
catalytic characteristics of these proteins has so far been
obtained due to the absence of a suitable expression
system.
In the current work, we studied a novel and low
abundant midgut proteinase from the beet armyworm,
S. exigua. The proteinase is characterized by sequence
comparison with related proteinases and detailed
analysis of recombinant expressed protein. A recombi-
nant baculovirus (Autographa californica multicapsid
nucleopolyhedrovirus, AcMNPV) containing a deletion
of the chitinase and cathepsin genes was employed for
the expression of a functional proteinase in insect cells.
Puried recombinant enzyme was characterized by its
ability to hydrolyze synthetic substrates, its kinetic
parameters and its susceptibility to different proteinase
inhibitors.
2. Material and methods
2.1. Proteinase substrates and inhibitors
Synthetic substrates Na-benzoyl-L-arginine p-nitroanilide
(BApNA), N-succinyl-alanine-alanine-proline-phenylala-
nine p-nitroanilide (SAAPFpNA), N-succinyl-alanine-
alanine-proline-leucine p-nitroanilide (SAAPLrNA),
N-succinyl-alanine-alanine-alanine p-nitroanilide (SAA
ApNA) and pyroglutamyl-phenylalanine-leucine p-ni-
troanilide (EFLpNA) were purchased from Bachem AG
(Bubendorf) and Sigma-Aldrich Chemie BV (Zwijn-
drecht). Proteinase inhibitors aprotinin, BowmanBirk
trypsin inhibitor (BBI), phenylmethylsulfonyl uoride
(PMSF), N-tosyl-L-phenylalanine chloromethyl ketone
(TPCK), ethylenediamine tetra-acetic acid (EDTA) and
antipain were purchased from Sigma-Aldrich Chemie
BV (Zwijndrecht). Stock solutions were prepared
according the suppliers specications.
2.2. Insect RNA isolation
S. exigua larvae were continuously reared on articial
diet at 28 1C as described before (Smits and Vlak, 1988).
RNA was isolated at different instars from whole larvae,
from the larval midgut, the adult gut, hemocytes, and
eggs. Larval midguts were pulled from the larvae after
cutting off the hindbody between the last two pseudoleg
pairs. Next, midguts were cut longitudinally with
scissors and washed in phosphate-buffered physiological
saline to remove the gut contents. Adult guts were
obtained by longitudinally cutting of the abdomen.
Although attention was given to remove all non-gut
tissues during dissections, minor contamination could
not be ruled out. For hemocyte isolation hemolymph
ARTICLE IN PRESS
S. Herrero et al. / Insect Biochemistry and Molecular Biology 35 (2005) 10731082 1074
was obtained from last instar larvae by a small incision
in the last pseudoleg. Hemolymph was collected and
mixed (1:1) with anticlotting solution (1.5 mM K
2
HPO
4
,
8 mM NaH
2
PO
4
, 1 mM CaCl
2
, 3 mM KCl, 0.5 mM
MgCl
2
, 0.3 mM NaCl) and the hemocytes were collected
by centrifugation for 5 min at 12,000g. Samples were
stored at 80 1C until further use or used directly for
RNA isolation. Samples for RNA isolation were homo-
genized or incubated (for hemocytes) in Tripure
TM
reagent (Roche, Mannheim) and RNA was subse-
quently isolated according to the protocol described by
the manufacturer.
2.3. Cloning of the SeCT34 gene
An expressed sequence tag (EST) with homology to
chymotrypsins was obtained from a suppression sub-
tractive hybridization (SSH) library from 5th instar
midgut of S. exigua. Library construction will be
described elsewhere (manuscript in preparation). The
EST fragment has a size of 270 nucleotides and covers
nucleotides 131400 in the nal ORF of the SeCT34
gene. Both the 3
0
and 5
0
cDNA fragments were amplied
using the SMART RACE-kit (Clontech, Palo Alto).
cDNAs produced from reverse transcribed mRNA
isolated from midguts of 5th instar larvae of S. exigua
were used to amplify 5
0
and 3
0
ends of the SeCT34 by a
nested PCR procedure. Primers for amplication were
designed based on the cDNA fragment obtained from
the subtractive library. For the 5
0
-end the primers were
5
0
-CCCATTGTGGATGCATGTGGTAGCCC-3
0
for
primary PCR and 5-ACCATCATCAGCTATGAAA
GAC-3
0
for the nested PCR. For the 3
0
-end the primers
were 5
0
-GCAGGCGGCTTATGTTGACTGCAGCC-3
0
and 5
0
-TGGAACTCAGGAGGCACCATGG-3
0
for the
primary and nested PCR, respectively. Amplied
cDNA-ends were puried using a QIAquick PCR
purication kit (Qiagen Benelux B.V., Venlo) and
ligated into pGEM-T Easy (Promega Benelux B.V.,
Leiden). Several clones were sequenced for each frag-
ment and assembled using the Seqman program
(DNAstar package, DNASTAR Inc., Madison).
2.4. Sequence analysis
Comparison of the deduced amino acid sequence of
the SeCT34 gene and phylogenetic reconstructions were
performed using the ClustalX program (Thompson
et al., 1997). Phylogenetic reconstruction was obtained
by the neighbor-joining method (Saitou and Nei, 1987)
together with bootstrap analysis using 100 replicates.
Kimura correction for multiple substitutions was
applied (Kimura, 1983). When specic residues in the
sequence are referred to, the bovine chymotrypsin
numbering is used (Brown and Hartley 1966). Presence
of a signal peptide was predicted using Signal P program
(Nielsen et al., 1997).
In order to simplify the phylogenetic reconstruction, a
total of 15 insect protease sequences was deployed in the
nal analysis, representing the branches for lepidopteran
trypsin and chymotrypsins known to be expressed in the
gut and dipteran chymotrypsins identied by BLAST
search. Other insect proteinases families appeared to be
more distantly related (not shown), and were left out for
clarity.
2.5. Expression analysis by reverse transcription-
polymerase chain reaction (RT-PCR)
The mRNA abundance of SeCT34 in different tissues
and larval instars was estimated by RT-PCR. Total
RNA from the different samples was isolated using
TriPure
TM
reagent. A total of 0.5 mg RNA was reverse
transcribed into cDNA using an oligo-dT primer and
SuperScript
TM
II reverse transcriptase (Invitrogen,
Breda). A total of 5 ml of a 1:5 dilution of cDNA were
used for PCR amplication. PCRs were carried out for
35 cycles of 20 s at 94 1C, 15 s at 541C and 60 s at 72 1C.
The primers employed were 5
0
-AGTCTTTCATAGCT
GATGATGG-3
0
(forward) and 5
0
-CTCCCTTGTCAC
CAATACTG-3
0
(reverse). Ribosomal RNA was used as
a control for the RNA concentration in the samples.
2.6. Generation of recombinant baculovirus
The full length open reading frame (ORF) of SeCT34
was amplied from cDNA from the midgut of last instar
larvae by PCR using a forward primer adding a BamHI
restriction site (5
0
-GAGGATCCGATTAAGTTTCTA
AATTCGAAAATGG-3
0
) and a reverse primer contain-
ing the coding sequence for a polyhistidine tag, a stop
codon and a HindIII restriction site (5
0
-GGAA
GCCTTAATGGTGATGGTGATGGTGGTCCTCAT
AGAGTGCCATGGTAGAC-3
0
). The resulting frag-
ment was cloned in pGemT-easy and sequenced. The
BamHI-HindIII fragment was recloned in plasmid
pFBD-GFP (Kaba et al., 2002) downstream of the
AcMNPV polyhedrin (ph) promoter to generate the
pFBD-GFP-SeCT34 vector. This plasmid, also contains
the GFP protein downstream of the AcMNPV p10
promoter to facilitate screening and tritation in insect
cells. Plasmid pFBD-GFP was employed subsequently
as a negative control.
To generate recombinant baculoviruses, Escherichia
coli DH10BAC cells containing the AcMNPV DCC
bacmid (a recombinant bacmid from which the chitinase
and v-cathepsin genes were deleted (Kaba et al.,
2004)) and the pMON7124 helper plasmid (Luckow
et al., 1993) were transformed with pFBD-GFP
and pFBD-GFP-SeCT34 plasmids. Putative recombi-
nant AcMNPV bacmids were selected by white/blue
ARTICLE IN PRESS
screening and conrmed by PCR. Isolated bacmid DNA
was used to transfect Spodoptera frugiperda Sf21 cells
(Vaughn et al., 1977) using Cellfectin (Invitrogen)
resulting in the recombinant viruses AcDCC-p10GFP
(control) and AcDCC-p10GFPph34. Recombinant
viruses were grown to high titer stocks using standard
procedures (King and Possee, 1992).
2.7. Baculovirus expression and protein purication
Sf21 cells were infected with a multiplicity of infection
(m.o.i.) of 20 and incubated for 48 h. After that medium
and cells were separated and analyzed for recombinant
protein production. Cells were lysed in PBS/Triton X-
100 0.2% and the cell extract was separated from
insoluble material by centrifugation. Proteins were
separated in a 13% SDS-PAGE gel and the presence
of recombinant protein was detected using antibodies
against the 6xHis-tag (BD biosciences) in a standard
Western-blot assay.
For protein production and purication, approxi-
mately 20 10
6
cells were infected with bacmids as
before in 75F tissue culture asks. After 48 h, medium
was collected, centrifuged to remove free cells in
suspension and ltered through a 0.2 mm lter. Protein
was puried from the medium by afnity chromato-
graphy using BD-TALONspin
TM
column (BD
bioscience) according to the user manual.
2.8. SeCT34 activity Assays
Activity experiments were always performed in
parallel with the puried SeCT34 and with the eluates
from the AcDCC-p10GFP infected cells as a negative
control. Immediately after protein purication, eluates
containing SeCT34 were employed for activity assays.
For the substrate specicity studies, 0.2 mg of BApNA,
SAAPFpNA, SAAPLrNA, SAAApNA or EFLpNA
(around 2 mM) were incubated with approximately
100 ng of puried SeCT34 in a nal volume of 200 ml
of 25 mM Tris-HCl pH 8.5 containing 25 mg/ml BSA.
Activity was measured as the increase in absorbance at
405 nm after 12 h. In order to determine the optimal pH
of SeCT34, the experiments were set up as before but
using 2 mM SAAPFpNA as substrate. The following
buffers were used to obtain a pH range: 25 mM sodium
acetate for pH 5.2, 25 mM Tris HCl for pH 610, and
25 mM glycine for pH 911. The effect of the different
proteinases inhibitors on the activity of SeCT34 was
measured in 25 mM glycine pH 11 with 25 mg/ml BSA.
For the inhibition assays, inhibitors were pre-incubated
with the enzyme for 10 min before adding the SAAPFp-
NA to the reaction. Inhibitors included aprotinin, BBI,
PMSF, TPCK, EDTA, and antipain (For the concen-
trations used see Table 1).
Kinetic parameters as catalytic constant (Kcat) and
constant of MichaelisMenten (Km) were calculated by
incubation of SeCT34 with different concentrations of
SAAPFpNA. Variation in the absorbance at 405 nm,
during the rst 3 h of the reaction, was used to estimate
the initial velocities for each substrate concentration.
Experimental values were tted to a MichaelisMenten
equation using GraphPad Prism (GraphPad software
Inc., San Diego). Since activity assays were performed in
the presence of BSA (which appeared to increase the
stability of the puried SeCT34, but at the same time
may act as a competitive substrate) kinetic parameters
are not absolutes values and reect the relative activity
at different conditions.
3. Results
3.1. Cloning and sequence analysis of SeCT34
An EST encoding an atypical chymotrypsin was
identied among ESTs from midgut tissue of 5th instar
larvae of S. exigua. Its complete cDNA was amplied by
RACE procedures, and analyzed (GenBank accession
no. AY820894). The cDNA, referred to as SeCT34, has
a total length of 1234 bp and contains an ORF of
846 bp, encoding a predicted protein of 281 amino acids
(Fig. 1). Analysis of the N-terminal region predicted the
presence of a 20-amino acid signal peptide, followed by
a pro-peptide of 10 amino acids. At the end of the pro-
peptide, most SPs show a consensus sequence RI(V/I)G,
which is cleaved after the arginine residue for activation
by a trypsin-like proteinase. In contrast, SeCT34 has a
phenylalanine residue instead of the arginine, suggesting
activation by chymotrypsin-like proteins. The mature
SeCT34 protein contains the canonic residues His57,
Asp102, and Ser195 (bovine chymotrypsin numbering),
ARTICLE IN PRESS
Table 1
Effect of proteinases inhibitors on the activity of rSeCT34. The
proteinase was used in a concentration of around 10 nM
Inhibitor Concentration (mM) % inhibition
a
Aprotinin 0.001 100
0.01 100
BBI 0.1 82
1 95
PMSF 1 10
10 46
TPCK 0.1 10
1 16
EDTA 1 0
10 10
Antipain 0.1 4
1 69
a
Relative activity compared with activity obtained using SAAPFp-
NA in absence of proteinase inhibitors.
A
R
T
I
C
L
E
I
N
P
R
E
S
S
Fig. 1. Multiple alignment of the predicted amino acid sequence for SeCT34 (AY820894) with bovine chymotrypsin (Bt: Bos taurus, P00766) and insect SP (Ct: Ctenocephalides felis, AAD21828.
Ae: Aedes aegypti, AF487334. Dm: Drosophila melanogaster, NP_732210. Si: Solenopsis invicta, 1EQ9A. Ha: Helicoverpa armigera, CAA72950. BmCT: Bombyx mori, mg-0778.). Highly conserved
motives containing each of the catalytic triad residues are boxed. The black arrow on the N-terminal part indicates the predicted activation site, the gray arrow the residue determining the trypsin/
chymotrypsin.activity.
S
.
H
e
r
r
e
r
o
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
0
7
3
1
0
8
2
1
0
7
7
which are universally conserved among SPs and
constitute the catalytic triad. The presence of a glycin
residue at position 189 has been found in insect
chymotrypsins, such as the re ant (Solenopsis invicta)
chymotrypsin (Botos et al., 2000). Therefore, the
SeCT34 was putatively classied as a chymotrypsin-like
protein.
The SeCT34 protein was aligned and compared with
representative SPs from different species (Fig. 1). Blast-
X analysis (Altschul et al., 1997) of SeCT34 against the
NCBI database did not show high homology to any
known lepidopteran SP. The highest homology was
found with a chymotrypsin from the cat ea, Ctenoce-
phalides felis and dipteran chymotrypsin-like proteins. A
blast search against the Bombyx mori Silkworm EST
database (Mita et al., 2003) revealed homology to a
cDNA fragment (mg0778), encoding a putative
chymotrypsin (BmCT0778 in this work). SeCT34 has
two insertions of approximately six amino acids relative
to bovine chymotrypsin, and to the dipteran chymo-
trypsins (Fig. 1). Insertions are also present in
BmCT0778 at these locations, although their size is
different. The regions around the catalytic-triad resi-
dues, which are conserved in most SPs (TAAHC
around His59, DIAL around Asp102), have also been
conserved in SeCT34. However, both SeCT34 and
BmCT0778 show a remarkable change in the conserved
GDSGGP region (around the catalytic Ser195) to
GDSGSA.
SeCT34 clearly forms a distinct group among the
lepidopteran SPs. This becomes obvious when the
protein is included in a phylogenetic tree with lepidop-
teran trypsins and chymotrypsins known to be expressed
in the midgut (Fig. 2). These proteinases, presumab-
ly involved in digestion of dietary protein, are
only distantly related to SeCT34, whereas the dipteran
chymotrypsins that came out of the homology
analysis appeared to be more closely related to it.
Only the B. mori protein BmCT0778 is located at the
same branch of the tree; this branch remains sepa-
rate when all known insect SP are included in the tree
(Fig. 2).
3.2. Expression analysis of SeCT34
Expression of SeCT34 was examined by RT-PCR on
RNA from S. exigua eggs, neonates, 2nd, 3rd and 4th
instar larvae, early and late 5th instar larvae, in midgut
tissue from 4th, early 5th and late 5th instar larvae and
from mature insects, and in hemocytes from larvae 1 day
into their 5th instar (Fig. 3). Expression was detected
exclusively in the midgut of late 5th instar larvae. Under
the conditions employed, we did not detect SeCT34
expression when RNA from the whole late 5th instar
larvae was used as a template, probably as a result of the
dilution of midgut RNA in the total body RNA.
3.3. Heterologous expression and purication of SeCT34
SeCT34 protein was expressed in Sf21 insect cells,
using the baculovirus-insect cells expression system. For
this purpose the SeCT34 gene was fused to the
polyhedrin promoter, and to a C-terminal 6xHis-tag
encoding DNA. The expression of the recombinant
SeCT34 (rSeCT34) protein was monitored by Western
blot analysis, using an antibody against the 6xHis tag.
At 48 h.p.i, rSeCT34 was detected in the medium as well
as in the cells (Fig. 4A). The rSeCT34 was puried from
the medium by 6xHis afnity chromatography, to a level
where less than 10% contaminant protein was detected
by silver-staining (Fig. 4B). Puried rSeCT34 protein
showed an estimated mobility of around 30 KDa, which
is close to the predicted molecular weight (29 KDa) of
the mature rSeCT34 protein on the basis of the gene
sequence, including the 6xHis-tag. The estimated yield
of puried rSeCT34 was around 100 mg/l of medium.
ARTICLE IN PRESS
0.1
100
100
94
100
SeCT34
BmCT0778
96
98
94
83
100
100
100 65
99
AiT (L)
PiT (L)
MsT (L)
HzT (L)
HaT (L)
HvCT (L) HaCT (L)
SfCT (L)
AeCT (L)
MsCT (L)
DsCT (D)
DmCT (D)
AaeCT (D)
AdCT (D)
AaCT (D)
Fig. 2. Unrooted phylogenetic tree derived from a ClustalX alignment
of selected insect trypsin and chymotrypsin-like proteins. Numbers on
the branches report the level of condence as determined by bootstrap
analysis (100 bootstrap replicates). T in the name indicates trypsin and
CT indicates chymotrypsin. Letter in parenthesis indicates the insect
order L for lepidopera, D for diptera. The proteins used in the tree are:
AdCT: Anopheles darlingi, AAD17494. AeCT: Anopheles aquasalis,
AAD17492. AaeCT: Aedes aegypti AAL93243. DmCT: Drosophila
melanogaster, NP_732210, Dp: Drosophila pseudoobscura, EAL27112.
HzT: Helicoverpa zea, AAF74742. PiT: Plodia interpunctella,
AAF24226. AiT: Agrotis ipsilon,, AAF74752. MsT: Manduca sexta,
T10109. HaT: Helicoverpa armigera, CAA72962. HaCT: Helicoverpa
armigera, CAA72966. SfCT: Spodoptera frugiperda, AAO75039.
HvCT: Heliothis virescens, AAF43709. MsCT: Manduca sexta,
AAA58743. AiCT: Agrotis ipsilon, AAF71516. BmCT0778: Bombyx
mori, mg-0778. The scale bar indicates an evolutionary distance of 0.1
amino acid substitutions per position in the sequence.
3.4. Characterization of the recombinant SeCT34
(rSeCT34) activity
Chymotrypsins are known to cleave peptide bonds
behind Phe, Tyr, Trp and Leu residues (Barrett et al.,
2003). To determine the substrate preference of
rSeCT34, the enzyme was incubated with ve different
synthetic substrates. Maximum activity was obtained
with SAAPFpNA, while hydrolysis of SAAPLpNA was
low (5% of the activity obtained with SAAPFpNA).
Under the conditions employed, rSeCT34 was not able
to hydrolyze SAAApNA (an elastase substrate), BAp-
NA (a trypsin substrate) and EFLpNA (a thiol-
proteinase substrate).
The pH optimum of activity was analyzed by
measuring the hydrolysis of SAAPFpNA in a pH range
from 5 to 11 (Fig. 5). rSeCT34 was more active at basic
pH values, with a maximum activity at pH 11. We could
not test higher pH values, as the substrate appeared to
be unstable at pH411.
Kinetic parameters were obtained at two different pH
values. Under our assay conditions (i.e. in the presence
of BSA, which may compete as a substrate but stabilizes
the enzyme), the Km value for the hydrolysis of
SAAPFpNA was around 4-fold higher at pH 11
(Km 6.2 mM) than at pH 8 (Km 1.6 mM). Substrate
turnover at pH 11 (Kcat 1.8 s
1
) was around 10-fold
higher than at pH 8 (Kcat 0.17 s
1
). Catalytic
efciency values (Kcat/Km) were around 3-fold higher
at pH 11 than at pH 8. The observed 3-fold higher
catalytic efciency at pH 11 is in agreement with the
differences previously found in the activity studies
performed at different pH range (Fig. 5).
Since lepidopteran midgut proteinases are targets for
plant proteinase inhibitors, rSeCT34 was also charac-
terized with regard to its sensitivity to different
proteinase inhibitors. Proteinaceous inhibitors such as
aprotinin and BBI almost fully inhibited the activity of
the rSeCT34 proteinase at all the concentrations tested.
The most active inhibitor was aprotinin, which showed
values of 100% inhibition at the lowest concentration
tested. In contrast, synthetic inhibitors such as PMSF
and Antipain could only inhibit around 50% of the
activity at the highest concentrations tested. Inhibitors
as TPCK and EDTA hardly affected the activity of
rSeCT34 even at the highest concentrations tested
(Table 1).
4. Discussion
4.1. Recombinant expression of lepidopteran gut
proteinases
In this study, we describe the characterization and
funtional expression of SeCT34, a novel chymotrypsin
expressed in the midgut of S. exigua. To our knowledge
ARTICLE IN PRESS
Fig. 3. RT-PCR of SeCT34 transcripts in different tissues and
development stages. rRNA refers to ribosomal RNA.
Fig. 4. Expression and purication of insect-cell expressed rSeCT34.
Panel A, detection of the 6xHis-tag from rSeCT34 by western analysis
in the medium and in the cell extract of Sf21 cells culture infected with
baculovirus expressing rSeCT34 (34) or infected with the control
baculovirus lacking rSeCT34 (C). Panel B, silver stained 12% SDS-
PAGE of rSeCT34 puried from the medium of cells infected with
baculovirus expressing rSeCT34 (34) or with a control baculovirus
lacking rSeCT34 (C). Crude, refers to crude protein extract loaded
onto the column and Pure, refers to the elution from the column of the
puried protein.
4 5 6 7 8 9 10 11 12
0
20
40
60
80
100
120
NaAc
Tris
Gly
pH
%

r
e
l
a
t
i
v
e

a
c
t
i
v
i
t
y
Fig. 5. Inuence of pH on the activity of rSeCT34. Different buffers
were employed to cover the whole pH range as indicated by different
marks (see also Section 2).
this is the rst report of the functional expression of a
recombinant midgut serine proteinase from insects,
despite the availability of many SP sequences from a
broad range of insect species. Sf21 cells, in which
rSeCT34 was expressed, are derived from S. frugiperda,
a lepidopteran species closely related to S. exigua Thus,
the applied recombinant protein expression system is
closely related to the insect species where SeCT34 was
isolated from. Another contribution to the successful
expression of rSeCT34 was possibly made by the use of
an improved baculovirus expression vector, from which
the viral cathepsin and chitinase genes had been removed
(Kaba et al., 2004). V-cathepsin is a cysteine proteinase
involved, in conjunction with the chitinase, in liquefac-
tion of the insect host cell at the end of the baculovirus
infection (Hawtin et al., 1997). Production and stability
of recombinant proteins has been described to be
enhanced when both genes had been eliminated (Kaba
et al., 2004; Berger et al., 2004). Therefore the system
used here may well be applicable to other lepidopteran
midgut SPs, though still some may proof difcult to
express without damage to the expression system.
However, in the case of SeCT34 we have not observed
any indications of such damage.
4.2. Features of SeCT34
SeCT34 is a representative of a novel subgroup of
lepidopteran chymotrypsins, which map in a distinct
branch of the lepidoptera SP phylogenetic tree (Fig. 2).
Characteristic for this sub-group seems to be the
substitution of GP-SA in the highly conserved
GDSGGP domain around the catalytic Ser195 residue.
These substitution occur in both SeCT34 and its
homologue in B. mori (BmCT0778). Some variation is
known to exist at these residues in the SP family from D.
melanogaster (Ross et al., 2003). Out of 148 members of
the SP family, 11 are different in either of these two
residues. Out of 6 members where the Gly197 position is
different, 4 have the change Gly-Ser. Similarly,
sequence analysis of human SP (subfamily S1A)
revealed that only two out of 79 proteins had changed
the Gly197 residue, both of them Gly-Ser, and from
the four members where Pro198 is different, three of
them have Pro-Ala (Yousef et al., 2004). These
observations suggest that the GP-SA substitution is
typical for the SeCT34 subgroup and is one of the very
rare variations allowed at this position that still yields a
functional protein. It is likely that phylogenetic compar-
ison of the SPs from other insect orders reveals the
presence of this sub-group in other orders. In the crystal
structure of bovine chymotrypsin and re ant chymo-
trypsin (Hynes et al., 1990; Botos et al., 2000) the
Gly197 residue (Ser in SeCT34) localizes adjacent to the
catalytic triad His57, Asp102, and Ser195, though it is
not in direct contact with the substrate. This suggests
that Gly197 may have a possible role in positioning of
the catalytic triad relative to the substrate, rather than in
positioning the substrate relative to the enzyme.
The rSeCT34 is a true chymotrypsin, as can be
deduced from its ability to hydrolyze SAAPFpNA.
However, in contrast to most vertebrate chymotrypsins
(Barrett et al., 2003) and invertebrate chymotrypsins
(Lee and Anstee, 1995; Valaitis, 1995), which hydrolyze
SAAPLpNA with similar efciency, rSeCT34 does not
digest the substrate having Leu at the P1 position very
well. This suggests that SeCT34 may have a specic
function, rather than being involved in general digestion
of dietary protein.
The activation site of SeCT34 is atypical. Most
proteinases that are expressed in the gut, both in man
and in insects, are activated by a trypsin-like activity,
which acts on an Arg residue at position 15 (bovine
chymotrypsin numbering) (Brown and Hartley, 1966).
This holds true for trypsins, carboxypeptidases and
chymotrypsins. SeCT34 is an exception to this rule, as it
carries a Phe in this position, suggesting that it is
activated by chymotrypsin rather than by a trypsin. Our
activity assays showed chymotrypsin activity for
SeCT34 without the need of pre-incubation with trypsin,
and the migration of rSeCT34 as a single band. This
suggests that the proteinase activates itself.
4.3. Physiological role of SeCT34
The role of SeCT34 in the midgut of S. exigua remains
unclear. The pH optimum of the recombinant enzyme
(at pH411) is very similar to that of the total gut
proteinase activity (Jongsma et al., 1996). This suggests
that it functions in a similar environment as the
proteinases involved in digestion of dietary protein, i.e.
the gut lumen. This is further supported by our
observation that both SeCT34 and the total gut
proteolytic activity of S. exigua are relatively sensitive
to proteinaceous inhibitors such as BBI and aprotinin
(Table 1; Jongsma et al., 1996).
The possible role of SeCT34 in insect gut physiology
can be inferred from the timing of its expression. The
SeCT34-encoding transcript could exclusively be de-
tected in the fth instar insect, just prior to pupation. To
gain further support for this apparent restricted expres-
sion of SeCT34, expression data for the B. mori
homologue, BmCT0778, available on the Internet were
searched (http://kaikocdna.dna.affrc.go.jp/page_pub.
html). BmCT0778 has only been identied in the
mg-B. mori cDNA library, which was obtained from
the midgut of larvae four days after molting to 5th
instar. No similar fragment has been found in 39 other
cDNA libraries obtained from different tissues and
developmental instars. Thus, both SeCT34 and
BmCT0778 seem to be exclusively expressed during the
transition from 5th instar larvae to pupae. Specic
ARTICLE IN PRESS
proteolytic events occur during this period, when the
midgut epithelium is replaced completely and the
material is recycled by the action of digestive proteinases
(Uwo et al., 2002). Expression levels of SeCT34 were
low in comparison with levels of other midgut protei-
nase genes (data not shown). The precise tuning of its
expression, its auto-activation and its relatively narrow
substrate specicity could mean that SeCT34 is involved
in the activation of other proteinases, which are
involved in midgut remodeling upon pupation. Knock-
out experiments should be carried out to conrm the
role of SeCT34 in this process. Although the specic role
of SeCT34 remains unclear, the functional expression of
SeCT34 in the baculovirus-insect cell system opens a
wide range of possibilities for the study of insect SPs.
Mutational studies could be applied to determine the
role of the different residues in the interaction with plant
proteinase inhibitors or in substrate specicity. Most
signicantly, we have demonstrated that the baculovirus
system is capable of expressing a lepidopteran midgut
SP, and this system should be tested with other
lepidopteran SP genes relevant to digestion of dietary
protein.
Acknowledgments
S. Herrero was supported by a Marie Curie fellowship
contract No. HPMF-CT-2002-01994 from the EU. R.
de Maagd was supported by Program subsidy 347 of the
Dutch Ministry of Agriculture, Nature Management
and Fisheries.
References
Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z.,
Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST:
a new generation of protein database search programs. Nucl. Acids
Res. 25, 33893402.
Barrett, A.J., Rawlings, N.D., Woessner, J.F., 2003. Handbook of
Proteolytic Enzymes. Academic Press, New York.
Berger, I., Fitzgerald, D.J., Richmond, T.J., 2004. Baculovirus
expression system for heterologous multiprotein complexes. Nat.
Biotechnol. 22, 15831587.
Botos, I., Meyer, E., Nguyen, M., Swanson, S.M., Koomen, J.M.,
2000. The structure of an insect chymotrypsin. J. Mol. Biol. 298,
895901.
Bown, D.P., Wilkinson, H.S., Gatehouse, J.A., 1997. Differentially
regulated inhibitor-sensitive and insensitive protease genes from the
phytophagous insect pest, Helicoverpa armigera, are members of
complex multigene families. Insect Biochem. Mol. Biol. 27,
625638.
Brown, J.R., Hartley, B.S., 1966. Location of disulphide bridges by
diagonal paper electrophoresis. The disulphide bridges of bovine
chymotrypsinogen A. Biochem. J. 101, 214228.
Cerenius, L., Soderhall, K., 2004. The prophenoloxidase-activating
system in invertebrates. Immunol. Rev. 198, 116126.
Forcada, C., Alcacer, E., Garcera, M.D., Martinez, R., 1996.
Differences in the midgut proteolytic activity of two Heliothis
virescens strains, one susceptible and one resistant to Bacillus
thuringiensis toxins. Arch. Insect Biochem. Physiol. 31, 257272.
Hawtin, R.E., Zarkowska, T., Arnold, K., Thomas, C.J., Gooday,
G.W., King, L.A., Kuzio, J.A., Possee, R.D., 1997. Liquefaction of
Autographa californica nucleopolyhedrovirus-infected insects is
dependent on the integrity of virus-encoded chitinase and cathepsin
genes. Virology 238, 243253.
Herrero, S., Oppert, B., Ferre, J., 2001. Different mechanisms of
resistance to Bacillus thuringiensis toxins in the Indianmeal Moth.
Appl. Environ. Microbiol. 67, 10851089.
Hynes, T.R., Randal, M., Kennedy, L.A., Eigenbrot, C., Kossiakoff,
A.A., 1990. X-ray crystal structure of the protease inhibitor
domain of Alzheimers amyloid beta-protein precursor. Biochem-
istry 29, 1001810022.
Ji, C., Wang, Y., Guo, X., Hartson, S., Jiang, H., 2004. A pattern
recognition serine proteinase triggers the prophenoloxidase activa-
tion cascade in the tobacco hornworm, Manduca sexta. J. Biol.
Chem. 279, 3410134106.
Johnson, R., Narvaez, J., An, G., Ryan, C., 1989. Expression of
proteinase inhibitors I and II in transgenic tobacco plants: effects
on natural defense against Manduca sexta larvae. Proc. Natl. Acad.
Sci. USA 86, 98719875.
Jongsma, M.A., Bakker, P.L., Peters, J., Bosch, D., Stiekema, W.J.,
1995. Adaptation of Spodoptera exigua larvae to plant proteinase
inhibitors by induction of gut proteinase activity insensitive to
inhibition. Proc. Natl. Acad. Sci. USA 92, 80418045.
Jongsma, M.A., Peters, J., Stiekema, W.J., Bosch, D., 1996.
Characterization and partial purication of gut proteinases of
Spodoptera exigua Hubner (Lepidoptera: Noctuidae). Insect
Biochem. Mol. Biol. 26, 185193.
Kaba, S.A., Nene, V., Musoke, A.J., Vlak, J.M., van Oers, M.M.,
2002. Fusion to green uorescent protein improves expression
levels of Theileria parva sporozoite surface antigen p67 in insect
cells. Parasitology 125, 497505.
Kaba, S.A., Salcedo, A.M., Wafula, P.O., Vlak, J.M., van Oers, M.M.,
2004. Development of a chitinase and v-cathepsin negative bacmid
for improved integrity of secreted recombinant proteins. J. Virol.
Methods 122, 113118.
Kimura, M., 1983. The Neutral Theory of Molecular Evolution.
Cambridge, England.
King, L.A., Possee, R.D., 1992. The Baculovirus Expression System: A
Laboratory Guide. Chapman & Hall, London.
Lee, M.J., Anstee, J.H., 1995. Endoproteases from the midgut of larval
Spodoptera littoralis include a chymotrypsin-like enzyme with an
extended binding-site. Insect Biochem. Mol. Biol. 25, 4961.
Luckow, V.A., Lee, S.C., Barry, G.F., Olins, P.O., 1993. Ef-
cient generation of infectious recombinant baculoviruses by site-
specic transposon-mediated insertion of foreign genes into a
baculovirus genome propagated in Escherichia coli. J. Virol. 67,
45664579.
Mazumdar-Leighton, S., Broadway, R.M., 2001a. Identication of six
chymotrypsin cDNAs from larval midguts of Helicoverpa zea and
Agrotis ipsilon feeding on the soybean (Kunitz) trypsin inhibitor.
Insect Biochem. Mol. Biol. 31, 633644.
Mazumdar-Leighton, S., Broadway, R.M., 2001b. Transcriptional
induction of diverse midgut trypsins in larval Agrotis ipsilon and
Helicoverpa zea feeding on the soybean trypsin inhibitor. Insect
Mita, K., Morimyo, M., Okano, K., Koike, Y., Nohata, J., Kawasaki,
H., Kadono-Okuda, K., Yamamoto, K., Suzuki, M.G., Shimada,
T., Goldsmith, M.R., Maeda, S., 2003. The construction of an EST
database for Bombyx mori and its application. Proc. Natl. Acad.
Sci. USA 100, 1412114126.
Nielsen, H., Engelbrecht, J., Brunak, S., von Heijne, G., 1997.
Identication of prokaryotic and eukaryotic signal peptides and
prediction of their cleavage sites. Protein Eng. 10, 16.
ARTICLE IN PRESS
Oppert, B., 1999. Protease interactions with Bacillus thuringiensis
insecticidal toxins. Arch. Insect Biochem. Physiol. 42, 112.
Oppert, B., Kramer, K.J., Beeman, R.W., Johnson, D., McGaughey,
W.H., 1997. Proteinase-mediated insect resistance to Bacillus
thuringiensis toxins. J. Biol. Chem. 272, 2347323476.
Rose, T., LeMosy, E.K., Cantwell, A.M., Banerjee-Roy, D., Skeath,
J.B., Di Cera, E., 2003. Three-dimensional models of proteases
involved in patterning of the Drosophila embryo. Crucial role of
predicted cation binding sites. J. Biol. Chem. 278, 1132011330.
Ross, J., Jiang, H., Kanost, M.R., Wang, Y., 2003. Serine proteases
and their homologs in the Drosophila melanogaster genome: an
initial analysis of sequence conservation and phylogenetic relation-
ships. Gene 304, 117131.
Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method
for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406425.
Smits, P.H., Vlak, J.M., 1988. Quantitative and qualitative aspects in
the production of a nuclear Polyhedrosis-virus in Spodoptera
exigua larvae. Ann. Appl. Biol. 112, 249257.
Southan, C., 2001. A genomic perspective on human proteases as drug
targets. Drug Discov. Today 6, 681688.
Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins,
D.G., 1997. The CLUSTAL_X windows interface: exible
strategies for multiple sequence alignment aided by quality analysis
tools. Nucl. Acids Res. 25, 48764882.
Uwo, M.F., Ui-Tei, K., Park, P., Takeda, M., 2002. Replace-
ment of midgut epithelium in the greater wax moth, Galleria
mellonela, during larval-pupal moult. Cell Tissue Res. 308,
319331.
Valaitis, A.P., 1995. Gypsy moth midgut proteinases: purication and
characterization of luminal trypsin, elastase and the brush border
membrane leucine aminopeptidase. Insect Biochem. Mol. Biol. 25,
139149.
Vaughn, J.L., Goodwin, R.H., Tompkins, G.J., McCawley, P.,
1977. The establishment of two cell lines from the insect
Spodoptera frugiperda (Lepidoptera; Noctuidae). In Vitro 13,
213217.
Volpicella, M., Ceci, L.R., Cordewener, J., America, T., Gallerani, R.,
Bode, W., Jongsma, M.A., Beekwilder, J., 2003. Properties of
puried gut trypsin from Helicoverpa zea, adapted to proteinase
inhibitors. Eur. J. Biochem. 270, 1019.
Yousef, G.M., Elliott, M.B., Kopolovic, A.D., Serry, E., Diamandis,
E.P., 2004. Sequence and evolutionary analysis of the human
trypsin subfamily of serine peptidases. Biochim. Biophys. Acta
1698, 7786.
Zavala, J.A., Patankar, A.G., Gase, K., Hui, D., Baldwin, I.T., 2004.
Manipulation of endogenous trypsin proteinase inhibitor produc-
tion in Nicotiana attenuata demonstrates their function as
antiherbivore defenses. Plant Physiol. 134, 11811190.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Acquisition, transformation and maintenance of plant pyrrolizidine
alkaloids by the polyphagous arctiid Grammia geneura
T. Hartmann
a,
, C. Theuring
a
, T. Beuerle
a
, E.A. Bernays
b
, M.S. Singer
c
a
Institut fur Pharmazeutische Biologie der Technischen Universitat Braunschweig, Mendelssohnstrasse 1, D-38106 Braunschweig, Germany
b
Department of Entomology, University of Arizona, P.O. Box 210088, Tucson, AZ 85721-0088, USA
c
Department of Biology, Wesleyan University, Hall-Atwater Labs, Rm. 259, Middletown, CT 06459, USA
Received 9 March 2005; accepted 6 May 2005
Abstract
The polyphagous arctiid Grammia geneura appears well adapted to utilize for its protection plant pyrrolizidine alkaloids of almost
all known structural types. Plant-acquired alkaloids that are maintained through all life-stages include various classes of macrocyclic
diesters (typically occurring in the Asteraceae tribe Senecioneae and Fabaceae), macrocyclic triesters (Apocynaceae) and open-chain
esters of the lycopsamine type (Asteraceae tribe Eupatorieae, Boraginaceae and Apocynaceae). As in other arctiids, all sequestered
and processed pyrrolizidine alkaloids are maintained as non-toxic N-oxides. The only type of pyrrolizidine alkaloids that is neither
sequestered nor metabolized are the pro-toxic otonecine-derivatives, e.g. the senecionine analog senkirkine that cannot be detoxied
by N-oxidation. In its sequestration behavior, G. geneura resembles the previously studied highly polyphagous Estigmene acrea.
Both arctiids are adapted to exploit pyrrolizidine alkaloid-containing plants as drug sources. However, unlike E. acrea, G. geneura
is not known to synthesize the pyrrolizidine-derived male courtship pheromone, hydroxydanaidal, and differs distinctly in its
metabolic processing of the plant-acquired alkaloids. Necine bases obtained from plant acquired pyrrolizidine alkaloids are re-
esteried yielding two distinct classes of insect-specic ester alkaloids, the creatonotines, also present in E. acrea, and the
callimorphines, missing in E. acrea. The creatonotines are preferentially found in pupae; in adults they are largely replaced by the
callimorphines. Before eclosion the creatonotines are apparently converted into the callimorphines by trans-esterication. Open-
chain ester alkaloids such as the platynecine ester sarracine and the orchid alkaloid phalaenopsine, that do not possess the unique
necic acid moiety of the lycopsamine type, are sequestered by larvae but they need to be converted into the respective creatonotines
and callimorphines by trans-esterication in order to be transferred to the adult stage. In the case of the orchid alkaloids, evidence is
presented that during this processing the necine base (trachelanthamidine) is converted into its 7-(R)-hydroxy derivative
(turneforcidine), indicating the ability of G. geneura to introduce a hydroxyl group at C-7 of a necine base. The creatonotines and
callimorphines display a striking similarity to plant necine monoesters of the lycopsamine type to which G. geneura is well adapted.
The possible function of insect-specic trans-esterication in the acquisition of necine bases derived from plant acquired alkaloids,
especially from those that cannot be maintained through all life-stages, is discussed.
Keywords: Grammia geneura (Lepidoptera; Arctiidae); Alkaloid sequestration; Alkaloid processing; Pyrrolizidine alkaloids; Insect alkaloids;
Creatonotines; Callimorphines; Chemical defense
1. Introduction
Among insects that sequester plant pyrrolizidine
alkaloids and utilize them for their own chemical defense,
the tiger moths (Lepidotpera: Arctiidae) represent an
impressive example. The ability to sequester pyrrolizidine
alkaloids from the larval diet is most parsimoniously
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.011
Corresponding author. Tel.: +49 5313 915681;

fax: +49 5313 918104.
E-mail address: t.hartmann@tu-bs.de (T. Hartmann).
inferred to have arisen at the ancestral node of the
subfamily Arctiinae (Weller et al., 1999; Conner and
Weller, 2004). Subsequent loss of alkaloid-use within the
Arctiinae appears to have occurred multiple times as have
switches from larval to adult alkaloid feeding.
The success of pyrrolizidine alkaloids as plant-
acquired defense compounds in various insect species
is attributed to a unique propertyan ability to exist in
two interchangeable forms: the pro-toxic free base
(tertiary amine) and its non-toxic N-oxide (Hartmann,
1999; Hartmann and Ober, 2000). All adapted insects so
far studied that recruit pyrrolizidine alkaloids from their
plant hosts have evolved strategies to avoid accumula-
tion of detrimental concentrations of the free bases in
metabolically active tissues. Pyrrolizidine alkaloid-
sequestering Arctiinae maintain the plant-acquired
alkaloids in the state of their N-oxides. They possess a
specic enzyme (senecionine N-oxygenase) localized in
the hemolymph that efciently converts any pro-toxic
free base into its non-toxic N-oxide (Lindigkeit et al.,
1997; Naumann et al., 2002). The acquisition of this
enzyme in ancestral Arctiinae appears to be a mechan-
istic prerequisite for pyrrolizidine alkaloid sequestra-
tion. A second mechanistic requirement for pyrrolizidine
alkaloid sequestration is the ability to recognize the
alkaloids or alkaloid-sources. It has long been known
that pyrrolizidine alkaloids are larval feeding stimulants
(Boppre , 1986; Schneider, 1987) but only recently arctiid
caterpillars have been shown to possess single sensory
neurons in both the lateral and medial styloconic sensilla
of the galeae that respond specically and sensitively
(threshold of response 10
12
10
9
M) to a variety of
pyrrolizidine alkaloids (Bernays et al., 2002a, b).
Among Arctiinae that are adapted to recognize, recruit
and detoxify pyrrolizidine alkaloids from their larval diets
at least three distinctive strategies exist: (i) monophagous
species that as larvae utilize specic host-plants as both
nutrient and alkaloid source, e.g. Tyria jacobaeae, feeding
on Senecio jacobaea (Asteraceae) or Utetheisa ornatrix
feeding on Crotalaria (Fabaceae); (ii) polyphagous species,
e.g. Creatonotos transiens, Estigmene acrea, or Grammia
geneura, that as larvae feed on a variety of different plant
species including the local range of pyrrolizidine alkaloid-
containing species; (iii) Among both types there are some
species like U. ornatrix, C. transiens or E. acrea that
possess androconial organs (coremata) in which they
produce and emit the pyrrolizidine alkaloid-derived male
courtship pheromone, hydroxydanaidal, while others like
T. jacobaeae and G. geneura do not possess coremata and
are not known to produce hydroxydanidal. These
differences may greatly affect the individual strategies to
deal with pyrrolizidine alkaloids. The pyrrolizidine alka-
loid specialist just needs to be adapted to the type of
alkaloids present in its host plant while polyphagous
species are opportunistically able to utilize a variety of
plant pyrrolizidine alkaloids from different sources and to
maintain them in the non-toxic state. In fact, we previously
showed that E. acrea is able to sequester, detoxify and
process pyrrolizidine alkaloids of almost any known
structural type with one exception: otonecine derivatives
(e.g. senkirkine) that cannot be detoxied by N-oxidation
(Hartmann et al., 2005). Senkirkine is neither sequestered
nor metabolized but tolerated. Moreover, E. acrea is able
to convert all kinds of retronecine and heliotridine esters
into insect-specic retronecine esters, the creatonotines,
which appear to be the common precursor for the
formation of the male pyrrolizidine alkaloid-signal hydro-
xydanaidal (Hartmann et al., 2003a, 2004b). The role of
hydroxydanaidal as a male alkaloid signal emitted from
scent brushes (coremata) has been most completely
elucidated by Thomas Eisner and his colleagues with U.
ornatrix (Eisner et al., 2002). During close-range pre-
copulatory behavior, males use the pheromone to signal
the females the amount of their pyrrolizidine alkaloid
load. Females can differentiate between males that contain
different quantities of hydroxydanaidal and appear to
favor males having higher levels (Conner et al., 1990;
Dussourd et al., 1991). At mating the male transmits a
portion of his alkaloids to the female during insemination.
At oviposition these alkaloids together with the females
own load are transmitted to the eggs (Dussourd et al.,
1988; Iyengar et al., 2001). E. acrea shows a similar
pheromone-affected mating behavior (Davenport and
Conner, 2003; Jordan et al., 2005) and male-to-female-
to-eggs alkaloid transfer (Hartmann et al., 2004a).
Like E. acrea, G. geneura inhabits arid savanna and
grasslands of the southwestern USA. In this paper we
show that this arctiid, like E. acrea, is well adapted to
exploit almost any naturally occurring pyrrolizidine
alkaloid containing plant as a drug source. To a great
extent the two arctiids show similar mechanisms of
alkaloid sequestration and processing but also display
distinct differences. Although G. geneura is not known
to synthesize pyrrolizidine-derived pheromones, insect-
specic pyrrolizidine alkaloids play an important role,
but the creatonotines, typical of E. acrea, are largely
replaced by the callimorphines. Our results show a
striking structural similarity of creatonotines and
callimorphines with plant monoesters of the lycopsa-
mine type that are maintained through all life-stages.
We therefore hypothesize that a fundamental function
of the insect-specic necine esters is to sustain the
transfer of pro-toxic pyrrolizidine alkaloid across
different life-stages of the insect.
2. Materials and Methods
2.1. Insects
Caterpillars (penultimate or nal instar larvae) of G.
geneura (Strecker) were collected from a eld population
ARTICLE IN PRESS
T. Hartmann et al. / Insect Biochemistry and Molecular Biology 35 (2005) 10831099 1084
where Senecio longilobus Benth. and Plagiobothrys
arizonicus (A.Gray) Greene ex A. Gray were the only
abundant alkaloid containing host plants. Caterpillar
cultures were reared on a wheat-germ-based articial
diet (Yamamoto, 1969). Larvae were raised individually
in 200-ml plastic cups containing a small cube of plain
diet (alkaloid-free) that was replaced daily. Fifth instar
larvae received a cube of diet (approximately
10 mm10 mm) containing approximately 1 mg of test
alkaloid(s) for 24 h in place of the plain diet. In most
cases the alkaloid meal was completely consumed within
24 h. Afterwards larvae were allowed to complete
development on the plain diet. Some larvae and pupae
(within 48 h after the start of pupation) were frozen for
alkaloid analysis. Pupae retained for obtaining adults
were sexed and individually kept in 200-ml cups. All
samples were preserved within 24 h of eclosion by
freezing. Samples allotted to alkaloid analysis were
lyophilized and kept in closed vials until analysis.
2.2. Exuviae from eld collected caterpillars of G.
geneura
In spring 2002, caterpillars from several eld sites
were opportunistically collected during one of the nal
three larval stages (Table 8). In most cases, any G.
geneura caterpillar found was collected. On one occasion
(Table 8, C), the collected individuals were chosen
haphazardly. These caterpillars were taken to the
laboratory and kept individually in 200-ml plastic cups
containing plain diet, as described above. The exuviae
molted from the stage of collection were saved in
Eppendorf tubes and stored at ambient laboratory
conditions. These exuviae were expected to contain
any pyrrolizidine alkaloids sequestered from host plants
eaten in nature.
2.3. Origin and preparation of pure pyrrolizidine
alkaloids and alkaloid mixtures
Pure pyrrolizidine alkaloids were prepared or ob-
tained as follows: retronecine by hydrolysis of mono-
crotaline (Carl Roth, Karlsruhe, Germany), heliotridine
by hydrolysis of heliotrine, sarracine (containing 5%
sarracinine) was isolated from Senecio silvaticus (Witte
et al., 1990), senkirkine (containing 3% retronecine
esters) was isolated from ower heads of Senecio vernalis
(Hartmann and Zimmer, 1986).
Puried alkaloid extracts were prepared from the
following plant sources: pyrrolizidine alkaloids of the
senecionine type: eld-grown Senecio congestus (shoots),
eld-grown S. jacobaea (ower heads), eld-grown S.
vernalis (ower heads after removal of senkirkine);
pyrrolizidine alkaloids of the lycopsamine type: eld-
grown Eupatorium cannabinum (inorescences), green-
house-grown Heliotropium indicum (inorescences);
pyrrolizidine alkaloids of the parsonsine type: in vitro-
grown plantlets of Parsonsia laevigata (Hartmann et al.,
2003b); pyrrolizidine alkaloids of the phalaenopsine
type (orchid alkaloids): commercially available Phalae-
nopsis hybrids (owers). The alkaloid extracts were
puried as follows: methanolic or aqueous acidic plant
extracts were evaporated, the residue dissolved in
1 MH
2
SO
4
and incubated with an excess Zn dust for
5 h to reduce the pyrrolizidine alkaloid N-oxides. Then
the solution was extracted three times with ethyl ether,
the organic phase was discarded and the aqueous
solution made basic (pH 11) with ammonia and
extracted three times with ethyl ether. The solvent was
evaporated and the residue saved and directly applied in
the feeding experiments.
The identity and purity of the individual pyrrolizidine
alkaloids was conrmed by gas chromatography
(GC)MS basing on their retention indices (RI),
molecular ions and mass fragmentation patterns in
comparison to reference compounds and our compre-
hensive data base. The quantitative composition of
alkaloid mixtures and total alkaloid contents were
determined by quantitative GC (Witte et al., 1993).
2.4. Alkaloid analysis
Single freeze-dried insects (larvae, pupae, adults) were
weighed and then ground in 0.22ml 1M HCl in a mortar,
extracted for 23h and then centrifuged. The pellet was
dissolved in a small volume of HCl and again extracted.
The combined supernatants were extracted with 2ml
dichloromethane, the aqueous phase was recovered, mixed
with excess of Zn dust and stirred for 3h at room
temperature for complete reduction of the pyrrolizidine
alkaloid N-oxides. Then the mixture was made basic with
25% ammonia and applied to an Extrelut (Merck) column
(size adapted to 1.4ml solution/g Extrelut). Pyrrolizidine
alkaloids (free bases) were eluted with dichloromethane
(6ml/g Extrelut). The solvent was evaporated, and the
residue dissolved in 10100 ml methanol prior to GC or
GCMS. Routine GC was performed as described
previously (Witte et al., 1993; Hartmann et al., 2004b).
Quantitative analyses were performed via the FID signals
with heliotrine or monocrotaline as internal standards.
The GCMS data were obtained with a Hewlett
Packard 5890A gas chromatograph equipped with a
30 m0:32 mm analytical column (ZB1, Phenomenex).
The capillary column was directly coupled to a triple
quadrupole mass spectrometer (TSQ 700, Finnigan).
The conditions applied were: Injector and transfer line
were set at 250 1C; the ion source temperature was
150 1C; the temperature program used was: 100 1C
(3 min)-310 1C at 6 1C/min. The injection volume was
1 ml. The split ratio was 1:20, the carrier gas ow was
1.6 ml min
1
He, and the mass spectra were recorded at
70 eV. CI mass spectra were recorded in the positive
ARTICLE IN PRESS
mode with the same GCMS system using ammonia as a
reagent gas; Ion source temperature was 1301C.
2.5. Identication of insect alkaloids
The creatonotines and isocreatonotines A and B and
the three callimorphines, i.e. callimorphine, homocalli-
morphine and deacetylcallimorphine were identied by
their characteristic RIs, molecular ions and mass
fragmentation patterns as described elsewhere (Hart-
mann et al., 2004b).
Callimorphine analogs like the 1,2-dihydrocallimor-
phines and 7-deoxy-1,2-dihydrocallimorphines were
tentatively identied by GCMS and the structures
subsequently conrmed by analysis of necine bases
obtained after hydrolysis. For hydrolysis of callimor-
phine analogs containing 1,2-unsaturated necine bases
puried extracts were kept in 15% ammonia for 2 days
at room temperature. Subsequently the samples were
dried, directly dissolved in N-Methyl-N-(trimethylsilyl)-
triuoro-acetamid (MSTFA) (Fluka) and heated to
75 1C. After 30 min the necine bases (i.e. platynecine,
turneforcidine, trachelanthamidine, isoretronecanol)
were analyzed by GCMS and identied by their RI-
values and mass fragmentation patterns (see data below)
in comparison to reference compounds.
The identity of 7-(S)-callimorphines (heliotridine es-
ters) was deduced as follows: (i) they showed the same
molecular ions and mass fragmentation patterns as the
respective R-congurated callimorphines (retronecine
esters) but differed in their RIs (Table 7); (ii) they were
only detected in feeding experiments with heliotridine;
(iii) hydrolysis of the respective alkaloid extracts (in 10%
NaOH at 100 1C for 2 h) revealed a mixture of
heliotridine and retronecine that were identied by their
characteristic RI-values (Table 7) and identical fragmen-
tation pattern in comparison to reference compounds.
GCMS properties of the novel callimorphine analogs:
(1S)-1,2-Dihdrocallimorphine (necine base: platyneci-
ne)(Fig. 3B): RI 2016; GC-EIMS, m/z (rel. int.): 299
([M]
+
, 11), 255 (32), 140 (18), 138 (7), 96 (16), 95 (1 0 0),
82 (78), 73 (8), 55 (10), 43(17).
(1R)-1,2-Dihdrocallimorphine (necine base: turnefor-
cidine)(Fig. 3B): RI 1975; GC-EIMS, m/z (rel. int.): 299
([M]
+
, 11), 255 (32), 140 (18), 138 (7), 96 (16), 95 (1 0 0),
82 (78), 73 (8), 55 (10), 43(17).
(1S)-1,2-Dihydrohomocallimorphine (necine base:
platynecine)(Fig. 3B): RI 2097; GC-EIMS, m/z (rel.
int.): 313 ([M]
+
, 9), 269 (33), 141 (8), 140 (20), 138 (7),
96 (27), 95 (1 0 0), 82 (78), 57 (26), 55 (11).
(1R)-1,2-Dihydrohomocallimorphine (necine base:
turneforcidine)(Fig. 3B): RI 2053; GC-EIMS, m/z (rel.
int.): 313 ([M]
+
, 9), 269 (33), 141 (8), 140 (20), 138 (7),
96 (27), 95 (1 0 0), 82 (78), 57 (26), 55 (11).
7-deoxy-(1R)-1,2-Dihdrocallimorphine (necine base:
trachelanthamidine)(Fig. 3C): RI 1833; GC-EIMS, m/z
(rel. int.): 283 ([M]
+
, 7), 125 (12), 124 (1 0 0), 122 (6), 95
(5), 83 (17), 82 (8), 73 (4), 55 (8),43 (9).
7-deoxy-(1R)-1,2-Dihydrohomocallimorphine (necine
base: trachelanthamidine)(Fig. 1C): RI 1913; GC-EIMS,
m/z (rel. int.): 297 ([M]
+
,4), 125 (13), 124 (1 0 0), 123 (3),
122 (4), 95 (4), 83 (17), 82 (7), 57 (10), 55 (7).
7-Chloromethoxy-(1S)-1,2-Dihydrohomocallimor-
phine (necine base platynecine): RI 2207;
GC-EIMS, m/z (rel. int.): 284 (8), 255 (54), 196 (10),
188 (13), 96 (23), 95 (1 0 0), 82 (75), 73 (12), 55 (14), 43
(22). GC-CIMS, m/z (rel. int.): 348 (100;
[M(
35
Cl)+H]
+
), 350 (32, [M(
37
Cl)+H]
+
).
7-Chloromethoxy-(1S)-1,2-Dihydrohomocallimor-
phine (necine base platynecine): RI 2282;
GC-EIMS, m/z (rel. int.): 269 (66), 188 (9), 97 (5), 96
(39), 95 (1 0 0), 83 (11), 82 (83), 57 (40), 55 (13), 41 (7).
GC-CIMS, m/z (rel. int.): 362 (100; [M(
35
Cl)+H]
+
),
364 (32, [M(
37
Cl)+H]
+
).
GCMS properties of the trimethylsilyl derivatives of
necine bases obtained by hydrolysis of 1,2-saturated
plant and insect derived pyrrolizidine alkaloids:
Trimethylsilyl-(-)-trachelanthamidine (obtained from
phalaenopsine and 7-deoxy-1,2-dihydrohomocallimor-
phine): RI(ZB1) 1350; EIMS, m/z (rel. int.): 213 (27,
[M]
+
), 212 (14), 198 (24), 185 (27), 124 (12), 122 (13),
110 (23), 84 (19), 83 (1 0 0), 82 (36).
Trimethylsilyl-(-)-isoretronecanol (obtained from
phalaenopsine and 7-deoxy-1,2-dihydrohomocallimor-
phine): RI(ZB1) 1377; EIMS, m/z (rel. int.): 213 (25,
[M]
+
), 212 (14), 198 (21), 185 (27), 110 (23), 84 (19), 83
(1 0 0), 82 (38), 73 (14), 55(13).
Di-trimethylsilyl-(-)-turneforcidine (obtained from in-
sects fed with phalaenopsine): RI(ZB1) 1569; EIMS, m/
z (rel. int.): 301 (7, [M]
+
), 286 (10), 212 (4), 211 (17), 187
(3), 186 (9), 185 (74), 83 (5), 82 (1 0 0), 73 (15).
Di-trimethylsilyl-(-)-platinecine (obtained from platy-
phylline and sarracine and callimorphine analogs of
insects fed with sarracine and platyphylline): EIMS, m/z
(rel. int.): RI(ZB1) 1611; EIMS, m/z (rel. int.): 301 (5,
[M]
+
), 286 (6), 211 (14), 186 (9), 185 (73), 147 (3), 122
(4), 83 (6), 82 (1 0 0), 73 (15).
3. Results
3.1. Sequestration and processing of macrocyclic
pyrrolizidine alkaloids
Extracts of pyrrolizidine alkaloids from three Senecio
species with structurally different alkaloid proles were
fed to larvae. We were particularly interested to see how
larvae deal with macrocyclic pyrrolizidine alkaloids
which contain unusual necine bases like platynecine
and otonecine. The alkaloids of S. jacobaea and S.
vernalis are all sequestered and transmitted almost
unaltered to the adult stage (Table 1). A distinct change
ARTICLE IN PRESS
in the relative pyrrolizidine alkaloid composition was
only observed with the two 15,20-epoxides jacobine
(Fig. 1A) and jacozine, which in comparison to the plant
prole are less abundant in the insects alkaloid prole.
Since the relative proportions of jacoline and jaconine,
the respective hydrolytic and chlorolytic derivatives of
jacobine, are clearly increased in comparison to their
dietary proportions, some degradation of the epoxide
during sequestration seems likely. Although an articial
degradation cannot be excluded, this appears unlikely
since degradation was neither observed under identical
extraction conditions with the articial diet nor in
analogous insect feeding experiments with E. acrea
(Hartmann et al., 2005).
Besides small amounts of the retronecine esters
senecionine/integerrimine, the dietary pyrrolizidine al-
kaloid mixture from S. congestus contains mainly their
platynecine analogs platyphylline/neoplatyphylline, and
senkirkine, the otonecine analog of senecionine.
Whereas the two macrocyclic platynecine esters are
sequestered and stored with almost the same efciency
as their retronecine analogs, senkirkine is entirely
excluded. Neither senkirkine itself nor insect-specic
otonecine esters are detectable in insect extracts.
Senkirkine (Fig. 1C) is as toxic as senecionine but
cannot be detoxied by N-oxidation (Lindigkeit et al.,
1997; Fu et al., 2004). To conrm the ability of G.
geneura to exclude senkirkine from being sequestered,
an additional feeding experiment with 97% pure
senkirkine was performed (Table 2). No traces of
senkirkine or potential metabolites were recovered from
the analyzed adults. However, the insects did contain
four retronecine esters that were present as impurities in
the senkirkine sample. One can calculate that larvae
ARTICLE IN PRESS
Fig. 1. Plant-acquired pyrrolizidine alkaloids sequestered and maintained by G. geneura through all life-stages comprise: (A) Various types of
macrocyclic retronecine esters, and (B) open-chain monoesters of the lycopsamine type. In the latter case adults preferentially contain alkaloids with
(7R)- and (3S)-conguration; alkaloids with opposite conguration are largely epimerized. (C) Macrocyclic otonecine esters that cannot form N-
oxides are neither sequestered nor metabolized.
A
R
T
I
C
L
E
I
N
P
R
E
S
S
Table 1
Proles of the pyrrolizidine alkaloids established by GCMS for G. geneura that as larvae (penultimate instar) had received about 1 mg per individual of the indicated plant derived alkaloid mixtures
added to the articial diet
Alkaloids recovered m/z [M
+
] RI Relative abundance (%)
Alkaloid mixture from Senecio jacobaea Alkaloid mixture from Senecio vernalis Alkaloid mixture from Senecio congestus
Diet Larvae
n 2
Males
n 3
Females
n 4
Diet Larvae
n 2
Males
n 4
Females
n 3
Diet Larvae
n 2
Males
n 6
Females
n 1
Plant acquired alkaloids
9-Angeloylplatynecine 5 2.570.5
Senecivernine 335 2283 73 74.570.5 5870.6 7173.8
Senecionine 335 2274 3 570 8.371.5 5.870.3 6 6.570.5 1170.5 7.370.3 3 1772 1470.8 14
Seneciphylline 333 2293 13 21.570.5 2870.8 2271.1 4 4.070 6.370.5 4.370.3
Spartioidine 333 2325 o1 170 1.370.3 170 3 3.070 3.570.3 2.071.0
Integerrimine 335 2335 3 670 7.770.3 7.070.7 10 1270 1570.3 1370.7 3 1471.5 1172.3 11
Unknown senecivernine derivative 349 2400 4 470 170 2.071.0
Platyphylline 337 2328 24 6272.5 6072.3 59
Neoplatyphylline 337 2354 2 470 4.070.4 4
Jacobine 351 2420 46 1570. 11.371.5 16.570.9
Jacozine 349 2440 9 2.570.5 1.270.4 1.770.3
Senkirkine 365 2450 59
Jacoline 369 2471 7 2072 2174.3 2371.4
Jaconine 387 2507 8 2372 1370.3 1570.7
Dehydrojaconine 385 2540 &lt 0.270.1
Eruciorine 351 2591 2 2.570.5 2.070.6 1.570.3
Creatonotines
Creatonotine B 269 1978 Tr Tr
Callimorphines
Desacetylcallimorphine 255 1821 0.270.1 0.270.1
Callimorphine 269 1972 3.771.7 3.570.5 4.371.4 2.771.8
Homocallimorphine 311 2033 0.570.3 1.470.6 1.370.6 Tr 0.470.5
(1S)-1,2-Dihydrocallimorphine 299 2015 5.571.0 8.0
(1S)-1,2-Dihydrohomocallimorphine 313 2096 3.771.1 3
Total alkaloid (mg/individual) 189753 227766 2437106 390736 186724 81752 4577 4476.1 42
Total alkaloid (mg/g dry wt) 1.370.5 2.470.8 1.470.6 2.770.2 1.970.1 0.770.2 0.370.1 0.470.07 0.2
T
.
H
a
r
t
m
a
n
n
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
0
8
3
1
0
9
9
1
0
8
8
accumulate about 50% of the trace amounts of these
alkaloids present in their larval food. No toxic or
detrimental effects of senkirkine were observed in the
experiment during further larval development, indicat-
ing that the larvae are well adapted to tolerate otonecine
derivatives present in their alkaloid meals.
In all feeding experiments callimorphines (Fig. 2B)
could be recovered as insect alkaloids from adults but
not larvae. Creatonotines (Fig. 2A) were only detected
in trace amounts in larvae and males fed on S. jacobaea
alkaloids. Insects fed on S. congestus alkaloids con-
tained 1,2-dihydrocallimorphines indicating insect-spe-
cic esterication of platynecine obtained from the
plant-acquired platyphyllines (Fig. 3B).
Pyrrolizidine alkaloid-containing species of the Apoc-
ynaceae often possess unique macrocyclic triesters.
Examples are 14-deoxyparsonsianidine and 14-deoxy-
parsonsianine (Fig. 1A) the major alkaloids of Parsonsia
laevigata. Larvae are able to sequester and store these
alkaloids (Table 3). It is interesting to note that 14-
deoxyparsonsianine, the less abundant pyrrolizidine
alkaloid in the larval diet, accumulates in adults as the
major component. The two pyrrolizidine alkaloids differ
in just one carbon atom (Fig. 1A). In adults the
callimorphines represent a considerable portion (15 to
38%) of total pyrrolizidine alkaloids.
3.2. Sequestration and processing of pyrrolizidine
alkaloids of the lycopsamine type
Alkaloids of the lycopsamine type are characterized by
their unique necic acid moiety, 2-isopropyl-2,3-dihydrox-
ybutyric acid. At least three stereoisomers of this rare
acid are known to occur in alkaloids of the lycopsamine
type: (-)-trachelanthic acid with (2R)(3S)-conguration
in indicine; (-)-viridioric acid, (2
0
S)(3S), in lycopsamine
and echinatine and (+)-trachelanthic acid, (2S)(3R), in
intermedine and rinderine (Fig. 1B). Alkaloids of this
type are typical for pyrrolizidine alkaloid-containing
species of the Boraginaceae, Apocynaceae and the tribe
Eupatorieae of the Asteraceae. For example, indicine and
lycopsamine (from Heliotropium indicum) were seques-
tered and maintained without discrimination (Table 4). It
is notable that the concentration of 3acetylindicine, an
alkaloid that is only detectable in trace amounts in the
larval diet and larval extract, is considerably increased in
adults; it is accompanied by trace amounts of 3-
acetyllycopsamine which does not occur in the larval diet.
Feeding of a puried alkaloid extract from Eupator-
ium cannabinum gave more complex results (Table 4).
Rinderine as a major alkaloid in the larval diet was
found at already decreased levels in larvae and only in
traces in adults which instead contained lycopsamine
and echinatine as major alkaloids. Obviously, alkaloids
with a 3S-conguration (Fig. 1B) are preferentially
transferred to the adult life-stage. While for larvae the
changed alkaloid composition could be accomplished by
uptake discrimination, this explanation can be excluded
for adults. In particular, the strong increase in the
lycopsamine level indicates an insect-specic epimeriza-
tion of (3R)-congurated alkaloids, probably accom-
panied by the known (see Chapter 3.4) epimerization of
(7S)-congurated alkaloids (Fig. 1B).
In addition, like in the experiment with indicine small
amounts of acetyl derivatives are detectable, which were
not present in the larval diet and thus must have been
formed by the insect. Interestingly, besides 3-acetyl
derivatives, 7-acety esters are detected.
In both feeding experiments considerable amounts of
callimorphines are detectable. In the experiment with H.
indicum alkaloids the insect-specic alkaloids account
for 1012%, while in the E. cannabinum experiment, the
callimorphines add up to 27% (males) and 50%
(females) of total alkaloids (Table 4).
ARTICLE IN PRESS
Table 2
Pyrrolizidine alkaloid prole established by GCMS for G. geneura that as larvae (penultimate instar) had received about 1 mg senkirkine per
individual added to the articial diet
Pyrrolizidine alkaloids recovered from insects m/z [M
+
Diet Larvae (n 2) Males (n 3) Females (n 4)
Senecivernine 335 2267 2 42.571.5 38.572.5 40.071.4
Senecionine 335 2275 1 28.071.0 33.571.5 32.070.9
Seneciphylline 333 2288 Tr 12.071.0 13.570.5 12.770.8
Integerrimine 335 2335 Tr 12.070 13.571.5 14.370.5
Senkirkine 365 2460 97 5.571.5
a
Nd Nd
Callimorphines
Homocallimorphine 311 2037 1.171.0 1.170.6
Total alkaloid (mg/individual 18.9710.8 14.371.3 12.872.8
Total alkaloid (mg/g dry wt) 0.0770.04 0.1670.02 0.0970.03
a
Most likely due to the gut content
3.3. Sequestration and metabolism of open-chain
platynecine and trachelanthamidine esters
Feeding of a dietary alkaloid mixture that contained
the open-chain platynecine diester sarracine (containing
5% of its (E)(Z)-isomer sarracinine) (Fig. 3B) (Table 5).
In contrast, adults did not contain even traces of
the plant-derived pyrrolizidine alkaloids but instead
stored the respective platyphylline analogs of creatono-
tines and callimorphines, i.e. (1S)-1,2-dihydrocreatono-
tines and (1S)-1,2-dihdyrocallimorphines (Table 5).
Hydrolysis of the insects alkaloids recovered from
ARTICLE IN PRESS
Fig. 2. Retronecine and heliotridine are converted into insect-specic monoesters. (A) Creatonotines are found in pupae and probably synthesized at
early stages of pupation, (B) callimorphines are found in adults and probably are synthesized shortly before eclosion at the expense of creatonotines,
and (C) (7S)-Congurated heliotridine is partly epimerized yielding (7R)-congurated retronecine and partly converted into callimorphine derivatives
with (7S)-conguration.
adults and GC-MS analysis of the necine base fraction
revealed the presence of platynecine as exclusive necine
base. The two chlorinated alkaloids are most likely
artifacts generated during treatment with dichloro-
methane.
Insects given the dietary mixture of T-phalaenopsine
(trachelanthamidine ester, 80%) and Is-phalaenopsine
(isoretronecanol ester, 20%) (Fig. 3C) did not, as
adults, contain even trace amounts of the dietary
pyrrolizidine alkaloids. Instead the respective 7-deso-
xy-1,2-dihydrocreatonotines and 7-desoxy-1,2-callimor-
phine were present (Table 6). Most interestingly
adults were found to contain as major alkaloids 1,2-
dihydrocallimorphine and 1,2-dihydrohomocallimor-
phine which account for more than 60% of total
pyrrolizidine alkaloids recovered from the insects.
The two compounds display mass fragmentation
patterns identical to those of the 1,2-dihydrocallimor-
phines identied after feeding of plant-acquired platy-
necine esters, i.e. S. congestus (Table 1) and sarracine
(Table 5) but show different RI values (Fig. 4).
Hydrolysis of the alkaloid mixtures recovered from
adults and analysis of the TMS-derivatives of the necine
base fraction revealed the presence a necine base with a
fragmentation pattern identical to that of platynecine
but with a different RI. It was identied as the
platynecine isomer turneforcidine with (1R)-congura-
tion like trachelanthamidine (Fig 3). Trachelanthami-
dine itself was identied in the same experiment
accompanied by only traces of its (1S)-congurated
isomer, i.e. isoretronecanol. This conrms, rstly, that
the alkaloids recovered from the insects have (1R)-
conguration (Table 6) and, secondly, that, G. geneura
must be able to hydroxylate the trachelanthamidine
moiety at C-7 (Table 6; Fig. 3B, C).
ARTICLE IN PRESS
Fig. 3. Formation of insect-specic necine esters with insect-specic
necic acids, i.e. creatonotic acids and callimorphic acids (A). (B)
Formation of 1,2-dihydro derivatives from plant-acquired platynecine,
and (C) formation of 7-deoxy-1,2-dihdyro derivatives from plant
acquired trachelanthamidine and insect-specic 7-hydroxylation of
trachelanthamidine yielding turneforcidine.
Table 3
Pyrrolizidine alkaloid proles established by GCMS for G. geneura that as larvae (penultimate instar) had received about 2 mg per individual of an
alkaloid mixture derived from in vitro cultivated Parsonsia laevigata plantlets added to the articial diet
+
14-Deoxyparsonsianine 425 2773 23 45.077.0 35.776.5 44.371.2
14-Deoxyparsonsianidine 439 2860 61 52.574.5 22.575.9 38.070.6
Heterophylline
a
453 2920 5 1.571.5
Parsonsianidine 455 2935 7
17-Methylparsonsianidine
a
469 2993 3
Creatonotines
Creatonotine B 269 1973 Tr 2.371.3 0.470.3
Callimorphines
Deacetylcallimorphine 255 1821 1.070.99 1.070.6
Callimorphine 297 1955 14.574.8 8.771.3
Homocallimorphine 341 2033 23.376.7 6.770.9
Total alkaloids (mg/individual) 37.2736.8 14.374.0 33.078.2
Total alkaloids (mg/g dry wt) 0.370.3 0.1170.07 0.270.06
a
Tentatively identied
3.4. Metabolism of retronecine and heliotridine:
formation of creatonotines and callimorphines
To study the specicity and temporal sequence of the
formation of insect-specic necine esters, retronecine
and heliotridine were fed with larval diet to G. geneura.
The results are summarized in Table 7. Pupae of
individuals that as larvae received retronecine contain,
besides a small proportion of residual retronecine, the
full set of creatonotines (Fig. 2A) but not even traces of
ARTICLE IN PRESS
Table 4
Proles of the pyrrolizidine alkaloids established by GCMS for G. geneura that as larvae (penultimate instar) had received about 1 mg per individual
of the indicated plant derived alkaloid mixtures added to the articial diet
Alkaloids recovered m/z [M
+
Alkaloid mixture from Eupatorium cannabinum Alkaloid mixture from Heliotropium indicum
Diet Larvae
n 4
Males
n 4
Females
n 2
Diet Larvae
n 2
Males
n 1
Females
n 5
Supinine 283 1967 8 5.070.4
Amabiline 283 1972 Tr 5.872.2
Indicine 299 2120 88 83.572.5 64 50.873.6
Intermedine 299 2131 3 1.870.6
Lycopsamine 299 2145 1 1.871.2 32.5713.9 3575 12 15.071.0 9 8.270.7
Rinderine 299 2151 60 36.576.6 Tr
Echinatine 299 2164 19 42.874.4 30.5711.2 2.570.5
3
0
-Acetylindicin 341 2182 Tr Tr 15 27.873.4
3
0
-Acetylrinderine 341 2210 9
7
0
-Acetyllycopsmaine 341 2210 5.071.8 0.670.2
7
0
-Acetylechinatine 341 2228 6.571.7 2.570.7 0.370.2
3
0
-Acetyllycopsamine 341 2239 Tr 2.370.5 7.570.5 Tr 1.570.4
3
0
-Acetylechinatine 341 2269 1.470.7 0.470.2
Creatonotines
Estigmine B 253 1830 Tr 0.870.3
Creatonotine A 255 1880 Tr
Creatonotine B 269 1973 Tr
Callimorphines
Isodeacetylcallimorphine 255 1814 0.370.1 1.071.0
Deacetylcallimorphine 255 1822 1.570.3 5.070
Callimorphine 297 1955 20.572.4 40.571.5 Tr 9 9.070.52
Homocallimorphine 5.372.4 5.573.5 3 1.670.4
Total alkaloid (mg/individual) 75.8716.5 47.378.5 58.5720.5 186759 105 165722
Total alkaloid (mg/g dry wt) 0.3370.09 0.3570.12 0.3570.15 1.1870.42 0.9 0.9870.09
Table 5
Pyrrolizidine alkaloid proles established by GCMS for G. geneura that as larvae (penultimate instar) had received about 1 mg per individual of
sarracine/sarracinine added to the articial diet
+
Sarracine 337 2390 95 56.072.0
Sarracinine 337 2401 5 10.1710.0
9-Angeloylplatynecine 239 1842 34.078.0
Creatonotines
(1S)-1,2-Dihydrocreatonotine A 257 1923 Tr Tr Tr
(1S)-1,2-Dihydrocreatonotine B 271 2032 Tr 11.974.2 Tr
Callimorphines
(1S)-1,2-Dihydrocallimorphine 299 2016 54.475.6 60
(1S)-1,2-Dihydrohomocallimorphine 313 2097 30.076.0 30
7-Chlormethoxy-(1S)-1,2-dihydrocallimorphine
a
347 2207 2.571.9 8
7-Chlormethoxy-(1S)-1,2-dihydrohomocallimorphine
a
361 2282 Tr 3
Total alkaloid (mg/individual) 7.775.3 6.872.5 27
Total alkaloid (mg/g dry wt) 0.03570.025 0.06170.023 0.17
a
Most likely artifacts generated during extraction with dichloromethane.
callimorphines. In contrast male and female adults were
found to contain the full set of callimorphines (Fig 2B)
and a reduced level of creatonotines. A comparison of
the absolute amounts of the two classes of insect-specic
retronecine esters clearly conrms that the callimor-
phines in adults must have been synthesized at the
expense of the creatonotines (Fig. 5).
Feeding of heliotridine with the larval diet revealed
the full pattern of callimorphines in adult males and
females. However, in the case of callimorphine and
homocallimorphine, in addition to the respective retro-
necine esters, two isomers with different RIs but
identical mol masses and mass fragmentation pattern
were detected and tentatively identied as the respective
7(S)-congurated esters called 7(S)-callimorphines (Fig.
2C). In the case of deacetylcallimorphine that, however,
account for less than 10% of the total callimorphines
only a single peak with an RI identical to the 7(R)-
congurated compound was detected, indicating either
insufcient resolution or absence of 7(S)-deacetycalli-
morphine. In males and females 78% and 48%,
respectively, of total alkaloids accounted for 7(R)-
callimorphines. Hydrolysis of total callimorphines of
both male and females and GC-MS of the resulting
necine bases revealed 69% retronecine and 31%
heliotridine. These proportions are similar to the 66%
retronecine and 34% heliotridine calculated from the
GC-MS data documented in Table 7.
The total amount of insect-specic pyrrolizidine alka-
loids recovered from adults is approximately vefold
higher in the retronecine experiment (Table 7) indicating a
less efcient utilization of heliotridine. In both experiments
females accumulated somewhat higher total amounts than
males but due to their higher body weight the alkaloid
concentrations was almost the same for both sexes.
3.5. Pyrrolizidine alkaloid analysis in exuviae of eld
collected larvae
Exuviae from eld-collected caterpillars varied in
their pyrrolizidine alkaloid content (Table 8). At one
ARTICLE IN PRESS
Table 6
Pyrrolizidine alkaloid proles established by GCMS for G. geneura that as larvae (penultimate instar) had received about 1 mg per individual of a
puried alkaloid mixture derived from a Phalaenopsis hybrid added to the articial diet
+
Diet Males n 2 Females n 1
T-Phalaenopsine (necine base trachelanthamidine, with 1(R)-conguration) 361 2522 81 Nd Nd
Is-Phalaenopsine (necine base isoretronecanol, with 1(S)-conguration) 361 2560 19 Nd Nd
Creatonotines
7-Deoxy-(1R)-1,2-dihydrocreatonotine A (necine base trachelanthamidine) 241 1674 5 8
7-Deoxy-(1R)-1,2-dihydrocreatonotine B (necine base trachelanthamidine) 255 1822 12.572.5 Tr
Callimorphines
7-Deoxy-(1R)-1,2-dihydrocallimorphine (necine base trachelanthamidine) 283 1833 20.578.5 31
7-Deoxy-(1R)-1,2-dihydrohomocallimorphine (necine base trachelanthamidine) 297 1913 Tr Tr
(1R)-1,2-Dihydrocallimorphine (necine base turneforcidine) 299 1975 39.579.5 37
(1R)-1,2-Dihydrohomocallimorphine (necine base turneforcidine) 313 2053 22.071.0 25
Total alkaloids (mg/individual) 8.571.5 5
Total alkaloids (mg/g dry wt) 0.1270.02 0.03
Nd not detected; Tr traces
Fig. 4. GCMS analysis of (1S)-1,2-dihydrocallimorphine (necine
base: platynecine) obtained from G. geneura adults that had received
sarracine with their larval diet (A). Analysis of (1R)-1,2-dihydrocalli-
morphine (necine base: turneforcidine) obtained from G. geneura
adults that had received the orchid alkaloid phalaenopsine with their
larval diet (B).
eld site (A) all 10 caterpillars were devoid of alkaloids,
at two eld sides (E and F) alkaloid-containing and
alkaloid-free caterpillars were found, and at three sites
(B, C, D) all specimens were found to have pyrrolizidine
alkaloids. The characteristic alkaloid patterns of the
alkaloid-positive individuals clearly indicated the kind
of pyrrolizidine alkaloid source: either S. longilobus
(Asteraceae) or Plagiobothrys arizonicus (Boraginaceae)
(Hartmann et al., 2004b). In one case, eld site B, two
individuals with trace amounts of creatonotine B as
exclusive alkaloids were found. In addition to the
summer annual, Crotalaria pumila, which was not yet
present at the time of sampling (March-April), S.
longilobus and P. arizonicus were the only two pyrroli-
zidine alkaloid-containing species found at the sites of
sampling.
4. Discussion
4.1. Larvae of G. geneura are adapted to exploit any
potential plant pyrrolizidine alkaloid source
In a previous study we demonstrated that the arctiid
E. acrea is well adapted to recruit pyrrolizidine alkaloids
from almost any plant source. The ingested alkaloids are
detoxied by N-oxidation, stored and partially trans-
formed into insect-specic creatonotines, the female-
specic creatonotine diesters (i.e., platyphorines) and
the male-specic mating pheromone hydroxydanaidal
(Hartmann et al., 2005). G. geneura shows the same
ARTICLE IN PRESS
Table 7
Metabolism of retronecine and heliotridine by G. geneura. Each individual (penultimate instar) received 1 mg retronecine or heliotridine with the
larval diet. Pupae and adults were sexed before analysis; m males, fm females. Pupae were preserved within 48 h after begin of pupation
Alkaloid recovered m/z[M
+
Retronecine Heliotridine
Pupae (m)
n 2
Pupae (fm)
n 3
Adults (m)
n 9
Adults (fm)
n 11
Adults (m)
n 5
Adults (fm)
n 3
Retronecine 155 1425 5.575.5 11.073.1 0.570.4
Heliotridine 155 1445
Creatonotines
Isocreatonotine A 255 1857 2.070.6
Creatonotine A 255 1878 3.770.9
Isocreatonotine B 269 1955 33.070 27.771.7 0.770.7 0.0670.05
Creatonotine B 269 1981 61.575.5 55.772.7 16.471.9 10.472.8
Total creatonotines 100 100 17.272.1 10.472.8
Callimorphines
Isodeacetylcallimorphine 255 1818 1.270.3 1.870.2 0.470.4 0.370.3
Deacetylcallimorphine 255 1825 5.970.5 6.970.8 3.271.5 3.372.4
Callimorphine 297 1956 69.871.6 75.272.7 41.4710.6 26.3711.3
(S)-Callimorphine 297 1986 16.276.1 46.7714.9
Homocallimorphine 311 2036 5.970.7 3.970.7 33.076.3 17.775.4
(S)-Homocallimorphine 311 2060 5.675.6 5.072.5
Total (R)-callimorphines 82.972.0 89.372.8 78.0710.2 48.3717.4
Total (S)-callimorphines 21.8710.4 51.7717.4
Total alkaloid (mg/individual) 32.075.0 70.674.5 56.377.7 97.0711.9 10.973.3 19.277.4
Total alkaloid (mg/g dry wt) 0.270 0.3770.03 0.6270.09 0.5670.09 0.1270.03 0.1270.05
Fig. 5. Recovery of creatonotines and callimorphines from sexed
pupae and adults of G. geneura that had received retronecine with their
larval diet. Pupae were preserved within 48 h after begin of pupation.
Within sexes the amounts of creatonotines were signicantly different
between pupae and adults, males P 0:00123, females P o0:0001
(t-test); the respective values of total insect pyrrolizidine alkaloids were
not signicantly different.
general adaptations: (i) recognition of pyrrolizidine
alkaloid-containing plants through phagostimulatory
taste receptor neurons specically dedicated to the
perception of pyrrolizidine alkaloids (Bernays et al.,
2002b); (ii) detoxication of ingested alkaloids by
specic N-oxidation indicating the presence of senecio-
nine N-oxygenase which appears to be present in any
arctiid adapted to pyrrolizidine alkaloids (Lindigkeit et
al., 1997; Naumann et al., 2002); (iii) partial or complete
hydrolysis of the various types of plant-acquired
pyrrolizidine alkaloids and subsequent transformation,
sometimes modication, of the resulting necine bases
into insect-specic alkaloids (Fig. 6).
The specicity of uptake and biochemical processing
of plant acquired pyrrolizidine alkaloids by G. geneura
largely corresponds to the pattern established for E.
acrea but also shows distinctive differences. Macrocyclic
retronecine diesters and triesters (Fig. 1) as those found
in species of the Asteraceae (tribe Senecioneae), the
Fabaceae (Crotalaria) and the Apocynaceae are seques-
tered and transmitted to adults in the same manner as
shown for E. acrea. The same accounts for alkaloids of
the prominent lycopsamine type (Fig. 1) found in
alkaloid-containing species of the Asteraceae (tribe
Eupatorieae), the Boraginaceae and some Apocynaceae.
A difference between the two arctiid species exists in
their ability to epimerize heliotridine, the 7S-epimer of
retronecine. Adults that as larvae had received helio-
tridine contain between about 2050% as insect-specic
heliotridine esters (Table 7) while in E. acrea heliotridine
was always completely epimerized (Hartmann et al.,
2005) yielding exclusively retronecine esters. A simple
explanation for this difference could be that E. acrea
males need an efcient 7S-epimerization for a proper
courtship pheromone biosynthesis since hydroxydanai-
dal has 7R-conguration (Schulz et al., 1993) while this
requirement does not apply for G. geneura.
In contrast to all tested macrocyclic ester alkaloids
and open-chain esters of the lycopsamine type, various
other open-chain esters (i.e., 9-angeloylplatynecine,
sarracine and phalaenopsines) are sequestered by larvae
but only transmitted to the adult life-stage after trans-
esterication into insect-specic pyrrolizidine alkaloids.
In the course of this trans-esterication G. geneura was
shown to convert a major proportion of the trache-
lanthamidine moiety of the orchid alkaloids into its 7-
hydroyl derivative (e.g., turneforcidine moiety). Thus,
the insect is not only able to epimerize the 7-hydroxyl
group but even to introduce it into the molecule. The
mechanism of this hydroxylation awaits elucidation.
Interestingly, E. acrea is not able to catalyze this
reaction, although it utilizes platynecine esters as
pheromone precursors (Hartmann et al., 2005).
ARTICLE IN PRESS
Table 8
Pyrrolizidine alkaloids in the exuviae of eld caught larvae (penultimate instar) of G
Parameter A n 10 B n 2 C n 12 D n 11 E n 9 F n 12
Pyrrolizidine alkaloids
mg / individual Nd 0.4570.15 5.1670.85 1.8470.60 2.6270.87 0.7070.10
mg / g dry weight 0.04370.018 0.53670.078 0.41170.105 0.17770.060 0.06070,010
Individuals with traces of alkaloids 0 0 0 0 2 4
Individuals devoid of alkaloids 10 0 0 0 1 6
Type of alkaloid prole Creatonotines Senecio Senecio Plagiobothrys Plagiobothrys
The eld sites A to F in south-eastern Arizona and date of sampling are: A Santa Rita Mountains, Gardner Canyon (20 March 2002); B Santa
Rita Mountains, Box Canyon (20 March 2002); C Patagonia Mountains, Harshaw Canyon (29 March 2002); D Patagonia Mountains,
Harshaw Road (7 April 2002); E Santa Catalina Mountains, Oracle (3 April 2002); F Rincon, Happy Valley (5 April 2002). The alkaloid
proles of the exuviae indicate larval host-plants, i.e. Senecio S. longilobus and Plagiobothrys P. arizonicus; in one case (B) only creatonotines
were detectable.
Percent of total alkaloids
0 20 40 60 80 100 120
Phalaenopsis sp.
Sarracine/sarracinine
Parsonsia laevigata
Heliotropium indicum
Eupatorium cannabinum
Senecio congestus
Senecio vernalis
Senecio jacobaea
Males
Females 12.5
9.9
2.1
10.4
4.6
4.2
17.4
12.6
5.5
5.8
27
6.8
5
8.5
13.1
30.4
Fig. 6. The percentage of insect pyrrolizidine alkaloids (creatonotines
plus callimorphines) of total pyrrolizidine alkaloids recovered from
adult females and males that as larvae had received various
pyrrolizidine alkaloid mixtures as indicated. Notice: Adults that as
larvae had received sarracine or phalaenopsine contain exclusively
insect alkaloids. The numbers alongside the columns give the
respective absolute amounts (mg) of insect alkaloids.
4.2. Is insect-specic trans-esterication the answer of
polyphagous arctiids to cope with the structural diversity
of plant acquired pyrrolizidine alkaloids?
Both E. acrea and G. geneura are able to specically
esterify a variety of necine bases derived from plant
acquired pyrrolizidine alkaloids. This led to the dis-
covery of at least two classes of insect-made pyrrolizi-
dine alkaloids, the callimorphines and the creatonotines.
The callimorphines contain 2-hydroxy-2-methylbuty-
ric acid as basic necic acid (Fig. 3A). This acid moiety
occurs either free or acetylated (dominating derivative)
or propionylated (Hartmann et al., 2004b). These three
callimorphic acids are only found as the ester moiety of
arctiid-specic pyrrolizidine alkaloids (Fig. 2B). Calli-
morphine, the retronecine-O
9
-ester with the acetylated
callimorphic acid was rst described as pyrrolizidine
alkaloid-metabolite from pupae of Tyria jacobaeae
(Aplin et al., 1968). Later its structure was elucidated
(Edgar et al., 1980) and the biosynthesis from plant-
derived retronecine demonstrated in T. jacobaeae
(Ehmke et al., 1990). Callimorphine has been identied
in a number of arctiids: Arctia caja (Aplin and Roths-
child, 1972), Callimorpha dominula (Edgar et al., 1980),
Gnophaela latipennis (LEmpereur et al., 1989), Hyalur-
ga syma (Trigo et al., 1993) and Creatonotos transiens
(Wink et al., 1988; Hartmann et al., 1990).
The creatonotines, which contain in place of calli-
morphic acids 2-hydroxy-3-methylbutanoic acid (creato-
notine A) or 2-hydroxy-3-methylpentanoic acid
(creatonotine B, the major compound) (Fig. 3A), were
rst identied as insect alkaloids in C. transiens adults
that with their larval diet had received retronecine or a
plant-derived pyrrolizidine alkaloid mixture (Hartmann
et al., 1990). Creatonotine A and B are usually
accompanied by their O
7
-esters (isocreatonotines) (Fig.
2A). In E. acrea exclusively creatonotines are found
(Hartmann et al., 2004b; 2005); in C. transiens they are
accompanied by trace amounts of callimorphine (Hart-
mann et al., 1990). In both species creatonotines are
considered direct pheromone precursors (Schulz et al.,
1993; Hartmann et al., 2003a). In E. acrea it has been
demonstrated that all plant-acquired pyrrolizidine alka-
loids that after hydrolysis yield retronecine or platynecine
are pheromone precursors (Hartmann et al., 2005). The
same is true for heliotridine esters after C-7 epimeriza-
tion. In any case esterication with creatonotic acids
appears to be the committed step. Pheromone formation
in males occurs at the expense of previously synthesized
creatonotines (Hartmann et al., 2003a; 2004a). G.
geneura, not known to produce hydroxydanaidal, synthe-
sizes creatonotines from retronecine like E. acrea.
However, in E. acrea the creatonotines are already
synthesized in the larval stage (Hartmann et al., 2004a)
while in G. geneura they are rst observed in the pupal
stage (Table 7) (Hartmann et al., 2004b). The most
intriguing difference between the two arctiid species is
that G. geneura transforms most of its creatonotines into
callimorphines during transition from the pupal to the
adult stage (Table 7, Fig. 5). In T. jacobaeae,which does
not form creatonotines, callimorphine is not detectable
before the pupal stage (Aplin et al., 1968; Aplin and
Rothschild, 1972). Its biosynthesis appears to be re-
stricted to the very early stages of pupation. Callimor-
phine is rst detectable in pre-pupae (Ehmke et al., 1990).
Since in G. geneura the creatonotines are found in young
pupae but not larvae, we assume that they are synthesized
at the early stages of pupation, like the callimorphines in
T. jacobaeae. The conversion of the creatonotines into the
callimorphines, the major insect alkaloids in adults, by
trans-esterication most likely occurs just before eclosion,
but this needs to be conrmed.
The present study together with the results of previous
work with E. acrea (Hartmann et al., 2003a; 2004b;
2005) provides the rst evidence on the functional
importance of the insect-specic pyrrolizidine alkaloids.
Both arctiid species sequester as larvae all kinds of plant
pyrrolizidine alkaloids. Apparently only macrocyclic
pyrrolizidine alkaloids and open-chain esters of the
lycopsamine type are maintained through all life-stages,
while other pyrrolizidine alkaloids need insect-specic
trans-esterication before transfer to the pupal and
adult stages (see 4.1.). With the exception of the
otonecine derivatives all tested classes of pyrrolizidine
alkaloids are subjected to partial or total trans-
esterication (Fig. 6). Thus, the insect-specic trans-
esterication provides a means to recover and salvage all
kinds of necine bases from plant acquired pyrrolizidine
alkaloids, especially those that cannot be transmitted to
later life-stages. Moreover, in E. acrea insect-specic
trans-esterication is the essential step to create creato-
notines as common precursor for the formation of the
male pyrrolizidine alkaloid-signal hydroxydanaidal
from all kinds of sequestered pyrrolizidine alkaloids
including retronecine, heliotridine and platynecine esters
(Hartmann et al., 2003a; 2004a, b; 2005).
The insect-specic creatonotines and callimorphines
appear to represent the only necine monoesters that, in
addition to the plant acquired pyrrolizidine alkaloids of
the lycopsamine type, are maintained through all life-
stages. This implies that the insect-made necic acids of
these alkaloids have common structural features allowing
their stable maintenance and transmission between life-
stages. Indeed, common structural features between plant
monoesters of the lycopsamine type and the insect-specic
monoesters exist (Fig. 7): (i) they all represent aliphatic
branched-chain 2-hydroxy acids; (ii) in the callimorphines
and the plant necic acids this hydroxyl groups is tertiary
hydroxyl (Fig. 7B); (iii) the branching of the carbon
skeletons of all three types of necic acids display
similarities. The most conspicuous difference between the
plant-specic and the insect-specic necic acids is the
ARTICLE IN PRESS
second hydroxyl group (at the 3-carbon) in the plant
acids. The stereochemistry of the 3-hydroxyl appears to be
important in plant-acquired alkaloids since in G. geneura
adults only (3S)-congurated monoesters are maintained,
i.e. lycopsamine, echinatine and indicine (Fig. 1B).
Rinderine the major pyrrolizidine alkaloid in E. cannabi-
num has (3R), (7S)-conguration. It is epimerized in both
positions yielding lycopsamine (Fig. 1B). Since the
inversion of conguration at C-7 is not total in G. geneura
(Table 7), echinatine accumulates in addition to lycopsa-
mine (Table 4). Epimerization of (3R)- and (7S)-
congurated alkaloids of the lycopsamine type in arctiids
is not unique. It has also been demonstrated in ithomiine
butteries, which as adults imbibe pyrrolizidine alkaloids
of the lycopsamine type mainly from Eupatorium and
Heliotropium species (Trigo et al., 1996). Although the
butteries sequester all kinds of lycopsamine stereoisomers
(see Fig. 1B) they maintain almost exclusively lycopsa-
mine. The reason for this is their ability to efciently
epimerize (3R)- and (7S)-congurated alkaloids (Trigo et
al., 1994). Even leaf-beetles of the neotropical genus
Platyphora, which are specialized on pyrrolizidine alka-
loids of the lycopsamine type, were found to convert
rinderine into intermedine and lycopsamine (Hartmann et
al., 2001). A pyrrolizidine alkaloid-sequestering Platyphora
clade radiated on single species of the three plant families,
Asteraceae tribe Eupatorieae, Apocynaceae and Boragi-
naceae (Termonia et al., 2002), which represent the only
families with species that contain pyrrolizidine alkaloids of
the lycopsamine type (Hartmann and Witte, 1995). Six
Platyphora species sequester pyrrolizidine alkaloids of the
lycopsamine type and concentrate them in the secretions of
their exocrine defense glands and all synthesize creatono-
tine A and few related mono and O
9
,O
7
-diesters with
insect-specic 2-hydroxy acids, e.g. lactic acid (Hartmann
et al., 2001; 2003b). The common pressure to invent a
necic acid that most properly meets the structural demands
of the necic acids of alkaloids of the lycopsamine type, to
which both arctiids and leaf-beetles are adapted, could be
the explanation for this intriguing biochemical conver-
gence. These mimics allow adapted insects to attain,
transmit and recycle necine bases from all kinds of
otherwise lost plant pyrrolizidine alkaloids. More experi-
mental evidence is needed to evaluate this general
hypothesis. Particularly, a complete elucidation of the
stereochemistry of the insect-made necic acids is required
for a precise structure-function comparison between plant
and insect necic acids. Moreover, additional feeding
experiments are needed to corroborate the assumed role
of the insect alkaloids.
4.3. Ecological aspects
As discussed above, G. geneura appears well adapted to
encounter and exploit any plant containing pyrrolizidine
alkaloids. Like E. acrea, Grammia larvae exploit alkaloid
plants primarily as a source for obtaining their chemical
defenses rather than for their use as foodmost feeding
generally occurs on plants without pyrrolizidine alka-
loids. Previous work shows that G. geneura larvae gain
resistance to parasitoids by eating a diet dominated by
the alkaloid-containing Senecio longilobus (Singer et al.,
2004a). This anti-parasitoid resistance was positively
associated with the concentration of sequestered pyrro-
lizidine alkaloids (Singer et al., 2004a). However, the
defensive benet of a diet dominated by Senecio comes at
the cost of reduced larval growth efciency (Singer et al.,
2004a). This same trade-off is demonstrated more clearly
in similar experiments with E. acrea (Singer et al., 2004b),
for which pyrrolizidine alkaloids themselves do not
appear to reduce larval performance (Hartmann et al.,
2005). We therefore suspect that G. geneura performance
is not negatively affected by the pyrrolizidine alkaloids,
but by other characteristics of Senecio. If true, this would
echo the nding in E. acrea that these caterpillars are
adapted to use pyrrolizidine alkaloid plants more as a
source of drugs than of high quality food.
Pyrrolizidine alkaloid-containing plants, such as
Senecio, Crotalaria, and Plagiobothrys, may be relatively
uncommon in the habitat (Singer and Stireman, 2001).
As such, G. geneura caterpillars were expected to vary in
the type and concentration of pyrrolizidine alkaloids
obtained from host plants. Indeed, this expectation was
supported in the present study by the analysis of exuviae
from eld-collected larvae (Table 8). Little can be said
about the possible role of pyrrolizidine alkaloids in G.
geneura courtship because nothing is known about the
mating behavior of this species. However, due to the
ARTICLE IN PRESS
Lycopsamine Type Creatonotine A Creatonotine B
N
O
H
O
OH
HO
OH
N
O
H
O
OH
HO
N
O
H
O
OH
HO
N
O
H
O
HO
O
O
Homocallimorphine
N
O
H
O
OH
HO
Deacetylcallimorphine
N
O
H
O
HO
O
O
Callimorphine
N
O
H
O
OH
HO
OH
Lycopsamine Type
(A)
(B)
Fig. 7. Structural similarity between the necic acid moiety of
pyrrolizidine alkaloids of the lycopsamine type and the insect-made
necic acids of the creatonotines and callimorphines. Structural
congruence is given in (red). The stereochemistry of the necic acids is
not given since it is still unknown for the callimorphines and needs to
be conrmed for the creatonotines.
uncertainty of acquiring pyrrolizidine alkaloids during
the larval stage, we expect the alkaloids to be transferred
from males to females during mating and incorporated
into eggs of the offspring as in E. acrea (Hartmann et
al., 2004a). This adult transfer of alkaloids allows a
female to gain pyrrolizidine alkaloids even if she did not
acquire them as a larva.
The present study suggests that a wide variety of
structural types of pyrrolizidine alkaloids are likely to be
functional in the ecological contexts described above.
1,2-Dihydropyrrolizidine alkaloids are assumed to be
non-toxic, nevertheless they are sequestered and main-
tained by G. geneura either per se (e.g., platyphylline) or
after insect-specic trans-esterication (e.g., sarracine).
E. acrea converts (aromatizes) the platynecine moiety to
hydroxydanaidal, whereas G. geneura even creates the
platynecine isomer turneforcidine (Fig. 3) by 7-hydro-
xylation. Obviously even the so-called non-toxic pyrro-
lizidine alkaloids are valuable for both insects. If we
speak of toxic pyrrolizidine alkaloids we restrict toxicity
to metabolic bioactivation of 1,2-unsaturated pyrrolizi-
dine alkaloids resulting in pyrrolic intermediates re-
sponsible for the well known cell toxicity, mutagenicity
and genotoxicity of pyrrolizidine alkaloids for verte-
brates and insects (Mattocks, 1986; Frei et al., 1992; Fu
et al., 2004; Hartmann et al., 2005). Probably pyrroli-
zidine alkaloids with 1,2-saturated necine bases possess
still unknown biological activities which are advanta-
geous for sequestering insects. There is only one report
indicating deterrent properties of 1,2-saturated pyrroli-
zidine alkaloids (Reina et al., 1997). In this context it is
important to recall that there are plant taxa, like
pyrrolizidine alkaloid-containing orchids (Hartmann
and Witte, 1995) or pyrrolizidine alkaloid-containing
Ipomoea species (Convolvulaceae) (Jenett-Siems et al.,
1998), that produce exclusively esters of 1,2-saturated
necine bases. If pyrrolizidine alkaloid-adapted larvae
sequester these pyrrolizidine alkaloids and specically
convert them into insect-specic pyrrolizidine alkaloids
by trans-esterication that can be maintained and
transmitted to all life-stages (see 4.2.) a functional
importance of these pyrrolizidine alkaloids is likely.
Acknowledgements
This work was supported by grants of the Deutsche
Forschungsgemeinschaft and Fonds der Chemischen
Industrie to T.H., and by the Center for Insect Science
(U. Arizona) through NIH Training Grant # 1 K12
Gm00708.
References
Aplin, R.T., Rothschild, M., 1972. Poisonous alkaloids in the body
tissue of the garden tiger moth (Arctia caja L.) (Lepidoptera) and
the cinnabar moth (Tyria jacobaeae L.). In: De Vries, A., Kochva,
E. (Eds.), Toxins of animal and plant origin. Gordon & Breach
Science Publication, New York, pp. 579595.
Aplin, R.T., Benn, M.H., Rothschild, M, 1968. Poisonos alkaloids in
the body tissues of the cinnabar moth (Callimorpha jacobaeae L).
Nature 219, 747748.
Bernays, E.A., Chapman, R.F., Hartmann, T, 2002a. A highly
sensitive taste receptor cell for pyrrolizidine alkaloids in the lateral
galeal sensillum of a polyphagous caterpillar, Estigmene acrea. J.
Comp. Physiol. A 188, 715723.
Bernays, E.A., Chapman, R.F., Hartmann, T., 2002b. A taste receptor
neuron dedicated to the perception of pyrrolizidine alkaloids in the
medial galeal sensillum of two polyphagous arctiid caterpillars.
Physiol. Entomol. 27, 110.
Boppre , M., 1986. Insects pharmacophagously utilizing defensive plant
chemicals (pyrrolizidine alkaloids). Naturwissenschaften 73, 1726.
Conner, W.E., Weller, S.J., 2004. A quest for alkaloids: the curious
relationship between tiger moths and plants containing pyrrolizi-
dine alkaloids. In: Carde , R.T., Millar, J.G. (Eds.), Advances in
Insect Chemical Ecology. University Press, Cambridge, pp.
248282.
Conner, W.E., Roach, B., Benedict, E., Meinwald, J., Eisner, T., 1990.
Courtship pheromone production and body size as correlates of
larval diet in males of the arctiid moth Utetheisa ornatrix. J. Chem.
Ecol. 16 (2), 543552.
Davenport, J.W., Conner, W.E., 2003. Dietary alkaloids and the
development of androconial organs in Estigmene acrea. J. Insect
Sci. Tucson 3:3, available online: insectscience.org/3.3.
Dussourd, D.E., Harvis, C.A., Meinwald, J., Eisner, T., 1991.
Pheromonal Advertisement of a Nuptial gift by a male moth
Utetheisa ornatrix. Proc. Natl. Acad. Sci. USA 88, 92249227.
Dussourd, D.E., Ubik, K., Harvis, C., Resch, J., Meinwald, J., Eisner,
T., 1988. Biparental defensive endowment of eggs with acquired
plant alkaloid in the moth Utetheisa ornatrix. Proc. Natl. Acad.
Sci. USA 85, 59925996.
Edgar, J.A., Culvenor, C.C.J., Cockrum, P.A., Smith, L.W, 1980.
Callimorphine: identication and synthesis of the cinnabar moth
metabolite. Tetrahedron Lett 21, 13831384.
Ehmke, A., Witte, L., Biller, A., Hartmann, T, 1990. Sequestration, N-
oxidation and transformation of plant pyrrolizidine alkaloids by
the arctiid moth Tyria jacobaeae L. Z. Naturforsch. 45c,
11851192.
Eisner, T., Rossini, C., Gonzalez, A., Iyengar, V.K., Siegler, M.V.S.,
Smedley, S.R., 2002. Paternal investment in egg defence. In: Hilker,
M., Meiners, T. (Eds.), Chemoecology of Insect Eggs and Egg
Deposition. Blackwell Publishing, Oxford, pp. 91116.
Frei, H., Lu thy, J., Bra uchli, J., Zweifel, U., Wurgler, F.E., Schlatter,
C, 1992. Structure/activity relationships of the genotoxic potencies
of sixteen pyrrolizidine alkaloids assayed for the induction of
somatic mutation and recombination in wing cells of Drosophila
melanogaster. Chem. Biol. Interact. 83, 122.
Fu, P.P., Xia, Q., Lin, G., Chou, M.W., 2004. Pyrrolizidine
alkaloidsgenotoxicity, metabolism enzymes, metabolic activa-
tion, and mechanisms. Drug Metab. Rev. 36, 155.
Hartmann, T, 1999. Chemical ecology of pyrrolizidine alkaloids.
Planta 207, 483495.
Hartmann, T., Ober, D., 2000. Biosynthesis and metabolism of
pyrrolizidine alkaloids in plants and specialized insect herbivores.
In: Leeper, F.J., Vederas, J.C. (Eds.), Topics in Current Chemistry:
BiosynthesisAromatic Polyketides, Isoprenoids, Alkaloids.
Springer, Berlin, pp. 207244.
Hartmann, T., Witte, L., 1995. Pyrrolizidine alkaloids: chemical,
biological and chemoecological aspects. In: Pelletier, S.W. (Ed.),
Alkaloids: Chemical and Biological Perspectives. Pergamon Press,
Oxford, pp. 155233.
ARTICLE IN PRESS
Hartmann, T., Zimmer, M., 1986. Organ-specic distribution and
accumulation of pyrrolizidine alkaloids during the life history of
two annual Senecio species. J. Plant Physiol. 122, 6780.
Hartmann, T., Biller, A., Witte, L., Ernst, L., Boppre, M., 1990.
Transformation of plant pyrrolizidine alkaloids into novel insect
alkaloids by arctiid moths (Lepidoptera). Biochem. Syst. Ecol. 18,
549554.
Hartmann, T., Theuring, C., Witte, L., Pasteels, J.M., 2001.
Sequestration, metabolism and partial synthesis of tertiary
pyrrolizidine alkaloids by the neotropical leaf-beetle Platyphora
boucardi. Insect Biochem. Mol. Biol. 31, 10411056.
Hartmann, T., Theuring, C., Bernays, E.A., 2003a. Are insect-
synthesized retronecine esters (creatonotines) the precursors of
the male courtship pheromone in the arctiid moth Estigmene acrea?
J. Chem. Ecol. 29, 26032608.
Hartmann, T., Theuring, C., Witte, L., Schulz, S., Pasteels, J.M.,
2003b. Biochemical processing of plant acquired pyrrolizidine
alkaloids by the neotropical leaf-beetle Platyphora boucardi. Insect
Hartmann, T., Theuring, C., Beuerle, T., Bernays, E.A., 2004a.
Phenological fate of plant-acquired pyrrolizidine alkaloids in the
polyphagous arctiid Estigmene acrea. Chemoecology 14, 207216.
Hartmann, T., Theuring, C., Beuerle, T., Ernst, L., Singer, M.S.,
Bernays, E.A., 2004b. Acquired and partially de novo synthesized
pyrrolizidine alkaloids in two polyphagous arctiids and the
alkaloid proles of their larval food-plants. J. Chem. Ecol. 30,
229254.
Hartmann, T., Theuring, C., Beuerle, T., Klewer, N., Schulz, S.,
Singer, M.S., Bernays, E.A., 2005. Specic recognition, detoxica-
tion and metabolism of pyrrolizidine alkaloids by the polyphagous
arctiid Estigmene acrea. Insect Biochem. Mol. Biol. 35, 391411.
Iyengar, V.K., Rossini, C., Eisner, T., 2001. Precopulatory assessment
of male quality in an arctiid moth (Utetheisa ornatrix): hydro-
xydanaidal is the only criterion of choice. Behav. Ecol. Sociobiol.
49, 283288.
Jenett-Siems, K., Schimming, T., Kaloga, M., Eich, E., Siems, K.,
Gupta, M.P., Witte, L., Hartmann, T., 1998. Pyrrolizidine
alkaloids of Ipomoea hederifolia and related species. Phytochem-
istry 47, 15511560.
Jordan, A.T., Jones, T.H., Conner, W.E., 2005. If youve got it, aunt
it: Ingested alkaloids affect corematal display behavior in the salt
marsh moth, Estigmene acrea. J. Insect Sci. 5:1, available online:
insectscience.org/5.1.
LEmpereur, K.M., Li, Y., Stermitz, F.R., 1989. Pyrrolizidine
alkaloids from Hackelia californica and Gnophaela latipennis, an
H. californica-hosted arctiid moth. J. Nat. Prod. 54, 360366.
Lindigkeit, R., Biller, A., Buch, M., Schiebel, H.M., Boppre , M.,
Hartmann, T., 1997. The two faces of pyrrolizidine alkaloids: the
role of the tertiary amine and its N-oxide in chemical defense of
insects with acquired plant alkaloids. Eur. J. Biochem. 245,
626636.
Mattocks, A.R., 1986. Chemistry and Toxicology of Pyrrolizidine
Alkaloids. Academic Press, London.
Naumann, C., Hartmann, T., Ober, D., 2002. Evolutionary recruit-
ment of a avin-dependent monooxygenase for the detoxication
of host plant-acquired pyrrolizidine alkaloids in the alkaloid-
defended arctiid moth Tyria jacobaeae. Proc. Natl. Acad. Sci. USA
99, 60856090.
Reina, M., Gonzalez, C.A., Gutierrez, C., Cabrera, R., Henriquez, J.,
Villarroel, L., 1997. Bioactive saturated pyrrolizidine alkaloids
from Heliotropium oridum. Phytochemistry Oxford 46, 845853.
Schneider, D., 1987. The strange fate of pyrrolizidine alkaloids. In:
Chapman, R.F., Bernays, E.A., Stoffolano, J.G. (Eds.), Perspec-
tives in chemoreception and behavior. Springer, pp. 123142.
Schulz, S., Francke, W., Boppre , M., Eisner, T., Meinwald, J., 1993.
Insect pheromone biosynthesis: stereochemical pathway of hydro-
xydanaidal production from alkaloidal precursors in Creatonotos
transiens (Lepidoptera, Arctiidae). Proc. Natl. Acad. Sci. USA 90,
68346838.
Singer, M.S., Stireman III, J.O., 2001. How foraging tactics determine
host-plant use by a polyphagous caterpillar. Oecologia 129, 98105.
Singer, M.S., Carriere, Y., Theuring, C., Hartmann, T., 2004a.
Disentangling food quality from resistance against parasitoids:
diet choice by a generalist caterpillar. Am. Nat. 164, 423429.
Singer, M.S., Rodrigues, D., Stireman III, J.O., Carriere, Y., 2004b.
Roles of food quality and enemy-free space in host use by a
generalist insect herbivore. Ecology 85, 27472753.
Termonia, A., Pasteels, J.M., Windsor, D.M., Milinkovitch, M.C,
2002. Dual chemical sequestration: a key mechanism in transitions
among ecological specialization. Proc. R. Soc. London B 269, 16.
Trigo, J.R., Witte, L., Brown Jr., K.S., Hartmann, T., Barata, L.E.S.,
1993. Pyrrolizidine alkaloids in the arctiid moth Hyalurga syma. J.
Chem. Ecol. 19, 669679.
Trigo, J.R., Barata, L.E.S., Brown Jr., K.S., 1994. Stereochemical
inversion of pyrrolizidine alkaloids by Mechanitis polymnia
(Lepidoptera: Nymphalidae: Ithomiinae): Specicity and evolu-
tionary signicance. J. Chem. Ecol. 20, 28832899.
Trigo, J.R., Brown, K.S., Henriques, S.A., Barata, L.E.S., 1996.
Qualitative patterns of pyrrolizidine alkaloids in ithomiine
butteries. Biochem. Syst. Ecol. 24, 181188.
Weller, S.J., Jacobson, N.L., Conner, W.E., 1999. The evolution of
chemical defences and mating systems in tiger moths (Lepidoptera:
Arctiidae). Bot. J. Linn. Soc. 68, 557578.
Wink, M., Schneider, D., Witte, L., 1988. Biosynthesis of pyrrolizidine
alkaloid-derived pheromones in the arctiid moth, Creatonotos
transiens: stereochemical conversion of heliotrine. Z. Naturforsch.
43c, 737741.
Witte, L., Ehmke, A., Hartmann, T., 1990. Interspecic ow of
pyrrolizidine alkaloids; from plants via aphids to ladybirds.
Naturwissenschaften 77, 540543.
Witte, L., Rubiolo, P., Bicchi, C., Hartmann, T., 1993. Comparative
analysis of pyrrolizidine alkaloids from natural sources by gas
chromatography-mass spectrometry. Phytochemistry 32, 187196.
Yamamoto, R.T., 1969. Mass rearing of the tobacco hornworm II.
Larval rearing and pupation. J. Econ. Entomol. 62, 14271431.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Molecular characterization and evolution of pheromone binding
protein genes in Agrotis moths
David Abraham, Christer Lo fstedt, Jean-Franc- ois Picimbon
Department of Ecology, Lund University, Ecology building, SE-22362, Sweden

Received 3 May 2005; accepted 6 May 2005
Abstract
Pheromone-binding proteins (PBPs) are soluble transporter proteins that increase the capture and the solubilization of
pheromone molecules in the lymph surrounding the olfactory receptors. A polymerase chain reaction-based method was used to
identify PBP genes in Agrotis species for an evolutionary genomic study of noctuid moth PBPs. From genomic DNA we determined
the structure of different PBP genes in the two closely related species, Agrotis ipsilon and A. segetum. In all, we clearly identied four
genes (Aips-1, Aips-2, Aseg-1 and Aseg-2) that represent two distinct PBP orthology groups. We found that the four genes have the
same exonintron structure and that they comprise three exons and two introns but differ in length mainly in the second intron. The
three exons of Aseg-2 and Aips-2 have the same lengths but both intron 1 and intron 2 differ in length between the genes. In contrast,
Aips-1 and Aseg-1 show dissimilarity only in the length of intron 2. Interestingly, introns 1 and 2 are inserted in the same positions in
the Aips-1, Aips-2, Aseg-1 and Aseg-2 genes. These ndings show that the Agrotis PBP genes have common ancestry and probably
originate from gene duplication before the speciation of ipsilon and segetum. We found that expression of Aips-1/Aseg-1 and Aips-2/
Aseg-2 is antennal-specic, but expression is not restricted to the male antennae.
Keywords: Olfaction; Noctuid; Agrotis ipsilon; A. segetum; PBP; Ancestry; Duplication
1. Introduction
In moths, the sense of olfaction and the use of volatile
pheromones to nd a suitable mate are crucial for
reproduction. Pheromones are detected by specic
olfactory receptor neurons located in specialized sensory
hairs or sensilla on the antenna. The proteinaceous
lymph surrounds the dendrites and separates the
pheromone receptors from the surface pores through
which the pheromone molecules penetrate the sensillae.
Moth pheromone compounds are largely hydrophobic.
The possibility that a pheromone molecule directly
enters the sensillum and quickly reaches the receptor
neuron is therefore very small; a solubilization process is
required. A class of small soluble proteins (1415 kDa) is
expected to mediate the solubilization of pheromone
molecules from the sensillar surface and their transport
to the receptor neurons. These proteins are named
pheromone-binding proteins (PBPs) (Vogt and Riddi-
ford, 1981; Vogt, 1987).
Moth PBPs studied so far have 3292% sequence
identity in the different species. The variability in PBPs
could be related to their function as carriers of
pheromone components (for reviews, see Leal, 2003;
Picimbon, 2003; Vogt, 2003). The PBPs from moths are
expressed in high concentrations in the lymph of
pheromone-sensitive sensilla (Steinbrecht et al., 1992,
1995; Maida et al., 2000; Zhang et al., 2001; Nardi et al.,
2003). Different types of PBPs coexist in several
Lepidopteran species including the noctuids and more
particularly the black cutworm moth, Agrotis ipsilon
(Picimbon and Gadenne, 2002). Only one PBP (Aseg-1)
has been reported in the closely related Agrotis species,
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.002
Corresponding author. Tel.: +46 46 222 9344;

fax: +46 46 222 4716.
E-mail address: jean-francois.picimbon@ekol.lu.se
(J.-F. Picimbon).
A. segetum (Prestwich et al., 1995; LaForest et al., 1999).
These two species use complex pheromone blends con-
sisting of at least (Z)-7-dodecenyl/(Z)-9-tetradecenyl/
(Z)-11-hexadecenyl acetates (Z7-12:OAc/Z9-14:OAc/
Z11-16:OAc) and (Z)-5-decenyl/(Z)-5-dodecenyl/(Z)-7-
dodecenyl/(Z)-9-tetradecenyl acetates (Z5-10:OAc/ Z5-
12:OAc/Z7-12:OAc/Z9-14:OAc), respectively (Lo fstedt
et al., 1985; Wu et al., 1995; Picimbon et al., 1997;
Gemeno and Haynes, 1998). In a recent study addres-
sing PBP variations in moths, a specic grouping of
PBPs has been proposed and two orthology groups of
noctuid PBPs, Grp1 and Grp2, have been documented.
The Grp1 and Grp2 PBPs from A. ipsilon, Aips-1 and
Aips-2, show about 45% identity between them and
each protein is characterized by specic amino acid
motifs. This may underlie specic functions or ligand
interactions (Picimbon and Gadenne, 2002).
We report here about the comparisons and evolution
of a pair of orthologous PBPs from the two Agrotis
species, A. ipsilon and A. segetum. We report the cloning
of a second PBP in A. segetum, Aseg-2. In addition, we
report the isolation of genomic DNA corresponding to
the Aips-1, Aips-2 and Aseg-2 genes (the Aseg-1
genomic region was reported previously by LaForest
et al., 1999). We have found similar nucleic acid identity
of exonic and intronic DNA sequences, in particular
that the Agrotis PBP genes differ in length but have the
same gene structure comprised of three exons and two
introns. Variability is found for intron 2 while intron 1 is
strictly conserved. An analysis of their deduced intro-
nexon structures reveals that in all four genes, each of
the two introns is inserted at an identical position,
suggesting a common PBP ancestor gene. We show a
phylogenetic analysis that supports the grouping of
noctuid PBPs into two orthology groups, Grp1 and
Grp2, and suggests that the PBP-1 and PBP-2 split
occurred before the diversication of Agrotis species.
We also show that after the two Agrotis species
diverged, each gene started specic accumulations of
mutations in the introns. Finally, we note that Aips-1/
Aseg-1 and Aseg-2, but not Aips-2, are expressed more
abundantly in male than female antennae.
2.1. Insect rearing and tissue collection
Antennae were collected from 17 days old adults of
the two Agrotis species reared in the laboratory at 23 1C,
70% r.h. and L17:D7 photoperiod. Some individuals
were kept frozen for extraction of genomic DNA. The
antennae and legs of 100 males and females were cut at
their bases and immediately frozen. Tissues were stored
at 80 1C until RNA or genomic DNA extractions.
2.2. Identication of PBP-encoding cDNA
The total RNAs were extracted using the TrizolTM
method (Life Technologies) in order to clone cDNAs
encoding PBP from Agrotis species. First strand cDNAs
were synthesized in vitro from adult antennal RNA
extracts according to the protocol detailed in Picimbon
et al. (2000). Antennal cDNAs served as templates in a
polymerase chain reaction (PCR)-based approach em-
ploying the Gene Amp PCR system 9600 (AB Applied
Biosystems). PCR experiments used sense degenerated
primers corresponding to regions highly conserved
across a selection of PBP sequenced to date in
combination with an oligo(dT) or anti-sense PBP-
specic primer. To isolate Agrotis PBP cDNAs, we
designed two classes of degenerated oligonucleotide
primers aimed at specic motifs for the PBPs: the pair of
primers Group1-1s (5
0
-TCGCARGAWRTYATKAAG-
3
0
, tuned to SQEIMK) and Group1-1as (5
0
-RACT
TCRGCCAAGACTTC-3
0
, tuned to EVLAEV), and
the pair of primers Group2-1s (5
0
-ATGACITCICA-
GRAGGTG-3
0
, tuned to MTSQE) and Group2-1as (5
0
-
NACNRVIGTCATDATYTC-3, tuned to EIMTAV).
Touchdown PCR reactions were performed as follows;
94 1C for 5 min, then 20 cycles of 94 1C for 30 s, 50 1C for
30 s and 72 1C for 1 min. During every cycle the
annealing temperature was lowered with 0.5 1C. These
20 cycles were followed by 15 cycles of 94 1C for 30 s,
40 1C for 30 s and 72 1C for 1 min followed by and
extension step of 72 1C for 5 min. Full-length clones
were obtained using the SMART technology (Clontech)
or complementary PCRs employing nested primers
Group2-1s and PBP2as (5
0
-TCYTTCCARAARTTR-
TA-3
0
encoding YNFWD/E). In all the different
experiments, the resulting PCR products were then gel
puried and prepared for cloning using the protocol
from geneclean (Qbiogene, Inc.). The GENECLEAN
products were ligated into the pGEM-Zf(+) plasmid
prepared by digestion with EcoRV and 3
0
terminal
thymidines added (pGEM-T Easy vector system I,
Promega). Recombinant plasmids were subjected to
sequencing using the BigDyeTM Terminator Cycle
Sequencing Ready Reaction mixture with Ampli-
TaqRDNA polymerase (ABI PRISM) and forward or
reverse pUC/M13 primers. Both sense and anti-sense
strands were sequenced using an ABI 310 or an ABI
3100 automated sequencer according to the manufac-
turers instructions (Perkin Elmer). Sequences were
examined in Sequence Navigator (ABI PRISM) or
BioEdit version 5.0.9 (Hall, 2001).
2.3. Reverse transcriptase polymerase chain reactions
(RT PCR)
Total RNA was extracted from the antennae and legs
of Agrotis males and females by following the TrizolTM
ARTICLE IN PRESS
D. Abraham et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11001111 1101
protocol (Life Technologies). Concentration and purity
of RNA (10 mg) were assessed by spectrophotometry
and electrophoresis on agarose/formaldehyde gels with
ethidium bromide staining. From each RNA sample,
500 ng were reverse transcribed using the Advantage R
One-Step RTPCR Kit and the suppliers recommenda-
tions (Clontech). The whole cDNA product was used as
a template for PCR. Reverse transcription and PCR
were carried out in the automated program: 50 1C for
1 h, 94 1C for 5 min, then 30 cycles of 94 1C for 30 s,
55 1C for 30 s and 68 1C for 1 min followed by 68 1C for
2 min. Control RTPCR experiments were performed
using specic Actin primers (Isopteran Actin-1s
CTCAGGCGATGGTGTCTC; Actin-1as GGGGTA
CATGGTGGTGC). Specic primer pairs were chosen
to amplify only Aips-1, Aips-2, Aseg-1 or Aseg-2
(Table 1). One-step RTPCR reactions were performed
as follows: 50 1C for 1 h, 94 1C for 5 min, then 30 cycles
of 94 1C for 30 s, 65 1C, 60 1C or 52 1C for 30 s and 68 1C
for 1 min followed by 68 1C for 2 min. Identication of
RTPCR products were obtained by direct sequencing.
2.4. Isolation and analysis of genomic DNA sequences
For genomic DNA extraction, single individuals from
A. ipsilon or A. segetum species were frozen and crushed
in liquid nitrogen. Genomic DNA was then extracted
according to Abraham et al. (2001). To identify the
genes encoding Aips-1, Aips-2 and Aseg-2, we used a
PCR approach employing specic oligonucleotide pri-
mers that matched the cDNA sequences or that are
based on the alignment of known PBP nucleotide
sequences (Table 2). Touchdown PCR reactions were
performed as described earlier. From the agarose gel
analyzing the PCR products, the most prominent bands
were excised from the gels and cleaned with geneclean
(Qbiogene, Inc.) according to the protocol. The pure
gene products were cloned and sequenced as described
in Picimbon et al. (2000).
2.5. Phylogenetic analysis of PBP sequences
The cDNA and genomic DNA sequences were
aligned using CLUSTAL1.8. The alignment was
subsequently checked manually and converted to the
Nexus format (Thompson et al., 1997). Maximum
parsimony was used to build strict consensus trees
(commands: Heuristic, TBR, random, nreps 1) in
PAUP4*d65 (Swofford, 1999). Branch support was
assessed by bootstrapping as implemented in
PAUP4*d65 (commands: random, TBR, MULPARS
and 1000 bootstraps; Felsenstein, 1985). In all trees,
only bootstrap values over 50% are shown (Figs. 2
and 5).
For the maximum parsimony trees of the PBPs of
moth species (amino acid tree length 891, CI 0.530, RI
0.681, RC 0.361, HI 0.470, cDNA tree length 2280, CI
0.328, RI 0.458, RC 0.150, HI 0.672), Lymantria PBPs
Ldis-1 and Ldis-2 were used as outgroup. For phyloge-
netic analysis, the amino acid sequences of all PBPs
reported so far were used (Vogt, 2003 and references
therein).
For the parsimony tree of intron (Aseg-1, Aseg-2,
Aips-1, Aips-2; Tree Length 642, CI 0.882, RC 0.618)
and gene sequences (Aseg-1, Aseg-2, Aips-1, Aips-2,
OnubPBP and Aper-1; Tree Length 3339, CI 0.818, RC
0.434), different alignments and outgroups had to be
tested (see Fig. 5). In the alignment for the gene analysis,
the rst 200 nucleotides in the alignment of intron 2
consisted only of Aips-1 and Aseg-1 intron 2. This could
mean that the other introns are shorter due to a 200
bases loss. The rst 200 bp of the intron 2 sequence
found only in the Aips-1 and Aseg-1 genes do not
represent a transposable element. In the rest of the
ARTICLE IN PRESS
Table 1
RTPCR primers for amplication of Agrotis PBP from total RNA
Gene name Acc. num. Primer name Sequence 5
0
3
0
Target region
Aips-1 AY301985 AipsPBP1s TCGCAGGAAATCATGAAGAAT SQEIMKN
AipsPBP1as AACTTCGGCCAAGACTTCGCC GEVLAEV
PBPGroup1s WSGCARGADRTHATKAA SQEIMK
PBP1as ACYTCIGCIARIACYTCICC GEVLAEV
Oligo(dT) GAGTTTTTTTTTTTTTTTTT Poly A
Aips-2 AY301986 AipsPBP2s TCGCAGGAGGTGGTCGCCAGC SQEVVSP
AipsPBP2as AACGGCCGTCATGATCTCCTCCAT MEEIMTAV
PBPGroup2s TCKMARGAVITIVTIRM SQEVVS
PBP2as GCIGTCATDATYTCYTCCAT MEEIMTAV
Oligo(dT) GAGTTTTTTTTTTTTTTTTT Poly A
Aseg-1 AF134253 AsegPBP1s TCGCAGGAAATCATGAA SQEIMK
AsegPBP1as AACTTCGGCCAAGACTTCG GEVLAEV
Aseg-2 AY301987 AsegPBP2s AAGGTGATCAAYAAYTT KVINNF
AsegPBP2as TACGGCTGTCATTATTT EEIMTA
alignment, the introns have conserved sites with gaps in
between. Similarly, the rst 300 nucleotides and last
1448 nucleotides were found only in the Aper-1 gene
nucleotide sequence. Because of the differences in length
of the introns, the sequences can be aligned in several
ways. We therefore constructed the alignment with a
different gap opening and extension values (gap opening
15gap extension 6.66, gap opening 10gap extension 3
and gap opening 3gap extension 1) and used these
different alignments for phylogenetic analysis with
different outgroups. Using different parameters or
assigning one of the introns or genes as outgroup did
not have an effect on the tree topology (data not
shown).
3. Results
3.1. Identication of cDNA sequences encoding Agrotis
PBPs
Based on our previous study, the two PBPs of A.
ipsilon appear to associate with two distinct subgroups
of PBPs (Picimbon and Gadenne, 2002). We therefore
designed degenerate primers to conserved regions
common to those respective subgroups (Group1 and
Group2) and used these to identify two orthologues in
A. segetum.
Using the pair of Group1 primers generated
600800 bps PCR products. Sequencing of about 20
positive clones in A. ipsilon resulted in the same
sequence identical to the protein Aips-1, Aips-1a, and
two additional sequences that differed from the others
only by a few nucleotides, Aips-1b and Aips-1c (Fig. 1).
The same differences were found analyzing additional
Aips-1 cDNA clones from either males or females (not
shown), suggesting the occurrence of multiple isoforms
of Aips-1 in both sexes. No change in the amino acid
sequence was found for Aseg-1 except for one amino
acid change (LaForest et al., 1999). Using the pair of
Group2 primers yielded 400800 bps-long PCR frag-
ments when tested with A. ipsilon or A. segetum cDNA.
Sequencing of cloned PCR products from A. ipsilon
identied multiple clones encoding Aips-2. Three
sequences that differ by two or ten amino acids were
found, Aips-2a, Aips-2b and Aips-2c. Sequencing of
cloned PCR products from A. segetum allowed the
identication of a novel PBP cDNA given the known
protein Aseg-1 sequence. The identied sequence
showed only about 50% identity to Aseg-1/Aips-1.
A considerable higher degree of similarity (7477%) was
found with Aips-2. Therefore, we identied a second
PBP in A. segetum, Aseg-2. The Aseg-2 cDNA clone is
426 nucleotides long and codes for a 142 amino acids-
long sequence. The mature protein Aseg-2 has a
molecular weight of 16.247 kDa and an isoelectric point
of 5.22 that is conserved among orthologs of this type of
protein. Indeed, this A. segetum PBP exhibited several
amino acid motifs that are also found in Aips-2 and
related proteins: 26-GEHIMQD-32, 61-LIGEDLQK-
MHH-70, 75-FAKSHGAD-82, 122-KMHELKWAPS-
MEV-134 and 136-EEIMTAV-142. As found for Aips-
2, three different Aseg-2 cDNA products were found:
Aseg-2a, Aseg-2b and Aseg-2c, which differ at specic
ARTICLE IN PRESS
Table 2
PCR primers for amplication of Agrotis PBP sequences from genomic DNA
Gene name Acc. Num. Primer name Sequence 5
0
3
0
Target region
Aips-1 AY301985 Group1-1s TCGCARGAWRTYATKAAG SQEIMK Exon1
Aips-1/Int2s TTCGCGAAGAAACATGG FAKKHG Exon2
Group1-2as RTANYCNTCNWYCCARAARTTRTA YNFWKEG Exon2
Group1-2s TAYAAYTTYTGGRGANGRNTAYG YNFWKEG Exon2
Group1gene-2s ACGAGGCGATGGCGAAGCAG EAMAKQ Exon3
Group1gene-2as CTGCTTCGCCATCGCCTCGT EAMAKQ Exon3
Aips-1/Int2as TTCGCCATTGCCTCGTCT DEAMAK Exon3
Group1-1as RACTTCRGCCAAGACTTC EVLAEV Exon3
Aips-2 AY301986 Group2gene-1As ATGACITCICAGRAGGTG MTSQEV Exon1
Aips-2/Int2 s TGAGGAGTTTGCTAAGA EEFAKS Exon2
Group2gene-1as GCKCCRTGDYTCTTRGC FAKKHG Exon2
Group2gene-2s GACGACGCCACAGCCAAGCAG DDATAKQ Exon3
Aips-2/Int2as TTGGCTACGGCTTCGTCT DEAVAK Exon3
Group2gene-2as ACNRVIGTCATDATYTC EIMTAV Exon3
Aseg-2 AY301987 Group2gene-1Bs ATGACITCICAGRAGGTG MTSQKV Exon1
Aseg2/Int2s GAGTACGCCAAGAAACAC LGLLKE Exon2
Group2gene-1as GCKCCRTGDYTCTTRGC FAKKHG Exon2
Group2gene-2s GACGACGCCACAGCCAAGCAG DDATAKQ Exon3
Aseg2/Int1as CTTGCCYRHGKCRTCGTC ALEIAK Exon3
Group2gene-2as ACNRVIGTCATDATYTC EIMTAV Exon3
ARTICLE IN PRESS
residue positions. Most of the variations are found at
the N-terminus and between amino acid 84100 and
107135 corresponding to a helices a1, a5 and a6 of the
PBP molecule, respectively (Sandler et al., 2000). The N-
terminal region identies two different subtypes: SQ-
VVSF (Aseg-2a) and SQKVINNF (Aseg-2b and Aseg-
2c). Aseg2c and Aips-2c display the residue Met111 and
the C-terminal motif 132-SMEVAM-137. The different
Aseg-2 sequences show about 7683% identity to
Aips-2s.
Therefore, the diversity of PBPs within Agrotis species
may depend not only on the number of PBP genes
within the genome but also on the number of allelic
variations that originate from each PBP gene as found in
many other species of moths (Vogt, 1987; Vogt and
Prestwich, 1988; Gyo rgyi et al., 1988; Newcomb et al.,
2002). In addition, our examination of Agrotis PBP
sequences conrms that the noctuid PBPs may separate
into two groups based on conserved sequence motif,
length and unique residues, Grp1 and Grp2 (Fig. 1).
The phylogenetic tree of moth PBPs suggests more
than two similarity groupings (at least three: Grp1 to 3)
and subdivision levels in specic similarity groups,
which perhaps represent gene duplications (Fig. 2).
We found that both nucleotide and amino acid trees
strongly support the specic grouping of Agrotis PBPs
with the other noctuid PBPs into the two orthology
groups Grp1 and Grp2. Indeed, we observed that the
proteins Aips-1, Aseg-1, Aips-2 and Aseg-2, together
with noctuid PBPs, form two separate groups with high
bootstrap support values: 93% and 99% support for
Grp1 (Aips-1, Aseg-1, Hzea-1, Hvir-1, Mbra-2) in the
trees based on amino acid, respectively, nucleotide-
based data. For Grp2 (Aseg-2, Aips-2, Hvir-2, Mbra-1,
Msex-2, EposPBP2, YcagPBP, Msex-3, Harm-3) there is
98% and 96% support based on amino acid, respec-
tively, nucleotide-based data. Msex-3 and Harm-3 group
with Grp2 proteins, but should be coded by different
genes; Msex-2 and Msex-3 should be the result of gene
duplication. Still, the proteins Msex-3 and Harm-3
contain the amino acid residues that are unique for
Grp2 proteins (see Fig. 1). The proteins Msex-3 and
Harm-3 may therefore represent a subgroup of Grp2
PBP (Grp2B) with different functions than Aips-2,
Aseg-2, Msex-2 and other Grp2 proteins (Grp2A).
Non-noctuid PBPs Aper-1, Aper-2, Aper-3, Apol-
1, Apol-2, Apol-3, Ldis-1, Ldis-2, Bmor-1, Msex-1,
SexiPBP, PgosPBP, OnubPBP, CfumPBP and AvelPBP
fall outside the groups Grp1 and Grp2 but do not form
a single group. One orthology group in Saturnids (Grp3:
Aper-1/Apol-1/Aper-2/Apol-2) is strongly supported by
both amino acid and nucleotide trees (bootstrap values:
100%). Tortricid CfumPBP/AvelPBP/EposPBP, satur-
nid Aper-3/Apol-3 and other proteins such as Ldis-1/
Ldis-2, Bmor-1, Msex-1, SexiPBP, PgosPBP and
OnubPBP probably constitute additional groups of
PBPs characterized by common motifs and possibly
suggesting group-specic functions.
3.2. Antennal and sex specicity of Agrotis PBPs
To expand our analysis of PBPs towards functional
localization, we analyzed in A. ipsilon and A. segetum
the tissue and sex distribution of PBPRNAs by
RTPCR (Fig. 3). Using pairs of primers specic to
Aips-1, Aseg-1, Aips-2 and Aseg-2, RTPCR products
of 400600 bps were amplied in both male and female
antennae but not in the legs. Sequencing identied the
RTPCR products as Aips-1, Aseg-1, Aips-2 and Aseg-
2, respectively. We therefore revealed antennal speci-
city for all four proteins Aips-1, Aseg-1, Aips-2 and
Aseg2. The Aips-1/Aseg-1 mRNAs appear more abun-
dant in male than in female antennae. In contrast, the
Aips-2 mRNAs appear equally expressed in both males
and females. In preliminary experiments, relative higher
levels of Aseg-2 mRNAs are found in males.
3.3. Structure of Agrotis PBP genes
Different PBPs may originate from different alleles or
alternative splice forms of the same gene but could also
originate from genes at different loci. We clearly
demonstrated that Aips-1/Aseg-1 and Aips-2/Aseg-2
are the translated products of two different PBP genes
and derive from common ancestry.
We cloned and sequenced from genomic DNA of A.
ipsilon two genes whose respective products were similar
to the previously identied cDNA sequences for Aips-1
and Aips-2 (Picimbon and Gadenne, 2002). We also
identied the Aseg-2 gene after cloning of Aseg-2 cDNA
from A. segetum. Comparisons between cDNA and
genomic sequences lead to the identication of coding
(Exon) and non-coding (Intron) regions in the genes
identied (Acc. Num. AY973626, AY973627,
AY973628). The structure of the genes encoding Aips-
1, Aips-2 and Aseg-2 was then compared to the structure
ARTICLE IN PRESS
Fig. 1. Alignment of amino acid sequences from Agrotis PBPs with related moth PBPs. Common amino acids in PBPs are indicated by a gradient of
shading. Highly conserved amino acids including the six cysteines (marked with asterisks) residues appear in black in a gray background. From the
alignment, two orthologous groups are identied: Grp1 (Aips-1a/Aips1b/Aips1c/Aseg-1/Mbra-2/Hzea-1/Hvir-1/Harm-1) and Grp2 (Aips-2a/Aips-
2b/Aips-2c/Aseg-2a/Aseg-2b/Aseg-2c/Mbra-1/Hvir-2/Msex-2/YcagPBP/Msex-3/Harm-3). The residues specic to Grp1 proteins are boxed with
black contour lines, Grp2 protein-specic residues are boxed with light gray contour lines. Hydropathy proles are shown for Agrotis PBPs. The
arrows indicate the positions of the primers designed to clone the genomic sequences encoding the PBPs from Agrotis (see Table 1). Primers were
based on the position of introns in the gene encoding Aseg-1 (LaForest et al., 1999). The arrowheads indicate the intron insertion sites.
of the gene encoding Aseg-1 (Acc. Num.
#AF134253AF134294, LaForest et al., 1999). This
allowed us to analyze the evolution of the PBP genes
within Agrotis species and to perform a comparative
study of moth PBP genes.
We identied two introns within the Aips-1 coding
region. They are located between Glu22 and Met23
(Intron1; 312 bp) and between Ala82 and Asp83
(Intron2; 1056 bp), and are remarkably different in
length. The genes encoding Aips-1 and Aips-2 display a
similar exonintron structure but differ in the length of
Intron2. The structure of the Aseg-2 gene appears very
similar to that of the Aips-2 gene but more distantly
related to the genes encoding Aips-1 and Aseg-1,
particularly with respect to the length of Intron2. Our
results show that the genes encoding Aips-1, Aips-2,
Aseg-1 and Aseg-2 all consist of three exons and two
introns. We found that when all four genes are aligned,
Exon1 appears much shorter than Exons 2 and 3 and the
two introns signicantly differ in length. The Aips-1/
Aseg-1 genes are generally much longer than the Aips-2/
Aseg-2 genes. The Exon1s and Exon2s have exactly the
same size (66 and 180 bp, respectively). A difference in
length is only found for Exon3. The third exons of Aseg-
2 and Aips-2 genes are three nucleotides shorter than the
third exons of Aips-1 and Aseg-1 genes. Those three
ARTICLE IN PRESS
100
94
88
86
100
100
60
83
88
99
100
89
96
76
65
73
50
99
Aips-1
Aseg-1
Hvir-1
Hzea-1
Harm-1
Ldis-1
Ldis-2
Bmor-1
Msex-1
OnubPBP
PgosPBP
SexiPBP
Apol-3
Aper-3
CfumPBP
AvelPBP
EposPBP
Aper-1
Apol-1
Aper-2
Apol-2
Mbra-2
Msex-3
Harm-3
EposPBP2
YcagPBP
Msex-2
Hvir-2
Mbra-1
Aips-2
Aseg-2
55
100
98
52
100
85
76
93
100
62
63
98
99
91
61
93
92
66
98
100
100
95
100
Ldis-1
Ldis-2
Bmor-1
Msex-1
SexiPBP
PgosPBP
OnubPBP
Apol-3
Aper-3
CfumPBP
AvelPBP
EposPBP
Aper-1
Apol-1
Aper-2
Apol-2
Hzea-1
Harm-1
Hvir-1
Mbra-2
Aseg-1
Aips-1c
Aips-1a
Aips-1b
Msex-3
Harm-3
YcagPBP
EposPBP2
Msex-2
Mbra-1
Hvir-2
Aips-2c
Aips-2a
Aips-2b
Aseg-2c
Aseg-2a
Aseg-2b
100
Grp3 Grp3
Grp1
Grp2
Grp1
Grp2
Fig. 2. Phylogenetic analysis of amino acid (left) and nucleotide (right) PBP sequences. Trees are with Ldis-1 and Ldis-2 as outgroup, Amino acid
tree length 891, CI 0.530, RI 0.681, RC 0.361, HI 0.470, cDNA tree length 2280, CI 0.328, RI 0.458, RC 0.150, HI 0.672. Proteins are
(name+accession number): Ldis-1, AAC47913; Ldis-2, genbank:AAC47914; Bmor-1, CAA64430; Msex-1, genbank:AAG50016; Aper-1,
AAB49502; Apol-1, CAA35592; Aper-2, CAA65603; Apol-2, CAB86718; SexiPBP, AAF06123; PgosPBP, AAF06141; Aips-1, AAP57457; Aseg-
1, AAD41275; Hvir-1, CAA65604; Hzea-1, AAC36315; Harm-1, CAC08212; Mbra-2, AAC05701; OnubPBP, AAD3944; CfumPBP, AAF06129;
AvelPBP, AAF06126; Apol-3, CAB86719; Aper-3, CAB86717; Aips-2, AAP57468; Aseg-2, AAP57469; Hvir-2, AAP57470, Mbra-1 AAC05702;
Msex-2, AAF16710; YcagPBP, AAF06143; EposPBP, AAL09027; EposPBP2, AAL05868;Msex-3, genbank:AAF16702; Harm-3, AAO16091) (Vogt,
2003; and references therein). The arrows indicate gene duplication events that may have given rise to Antheraea (Grp3) and Agrotis PBPs (Grp1 and
Grp2), respectively. A particular duplication may have given rise to proteins of Grp1 and Grp2 (bold arrow).
nucleotides code for the extra amino acid found in Aips-
1 and Aseg-1 between the third and fourth cysteine.
Therefore, we show that the exon3 pattern between the
genes Aips-1 and Aseg-1 is identical and differs slightly
with respect to Exon3 of Aips-2 and Aseg-2 genes. The
main differences are found in the intron sequences. In all
four genes intron1 is of about the same size
(312458 bp). In contrast, intron2 is signicantly longer
in Aseg-1 and Aips-1 (993 and 1056 bp) than in Aseg-2
and Aips-2 (512 and 564 bp). However, the two introns
are always located at the same position, suggesting a
common PBP ancestor gene (Fig. 4).
3.4. Structure and evolution of introns in Agrotis PBP
genes
The introns in both Agrotis PBP genes are AT-rich;
adenosine and thymidine represent about 70% of the
intron sequence. Overall, intron1 appears to be more
conserved than intron2. The Aips-1 and Aips-2 gene
ARTICLE IN PRESS
Fig. 3. Sex and tissue-specicity of Agrotis PBPs. Reverse transcriptasePCR analysis of total RNA (500 ng) from male antennae, female antennae
and legs in Agrotis ipsilon and A. segetum. Control experiments using actin primers show the same amplicon (0.5 kb) in all samples. For RNA control,
10 mg of each total RNA sample were analyzed on a formaldehyde gel. After staining by ethidium bromide, equivalent amounts of ribosomal RNA
are shown. The RTPCRs were carried out using oligonucleotide primers specic to Aips-1, Aips-2, Aseg-1 and Aseg-2, respectively. Using
TITANIUM Taq Reverse Transcriptase shows punctate amplication of PBP products in both male and female antennae. The RTPCR bands are
approximately 0.60.4 kb.
Aseg-1
Aseg-1 gene 1740 base pairs
Ala82 Glu22
5
3
5 3
Aips-1
Aips-1 gene 1797 base pairs
Ala82 Glu22
5
3
5 3
Aips-2
Aips-2 gene 1448 base pairs
Ala82 Glu22
5
3
5 3
Aseg-2
Aseg-2 gene 1121 base pairs
Ala82 Glu22
5 3
5 3
Fig. 4. Structure of Agrotis PBP genes. All the four genes Aseg-1, Aips-1, Aips-2 and Aseg-2 from A. segetum and A. ipsilon exhibit the same
structure characterized by 3 exons2 introns. Regions in black correspond to the protein coding sequences. The introns are shown in light-shaded
regions. Introns 1 and 2 are inserted in the same positions in all four genes. Indicated length shows the gene length variability.
share overall 37% similarity with each other. The
two introns share about 4648% identity. A similar
degree of identity is found analyzing the amino acid
composition of the exons; exon 2 and 3 show about
4748% identity, while exon 1 exhibits only 41%
identity. The same observations were made comparing
the introns and the exons of the PBP genes from A.
segetum. Apparently, the degrees of similarity observed
between the coding and non-coding regions of PBP
genes are in the same range. We have found no
particular sequence motifs (i.e. potential transcription
factor-binding sites) that are shared between all of the
intron 1 or intron 2 sequences.
Interestingly, the introns of both Aips-1/Aseg-1 and
Aips-2/Aseg-2 seem to differ not only in length but also
in nucleotide content. To support this hypothesis, we
examined the phylogenetic distribution of introns based
on the alignment of the four Agrotis PBP genes using the
maximum parsimony method. The whole phylogenetic
analysis of the introns of Agrotis PBP genes supported a
specic grouping of introns, Aips-1intron1/Aseg-1in-
tron1, Aips-1intron2/Aseg-1intron2, Aips-2intron1/
Aseg-2intron1, and Aips-2intron2/Aseg-2intron2 (Fig.
5A). We have analyzed several intron alignments
constructed with different gap opening and extension
penalties. The phylogenetic analysis was performed with
and without an outgroup (Aseg-1 intron1, Aips-2
intron2 and Aseg-1 intron2). The resulting trees had
similar topologies. The bootstrap values were the same
in all trees (100%) (not shown). The collected data
suggest that Aips-1/Aseg-1 and Aips-2/Aseg-2 genes
existed in the common ancestor species of A. ipsilon and
A. segetum and that, after these two species diverged,
each started accumulating mutations in the introns.
To analyze the inuence of evolution on the PBP
genes, we analyzed the alignment of genomic sequences
encoding different moth PBP genes with the maximum
parsimony method, which resulted in a tree in which ve
groups are dened, Aseg-1/Aips-1, Aseg-2/Aips-2,
OnubPBP, Aper-1 and Msex-1 (Fig. 5B). This grouping
clearly shows that the introns of different PBP genes are
different, suggesting that they have probably evolved
independently.
4. Discussion
Our detailed investigation of PBPs in A. ipsilon and A.
segetum shows that the two Agrotis PBPs are the result
of gene duplication before the emergence of the two
noctuid species from a common ancestor.
We have identied a cDNA coding for a novel PBP in
A. segetum (Aseg-2) very similar to Aips-2 cloned from
A. ipsilon (Picimbon and Gadenne, 2002). For the
turnip moth, A. segetum, it has been shown earlier that
Aseg-1 is related to Aips-1 and other known PBPs,
ARTICLE IN PRESS
Aseg-1
intron 2
Aips-2
intron 1
Aseg-2
intron 1
Aseg-2
Aips-2
Aseg-1
intron 1
intron 2
intron 2
intron 1
Aips-1
intron 2
Aips-1
100
100
100
100
100
Msex-1
Aseg-1
Aips-1
OnubPBP
Aper-1
Aseg-2
Aips-2
60
100
79
100
(A)
(B)
Fig. 5. Phylogenetical analysis of introns from PBPs. Unrooted
maximum parsimony tree based on intron sequences from Grp1 and
Grp2 PBP genes in A. segetum and A. ipsilon (A) and the aligned genes
for Aseg-1, Aseg-2, Aips-1, Aips-2, OnubPBP, Aper-1 and Msex-1
(Krieger et al., 1991; Willett and Harrison, 1999; Vogt et al., 2002) (B).
The cluster of introns of Agrotis PBP genes is specic and falls outside
the different groups of PBP genes from moths.
Hel-1, Hzea-1, Harm-1 and Mbra-2 from A. ipsilon,
Heliothis virescens, H. zea, H. armigera and Mamestra
brassicae, respectively (Krieger et al., 1993; Mabe` che-
Coisne et al., 1998; Callahan et al., 2000; Picimbon and
Gadenne, 2002). It appears that most characteristic for
Aips-1/Aseg-1, Aips-2/Aseg-2 and related proteins are
the C-terminal sequences (GEVLAEV for Aips-1/Aseg-
1 and EEIMTAV for Aips-2/Aseg-2) that may have
functional importance. In Drosophila OS-F and OS-E
divergent regions have been found at the C-terminus
(Hekmat-Scafe et al., 2000). The C-terminus of the PBP
from the honey bee Apis mellifera folds into the protein
and forms a wall of the internal hydrophobic cavity,
suggesting direct interactions with specic ligands
(Lartigue et al., 2004). In contrast, the C-terminus is
exible and unstructured in the PBPs from Bombyx mori
and the cockroach Leucophea maderae (Sandler et al.,
2000; Lartigue et al., 2003). In B. mori, the C-terminus is
an a-helix that closes the ligand binding pocket and
therefore participates in the exclusion of the ligand
(Horst et al., 2001). The C-terminus could well play a
role in promoting the formation of proteinprotein
complexes as described for receptor molecules (Dosil
et al., 2000).
More interestingly, the existence of two PBPs in both
A. ipsilon and A. segetum is in agreement with the
existence of two orthologous groups of PBPs in
noctuids, Grp1 and Grp2, as suggested by Picimbon
and Gadenne (2002). A PBP related to Aips-2 and Aseg-
2 exists on H. virescens (Hvir-2: Acc. Num. AAP57470,
Picimbon, unpublished). Therefore, despite differences
in pheromone systems, the same PBP prole (e.g. PBPs
representing the two specic orthologous groups Grp1
and Grp2) seems to be maintained in the different
noctuid species.
The tree of nucleotide sequences tends to explain the
existence of the two Agrotis PBPs as a consequence of
duplication events of a common ancestor gene. Two
duplications may have occurred before the divergence of
the different Agrotis species. An early duplication gave
rise to Grp1 and Grp2 proteins; a late-duplication gave
rise to the sphingid proteins Msex-2 and Msex-3 and to
other Grp2 PBPs. This late-duplication may have
occurred in both noctuids and sphingids as a PBP
related to Msex-3 is found in H. armigera (Harm-3) or it
may have happened before the divergence of sphingids
and noctuids. A similar duplication pattern seems to
have occurred in saturnids. Branching within the
orthologous groups of PBPs is such that the saturnid
PBPs Apol-1 and Apol-2 from Antheraea polyphemus
seem to originate from a late duplication that is a
common event in saturnid species. The PBPs Aper-1 and
Aper-2 from A. pernyi have similar origins. While Aper-1,
Aper-2, Apol-2 and Apol-1 might derive from the same
duplication, Aper-3 and Apol-3 might result from a more
ancient duplication that occurred before the separation of
Antheraea species (see Fig. 2). The large number of gene
duplications suggested among the groups of PBPs is
typical for an evolutionary active gene family.
Analysis of tissue and sex distribution of Agrotis PBPs
in A. ipsilon and A. segetum shows that both proteins
Aips/Aseg-1 and Aips/Aseg-2 are expressed in male and
female antennae as found for PBPs from many different
species of moths (Vogt et al., 1991; Krieger et al., 1993,
1996; Callahan et al., 2000; Picimbon and Gadenne,
2002). The PBPs Aips-1 and Aseg-1 are highly expressed
in males. In contrast, the proteins Msex-2 and Msex-3
that are related to Aips-2 are expressed at about the
same level in female and male antennae with expression
in a central subset of general-odorant sensilla (Robert-
son et al., 1999). In males, Msex-2 and Msex-3 are found
to be co-expressed (Nardi et al., 2003). The PBP Aseg-2
seems to be more expressed in males similarly to Aips-1,
Aseg-1, Hvir-1 and Hzea-1 (Krieger et al., 1993;
Callahan et al., 2000; this study). Like the two PBPs
from Lymantria dispar (Ldis-1 and Ldis-2), the two
Agrotis PBPs could well have different binding afnities
for pheromone compounds (Vogt et al., 1989; Plettner et
al., 2000). However, the most intriguing is perhaps the
appearance of both PBP-1 and PBP-2 in Agrotis
females. This may indicate that not all PBPs are
involved in pheromone detection or that pheromone
malefemale and femalefemale interactions take place
in the noctuid species (Hillier and Vickers, 2004). A
cautious analysis of the temporal and spatial expression
pattern of PBPs in noctuid females is necessary to
elucidate the function of these proteins in both sexes.
The two Agrotis PBPs clearly represent two distinct
genes as found for Lymantrid PBPs (Merritt et al., 1998).
The Aips-1 and Aips-2 are encoded by independent
genes since they differ in intron size and coding sequence
and have derived from a common ancestor considering
that their three exons are of the same length (Exon3 is
only three nucleotides longer in Aips-1) and that exon/
intron boundaries are conserved. Introns within the
coding region of Aips-1 and Aips-2 are located between
Glu22 and Val/Met23 and between Ala82 and Asp83,
respectively. The gene duplication that gave rise to
Agrotis PBPs must have happened before the divergence
of the Agrotis species. The Aseg-1 and Aips-1 genes have
similar intron lengths and show a high degree of identity.
The characterization of the gene encoding Aseg-2 in A.
segetum afrms our presumption that the PBP genes
from Agrotis share similarity with each other. We found
that the rst intron from the genes Aips-1 and Aseg-1 is
closer related to the second intron from the genes Aips-2
and Aseg-2 and vice versa. To conrm that this is unique
for Agrotis or is valid for all noctuid PBP genes, more
PBP gene introns from noctuid species will have to be
included. Nevertheless, the phylogenetic analysis based
on the introns shows clearly that the Agrotis genes within
each group are related to each other.
ARTICLE IN PRESS
We also found that both Aseg-2 and Aips-2 are
shorter than Aseg-1 and Aips-1. All four genes display
the same pattern of exon/intron boundaries and the
introns in all four genes are A-T-rich, which is probably
required for splicing and polyadenylation similar to the
mechanism in for instance Angiosperms (Luehrsen and
Walbot, 1994). All these observations clearly indicate
that Agrotis PBP-1 and PBP-2 genes are independent,
but originate from the duplication of a common
ancestral gene. Following the duplication of Aips-1/
Aseg-1 and Aips-2/Aseg-2 genes, other genetic events
such as DNA inclusion or deletion seem to have
occurred since the Aseg-2 gene appears to be shorter
than the Aips-2 gene, in particular at the level of the
second intron (intron2). While the intron2s of Aips-2
and Aseg-2 have marked differences, the introns 1 of
Aips-1 and Aseg-1 are very similar both in size and
nucleotide composition. Signicant amount of intron
differentiation is found in the PBP gene from Ostrinia
nubilalis, OnubPBP (Willett and Harrison, 1999). There-
fore, introns from PBP genes except Aips-1 and Aseg-1
seem to have evolved independently from each other.
The functional importance of introns remains to be
elucidated. Thus, it is now of interest to identify more
PBP genes in particular in noctuids and to determine
whether PBPs within and across species display not only
genetic but also functional differences. Such a study may
reveal whether evolution inuences also the presumably
silent intronic structures, together with the regions
coding for the functional elements of the protein.
Acknowledgements
We thank Erling Jirle and Karin Johnson for rearing
of Agrotis species. We also thank Dr. Christophe
Gadenne (INRA, Villenave dOrnon, France) for the
gift of the Agrotis ipsilon strain. This work was
supported by a stipendium from the Royal Physio-
graphic Society to DA, research grants from the Knut
and Alice Wallenberg foundation and the Swedish
research councils (NFR, VR) to CL, and research
grants from VR and the Crafoord foundation to JFP.
References
Abraham, D., Ryrholm, N., Wittzell, H., Scoble, M.J., Holloway, J.,
Lo fstedt, C., 2001. Molecular phylogeny of the subfamilies in
Geometridae (Geometroidea: Lepidoptera). Mol. Phyl. Evol. 20,
6577.
Callahan, F.E., Vogt, R.G., Tucker, M.L., Dickens, J.C., Mattoo,
A.K., 2000. High level expression of male specic pheromone
binding proteins (PBPs) in the antennae of female noctuiid moths.
Dosil, M., Schandel, K.A., Gutpa, E., Jenness, D.D., Konopka, J.B.,
2000. The C-terminus of the Saccharomyces cerevisiae alpha-factor
receptor contributes to the formation of preactivation complexes
with its cognate G protein. Mol. Cell. Biol. 20, 53215329.
Felsenstein, J., 1985. Condence limits on phylogenies: an approach
using the bootstrap. Evolution 39, 783791.
Gemeno, C., Haynes, K.F., 1998. Chemical and behavioral evidence
for a third pheromone component in a North American population
of the black cutworm moth, Agrotis ipsilon. J. Chem. Ecol. 26,
329342.
Gyo rgyi, T.K., Roby-Shemkovitz, A.J., Lerner, M.R., 1988. Char-
acterization and cDNA cloning of the pheromone-binding protein
from the tobacco hornworm, Manduca sexta: A tissue-specic
developmentally regulated protein. Proc. Natl. Acad. Sci. USA 85,
98519855.
Hall, T., 2001. BioEdit Version 5.0.9. Publisher, NC State University
(http://jwbrown.mbio.ncsu.edu/BioEdit/bioedit.html).
Hekmat-Scafe, D.S., Dorit, R.L., Carlson, J.R., 2000. Molecular
evolution of odorant-binding protein genes OSF and OSE in
Drosophila. Genetics 155, 117127.
Hillier, N.K., Vickers, N.J., 2004. The role of Heliothine hairpencil
compounds in female Heliothis virescens (Lepidoptera: Noctuidae)
behavior and mate acceptance. Chem. Senses 29, 499511.
Horst, R., Damberger, F., Luginbu hl, P., Gu ntert, P., Peng, G.,
Nikonova, L., Leal, W.S., Wu thrich, K., 2001. NMR structure
reveals intramolecular regulation mechanism for pheromone
binding and release. Proc. Natl. Acad. Sci. USA 25, 1437414379.
Krieger, J., Raming, K., Breer, H., 1991. Cloning of genomic and
complementary DNA encoding insect pheromone binding proteins:
evidence for microdiversity. Biochem. Biophys. Acta 1088, 277284.
Krieger, J., Ganssle, H., Raming, K., Breer, H., 1993. Odorant binding
proteins of Heliothis virescens. Insect Biochem. Mol. Biol. 23,
449456.
Krieger, J., von Nickisch-Roseneg, E., Mameli, M., Pelosi, P., Breer,
H., 1996. Binding Proteins from the Antennae of Bombyx mori.
LaForest, S., Prestwich, G.D., Lo fstedt, C., 1999. Intraspecic
nucleotide variation at the pheromone binding protein locus in
the turnip moth, Agrotis segetum. Insect. Mol. Biol. 8, 481490.
Lartigue, A., Gruez, A., Spinelli, S., Rivie` re, S., Brossut, R., Tegoni,
M., Cambillau, C., 2003. The crystal structure of a cockroach
pheromone binding protein suggests a new ligand binding and
release mechanism. J. Biol. Chem. 278, 3021330218.
Lartigue, A., Gruez, A., Briand, L., Blon, F., Bezirard, V., Walsh, M.,
Pernollet, J.C., Tegoni, M., Cambillau, C., 2004. Sulfur single-
wavelength anomalous diffraction crystal of a pheromone-binding
protein from the honeybee Apis mellifera L. J. Biol. Chem. 279,
44594464.
Leal, W.S., 2003. Proteins that make sense. In: Blomquist, G.J., Vogt,
R.G. (Eds.), Insect Pheromone Biochemistry and Molecular
Biology. Academic Press, New York, USA, pp. 447506.
Lo fstedt, C., Linn Jr., C.E., Lo fqvist, J., 1985. Behavioural responses
of male turnip moths Agrotis segetum to sex pheromone in a ight
tunnel and in the eld. J. Chem. Ecol. 11, 12091223.
Luehrsen, K.R., Walbot, V., 1994. Intron creation and polyadenyla-
tion in maize are directed by AU-rich RNA. Genes Dev. 8,
11171130.
Mabe` che-Coisne, M., Jacquin-Joly, E., Franc-ois, M.C., Nagnan-Le
Meillour, P., 1998. Molecular cloning of two pheromone binding
proteins in Mamestra brassicae. Insect Biochem. Mol. Biol. 28,
815818.
Maida, R., Krieger, J., Gebauer, T., Lange, U., Ziegelberger, G., 2000.
Three pheromone binding proteins in olfactory sensilla of the two
silkmoth species Antheraea polyphemus and Antheraea pernyi. Eur.
J. Biochem. 267, 28992908.
Merritt, T.J.S., LaForest, S., Prestwich, G.D., Quattro, J.M., Vogt,
R.G., 1998. Patterns of gene duplication in lepidopteran pher-
omone binding protein. J. Mol. Evol. 46, 272276.
ARTICLE IN PRESS
Nardi, J.B., Miller, L.A., Walden, K.K.O., Rovelstad, S., Wang, L.,
Frye, J.C., Ramsdell, K., Deem, L.S., Robertson, H.M., 2003.
Expression patterns of odorant-binding proteins in antennae of the
moth Manduca sexta. Cell Tissue Res. 313, 321333.
Newcomb, R.D., Sirey, T.M., Rassam, M., Greenwood, D.R., 2002.
Pheromone binding proteins of Epiphyas postvittana (Lepidoptera:
Tortricidae) are encoded at a single locus. Insect Biochem. Mol.
Biol. 32, 15431554.
Picimbon, J.F., 2003. Biochemistry and evolution of CSP and OBP
proteins. In: Blomquist, G.J., Vogt, R.G. (Eds.), Insect Pheromone
Biochemistry and Molecular Biology. Academic Press, New York,
USA, pp. 539566.
Picimbon, J.F., Gadenne, G., 2002. Evolution in noctuid pheromone
binding proteins: identication of PBP in the Black cutworm moth,
Agrotis ipsilon. Insect Biochem. Mol. Biol. 32, 839846.
Picimbon, J.F., Gadenne, C., Be card, J.M., Cle ment, J.L., Sreng, L.,
1997. Sex pheromone of the French black cutworm moth, Agrotis
ipsilon (Lepidoptera: Noctuidae): Identication and regulation of a
multicomponent blend. J. Chem. Ecol. 23, 211230.
Picimbon, J.F., Dietrich, K., Angeli, S., Scaloni, A., Krieger, J., Breer,
H., Pelosi, P., 2000. Purication and molecular cloning of
chemosensory proteins from Bombyx mori. Arch. Insect. Biochem.
Physiol. 44, 120129.
Plettner, E., Lazar, J., Prestwich, E.G., Prestwich, G.D., 2000.
Discrimination of Pheromone enantiomers by two pheromone
binding proteins from the Gypsy moth Lymantria dispar. Biochem-
istry 39, 89538962.
Prestwich, G.D., Du, G., LaForest, S., 1995. How is pheromone
specicity encoded in proteins? Chem. Senses 20, 461469.
Robertson, H.M., Martos, R., Sears, C.R., Todres, E.Z., Walden,
K.K.O., Nardi, J.B., 1999. Diversity of odourant binding proteins
revealed by an expressed sequence tag project on male Manduca
sexta moth antennae. Insect Mol. Biol. 8, 501518.
Sandler, B.H., Nikonova, L., Leal, W.S., Clardy, J., 2000.
Sexual attraction in the silkworm moth: structure of the
pheromone-binding proteinbombykol complex. Chem. Biol. 7,
143151.
Steinbrecht, R.A., Ozaki, M., Ziegelberger, G., 1992. Immunocyto-
chemical localization of pheromone binding protein in moth
antennae. Cell. Tissue Res. 270, 287302.
Steinbrecht, R.A., Laue, M., Ziegelberger, G., 1995. Immunolocaliza-
tion of pheromone-binding protein and general odorant-binding
protein in olfactory sensilla of the silk moths Antheraea and
Bombyx. Cell. Tissue Res. 282, 287302.
Swofford, D.L., 1999. PAUP*. Phylogenetic Analysis Using Parsi-
mony (*and Other Methods). Version 4. Sinauer Associates,
Sunderland, MA.
D.G., 1997. The CLUSTAL X windows interface: exible
tools. Nucl. Acids Res. 25, 48764882.
Vogt, R.G., 1987. The molecular basis of pheromone reception: its
inuence on behavior. In: Prestwich, G.D., Blomquist, G.J. (Eds.),
Pheromone Biochemistry. Academic Press, New York,
pp. 385431.
Vogt, R.G., 2003. Biochemical diversity of odor detection: OBPs,
ODEs and SNMPs. In: Blomquist, G.J., Vogt, R.G. (Eds.), Insect
Pheromone Biochemistry and Molecular Biology. Academic Press,
New York, USA, pp. 391445.
Vogt, R.G., Prestwich, G.D., 1988. Variation in olfactory proteins:
evolvable elements encoding insect behavior. Olfaction and taste
IX. Annals NY. Acad. Sci. 510, 689691.
Vogt, R.G., Riddiford, L.M., 1981. Pheromone binding and inactiva-
tion by moth antennae. Nature 293, 161163.
Vogt, R.G., Ko hne, A.C., Dubnau, J.T., Prestwich, G.D., 1989.
Expression of pheromone binding proteins during antennal
development in the gypsy moth Lymantria dispar. J. Neurosci. 9,
3323346.
Vogt, R.G., Prestwich, G.D., Lerner, M., 1991. Odorant-binding
protein subfamilies associate with distinct classes of olfactory
receptor neurons in insects. J. Neurobiol. 22, 7484.
Vogt, R.G., Rogers, M.E., Franco, M.D., Sun, M., 2002. A
comparative study of odorant binding protein genes: differential
expression of the PBP1-GOBP2 gene cluster in Manduca sexta
(Lepidoptera) and the organization of OBP genes in Drosophila
melanogaster (Diptera). J. Exp. Biol. 205, 719744.
Willett, C.S., Harrison, R.G., 1999. Insights into genome differentia-
tion: pheromone-binding protein variation and population history
in the European corn borer (Ostrinia nubilalis). Genetics 153,
17431751.
Wu, W.Q., Lo fstedt, C., Hansson, B., 1995. Electrophysiological and
behavioural evidence for a fourth component in the turnip moth
Agrotis segetum. Physiol. Entomol. 20, 8192.
Zhang, S.G., Maida, R., Steinbrecht, R.A., 2001. Immunolocalization
of odorant-binding proteins in noctuid moths (Insecta, Lepidop-
tera). Chem. Senses 26, 885896.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
The BmChi-h gene, a bacterial-type chitinase gene of Bombyx mori,
encodes a functional exochitinase that plays a role in the chitin
degradation during the molting process
Takaaki Daimon
a
, Susumu Katsuma
a,b
, Masashi Iwanaga
a,c
,
WonKyung Kang
d
, Toru Shimada
a,
a
Department of Agricultural and Environmental Biology, Graduate School of Agricultural and Life Sciences,
The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657, Japan
b
Department of Drug Discovery Science, Graduate School of Pharmaceutical Sciences, Kyoto University,
Yoshida Shimoadachi-cho 46-29, Sakyo-ku, Kyoto 606-8501, Japan
c
New Frontiers Research Laboratory, Toray Industries, Inc., Tebiro 1111, Kamakura, Kanagawa 248-8555, Japan
d
Molecular Entomology Laboratory, RIKEN, Hirosawa 2-1, Wako, Saitama 351-0198, Japan
Received 27 March 2005; received in revised form 6 May 2005; accepted 10 May 2005
Abstract
The silkworm, Bombyx mori, has been recently demonstrated to contain a bacterial-type chitinase gene (BmChi-h) in addition to
a well-characterized endochitinase gene (BmChitinase). The deduced amino acid sequence of BmChi-h showed extensive structural
similarities with chitinases from bacteria such as Serratia marcescens chiA and baculoviruses (v-CHIA). Bacterial-type chitinase
genes have not been found from any eukaryotes and viruses except for lepidopteran insects and lepidopteran baculoviruses. Thus, it
was suggested that BmChi-h may be derived from a bacterial or baculovirus chitinase gene via horizontal gene transfer.
In this report, we investigated the biological function of BmChi-h. Our enzymological study indicated that a chitinase encoded by
BmChi-h has exo-type substrate preference, which is the same as S. marcescens chiA and v-CHIA, and different from BmChitinase,
which has endo-type substrate preference. An immunohistochemical study revealed that BmChi-h localizes in the chitin-containing
tissues during the molting stages, indicating that it plays a role in chitin degradation during molting. These results suggest that
BmChi-h (exochitinase) and BmChitinase (endochitinase) may catalyze a native chitin by a concerted mechanism. Cloning and
comparison of BmChi-h orthologues revealed that bacterial-type chitinase genes are highly conserved among lepidopteran insects,
suggesting that the utilization of a bacterial-type chitinase during the molting process may be a general feature of lepidopteran
insects.
Keywords: Chitinase; Horizontal gene transfer; Bombyx mori; Baculovirus; Molting
1. Introduction
Chitinases (EC 3.2.1.14) catalyze the hydrolysis of the
b-1, 4-N-acetyl-D-glucosamine linkages in chitin poly-
mers. Chitinases have been detected in many organisms
and play important roles in each organism, such as
molting of the exoskeleton in insects and crustaceans,
cell growth and division in fungi, degradation of chitin
for nutrients in bacteria, and self-defense in plants
(Kramer and Muthukrishnan, 1997; Merzendorfer and
Zimoch, 2003). Chitinases from some pathogens of
insects are also utilized for enhancing their pathogeni-
city (Kramer and Muthukrishnan, 1997; Merzendorfer
and Zimoch, 2003). Some entomopathogenic microbes
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.005

Fax: +81 3 5841 8130.
E-mail address: shimada@ss.ab.a.u-tokyo.ac.jp (T. Shimada).
produce chitinases for penetration into the host body
(Kramer and Muthukrishnan, 1997; Gooday, 1999), and
baculoviruses, large double-stranded DNA viruses that
infect mainly lepidopteran insects, produce chitinase to
liquefy the host body after death due to infection
(Hawtin et al., 1995, 1997).
In insects, chitinases are believed to be induced stage-
and tissue-dependently to degrade the chitin in the
exoskeleton and peritrophic membrane (Kramer and
Muthukrishnan, 1997; Merzendorfer and Zimoch,
2003). Chitinases of lepidopteran insects have been
puried from two lepidopteran insects, Manduca sexta
(MsChitinase) and Bombyx mori (BmChitinase), and
their biochemical properties have been well character-
ized (Kramer and Muthukrishnan, 1997; Merzendorfer
and Zimoch, 2003). Chitinase genes encoding these two
enzymes have been successfully cloned, and their
orthologues have also been isolated from several
lepidopteran insects (Kramer et al., 1993; Kim et al.,
1998; Mikitani et al., 2000; Shinoda et al., 2001; Zheng
et al., 2002). These chitinases are extensively glycosy-
lated and produce several smaller forms by proteolysis
(Kramer and Muthukrishnan, 1997; Merzendorfer
and Zimoch, 2003). An immunohistochemical study of
CfChitinase (Zheng et al., 2003), an orthologue of
MsChitinase and BmChitinase cloned from the fall
armyworm, Choristoneura fumiferana, clearly proved
that these chitinases participate in chitin degradation
during the molting process (Kramer and Muthukrish-
nan, 1997; Merzendorfer and Zimoch, 2003).
Chitinases have been shown to be a multi-gene family
in lepidopteran insects (Daimon et al., 2003) as well as in
dipteran insects (de la Vega et al., 1998; Zhu et al.,
2004). We have previously cloned a novel chitinase gene
(BmChi-h) from the silkworm, B. mori, which belonged
to a bacterial-type chitinase gene family (Daimon et al.,
2003). BmChi-h showed extensive homology with
chitinase genes of bacteria and baculoviruses (Daimon
et al., 2003). Interestingly, it has been proposed that
baculovirus might acquire the chitinase gene from a
bacterium via horizontal gene transfer (Hawtin et al.,
1995; Hughes and Friedman, 2003). Accumulating data
suggest that baculoviruses have evolved through captur-
ing genes and transposable elements from their host
insects or other co-infecting viruses (Miller and Miller,
1982; Lerch and Friesen, 1992; Jehle et al., 1998;
Handler, 2002; Hughes and Friedman, 2003). For
example, some genes like ecdysteroid UDP-glucosyl-
transferase (egt), inhibitor of apoptosis (iap), and
ribonucleotide reductase 1 (rr1) of baculoviruses show
extensive similarities to those of host insects (OReilly,
1997; Hughes and Friedman, 2003; Kamita et al., 2005),
suggesting the horizontal gene transfer between the host
insect and the baculovirus. From these points of view,
the isolation of the BmChi-h gene has a great
signicance. Phylogenetic analysis demonstrated that
BmChi-h is also originated from bacteria. Therefore,
horizontal gene transfer of the chitinase gene might have
occurred among a bacterium, a baculovirus, and a
lepidopteran insect (Daimon et al., 2003).
It has been reported that BmChi-h mRNA was
specically expressed at chitin-containing tissues during
the molting stage with the same expression prole as
BmChitinase (Daimon et al., 2003), suggesting that the
protein encoded by BmChi-h plays a role in chitin
degradation at molting stages. In this study, we
described the functional characterization of BmChi-h.
Enzymological and histological studies were performed
to investigate the biological roles of BmChi-h. Successful
cloning and high sequence similarity demonstrated that
BmChi-h orthologues are present in broad taxa of
lepidopteran insects and might also play a role in the
molting process.
2. Experimental procedures
2.1. Insects and cell lines
The silkworm was reared as described (Daimon et al.,
2003). Samia cynthia and Antheraea pernyi were main-
tained at the University of Tokyo and reared on castor
leaves and sawtooth oak leaves, respectively. Agrius
convolvuli, Pseudaletia separata, and Ostrinia furnacalis
were kind gifts from Kenji Kiguchi (Shinshu University,
Japan), Shogo Matsumoto (RIKEN, Japan), and
Sadahiro Tatsuki (University of Tokyo), respectively.
The Spodoptera frugiperda Sf9 cell line was maintained
in a TC-100 medium containing 10% fetal bovine serum
as described previously (Maeda, 1989).
2.2. Construction of recombinant viruses
The recombinant Autographa californica nucleopoly-
hedroviruses (AcNPVs) were constructed using a Bac-
to-Bac system (Invitrogen). The coding region of
BmChi-h with a His-tag sequence at the C-terminus
was PCR-amplied with primers pBac1Chi-hF1 and
pBac1Chi-hR1 (Table 1) using the EST clone wdS20876
(Mita et al., 2003) as a template. The PCR product was
cloned into the cloning sites (BamHI and XbaI) of the
pFastBac1 vector (Invitrogen). To construct mutants,
the coding region was generated by two-step PCR-based
mutagenesis. In D306N, rst PCR was performed with
primer sets pBac1Chi-hF1 and D306N-R1, and D306N-
F1 and pBac1Chi-hR1, respectively. The PCR products
were mixed and used as templates for the second PCR
using primers pBac1Chi-hF1 and pBac1Chi-hR1, and
the product was then cloned into pFastBac1. In D310Q
and D306NE310Q, the coding regions were PCR-
amplied as described in D306N using primer sets
E310Q-F1 and E310Q-R1, and D306NE310Q-F1 and
ARTICLE IN PRESS
T. Daimon et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11121123 1113
D306NE310Q-R1, respectively. Recombinant AcNPVs
were propagated by following the suppliers protocol.
The correct construction of recombinant viruses was
conrmed by PCR and DNA sequencing.
2.3. Expression and purication of the BmChi-h protein
Monolayers of Sf9 cells cultured in a 150 mm dish
were infected with recombinant viruses at a multiplicity
of infection (MOI) of 5. After 72 h, the culture medium
was centrifuged at 2000 g for 5 min, and the supernatant
was collected. Protein purication was performed using
nickel chromatography. The medium was concentrated
from 300 to 20 ml using Centriprep YM-30 (Amicon),
dialyzed against a binding buffer (20 mM sodium
phosphate, pH 7.4, 500 mM NaCl, 10 mM imidazole)
overnight at 4 1C, and loaded onto a HiTrap column
(Amersham). After washing with 5 ml of a wash buffer
series including three buffers (20, 30, and 40 mM
imidazole), elution was performed with an elution buffer
containing 200 mM imidazole. The eluent was dialyzed
against a 20 mM sodium phosphate buffer (pH 7.4) and
stored at 4 1C or 80 1C until use. For N-terminal
sequencing, puried BmChi-h was blotted onto a PVDF
membrane (Immobilon-P, Millipore) and subjected to
amino acid sequencing using the HP G1005A protein
sequencing system (Hewlett-Packard).
2.4. Enzyme assay
Chitinolytic activity was measured using three kinds
of substrates: the glycol chitin, the native chitin
oligosaccharides [(GlcNAc)
16
, Wako], and the uor-
escent oligosaccharides, 4-methylumbelliferyl N-acetyl
oligosaccharides [MU-(GlcNAc)
13
, Sigma].
In the glycol chitin assay, chitinolytic activity
was detected using the method of Trudel and Asselin,
(1989). In the native chitin oligosaccharide assay, the
reaction (20 ml) containing puried BmChi-h (4500 ng)
and 10 mg of substrates [(GlcNAc)
16
] in a 50 mM
citrate/phosphate buffer (pH 6.2) was incubated for
various times at 37 1C. The hydrolysis products were
separated by thin-layer chromatography (TLC) using
a silica gel 60 plate (Merck) and a solvent system
of isopropyl alcohol:ethanol:water (5:2:1). The plates
were developed by spraying the plates with 10%
sulfuric acid in ethanol followed by heating at 120 1C
for 1020 min to detect dark spots. In the uorescent
substrate assay, the reaction (50 ml) contained 12.5 ml
of 500 mM substrate [MU-(GlcNAc)
13
], 12.5 ml of a
200 mM BrittonRobinsons wide range buffer (contain-
ing the same molar of acetic, phosphoric, and
boric acid and adjusted to desirable pH values with
NaOH), and 25 ml of enzyme (0.05 mg) diluted with
distilled water. After incubation for 210 min at
25 1C, 1.95 ml of a 0.15 M glycineNaOH buffer (pH
10.6) was added to stop the reaction. The released
uorescent methylumbelliferone (MU) was measured
using a uorescent spectrophotometer (FP-777, JASCO)
with excitation at 360 nm and emission at 450 nm.
The released MU was linearly increased with the
reaction time under the condition described above.
The enzyme activity unit was nmol product (MU)/mg
enzyme/min.
ARTICLE IN PRESS
Table 1
PCR primers used in this study
Name Nucleotide sequence (5
0
to 3
0
)
d-F1 GGKGGBGATGGCATCAAYGACAG
d-R1 AGTCHACHACDCCGTCYTGCCA
ScGSP1 GGTGGTCCGCAAGACGGTGATGTCTACGTG
ScGSP2 GCACGCCAAAGTCCGTCGTGTAGGTCTCT
ApGSP1 GCGGAGACTGGACGGAAATACGAACTCAC
ApGSP2 GGACGCCTTGTGTCAGAAGGAATCGTACTC
OfGSP1 GTAGGGCTCGTTCCAGGATGACAAACC
OfGSP2 GGGACAAGATCCAGGTCGTCGACTACAG
AcGSP1 GGGTTAGCTCCCTTACCACCAGGGAACTC
AcGSP2 CGTCGGTGTTGCGATGTATGGACGTGGT
PsGSP1 CTGAGGCAACGAAGCGGCTTCGCTTTGT
PsGSP2 CGTTGGTGTTGCCATGTACGGACGAGGCT
pBac1Chi-hF1 CGCGGATCCAGTTATGGGGCATCGGATGAT
pBac1Chi-hR1 GCTCTAGATCAATGGTGATGGTGATGATGACCAGCGGAATTTCCTAAGCCCAT
D306N-F1 CGATGGTGTCAACATTGATTGGGAGTTTCCAGGTG
D306N-R1 CACCTGGAAACTCCCAATCAATGTTGACACCATCG
E310Q-F1 GATGGTGTCGACATTGATTGGCAGTTTCCAGGTGG
E310Q-R1 CCACCTGGAAACTGCCAATCAATGTCGACACCATC
D306NE310Q-F1 CTTCGATGGTGTCAACATTGATTGGCAGTTTCCAGGTGGC
D306NE310Q-R1 GCCACCTGGAAACTGCCAATCAATGTTGACACCATCGAAG
2.5. Immunoblot and immunohistochemistry
The expression of recombinant proteins was analyzed
by SDS-PAGE (Laemmli, 1970) and Western blot. The
protein concentration was determined using the Coo-
massie Plus Protein Assay Reagent (Pierce) with bovine
serum albumin as a standard. Western blot was
performed as follows: after SDS-PAGE, the proteins
were blotted onto a nitrocellulose membrane (Hybond-
ECL, Amersham) using a semi-dry blotting apparatus
(Bio-Rad) at 2 mA/cm
2
for 30 min in a CAPS buffer
(10 mM CAPS, 0.3 mM DTT, 10% methanol, pH 10.6).
The membrane was incubated for 1 h in a TBS-T buffer
containing 2% skim milk (20 mM Tris-HCl pH 8.0,
150 mM NaCl, and 0.05% Tween 20) followed by
overnight incubation at 4 1C with the Penta-His anti-
body (QIAGEN) diluted 1:2000 in TBS-T. The blot was
incubated with the goat anti-mouse horseradish perox-
idase-conjugated secondary antibody (Zymed labora-
tory) diluted 1:3000 in TBS-T for 1 h at room
temperature. The blot was rinsed twice for 10 min in
TBS-T between incubation and before signal detection.
Signals were detected with the ECL Kit (Amersham)
using the LAS1000 Plus imaging system (Fuji Film).
An antiserum against recombinant BmChi-h was
raised in rabbits (QIAGEN). For western analysis,
frozen tissues from the silkworm strain F1 hybrid
Kinshu Showa were homogenized in a 10 mM
phosphate buffer containing a protein inhibitor cocktail
(Roche) using liquid nitrogen. Each protein sample
(2 mg) was electrophoresed by SDS-PAGE and blotted
onto a nitrocellulose membrane as described above.
An antiserum against BmChi-h was diluted 1:50000 in
TBS-T. Immunohistochemistry was performed as fol-
lows: larvae at the feeding stage (2 days of 4th-instar
larvae) and the larvallarval molting stage (5 days of
4th-instar larvae, in which head cap slippage was
observed) were xed with an FAA solution (4%
formaldehyde, 0.5% acetic acid, and 45% ethanol)
overnight at 4 1C. Samples were dehydrated through a
grade ethanol series, embedded in parafn, and then cut
into 10 mm thick serial sections. The primary antibody
was anti-BmChi-h at a dilution of 1:200, and the second
antibody was an AlexaFluor488-labeled goat anti-rabbit
IgG F(ab)
2
fragment (Molecular Probes) at a dilution of
1:200. The slides hydrated through ethanol series were
incubated with TBS-Ca (TBS containing 1 mM CaCl
2
)
for 5 min and then with a blocking buffer (TBS-Ca
containing 5% skim milk) for 1 h at 25 1C. After
blocking, the slides were incubated with the primary
antibody diluted in TBS-Ca containing 5% skim milk
for 1 h at 25 1C and then with the secondary antibody
diluted in TBS-Ca containing 5% skim milk at 25 1C
and subsequently counter-stained with a 4
0
, 6-diamidi-
no-2-phenylindole dihydrochloride solution (DAPI,
Wako) at a dilution of 1:1000 in TBS-Ca for 5 min.
Fluorescence was observed under a uorescence micro-
scope (Olympus BX51) and photographed.
2.6. Cloning of BmChi-h orthologues from lepidopteran
insects
Orthologous genes of BmChi-h were cloned from ve
species of lepidopteran insects (S. cynthia, A. pernyi, O.
furnacalis, A. convolvuli, and P. separata) by polymerase
chain reaction (PCR) followed by genome-walking PCR
(in S. cynthia and A. pernyi) or by 5
0
- and 3
0
-rapid
amplication of cDNA ends (RACE) (in O. furnacalis,
A. convolvuli, and P. separata). First, partial fragments
of BmChi-h orthologues were PCR-amplied using
ARTICLE IN PRESS
Fig. 1. Expression and purication of the His-tagged BmChi-h protein.
Recombinant His-tagged BmChi-h protein was expressed using a
baculovirus expression system. BmChi-h was puried from the medium
of virus-infected cells by nickel chromatography. Protein samples were
electrophoresed and stained with CBB (A), and the same samples were
analyzed by Western blot with the anti-His antibody (B). The
molecular weights of the protein standards are shown on the left,
and the estimated molecular weight of BmChi-h is shown with an
arrowhead. Samples are as follows: lane 1, medium of infected cells;
lane 2, proteins not binding to the nickel column; lanes 3 and 4, wash
fraction (30 mM and 40 mM imidazole, respectively); lanes 59, eluate
fraction (200mM imidazole); M, protein standard. (C) Sequence
comparisons between the N-terminal sequences of BmChi-h and the
puried protein. (D) Zymogram analysis of chitinolytic activity: lane 1,
BSA (1 mg); lane 2, S. marcescens chitinase (0.01 U; Sigma); lane 3,
puried BmChi-h (1 mg).
degenerate primer sets d-F1 and d-R1 (Table 1) with the
genomic DNA of each species as the template. Genomic
DNAs were prepared as described previously (Daimon
et al., 2003). Next, genome-walking PCR or 5
0
- and 3
0
-
RACE were performed using gene-specic primer sets
for each orthologue. The template total RNA for the
RACE experiment was extracted from the epidermis at
the molting stages as described (Daimon et al., 2003).
Genome-walking PCR was performed using the BD
Genome Walker Universal Kit (BD Biosciences),
and RACE experiments were performed using a BD
ARTICLE IN PRESS
Fig. 2. Enzymatic properties of recombinant BmChi-h protein. (A and
B) TLC analysis of the substrate specicity and reaction product
prole of BmChi-h using native chitin oligosaccharide substrates
[(GlcNAc)
16
]. Digestion products were analyzed by TLC. Lanes 16
contain the digestion products obtained after incubating puried
BmChi-h with the mono- to hexamer of GlcNAc, respectively, for 6 h
(A, full digestion) or 30 min (B, partial digestion). The positions of the
oligomers are indicated at right. (C) The TLC pattern for reaction
products of (GlcNAc)
6
. The reaction mixture (20 ml) containing 10 mg
of (GlcNAc)
6
in 50 mM citrate/phosphate buffer (pH 6.2) was
incubated with various amounts of puried BmChi-h for 15 min and
the reaction products were analyzed. Lanes 18 indicate 500, 250, 125,
62.5, 31.3, 15.6, 7.8, and 3.9 ng of the enzyme in the reaction,
respectively (1:1 dilution series). (D) Enzyme assays using uorescent
chitin substrates, MU-(GlcNAc)
n
(n 123). The enzyme activity unit
was nmol product (MU)/mg enzyme/min. The bars indicate the
mean7SD (n 3).
D D E
WT: 302 FDGVDIDWEFP 312
D306N: 302 FDGVNIDWEFP 312
E310Q: 302 FDGVDIDWQFP 312
D306NE310Q: 302 FDGVNIDWQFP 312
0
1
2
3
4
5
6
7
3 7 11
pH
WT
D306N
E310Q
D306NE310Q
WT
D306N
E310Q
D306NE310Q
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
V

(
n
m
o
l
/
g
/
m
i
n
)
V

(
n
m
o
l
/
g
/
m
i
n
)
9 5
3 7 11
pH
9 5
WT
D306N
E310Q
D306NE310Q
(B)
(C)
(A)
Fig. 3. pH proles of wt and mutant forms of BmChi-h. (A) Amino
acid sequences of active sites. The substituted amino acid residues are
shown in bold. (B and C) Comparison of pH proles for D306N (n),
E310Q (&), and D306NE310Q (x) with wt BmChi-h (K). Chitinase
activities were determined over the range of pH 4.011.0 using MU-
(GlcNAc)
2
(B) and MU-(GlcNAc)
3
(C) as substrates. The points
indicate the mean7SD (n 3).
SMART RACE cDNA Amplication Kit (BD Bio-
sciences) following the manufacturers protocol. Clon-
ing of PCR products and DNA sequencing were
performed as described previously (Daimon et al., 2003).
2.7. Sequence comparison and phylogenetic analysis
The nucleotide sequences of the ORFs of bacterial-
type chitinase genes were aligned using the CLUSTALX
program (Thompson et al., 1997), and the nucleotide
differences (K) at both synonymous (K
S
) and nonsynon-
ymous sites (K
N
) were calculated using the MEGA
program version 2.1 (Kumar et al., 2001). Phylogenetic
analysis was conducted as previously described (Daimon
et al., 2003). Briey, the amino acid sequences of
catalytic domains were aligned with the CLUSTALX
program (Thompson et al., 1997), and phylogenetic
trees were constructed with the neighbor-joining meth-
ods using the MEGA program version 2.1 (Kumar
et al., 2001).
ARTICLE IN PRESS
Fig. 4. Western blot and immunohistochemistry of BmChi-h. (A) Western blot of BmChi-h. Proteins were extracted from fatbody (FB), midgut
(MG), integument (I), hemolymph (H), and molting uid (MF) of B. mori larvae at the feeding and molting stages and analyzed by Western blot. (B,
C, and D) Immunohistochemistry of BmChi-h. Cross sections at the feeding (B) and molting stages (C and D) were used. The bright-green
uorescence indicates the presence of BmChi-h, and the nucleus is counter-stained with DAPI. M, muscles; EP, epidermal cells; T, trachea; NC, new
cuticle; OC, old cuticle; MF, molting uid. The bar indicates 50 mm.
2.8. Sequence deposition
The nucleotide sequences have been deposited in the
DDBJ/GenBank/EMBL database with the following
accession numbers: ApChi-h, AB201279; ScChi-h,
AB201280; OfChi-h, AB201281; AcChi-h, AB201282;
and PsChi-h, AB201283.
3. Results
3.1. Expression and purication of BmChi-h protein using
a baculovirus expression system
To characterize the substrate specicity, recombinant
His-tagged BmChi-h protein was expressed using a
baculovirus expression system (Fig. 1). The protein of
approximately 58 kDa was puried by nickel-chelating
chromatography (Fig. 1A). The presence of the His-tag
sequence was conrmed by Western blot (Fig. 1B). The
N-terminal sequence of this puried protein corre-
sponded to that of BmChi-h but started at the residues
of Gly 21, indicating that the signal peptide of the
puried protein was cleaved (Fig. 1C). Chitinase activity
was measured by zymogram analysis (Fig. 1D). After
electrophoresis in native polyacrylamide gels containing
the glycol chitin, the gels were incubated to allow
substrate hydrolysis by chitinases and then stained with
a uorescent dye that binds the substrate. Brightly
uorescent areas represent regions with no active
enzyme, whereas dark areas represent regions where
enzyme activity degraded the substrate. Chitinolytic
activity was observed in the puried protein and
S. marcescens chitinase (Sigma) whereas it was not
observed in the BSA (Fig. 1D), indicating that BmChi-h
possesses chitinolytic activity. Therefore, we used this
puried protein for further examination of the enzy-
matic properties.
3.2. Enzymatic properties of BmChi-h
The activity of BmChi-h was assessed with native
chitin oligosaccharide (GlcNAc)
16
and uorescent
chitin oligosaccharide MU-(GlcNAc)
13
(Fig. 2).
In the native chitin oligosaccharide assay (Fig. 2AC),
BmChi-h did not hydrolyze GlcNAc or (GlcNAc)
2
but
was able to hydrolyze longer substrates (GlcNAc)
36
(Fig. 2A and B). BmChi-h produced predominantly
(GlcNAc)
2
and very little GlcNAc (Fig. 2A and B).
Fig. 2C shows that BmChi-h primarily converts
(GlcNAc)
6
into (GlcNAc)
2
and (GlcNAc)
4
[subse-
quently converted to (GlcNAc)
2
]. These results indicate
that BmChi-h is an exochitinase. However, a small
amount of (GlcNAc)
3
was detected in the digestion
product of (GlcNAc)
6
(Fig. 2B and C), suggesting that
BmChi-h possesses some endochitinase activity.
The uorescent chitin oligosaccharide assay also
showed that BmChi-h did not have activity against
MU-(GlcNAc) but hydrolyzed MU-(GlcNAc)
23
(Fig. 2D). BmChi-h showed nine fold higher activity
against MU-(GlcNAc)
2
than MU-(GlcNAc)
3
, indicating
that BmChi-h is an exochitinase. This exo-type substrate
preference was similar to that of bacterial and baculo-
virus chitinases (Hawtin et al., 1995; Brurberg et al.,
1996; Shinoda et al., 2001) and different from BmChi-
tinase, which showed endo-type substrate preference
(Kramer and Muthukrishnan, 1997; Shinoda et al.,
2001; Merzendorfer and Zimoch, 2003).
The pH proles of BmChi-h were determined using
uorescent chitin oligosaccharide MU-(GlcNAc)
23
as
substrates (Fig. 3). To do this, the amino acid residues of
the chitinase active site were substituted as shown in
Fig. 3A. In the MU-(GlcNAc)
2
assay, it was shown that
wt BmChi-h kept 50% activity in a broad pH range
(4.08.0) with a pH optimum of 6.0 (Fig. 3B). In
contrast, the mutant D306N showed very low activity at
the acidic pH range and no activity at the neutral and
alkaline pH ranges. Neither E310Q nor D306NE310Q
showed any activity at any pH values. Similar results
were obtained in the MU-(GlcNAc)
3
assay (Fig. 3C).
3.3. Expression and localization of BmChi-h in vivo
Previously, Northern blot analysis demonstrated that
BmChi-h mRNA is expressed specically at the molting
stages and in chitin-containing tissues (Daimon et al.,
2003). To further investigate, the expression and
localization of the BmChi-h protein were examined at
the larval feeding stage and the larvallarval molting
stage. Western blot analysis using some tissues at the
ARTICLE IN PRESS
96
98
98
52
89 %
89 %
89 %
73 %
85 %
85 %
Homology
555
555
555
563
555
553
555
Size
(aa)
Difference (K)
K
N KS
0.102 0.681
0.108 0.749
0.0749 0.686
0.0642 0.739
0.0704 0.797
(Identity)
(vs. BmChi-h)
ApChi-h (Antheraea pernyi )
ApChi-h (Agrius convolvuli )
ScChi-h (Samia cynthia)
BmChi-h (Bombyx mori )
PsChi-h (Pseudaletia separata)
OfChi-h (Ostrinia furnacalis)
Serratia marcescens ChiA
Fig. 5. Sequence comparisons of bacterial-type chitinases from
lepidopteran insets and S. marcescens chiA. A phylogenetic tree was
constructed using the total protein sequences. The size of the predicted
protein (aa) and the sequence homology (%) to BmChi-h are shown.
The nucleotide differences at both synonymous (K
S
) and nonsynon-
ymous sites (K
N
) were calculated. Only the nucleotide differences to
BmChi-h are shown here.
ARTICLE IN PRESS
Fig. 6. Alignment of the amino acid sequences of bacterial-type chitinases. BmChi-h orthologues were cloned from S. cynthia, A. pernyi, A.
convolvuli, P. separata, and O. furnacalis and designated as ScChi-h, ApChi-h, AcChi-h, PsChi-h, and OfChi-h, respectively. (A) Total amino acid
sequences were aligned with the ClustalX program and shaded using the Boxshade program. Serratia marcescens chiA (SwissProt: P07254) and
AcNPV chiA (SwissProt: P41684) were included for comparison. The positions at the degenerate primer are shown by arrow and the putative active
site is underlined.
larval feeing stage and the larvallarval molting stage
showed that BmChi-h was expressed strongly in the
epidermis and molting uid during the molting stage but
was not expressed during the feeding stage (Fig. 4A). In
addition, an immunohistochemical study showed that
no signals were detected at the feeding stage (Fig. 4B),
whereas apparent signals were observed at the larval
larval molting stage (Fig. 4C and D). Strong signals for
BmChi-h were observed in old cuticle and molting uid,
and weak signals were observed in the cuticle of the
trachea. Taken together with the results of enzymolo-
gical studies, these results strongly indicated that
BmChi-h plays a role in chitin degradation at the
molting stages.
3.4. Cloning and sequence comparisons of the bacterial-
type chitinase genes from lepidopteran insects
To investigate whether bacterial-type chitinase genes
were present in the genome of lepidopteran insects other
than B. mori, we tried to clone the BmChi-h orthologues
from ve species of lepidopteran insects (Anthraea
pernyi, Saturniidae; Samia cynthia, Saturniidae; Agrius
convolvuli, Sphingidae; Pseudaletia separata, Noctuidae;
and Ostrinia furnacalis, Pyralidae). By 5
0
- and 3
0
-RACE
(A. convolvuli, P. separata, and O. furnacalis) or
genome-walking PCR (A. pernyi and S. cynthia),
bacterial-type chitinase genes were successfully cloned
from all ve lepidopteran species (Fig. 5).
Their predicted amino acid sequences shared exten-
sive identities with that of BmChi-h (85%89%) (Fig. 5).
The nucleotide differences at synonymous sites (K
S
)
were shown to be saturated ( 0.75) between chitinases
from different families. In contrast, the nucleotide
differences at non-synonymous sites (K
N
) were less than
0.1, suggesting that nucleotide substitutions at synon-
ymous sites have been limited by functional constraints
(Fig. 5). All of them were predicted to be a secretory
protein by the PSORT program (Nakai and Kanehisa,
1992). Domain structures including the active site
signature of family 18 chitinases (Henrissat, 1991,
1999) were also highly conserved (Fig. 6). Comparison
of domain structures showed that domain structures of
these bacterial-type chitinases were identical to that
of BmChi-h (Figs. 6 and 7). The genomic structure of
BmChi-h has shown that there was no intron in BmChi-
h, while three isoforms containing different 5
0
-UTR
generated from different promoters were present (Dai-
mon et al., 2003). The genome structures of all the
bacterial-type chitinase genes cloned in this report
seemed to be similar to that of BmChi-h. We found no
introns but identied several isoforms in 5
0
-UTR of
PsChi-h and OfChi-h (data not shown). Although the
biological signicance of 5
0
-UTR isoforms is not clear,
these data indicate that not only the encoded protein but
also the genome structure of the bacterial-type chitinase
gene are highly conserved among lepidopteran insects.
Phylogenetic analysis showed that these chitinase
genes of lepidopteran insects and baculoviruses be-
longed to bacterial lineages with a strong bootstrap
support, suggesting that they originated from bacteria
and have been maintained through evolution since they
transferred laterally (Fig. 8).
4. Discussion
In the silkworm genome project, a large-scale cDNA
database has been constructed (Mita et al., 2003), and
whole genome sequencing of B. mori has been
performed (Mita et al., 2004). An interesting nding is
that there are genes showing signicant homology with
bacterial or baculoviral genes (Mita et al., 2003;
Goldsmith et al., 2004). The BmChi-h gene, a bacter-
ial-type chitinase gene of B. mori, is one of them
(Daimon et al., 2003). In this study, we demonstrated
that BmChi-h encodes an exochitinase and plays a role
in chitin degradation at the molting stages.
Chitinases (EC 3.2.1.14) can be classied into two
major categories: endochitinases and exochitinases
(Robbins et al., 1988). Endochitinases cleave glycosidic
linkages randomly along the chitin chain and eventually
produce short oligomers of GlcNAc. Exochitinases, also
called chitobiosidase, cleave off chitobiose [(GlcNAc)
2
]
from (GlcNAc)
n
. The recombinant BmChi-h expressed
by a baculovirus expression system showed that BmChi-
h is a chitinolytic enzyme and a secretory protein
(Fig. 1). We propose that BmChi-h is an exochitinase for
the following reasons: (1) BmChi-h does not hydrolyze
(GlcNAc)
2
and MU-(GlcNAc), indicating that BmChi-h
is not a 14-b-N-acetylglucosaminidase (Fig. 2A, B
and D); (2) BmChi-h hydrolyzes MU-(GlcNAc)
2
faster
than MU-(GlcNAc)
3
(Fig. 2D); (3) when (GlcNAc)
6
is
used as a substrate, BmChi-h produces predominantly
(GlcNAc)
4
and (GlcNAc)
2
as an initial product
ARTICLE IN PRESS
Fig. 7. Schematic representation of the domain structures of bacterial-
type chitinases (BmChi-h and its orthologues) and endochitinases of
lepidopteran insects (BmChitinase and its orthologues).
ARTICLE IN PRESS
OpNPV
AcNPV
XcGV
ApChi-h
ScChi-h
AcChi-h
BmChi-h
OfChi-h
PsChi-h
S. marcescens ChiA
E. agglomerans
A. hydrophila
Alteromonas sp.
V. cholerae
S. plicatus
B. circulans
S. marcescens ChiB
C. immitis
A. album
T. harzianum
O. aries
B. taurus
H. sapiens
M. musculus
B. malayi
C. elegans
A. aegypti
B. mori
M. sexta 100
100
100
100
42
90
69
100
69
100
96
86
46
99
100
100
100
80
95
82
93
99
100
76
96
65
0.1
Baculoviruses
Lepidopteran
insects
Bacteria
Fungi
Animals
Fig. 8. Phylogenetic analysis of family 18 chitinases. The amino acid sequences of the catalytic domain (about 400 aa) were extracted and aligned
with the ClustalX program, and pair-wise distances were calculated based on the blosum matrix. The phylogenetic tree was constructed by the
neighbor-joining method using the MEGA2 program package. Bootstrap values after 1000 replications are shown. Sequence used for analysis:
Autographa californica nuclear polyhedrosis virus (SwissProt: P41684), Orgyia pseudotsugata multicapsid polyhedrosis virus (SwissProt: O10363),
Xestia c-nigrum granulovirus (AF162221), Serratia marcescens ChiA (SwissProt: P07254), Enterobacter agglomerans (U59304), Aeromonas hydrophila
(AF251793), Alteromonas sp. (SwissProt: P32823), Vibrio cholerae (PIR: D82510), Streptomyces plicatus Chi63 (SwissProt: P11220), Bacillus circulans
ChiA1 (SwissProt: P20533), S. marcescens ChiB (SwissProt: P11797), Coccidioides immitis (SwissProt: P54196), Aphanocladium album (SwissProt:
P32470), Trichoderma harzianum (SwissProt: P48827), Caenorhabditis elegans CHT-1 (SwissProt: Q11174), Brugia malayi (SwissProt: P29030), Mus
musculus CHI3L3 (SwissProt: Q61362), Ovis aries (SwissProt: Q28542), Bos taurus (SwissProt: Q28042), Homo sapiens CHI3L2 (SwissProt: Q15782),
Aedes aegypti (AF026491), M. sexta (SwissProt: P36362), and B. mori (AB048355).
(Fig. 2C). This exo-type substrate preference is the same
as those of S. marcescens chiA and baculovirus
chitinases (Hawtin et al., 1995; Brurberg et al., 1996;
Shinoda et al., 2001) but different from that of
BmChitinase, which showed endo-type substrate pre-
ference (Kramer and Muthukrishnan, 1997; Shinoda et
al., 2001; Merzendorfer and Zimoch, 2003). The TLC
pattern for the reaction products of (GlcNAc)
6
showed
minor (GlcNAc)
3
spots in the course of the enzyme
reaction (Fig. 2B and C). This suggests that BmChi-h is
an exochitinase with some endochitinase activity.
Similarly, both S. marcescens chiA and baculovirus
chitinase have been demonstrated to possess some
endochitinase activity (Brurberg et al., 1996; Hawtin et
al., 1995). The chitinase gene has been cloned from
many lepidopteran insects and showed endo-type sub-
strate preference in all cases (Kramer et al., 1993;
Kramer and Muthukrishnan, 1997; Kim et al., 1998;
Mikitani et al., 2000; Shinoda et al., 2001; Zheng et al.,
2002; Merzendorfer and Zimoch, 2003). Therefore, this
is the rst report to describe the molecular cloning and
characterization of exochitinase from lepidopteran
insects.
In our previous studies, BmChi-h mRNA was
observed in a tissue- and stage-specic manner (Daimon
et al., 2003). BmChi-h mRNA was expressed specically
at chitin-containing tissues during the molting stage, and
its expression pattern was almost identical to that of
BmChitinase (Daimon et al., 2003). In Western blot
analysis, the BmChi-h protein was observed in the
integument and molting uid at the molting stage
(Fig. 4A), consistently with the result of Northern blot
analysis (Daimon et al., 2003). Immunohistochemical
analysis revealed that BmChi-h localized in the old
cuticle, molting uid, and trachea (Fig. 4BD). This
spatial and temporal expression pattern seems to be
closely similar to that of CfChitinase, a BmChitinase
orthologue from Choristoneura fumiferana (Zheng et al.,
2003). These data strongly indicate that both BmChi-h
and BmChitinase degrade the chitin of the exoskeleton
during the molting stage. As a strong synergic effect was
observed when chitinolytic enzymes of different sub-
strate specicity were combined (Fukamizo and Kra-
mer, 1985a, b; Brurberg et al., 1996; Suzuki et al., 2002),
BmChi-h (exochitinase) may catalyze the native chitin by
a concerted mechanism with BmChitinase (endochiti-
nase) and b-N-acetylglucosaminidase (EC 3.2.1.30)
(Fukamizo and Kramer, 1985a, b). Taken together,
these ndings lead to the conclusion that the biological
role of the bacterial-type chitinases of lepidopteran
insects is chitin degradation for molting.
To see whether the bacterial-type chitinase is encoded
by lepidopteran insects other than B. mori, we cloned
the BmChi-h orthologues from ve species of lepidop-
teran insects. The BmChi-h orthologues were success-
fully cloned from all species examined (Fig. 5). Sequence
analysis strongly indicated that all of them encoded
functional enzymes belonging to the family of 18
chitinases (Figs. 5 and 6). Since BmChi-h orthologues
were cloned from species of the superfamily Noctuoidea
(P. separata) and Pyraloidea (O. furnacalis), which are
phylogenetically distant from Bombycoidea, it is likely
that the bacterial-type chitinase gene is encoded by
broad taxa of lepidopteran insects. In the phylogenetic
analysis, the bacterial-type chitinase genes of lepidop-
teran lineages were shown to be monophyletic (Fig. 7).
Therefore, an ancestral lepidopteran species might have
acquired the chitinase gene from a bacterium or a
baculovirus.
The nucleotide substitutions at the ORFs of the
bacterial-type chitinase gene of lepidopteran insects
have already been saturated (Fig. 6B). Therefore, it
seems difcult to obtain denitive evidence for hor-
izontal transfer at this point. One possible way would be
to investigate the species distribution of bacterial-type
chitinase genes in the genome of broader taxa of
arthropods and viruses. Accumulating genome data of
microbes and viruses and, especially, the complete
genome sequences of honeybee and Tribolium, which
will be available in a few years, would provide further
clues in this respect.
Acknowledgements
We are grateful to K. Kiguchi (Shinshu University,
Japan), S. Matsumoto (RIKEN, Japan), and S. Tatsuki
(Univ. of Tokyo) for providing specimens of
A. convolvuli, P. separata, and O. furnacalis, respec-
tively. We also thank N. Omuro (Univ. of Tokyo) for
her technical assistance. This work was nancially
supported by PROBRAIN (T.S.), the Insect Technology
Project, MAFF (T.S.), and Grants-in-Aid for Scientic
Research, MEXT/JSPS (T.S. and T.D.).
References
Brurberg, M.B., Nes, I.F., Eijsink, V.G., 1996. Comparative studies of
chitinases A and B from Serratia marcescens. Microbiology 142,
15811589.
Daimon, T., Hamada, K., Mita, K., Okano, K., Suzuki, M.G.,
Kobayashi, M., Shimada, T., 2003. A Bombyx mori gene, BmChi-h,
encodes a protein homologous to bacterial and baculovirus
chitinases. Insect Biochem. Mol. Biol. 33, 749759.
de la Vega, H., Specht, C.A., Liu, Y., Robbins, P.W., 1998. Chitinases
are a multi-gene family in Aedes, Anopheles and Drosophila. Insect
Mol. Biol. 7, 233239.
Fukamizo, T., Kramer, K.J., 1985a. Mechanism of chitin oligosac-
charide hydrolysis by the binary enzyme chitinase system in insect
molting uid. Insect Biochem. 15, 17.
Fukamizo, T., Kramer, K.J., 1985b. Mechanism of chitin hydrolysis
by the binary chitinase system in insect molting uid. Insect
Biochem. 15, 141145.
ARTICLE IN PRESS
Goldsmith, M.R., Shimada, T., Abe, H., 2004. The genetics and
genomics of the silkworm, Bombyx mori. Annu. Rev. Entomol. 50,
71100.
Gooday, G.W., 1999. Aggressive and defensive roles for chitinases. In:
Jolles, P., Muzzarelli, R.A.A. (Eds.), Chitin and Chitinases.
Birkhauser Verlag, Basel, Switzerland, pp. 157169.
Handler, A.M., 2002. Use of the piggyBac transposon for germ-line
transformation of insects. Insect Biochem. Mol. Biol. 32,
12111220.
Hawtin, R.E., Arnold, K., Ayres, M.D., Zanotto, P.M., Howard, S.C.,
Gooday, G.W., Chappell, L.H., Kitts, P.A., King, L.A., Possee,
R.D., 1995. Identication and preliminary characterization of a
chitinase gene in the Autographa californica nuclear polyhedrosis
virus genome. Virology 212, 673685.
Hawtin, R.E., Zarkowska, T., Arnold, K., Thomas, C.J., Gooday,
G.W., King, L.A., Kuzio, J.A., Possee, R.D., 1997. Liquefaction of
Autographa californica nucleopolyhedrovirus-infected insects is
dependent on the integrity of virus-encoded chitinase and cathepsin
genes. Virology 238, 243253.
Henrissat, B., 1991. A classication of glycosyl hydrolases based on
amino acid sequence similarities. Biochem. J. 280, 309316.
Henrissat, B., 1999. Classication of chitinases modules. In: Jolles, P.,
Muzzarelli, R.A.A. (Eds.), Chitin and Chitinases. Birkhauser
Verlag, Basel, Switzerland, pp. 137156.
Hughes, A.L., Friedman, R., 2003. Genome-wide survey for genes
horizontally transferred from cellular organisms to baculoviruses.
Mol. Biol. Evol. 20, 979987.
Jehle, J.A., Nickel, A., Vlak, J.M., Backhaus, H., 1998. Horizontal
escape of the novel Tc1-like lepidopteran transposon TCp3.2 into
Cydia pomonella granulovirus. J. Mol. Evol. 46, 215224.
Kamita, S.G., Nagasaka, K., Chua, J.W., Shimada, T., Mita, K.,
Kobayashi, M., Maeda, S., Hammock, B.D., 2005. A baculovirus-
encoded protein tyrosine phosphatase gene induces enhanced
locomotory activity in a lepidopteran host. Proc. Natl. Acad. Sci.
USA 102, 25842589.
Kim, M.G., Shin, S.W., Bae, K.S., Kim, S.C., Park, H.Y., 1998.
Molecular cloning of chitinase cDNAs from the silkworm, Bombyx
mori and the fall webworm, Hyphantria cunea. Insect Biochem.
Mol. Biol. 28, 163171.
Kramer, K.J., Muthukrishnan, S., 1997. Insect chitinases: molecular
biology and potential use as biopesticides. Insect Biochem. Mol.
Biol. 27, 887900.
Kramer, K.J., Corpuz, L., Choi, H.K., Muthukrishnan, S., 1993.
Sequence of a cDNA and expression of the gene encoding
epidermal and gut chitinases of Manduca sexta. Insect Biochem.
Mol. Biol. 23, 691701.
Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. MEGA2:
molecular evolutionary genetics analysis software. Bioinformatics
17, 12441245.
Laemmli, U.K., 1970. Cleavage of structural proteins during assembly
of head of bacteriophage-T4. Nature 227, 680685.
Lerch, R.A., Friesen, P.D., 1992. The baculovirus-integrated retro-
transposon TED encodes gag and pol proteins that assemble into
viruslike particles with reverse transcriptase. J. Virol. 66,
15901601.
Maeda, S., 1989. Gene tranfer vectors of a baculovirus, Bombyx mori
nucleopolyhedrovirus, and their use for expression of foreign genes
in insect cells. In: Mitsuhasi, J. (Ed.), Invertabrate Cell System
Application. CRC Press, Boca Raton, pp. 167181.
Merzendorfer, H., Zimoch, L., 2003. Chitin metabolism in insects:
structure, function and regulation of chitin synthases and
chitinases. J. Exp. Biol. 206, 43934412.
Mikitani, K., Sugasaki, T., Shimada, T., Kobayashi, M., Gustafsson,
J.A., 2000. The chitinase gene of the silkworm, Bombyx mori,
contains a novel Tc-like transposable element. J. Biol. Chem. 275,
3772537732.
Miller, D.W., Miller, L.K., 1982. A virus mutant with an insertion of a
copia-like transposable element. Nature 299, 562564.
Mita, K., Morimyo, M., Okano, K., Koike, Y., Nohata, J., Kawasaki,
H., Kadono-Okuda, K., Yamamoto, K., Suzuki, M.G., Shimada,
T., Goldsmith, M.R., Maeda, S., 2003. The construction of an EST
database for Bombyx mori and its application. Proc. Natl. Acad.
Sci. USA 100, 1412114126.
Mita, K., Kasahara, M., Sasaki, S., Nagayasu, Y., Yamada, T.,
Kanamori, H., Namiki, N., Kitagawa, M., Yamashita, H.,
Yasukochi, Y., Kadono-Okuda, K., Yamamoto, K., Ajimura,
M., Ravikumar, G., Shimomura, M., Nagamura, Y., Shin, I.T.,
Abe, H., Shimada, T., Morishita, S., Sasaki, T., 2004. The genome
sequence of silkworm, Bombyx mori. DNA Res. 11, 2735.
Nakai, K., Kanehisa, M., 1992. A knowledge base for predicting
protein localization sites in eukaryotic cells. Genomics 14, 897911.
OReilly, D.R., 1997. Auxiliary genes of baculoviruses. In: Miller, L.K.
(Ed.), The Baculoviruses. Plenum, New York, pp. 267300.
Robbins, P.W., Albright, C., Beneld, B., 1988. Cloning and
expression of a Streptomyces plicatus chitinase (chitinase-63) in
Escherichia coli. J. Biol. Chem. 263, 443447.
Shinoda, T., Kobayashi, J., Matsui, M., Chinzei, Y., 2001. Cloning
and functional expression of a chitinase cDNA from the common
cutworm, Spodoptera litura, using a recombinant baculovirus
lacking the virus-encoded chitinase gene. Insect Biochem. Mol.
Biol. 31, 521532.
Suzuki, K., Sugawara, N., Suzuki, M., Uchiyama, T., Katouno, F.,
Nikaidou, N., Watanabe, T., 2002. Chitinases A, B, and C1 of
Serratia marcescens 2170 produced by recombinant E. coli:
Enzymatic properties and synergism on chitin degradation. Biosci.
Biotech. Biochem. 66, 10751083.
D.G., 1997. The CLUSTAL_X windows interface: Flexible
tools. Nucleic Acids Res. 25, 48764882.
Trudel, J., Asselin, A., 1989. Detection of chitinase activity after
polyacrylamide gel electrophoresis. Anal. Biochem. 178, 362366.
Zheng, Y., Zheng, S., Cheng, X., Ladd, T., Lingohr, E.J., Krell, P.J.,
Arif, B.M., Retnakaran, A., Feng, Q., 2002. A molt-associated
chitinase cDNA from the spruce budworm, Choristoneura fumifer-
ana. Insect Biochem. Mol. Biol. 32, 18131823.
Zheng, Y.P., Retnakaran, A., Krell, P.J., Arif, B.M., Primavera, M.,
Feng, Q.L., 2003. Temporal, spatial and induced expression
of chitinase in the spruce budworm, Choristoneura fumiferana.
J. Insect Physiol. 49, 241247.
Zhu, Q., Deng, Y., Vanka, P., Brown, S.J., Muthukrishnan, S.,
Kramer, K.J., 2004. Computational identication of novel
chitinase-like proteins in the Drosophila melanogaster genome.
Bioinformatics 20, 161169.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Effect of chloroquine on the expression of genes involved in the
mosquito immune response to Plasmodium infection
P. Abrantes
, L.F. Lopes, V.E. do Rosa rio, H. Silveira

Centro de Malaria e Outras Doenc- as Tropicais/ UEI de Malaria, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa, Rua da
Junqueira, 96, 1349-008 Lisboa, Portugal
Received 7 December 2004; received in revised form 11 May 2005; accepted 17 May 2005
Abstract
Chloroquine has been described to increase Plasmodium infectivity to the mosquito vector and is known to affect the vertebrate
host immune response including during malarial infection. Although knowledge of the mosquito immune response has recently
improved, nothing is known about the impact of chloroquine on mosquito immunity. In order to characterize the inuence of
chloroquine on the mosquito immune system, we have analyzed the effect of chloroquine on Anopheles gambiae (i) serine proteases
and (ii) antimicrobial peptide gene expression, in uninfected and Plasmodium berghei infected mosquitoes, using real-time PCR. We
have demonstrated for the rst time that mosquitoes fed on chloroquine-treated mice showed a signicant down regulation of some
immune-related genes. This effect was independent of midgut bacterial burden. These results suggest that chloroquine might act on
the Anopheles serine proteases cascade, interfering with signal transduction pathways and at a transcriptional activation level.
Keywords: Chloroquine; Anopheles gambiae; Immune response; Real-time PCR; Plasmodium berghei
1. Introduction
Malaria is a vector-borne disease that still remains, to
our days, as one of the major causes of mortality
worldwide. The Plasmodium parasite life cycle is
complex and involves two hosts: a vertebrate host and
a mosquito vector. In Africa, where most deaths occur,
malaria is transmitted mainly by members of the
Anopheles gambiae complex. Strategies aimed at inter-
rupting parasite transmission by the mosquito vector are
thus an essential component of malaria control.
During its passage through the mosquito, the parasite
triggers a robust innate immune response against
malaria infection (Dimopoulos et al., 2001), that
in some cases is capable of controlling parasite
development.
The innate immune response of the mosquito consists
of humoral and cellular defense mechanisms. Humoral
immune responses involve the recognition of micro-
organisms by pattern recognition receptors (PRRs)
molecules, resulting in the activation of serine protease
cascades that in turn can activate defense reactions such
as melanotic encapsulation or initiate intracellular
immune signaling pathways which regulate the tran-
scription of antimicrobial peptide (AMP) genes (Dimo-
poulos, 2003). An increasing number of serine proteases
and antimicrobial peptides have been isolated and
characterized in A. gambiae. Changes in their transcript
abundance in response to bacterial and Plasmodium
infection have been demonstrated (Richman et al., 1997;
Dimopoulos et al., 1997, 1998; Gorman et al., 2000;
Vizioli et al., 2000, 2001). Components of the blood
meal are also known to interfere with parasite develop-
ment in the mosquito vector. Presence of the antimalar-
ial drug chloroquine has been associated with an
increase in infectivity of Plasmodium parasites to the
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.003
Corresponding author. Tel.: +351 213622458;

fax: +351 213622458.
E-mail address: patriciaabrantes@ihmt.unl.pt (P. Abrantes).
mosquito. Mosquitoes fed with Plasmodium falciparum
and P. berghei gametocytes in the presence of serum
from chloroquine-treated human volunteers showed a
45-fold increase in the mean oocyst number (Hogh et
al., 1998). The same effect was also observed in
infections with P. yoelii nigeriensis N67, which displayed
enhanced infectivity to mosquitoes when mice were
treated with chloroquine 12 h before feeding (Ichimori et
al., 1990). In these studies, enhanced infectivity to the
mosquito was not associated with increased gametocyte
number, suggesting that it was caused by the inter-
ference of chloroquine with the capability of the
mosquito to respond to the infection. Several studies
have also reported the effect of chloroquine on the
vertebrate host immune response during malaria infec-
tion (e.g. Rosa et al., 1999). Moreover chloroquine is
known to (i) inhibit intralysosomal degradation of
proteins (Schultz and Gilman, 1997); (ii) enhance fungal
protease activity (Staszczak et al., 2000); (iii) block the
nuclear factor-kappaB (NF-kB) induction by Interleu-
kin-1b in epithelial cells [IL-1 receptor belongs to super
family TLR (toll like receptor) (Bonizzi et al., 1997)],
(iv) inhibit tumor necrosis factor a (TNF-a) release from
human and murine cells (Jeong and Jue, 1997), (v)
inhibit nitric oxide synthase expression in murine
macrophages (Park et al., 1999) and (vi) inhibit pH-
dependent steps of the replication of several viruses
(Savarino et al., 2003).
In this study, we have characterized, for the rst time,
the effect of chloroquine on the mosquito immune
response at two levels: (i) serine proteases and (ii)
antimicrobial peptide gene expression. For that, seven
A. gambiae gene products were analyzed: four serine
proteases (Sp14D2, AgSP24D, ISPL5 and 1644280),
and three antimicrobial peptides (gambicin, defensin and
cecropin).
2. Materials and Methods
2.1. Mosquitoes
A. gambiae s.s. (Suakoko strain) mosquitoes were
reared at 25 1C and 75% humidity with a 12-h light/dark
cycle. Adult mosquitoes were maintained on a 10%
glucose solution.
2.2. Chloroquine treatment
Female BALB/c mice (Mus musculus) were injected
intraperitoneally (i.p.) with two doses of chloroquine (10
and 50 mg/kg), 17 h prior mosquito blood feeding.
Female mosquitoes, 56 days post-emergence and
deprived from sugar for 12 h prior feeding, were fed
on chloroquine treated and untreated mice for 2 h.
Unfed mosquitoes were discarded.
Mosquitoes were collected 24 h post-blood feeding
(p.bf.). Batches of 35 midguts were dissected in cold
DEPC-treated phosphate-buffered saline (PBS) and
processed for RNA preparation. A set of unfed female
mosquitoes was used as control. Six independent
experiments were performed.
2.3. Midgut bacterial quantication
In order to conrm that chloroquine had no
bactericidal effect on midgut bacteria, one dose
(50 mg/kg) of chloroquine was administered i.p. in mice
and the same procedure of chloroquine treatment was
followed.
Batches of 50 treated and untreated mosquito midguts
were externally sterilized in 70% ethanol for 5 min,
rinsed in PBS and dissected under sterile conditions.
Pools of ve midguts were homogenized in 50 ml of
PBS. Several dilutions were made and 10 ml of each
dilution was streaked on Columbia III Agar with 5%
Sheep Blood (Becton Dickinson, MD, USA). Plates
were incubated at 25 1C under aerobic conditions.
Bacterial counts (colony-forming units [cfu]/midgut)
were performed after 48 h in three independent
experiments.
2.4. Antibiotic treatment
The large spectrum sulfonamide antibiotic Sulfutrim
s
(Trimethoprim/Dimerazol) from HELSINN was used
to test the effect of midgut bacterial reduction by
antibiotic treatment in the immune response of A.
gambiae. One dose T/D (2.5/12.5 mg Kg
1
) was admi-
nistered i.p. in mice and the same procedure as for
chloroquine treatment was followed. Three independent
experiments were performed.
2.5. Plasmodium berghei infection and chloroquine
treatment
Female BALB/c mice were inoculated intraperitonally
with 10
7
P. berghei Anka parasitized red blood cells per
mililiter. The course of infection was determined by
Giemsa-stained blood lms prepared from tail blood.
When the parasitaemia reached 5%, two doses of
chloroquine (10 and 50 mg/kg), were administered
intraperitonally, as previously described.
Seventeen hours after chloroquine treatment, game-
tocitaemia and parasite exagellation were conrmed
and 56 days post-emergence female mosquitoes, were
naturally fed on chloroquine treated and untreated
infected mice. Unfed females were used as control.
Mosquito were collected 24 h post-blood feeding and
batches of 35 midguts were dissected in cold DEPC-
treated PBS.
ARTICLE IN PRESS
P. Abrantes et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11241132 1125
The remaining mosquitoes were maintained thereafter
at 19/21 1C to allow P. berghei development. Three
independent experiments were performed.
2.6. RNA extraction and cDNA synthesis
Total RNA from each group of mosquitoes, was
extracted with TRIzol
s
(Invitrogen- Life Technologies,
Barcelona, Spain) following the manufacturers instruc-
tions. For cDNA synthesis, 1 mg of total RNA was
treated with DNAse I (Invitrogen- Life Technologies)
and reverse transcribed with M-MLV-RT (Invitrogen-
Life Technologies) in the presence of oligo(dt)15 primer
(Roche Molecular Biochemicals).
2.7. Quantitative real-time PCR analysis
Quantitative real-time PCR was performed to analyze
relative levels of A. gambiae immune response gene
transcripts after chloroquine treatment. Sequences from
insect serine proteases (Sp14D2, AgSP24D, ISPL5, and
1644280) and antimicrobial peptides (gambicin, defensin
and cecropin) previously described to be involved in
immune response against Plasmodium infection (Dimo-
poulos et al., 1997; Gorman et al., 2000, Vizioli et al.,
2000, 2001) were obtained from GenBank (accession
numbers: AF117749, U21917, AJ000675, Z69978,
AJ237664, AF063402 and AF200686, respectively) and
used to generate specic primers. The sequences of
the primers used in the experiments were: Sp14D2-F
(5
0
-TGGGGCCAGACGGAAAACAGT-3
0
) and Sp14
D2-R (5
0
-CCGCGGCACGAGTCCTTACCC-3
0
); AgSP
24D-F (5
0
-TGGCCCGAGTAATAACGCACGAG-3
0
)
and AgSP24D-R (5
0
-TACATACGCCCCAGCCCGAG
ATAA-3
0
); ISPL5-F (5
0
-CTTAACAACATTGCCGTGC
TGGAG-3
0
) and ISPL5-R (5
0
- ATATCTGCGTCGGTG
GTGCGTTCT-3
0
); 1644280-F (5
0
-GCCGGCGAGCAC
GACTTCAG-3
0
) and 1644280-R (5
0
-CGGTTCCGGCA
GCGAGAC-3
0
); gambicin-F (5
0
-GCATCGGGGCACG
CTACTGT-3
0
) and gambicin-R (5
0
-GGTCCTGCCGAT
GATGGT TCC-3
0
); defensin-F (5
0
-GTACCATTGCCG
TTGTGCTG -3
0
) and defensin-R (5
0
- GATAGCGGC
GAGCGATACAG-3
0
); cecropin-F (5
0
-CAACCCAGA
GACCAACCAACCAC-3
0
) and cecropin-R (5
0
-ACTGC
CAGCACGACAAAGATGAAG-3
0
).
Quantitative analysis of the expression of these genes
was done by real-time PCR with qPCR core kit for
SYBR Green (Eurogentec, S.A., Seraing, Belgium)
using a Gene Amp 5700 Sequence Detector (Applied
Biosystems). Final concentrations of reagents were
1 reaction buffer, 3.5 mM Mgcl
2
, 200 mM dNTPs,
0.3 mM of primer concentrations, 0.025 U/ml of Hot
GoldStar enzyme and 1/66,000 of Sybr green for a nal
volume of 20 ml. One microliter (ca.0.02%) of cDNA
was used as template. Cycle conditions were: an initial
denaturation step at 95 1C for 10 min, followed by 40
cycles of 95 1C for 15 s, 60 1C for 1 min.
Ribosomal protein S7 gene (Salazar et al., 1993) was
used for data normalization (S7-F: 5
0
-CCTGGAGCTG-
GAGATGAACT-3
0
and S7-R: 5
0
-CGGCGCTCGGCA
ATGAACAC-3
0
).
2.8. Conrmation of P. berghei parasites on A. gambiae
midgut
In order to conrm the presence of P. berghei
parasites committed to sporogonic development, we
analyzed the expression of the ookinete surface protein
Pbs25 (Tsuboi et al., 1997) in A. gambiae midguts
dissected at 24 h post-feeding by real-time PCR. For this
purpose, two specic primers were designed (Pbs25-F:
5
0
-TATAACGCAGCAATTTCACCAA-3
0
and Pbs25-
R: 5
0
-ATTTTTCATTACATCGGCATTCT-3
0
). Real-
time PCR conditions followed those used for the
mosquito genes.
The remaining mosquitoes were maintained to allow
P. berghei development, oocysts detection and quanti-
cation on day 10 post-blood feeding.
2.9. Statistical analysis
Data analysis was performed using SPSS v 9.0 for
Windows (SPSS, Inc.). The Students t-test was used to
compare the mean levels of expression measured in each
mosquito treatment: unfed vs. untreated, chloroquine
treated (10 mg/kg) vs. untreated, chloroquine treated
(50 mg/kg) vs. untreated and chloroquine treated with
10 vs. 50 mg/kg. The Wilcoxon signed rank test was used
to compare the bacterial counts per midgut in chlor-
oquine-treated and untreated mosquitoes and the
number of oocysts per midgut on day 10 post-blood
feeding in each mosquito group: chloroquine treated
(10 mg/kg) vs. untreated, chloroquine treated (50 mg/kg)
vs. untreated and chloroquine treated with 10 vs. 50 mg/
kg. Statistically signicant differences were considered
for a po0:05.
3. Results
3.1. Expression analysis of A. gambiae transcripts after
chloroquine treatment
The levels of expression of seven immune-related
genes (Sp14D2, AgSP24D, ISPL5, 1644280, gambicin,
defensin and cecropin) were measured in the midguts of
A. gambiae mosquitoes. Two factors were considered
when measuring levels of gene expression: the effect of
blood feeding and the presence of chloroquine in the
blood meal. There was an overall increase of expression
ARTICLE IN PRESS
associated with blood feeding, which was suppressed by
the presence of chloroquine (Fig. 1).
The increase of expression observed in blood fed
mosquitoes, compared to unfed mosquitoes, ranged
320 times in the genes studied. When mosquitoes were
fed on mice that were treated with chloroquine, we
observed a 1.512.1-fold decrease in gene expression
when compared with untreated mosquitoes. In some
cases, expression levels were set back to values
approximately similar to those observed in unfed
mosquitoes. Gambicin, defensin and Sp14D2 expression
levels were signicantly lower in mosquitoes fed on the
higher dose of chloroquine (50 mg/kg) when compared
to mosquitoes fed on a blood meal without chloroquine.
Antimicrobial peptides showed a more pronounced
reduction of expression in the chloroquine-treated
mosquitoes than the serine proteases analyzed.
3.2. Midgut bacterial quantication
Although the amounts of chloroquine used in our
study, are considerably lower than those described to
have antibacterial effect, we decided to test if the
suppression observed with chloroquine treatment could
be related to a decrease in mosquito midgut bacterial
number. We observed considerable individual varia-
tion in bacterial counts within each group of mos-
quitoes (treated and untreated), ranging from 3 10
4
to 1.24 10
6
cfu/midgut in chloroquine treated, and
1 10
4
to 2.21 10
6
cfu/midgut in untreated mosqui-
toes. No signicant differences were observed between
bacterial counts in the chloroquine treated (median
5.1x10
5
cfu/midgut) and untreated (median 4.1 10
5
cfu/midgut) mosquitoes.
3.3. Expression analysis of A. gambiae transcripts after
antibiotic treatment
In order to conrm that effect of chloroquine
treatment on the immune system was not mediated
through antimicrobial activity, mosquitoes were fed on
BALB/c mice treated with a large spectrum antibiotic
recommended for veterinary treatment.
Once again, an increment of gene expression induced
by blood ingestion was observed (Fig. 2) thereby
reinforcing the data obtained with the chloroquine
experiment (Fig 1). Real-time PCR results revealed that
the antibiotic treatment did not signicantly reduce the
expression of the transcripts analyzed. The same effect
was observed in an extra experiment in which a 2-fold
higher dose (5.0/25.0 mg Kg
1
) of antibiotic was used
(data not shown). These data, together with the
observations on midgut bacterial counts of chloro-
quine-treated and untreated mosquitoes indicate that
the effect of chloroquine on the immune response was
independent of antimicrobial activity.
3.4. Expression analysis of A. gambiae transcripts
following P. berghei infection and chloroquine treatment
This part of the work, aimed at investigating the effect
of P. berghei infection in the chloroquine modulation of
ARTICLE IN PRESS
Sp14D2
0.1
1
10
100
Unfed B chl10 chl50
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
*
** *
AgsP24D
0.1
1
10
100
Unfed B chl10 chl50
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
ISPL5
0.1
1
10
100
Unfed B chl10 chl50 Unfed B chl10 chl50
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
Gambicin
Unfed B chl10 chl50
Treatment
Unfed B chl10 chl50
Treatment
Unfed B chl10 chl50
Treatment
* **
Defensin
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
* **
Cecropin
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
1644280
0.1
1
10
100
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
(B)
(A)
Fig. 1. Effect of chloroquine on the transcript abundance of (A) serine proteasesSp14D2, AgSP24D, ISPL5 and 1644280 and (B) antimicrobial
peptidesgambicin, defensin and cecropin in mosquito midguts 24 h after blood feeding. Unfed, non-blood fed control; B, blood fed on untreated
mice; chl10, blood fed on mice previously (17 h) treated with 10 mg/kg of chloroquine; chl50, blood fed on mice treated with 50 mg/kg of chloroquine.
Data was normalized using A. gambiae ribosomal protein S7 gene expression levels. Values represent the mean and standard deviation of six
independent experiments (signicant differences po0.05 between *Unfed and B, and **B and chl10 or chl50).
the mosquito immune system observed in uninfected
mosquitoes.
A general increase (422-fold) in the expression
levels of the serine proteases and the antimicro-
bial peptides monitored in this study was again
associated with blood feeding. Chloroquine treatment
was once again responsible for a general 1.59.8-
fold decrease of gene expression (Fig. 3). The only
exceptions were seen in the expression of Sp14D2
at low dose treatment (10 mg/kg), defensin at high
dose treatment (50 mg/kg) and 1644280 at both doses
treatment.
ARTICLE IN PRESS
(A)
AgsP24D
0.1
1
10
100
Unfed B A1 Unfed B A1
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
1644280
Treatment
Unfed B A1
Treatment
Unfed B A1 Unfed B A1
Treatment
Gambicin
Treatment
Defensin Cecropin
(B)
Fig. 2. Effect of large spectrum antibiotic on the transcript abundance of (A) serine proteasesAgSP24D and 1644280 and of (B) antimicrobial
peptidesgambicin, defensin and cecropin, at day one (24 h post-blood feeding). Unfed, non-blood fed control; B, blood fed on untreated mice; A1,
blood fed on mice previously (17 h) treated with 2.5/12.5 mgKg
1
trimethoprim/dimerazol. Data was normalized using A. gambiae ribosomal protein
S7 gene expression levels. Values represent the mean and standard deviation of three independent experiments.
Sp14D2
0.1
1
10
100
0.1
1
10
100
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
Unfed BPb Chl
10Pb
Chl
50Pb
Treatment
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
AgsP24D
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
0.1
1
10
100
A
r
b
i
t
r
a
r
y

U
n
i
t
s
ISPL5
*
Gambicin
*
Defensin
*
Cecropin
*
1644280
*
(B)
(A)
Fig. 3. Effect of chloroquine on the transcript abundance of (A) serine proteasesSp14D2, AgSP24D, ISPL5 and 1644280 and of (B) antimicrobial
peptidesgambicin, defensin and cecropin on Plasmodium berghei infected mosquitoes, at day one (24 h post-blood feeding). Unfed, non-blood fed
control; BPb, blood fed with P. berghei parasites; chl 10Pb, blood fed on P. berghei infected mice previously (17 h) treated with 10 mg/kg of
chloroquine; chl 50Pb, blood fed on P. berghei infected mice previously (17 h) treated with 50 mg/kg of chloroquine. Data was normalized using A.
gambiae ribosomal protein S7 gene expression levels. Values represent the mean and standard deviation of three independent experiments (signicant
differences po0.05 between *Unfed and BPb).
When we compared the effect of chloroquine on
infected and uninfected mosquitoes (Figs. 1 and 3), a
less pronounced down-regulation in infected chloro-
quine-treated mosquitoes was observed, suggesting that
the response triggered by Plasmodium infection can to
some extent bypass the suppression effect exerted by
chloroquine.
3.5. Conrmation of P. berghei parasites presence in A.
gambiae midgut
In order to guarantee that all P. berghei life stages
were normally developing in all the experiments, several
stages of the parasite were monitored.
Presence of gametocytes and parasite exagellation in
treated and untreated mice was conrmed by optical
microscopy prior to mosquito blood meal. P. berghei
chloroquine-treated mice presented lower gametocitae-
mia than untreated mice. The geometric mean of
gametocitaemia varied from 0.91% for untreated,
0.08% for 10 mg/kg chloroquine and 0.08% for 50 mg/
kg chloroquine-treated mosquitoes.
The presence of P. berghei that were committed to
sporogonic development was conrmed 24 h post-feed-
ing by the analysis of the transcripts that code for the
ookinetes surface protein Pbs25. As expected, expres-
sion was detected in P. berghei blood fed mosquitoes but
no transcripts were observed in unfed mosquitoes or in
mosquitoes fed on uninfected blood (data not shown).
No signicant differences were observed between Pbs25
expression levels on P. berghei chloroquine-treated and
untreated mosquitoes. The oocyst burden in the midguts
was measured 10 days post-blood feeding (Table 1).
Mosquitoes treated with the lower dose of chloroquine
(10 mg/kg) presented the highest median intensity of
infection when compared with the other two groups,
which had similar oocyst counts.
4. Discussion
This is the rst report on the effect of the anti-
malarial drug chloroquine, a 4-aminoquinoline, on the
immune response of A. gambiae. The levels of expression
of seven immune-related genes (Sp14D2, AgSP24D,
ISPL5, 1644280, gambicin, defensin and cecropin) pre-
viously described as being differentially transcribed
during Plasmodium infection, were analyzed in the
midguts of A. gambiae by real-time PCR. We observed
an overall increase of gene expression associated with
blood feeding, which was suppressed by the presence of
chloroquine.
The transcriptional activation after a blood meal,
probably associated with the metabolic needs of the
blood meal digestion, is consistent with previous
transcription studies carried out for most serine
proteases and antimicrobial peptides (Mu ller et al.,
1993; Ribeiro, 2003).
Activation of immune-related genes could also be a
consequence of an increase in midgut bacterial numbers,
as blood meals boost midgut ora (Pumpuni et al.,
1996). These authors also observed that midgut bacteria
at high concentration can inhibit sporogonic develop-
ment and suggested that bacteria could trigger defense
mechanisms. Therefore alterations in midgut bacterial
numbers could induce differential immune-related gene
transcription. The presence of chloroquine in the blood
meal abrogated the transcriptional activation observed
with blood meal intake, which could be a consequence
of chloroquine antimicrobial activity. In fact it is known
that high concentrations of chloroquine can exert a
bactericidal effect (Ciak and Hahn, 1966). However, a
bactericidal effect of chloroquine in our experiments was
ruled out, since we observed similar midgut bacterial
counts in treated and untreated mosquitoes and no
signicant impact of antibiotic activity on serine
proteases and antimicrobial peptides expression was
detected with the antibiotic doses tested. Higher doses
than those used here would probably have similar effect.
These results suggest that the down-regulation of gene
expression seen after chloroquine treatment was inde-
pendent of antimicrobial activity indicating a direct
effect of chloroquine on the mosquito immune system.
Taken together, our results suggest that chloroquine
might act directly on A. gambiae serine proteases
cascade, interfering at signal transduction pathways
and transcriptional activation level and thus leading to
the down-regulation of serine proteases and antimicro-
bial peptides.
Although chloroquine mechanisms of action have
been largely studied, many questions still exist and
ARTICLE IN PRESS
Table 1
Effect of chloroquine treatment on P. berghei oocyst formation
Mosquitoes (n)
a
# oocysts/
midgut
b
(min.max.)
Blood fed on untreated mice (n 45) 5.5 (1116)
Blood fed on mice treated with 10 mg/kg
chloroquine (n 33)
9.5 (1132)
c
Blood fed on mice treated with 50 mg/kg
chloroquine (n 46)
4.0 (133)
Treated (10 and 50 mg/kg) and untreated mosquitoes were dissected on
day 10 post-blood feeding and the number of oocysts per midgut
determined.
a
Number of infected mosquitoes used for determination of oocyst
counts.
b
Values represent the median, the minimum and the maximum
number of oocysts per midgut.
c
Signicant differences (po0.05) between 10 mg/kg chloroquine
treatment and 50 mg/kg chloroquine treatment.
nothing or very little is known about the chloroquine
mechanisms of action in insects.
Most of the inhibition effects related to chloroquine
treatment are attributed to a perturbation of intrave-
sicles acidication. Seyama et al. (2003) observed this
effect at the serine proteases level. Indeed, they reported
that chloroquine pretreatment completely blocked the
intracellular generation of trypsinogen activation pep-
tide (TAP) in cerulein-induced pancreatitis, mediated by
neutralization of acidic subcellular compartments. In A.
gambiae at least seven trypsinogens have been described
(Mu ller et al., 1993). In our study, all serine proteases
are trypsin-like, and were described to be involved in
insect innate immune response (Hoffmann et al., 1999;
Dimopoulos, 2003), mainly as prophenoloxidase (PPO)-
activating enzymes.
Considering the above result, we think that inter-
ference with endosomes pH, mediated by chloroquine,
represents one possible explanation for the down-
regulation of mosquito serine proteases and antimicro-
bial peptides expression observed in our study. Pertur-
bation of endosomal pH can interfere with TAP
activation (or other factor activation peptide), and
consequently with trypsin synthesis. This will have
consequences in the activation of intracellular signaling
pathways and the subsequent transcription of effector
genes, such as antimicrobial peptides.
Other inhibition effects related to chloroquine treat-
ment appear to be independent of a chloroquine-lyso-
somotropic mechanism of action. In human peripheral
mononuclear blood cells, chloroquine inhibits TNF-a
gene expression involving a nonlysosomotropic mechan-
ism (Weber and Levitz, 2000). In addition, chloroquine
blocks the activation of NF-kB induced by IL-1b in
epithelial cells, by a mechanism involving the acid
sphingomyelinase pathway (Bonizzi et al., 1997).
Given that chloroquine affects NF-kB activity, a
transcription factor that is rapidly inducible in most cell
types following a large variety of stimuli (Bonizzi et al.,
1997), and that NF-kB is a pivotal transcription factor
of innate immune responses (Delhase et al., 2000; Sacks
and Sher, 2002), we can thus presume that chloroquine
interference at NF-kB activity can constitute another
possible explanation for the reduction of expression
observed. Interference of chloroquine at NF-kB trans-
activators (or at other nuclear factor(s) binding activity)
could have blocked the activation of mosquito-specic
signaling transduction pathways (Toll, Imd and STAT),
and in the end, affected the activation of antimicrobial
peptides, such as the ones studied (gambicin, defensin
and cecropin).
The fact that 1644280 expression was not affected by
chloroquine treatment, and that its sequence is closer to
that of mammalian serine protease families than that of
insect families (Dimopoulos et al., 1996) suggests that
1644280 is located upstream of the chloroquine affected
pathway or that a different pathway (not affected in the
same way by chloroquine) is regulating the expression of
this serine protease.
Differences between oocyst counts were similar
between treated and untreated mosquitoes (p40:05),
but gametocyte counts were one log magnitude different
between them. This might have implications on the
outcome of midgut infection. Therefore, if we divide the
oocyst counts by gametocitaemia we will obtain a ratio
that might give a better view of the parasite infection
success in the mosquitoes. These ratios varied from 6.04
in untreated mosquitoes to 50 or 118.75 in treated
mosquitoes, showing that infection success was higher in
treated mosquitoes.
The effect of the higher dose of chloroquine (50 mg/
kg) on mosquitos infectivity was interesting, as we
would expect that the dose-dependent down-regulation
observed in the majority of the genes studied would
correspond to an increase in oocyst burden with the
higher dose of chloroquine (as seen with the lower dose).
However, if we take into account the proportion of
parasites ingested by the mosquito (less gametocytes
ingested by the higher dose chloroquine-treated mos-
quitoes than by untreated mosquitoes) and those that
successfully developed into oocysts, we can presume
that, despite apparently similar, mosquitoes treated with
the 50 mg/kg dose of chloroquine had higher infectivity
than untreated mosquitoes.
The signicant difference of mosquito infectivity
observed between the two chloroquine doses had a
parallel on the defensin expression trend. Defensin
constituted an exception and presented higher levels of
expression in the higher dose of chloroquine when
compared with the lower dose (Fig. 3). The antimicro-
bial peptide defensin has been shown to be induced upon
Plasmodium infection (Richman et al., 1997) and might
be induced by the presence of gametocytes in the blood
meal (Tahar et al., 2002). Furthermore it seems to exert
its activity differentially in different life cycle stages of
the parasite (Dimopoulos et al., 1997). In vitro para-
siticidal effects of defensin have also been reported
(Shahabuddin et al., 1998). Nevertheless, we believe that
this cannot be the only mechanism to justify the
differences in infectivity observed, as silencing of the
defensin gene has little impact on oocyst formation
(Blandin et al., 2002).
Sp14D2 is another gene that seems to have an
opposite pattern of expression in the two dosages.
Lower chloroquine dose increased gene expression in P.
berghei infected mosquitoes, while at a higher dose a
decrease in expression levels was observed (even not
signicantly different). As described in the literature,
Sp14D2 is slightly up-regulated after Plasmodium
infection, it has features of a PPO activating enzyme
and is probably involved in melanization (Gorman
et al., 2000). Other effector genes not covered by our
ARTICLE IN PRESS
analysis might also be affected differentially by the
higher dose of chloroquine in the same way as defensin
and Sp14D2 thus contributing to the infectivity results
observed by us and others (Ramkaran and Peters, 1969).
In conclusion, our work provides the rst description
of the effect of chloroquine on the immune response of
A. gambiae. We have shown that chloroquine down-
regulates the expression of immune-related serine
proteases and antimicrobial peptides independently of
midgut bacterial numbers. These results suggest that
chloroquine might act on Anopheles serine proteases
cascade, interfering at signal transduction pathways
and at the transcriptional activation level. Understand-
ing the mechanisms of chloroquine action in mosquito
vector and vertebrate hosts will reveal biological
details that can be fruitful for novel malaria control
strategies such as those based in transmission-blocking
vaccines.
Acknowledgements
We thank Prof. Paulo Almeida (UEI Entomologia
Me dica/IHMT) for providing the mosquitoes and Paula
Silva, Isabel Carvalho and Ana Gabriel for their
assistance with mosquito rearing. We are also grateful
to members of our laboratory for helpful discussions
and suggestions. This work was supported by Fundac-a o
para a Cie ncia e a Tecnologia grants to P. Abrantes
(SFRH/BD/6346/2001) and L.F. Lopes (SFRH/BD/
10202/2002) and research funds from projects POCTI/
35815/MGI/2000 and POCTI/MGI/44905/2002.
References
Blandin, S., Moita, L.F., Kocher, T., Wilm, M., Kafatos, F.C.,
Levashina, E.A., 2002. Reverse genetics in the mosquito Anopheles
gambiae: targeted disruption of the Defensin gene. EMBO Rep 3
(9), 852856.
Bonizzi, G., Piette, J., Merville, M.P., Bours, V., 1997. Distinct signal
transduction pathways mediate nuclear factor-kappaB induction
by IL-1beta in epithelial and lymphoid cells. J. Immunol. 159, 5264.
Ciak, J., Hahn, F.E., 1966. Chloroquine: mode of action. Science 151,
347349.
Delhase, M., Li, N., Karin, M., 2000. Kinase regulation in
inammatory response. Nature 406, 367368.
Dimopoulos, G., 2003. Insect immunity and its implication in
mosquitomalaria interactions. Cellular Microbiol 5 (1), 314.
Dimopoulos, G., Muller, H-M., Levashina, E., Kafatos, F.C., 2001.
Innate immune defense against malaria infection in the mosquito.
Curr. Opin. Immunol. 13, 7988.
Dimopoulos, G., Richman, A., Della Torre, A., Kafatos, F.C., Louis,
C., 1996. Identication and characterization of differentially
expressed cDNAs of the vector mosquito, Anopheles gambiae.
Proc. Natl. Acad. Sci. USA 93, 13,06613,071.
Dimopoulos, G., Richman, A., Mu ller, H-M., Kafatos, F.C., 1997.
Molecular immune responses of the mosquito Anopheles gambiae
to bacteria and malaria parasites. Proc. Natl. Acad. Sci. USA 94,
11,50811,513.
Dimopoulos, G., Seeley, D., Wolf, A., Kafatos, F.C., 1998. Malaria
infection of the mosquito Anopheles gambiae activates immune-
responsive genes during critical transitions stages of the parasite life
cycle. EMBO J 17, 61156123.
Gorman, M.J., Andreeva, O.V., Paskewitz, S.M., 2000. Molecular
characterization of ve serine protease genes cloned from
Anopheles gambiae hemolymph. Insect. Biochem. Mol. Biol. 30,
3546.
Hoffmann, J.A., Kafatos, F.C., Janeway, C.A., Ezekowitz, R.A.B.,
1999. Phylogenetic perspectives in innate immunity. Science 284,
13131318.
Hogh, B., Gamage-Mendis, A., Butcher, G.A., Thompson, R.,
Begtrup, K., Mendis, C., Enosse, S.M., Dgedge, M., Barreto, J.,
Eling, W., Sinden, R.E., 1998. The differing impact of chloro-
quine and pyrimethamine/sulfadoxine upon the infectivity of
malaria species to the mosquito vector. Am. J. Trop. Med. Hyg.
58, 176.
Ichimori, K., Curtis, C.F., Targett, G.A., 1990. The effects
of chloroquine on the infectivity of chloroquine-sensitive and
-resistant populations of P. yoelii nigeriensis to mosquitoes.
Parasitology 100, 377.
Jeong, J.Y., Jue, D.M., 1997. Chloroquine inhibits processing of tumor
necrosis factor in lipopolysaccharide-stimulated RAW 264.7
macrophages. J. Immunol. 158, 4901.
Mu ller, H.M., Crampton, J.M., della Torre, A., Sinden, R., Crisanti,
A., 1993. Members of a trypsin gene family in Anopheles
gambiae are induced in the gut by blood meal. EMBO J 12,
28912900.
Park, Y.C., Pae, H.O., Yoo, J.C., Choi, B.M., Jue, D.M., Chung,
H.T., 1999. Chloroquine inhibits inducible nitric oxide synthase
expression in murine peritoneal macrophages. Pharmac. Toxicol.
85 (4), 188191.
Pumpuni, C.B., Demaio, J., Kent, M., Davis, J.R., Beier, J.C., 1996.
Bacterial population dynamics in three Anopheline species: the
impact on Plasmodium sporogonic development. Am. J. Trop.
Med. Hyg. 54 (2), 214218.
Ramkaran, A.E., Peters, W., 1969. Infectivity of chloroquine resistant
Plasmodium berghei to Anopheles stephensi enhanced by chlor-
oquine. Nature 223 (206), 635636.
Ribeiro, J.M.C., 2003. A catalogue of Anopheles gambiae transcripts
signicantly more or less expressed following a blood meal. Insect.
Richman, A.M., Dimopoulos, G., Seeley, D., Kafatos, F.C., 1997.
Plasmodium activates the innate immune response of Anopheles
gambiae mosquitoes. EMBO J 16, 61146119.
Rosa, R., Silveira, H., Seixas, E., Rola o, N., Santos-Gomes, G.,
Rosa rio, V., 1999. The effect of chloroquine on the production of
interferon-g, interleukin IL-4, IL-6, and IL-10 in Plasmodium
chabaudi chabaudi in infected C57BL6 mice. J. Parasitol. 85 (5),
956960.
Sacks, D., Sher, A., 2002. Evasion of innate immunity by parasitic
protozoa. Nat. Immunol. 3 (11), 10411047.
Salazar, C.E., Mills-Hamm, D., Kumar, V., Collins, F.H., 1993.
Sequence of a cDNA from the mosquito Anopheles gambiae
encoding a homologue of human ribosomal protein S7. Nucl.
Acids Res.
Savarino, A., Boelaert, J.R., Cassone, A., Majori, G., Cauda, R., 2003.
Effects of chloroquine on viral infections: an old drug against
todays diseases? Lancet 3, 722727.
Schultz, K.R., Gilman, A.L., 1997. The lysosomotropic amines,
chloroquine and hydroxychloroquine: a potentially novel therapy
for graft-versus-host disease. Leuk. Lymphoma 24, 201.
Seyama, Y., Otani, T., Matsukura, A., Makuuchi, M., 2003. The pH
modulator chloroquine blocks trypsinogen activation peptide
generation in Cerulein-induced Pancreatitis. Pancreas 26 (1),
1517.
ARTICLE IN PRESS
Shahabuddin, M., Fields, I., Bulet, P., Hoffmann, J.A., Miller, L.H.,
1998. Plasmodium gallinaceum: differential killing of some mosqui-
to stages of the parasite by insect defensin. Exp. Parasitol. 89 (1),
103112.
Staszczak, M., Zdunek, E., Leonowicz, A., 2000. Studies on the role of
proteases in the White-rot fungus Trametes versicolor: effect of
PMSF and chloroquine on ligninolytic enzymes activity. J. Basic
Microb. 40 (1), 5163.
Tahar, R., Boudin, C., Thiery, I., Bourgouin, C., 2002. Immune
response of Anopheles gambiae to the early sporogonic stages of the
human malaria parasite Plasmodium falciparum. EMBO J 21 (24),
66736680.
Tsuboi, T., Cao, Y.M., Kaslow, D.C., Shiwaku, K., Torii, M.,
1997. Primary structure of a novel ookinete surface protein
from Plasmodium berghei. Mol. Biochem. Parasitol. 85 (1),
131134.
Vizioli, J., Bulet, P., Charlet, M., Lowenberger, C., Blass, C., Mu ller,
H.-M., Dimopoulos, G., Hoffmann, J., Kafatos, F.C., Richman,
A., 2000. Cloning and analysis of a cecropin gene from the
malaria vector mosquito, Anopheles gambiae. Insect. Mol. Biol. 9
(1), 7584.
Vizioli, J., Bulet, P., Hoffmann, J.A., Kafatos, F.C., Mu ller, H-M.,
Dimopoulos, G., 2001. Gambicin: a novel immune responsive
antimicrobial peptide from the malaria vector Anopheles gambiae.
Proc. Natl. Acad. Sci USA 94 (23), 12337.
Weber, S.M., Levitz, S.M., 2000. Chloroquine interferes with
Lipopolysaccharide-induced TNF-a gene expression by a non-
lysosomotropic mechanism. J. Immunol. 165, 15341540.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Accumulation of 23 kDa lipocalin during brain development and
injury in Hyphantria cunea
$
Hong Ja Kim
a
, Hyun Jeong Je
a
, Hyang Mi Cheon
a,1
, Sun Young Kong
a,2
, JikHyun Han
b
,
Chi Young Yun
b
, Yeon Su Han
c
, In Hee Lee
d
, Young Jin Kang
e
, Sook Jae Seo
a,
a
Division of Applied Life Science, Gyeongsang National University, Jinju 660-701, Republic of Korea
b
Department of Biology, Daejeon University, Daejeon
c
Department of Agricultural Biology, Chonnam National University, Gwangju
d
Department of Life Science, Hoseo University, Asan
e
Department of Pharmacology, Yeungnam University, Daegu, Republic of Korea
Received 6 May 2005; accepted 17 May 2005
Abstract
The cDNA corresponding to a novel lipocalin was identied from the fall webworm, Hyphantria cunea. This lipocalin cDNA
encodes a 194 residue protein with a calculated molecular mass of 23 kDa. Sequence analyses revealed that the 23 kDa lipocalin
cDNA is most similar to Drosophila lazarillo, human apolipoprotein D, and Bombyrin. Northern blot analyses showed that 23 kDa
lipocalin transcript is expressed in the whole body only in 4- and 6-day-old pupae. By Western blot analysis it was conrmed that
23 kDa lipocalin is mainly accumulated in brain and subesophageal ganglion, though it is detected in a small amount in fat body and
epidermis of Hyphantria cunea. The accumulation of 23 kDa lipocalin in brain tissue was upregulated in response to injury. The
putative function of 23 kDa lipocalin in brain is discussed.
Keywords: Hyphantria cunea; 23 kDa lipocalin; Brain; Injury
1. Introduction
The lipocalins are small proteins with molecular
masses averaging 20 kDa that transport hydrophobic
compounds. They are remarkably diverse at the
sequence level with pairwise comparisons often falling
below 20% and yet they have highly conserved
structures (Ganfornina et al., 2000). There are three
conserved sequence motifs called structurally conserved
regions (SCRs) that have been proposed as a prerequi-
site for a protein to be considered a lipocalin (Flower et
al., 1993). Besides SCRs, from one to three disulde
bridges, that are known to constrain the overall
structure by stabilizing the N- and C-terminal regions
of the protein, are found in most lipocalins (Ganfornina
et al., 2000).
Although they have been classied mainly as transport
proteins, it is now clear that members of the lipocalin
family fulll a variety of different functions with roles in
retinal transport, cryptic coloration, olfaction, phero-
mone transport, enzyme synthesis of prostaglandins, and
cell regulation (Flower, 1996). A number of lipocalins
act in insect coloration. The bluegreen coloration of
certain lepidopteran larvae is due in part to the presence
of blue pigment that is usually associated with the blue
pigment binding proteins. Blue pigment binding proteins
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.004
$
This work was supported by ABRL program of Korea Science
Engineering Foundation (R14-2002-056-01001-0) and Brain Korea 21
program of Gyeongsang National University.

fax: +82 55 754 0086.
E-mail address: sookjae@gshp.gsnu.ac.kr (S.J. Seo).
1
Present address: Department of Entomology, University of
California, Riverside, CA 92521, USA.
2
Present address: Department of Orthopaedic Surgery, Yonsei
University, Seoul, Republic of Korea.
in insect hemolymph have been identied from Manduca
sexta (Riley et al., 1984; Goodman et al., 1985), Locusta
migratoria (Chino et al., 1983), Heliothis zea (Hauner-
land and Bowers, 1986), Pieris brassicae (Huber et al.,
1987a), Spodoptera litura (Yoshiga and Tojo, 1995), and
Samia synthia ricini (Saito, 1998). The bilin binding
protein of Pieris brassicae has been crystallised (Huber et
al., 1987a, b) and its complete amino acid sequence has
been determined (Suter et al., 1988). In Manduca sexta,
two different isoelectric forms have been conrmed
(Riddiford et al., 1990) and their clones are characterized
(Li and Riddiford, 1994).
Recently, several neural lipocalins were reported,
especially present in the insect central nervous system
(Sa nchez et al., 1995, 2000; Sakai et al., 2001). Lazarillo
in the grasshopper has been reported as a guide of
pioneer neurons during nervous system development
(Ganfornina et al., 1995, 2000; Sa nchez et al., 2000). The
AcP cells, anterior commissural pioneer neurons are
misrouted when Lazarillo function is specically per-
turbed by the treatment with Lazarillo monoclonal
antibody (Sa nchez et al., 1995). Also, two Drosophila
neural and glial Lazarillos have been found in a subset
of axons and glial cells of the central nervous system
(Sa nchez et al., 2000). In lepidoptera, neural lipocalins,
Bombyrin and Gallerin, have been isolated from pupal
brain extracts of Bombyx mori (Sakai et al., 2001) and
Galleria mellonella (Filippov et al., 1995), respectively,
but their exact function was not reported.
We report here the molecular characterization of a
new lipocalin found in Hyphantria cunea, which was
discovered as an unintended PCR product from a study
of yolk protein. This lipocalin is mainly expressed in the
developing brain. The accumulation of this lipocalin was
also found in the fat body, epidermis, and subesopha-
geal ganglion, in addition to the brain which we discuss
in this report.
2. Materials and methods
2.1. Animals
Fall webworms, Hyphantria cunea, were obtained
from a colony in the laboratory of insect conservation,
Department of Sericulture and Entomology, National
Institute Agricultural Science and Technology, Suwon,
Republic of Korea. They were reared on an articial diet
(Ito & Tanaka, 1960) at 27 1C and 75% relative
humidity with a photoperiod of 16L:8D. Under the
laboratory conditions, the pupal stage lasts 10 days.
2.2. Collection and processing of hemolymph and tissues
Larval hemolymph was collected into cold test tubes
by cutting prothoracic leg. Pupal and adult hemolymph
was collected into cold test tubes by puncturing the
intersegment of abdomen with a needle. A few crystals
of phenylthiourea were added to each hemolymph
sample to prevent melanization. Hemolymph was
centrifuged at 10,000g for 10 min at 4 1C to remove
hemocytes and cell debris. The supernatant was stored
at 70 1C until used.
Brain, esophageal ganglion, fat body, and epidermis
were dissected in cold phosphate buffered saline (PBS,
0.02 M sodium phosphate, pH 7.2, 0.15 M NaCl) and
homogenized in homogenation buffer (50 mM Tris-HCl,
pH 8 containing 1 mM PMSF, 1 mM dithiothreitol,
1 mM EDTA, and 0.01% [W/V] 1-phenyl 2-thiourea).
The homogenate was centrifuged at 12,000g for 20 min
and the supernatant was used for electrophoresis.
2.3. Injury experiment
Twenty brains from 5-day-old pupae were used for
each group. Prior to the dissection the insects were
anesthized at 4 1C for 1 h and carefully dissected in
Grace medium (GibcoBRL, NY) at 4 1C to isolate the
brains without damage under a dissecting microscope.
The isolated brains were minced with a scalpel blade
under a dissecting microscope and incubated in a sterile
dish at 25 1C for 12 and 24 h, respectively. After
incubation they were sonicated, centrifuged, and the
supernatant was used for Western blot. Triplicate
cultures of brains were individually processed for
Western blot.
2.4. Isolation of RNA
Total RNA was isolated from tissues using the
RNeasy mini kit (Qiagen Inc. Chatsworth, USA)
according to the protocol recommended by the manu-
facturer. All RNA samples were evaluated in agarose
gels to ensure that they contained intact rRNA and were
free of contaminating DNA.
2.5. RT-PCR
Total RNA was isolated from whole bodies of late
pupae and 5 mg aliquot of RNA preparation was reverse
transcribed by the superscript II reverse transcriptase
(Gibco BRL) using oligo (dT)1218 primers in a
reaction volume of 25 ml. The reaction mixture was
incubated at 70 1C for 5 min, 42 1C for 1 h, and then at
94 1C for 1 min. The obtained cDNA was used in the
PCR. Sequences of primers used were 5
0
-CAYCAR-
GARAAYGGNCARATHTTY-3
0
(forward) and 5
0
-
YTTRTANGTRTAYTCRTT-3
0
(reverse). PCR was
performed for 35 cycles using cycling conditions of
94 1C for 30 s, 47 1C for 1 min, and 72 1C for 1 min. The
resulting 530-bp PCR product was separated on a 1%
agarose gel. This fragment was excised from the agarose
ARTICLE IN PRESS
H.J. Kim et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11331141 1134
gel, puried, ligated into a pGEM T-easy vector
(Promega, WI, USA), followed by transformation into
JM109 competent cells (Promega, WI, USA).
2.6. cDNA library screening
The l ZAPII cDNA library of H. cunea pupal and
adult whole bodies was used (Cheon et al., 2003).
Plaques (5 10
4
) were transferred to membrane lters
(Hybond N+; Amersham, Bioscience, UK) and
screened using random primed labeling with dig-high
prime (Boehringer Mannheim, Germany). After prehy-
bridization at 65 1C for 1 h, the membranes were
hybridized at 65 1C for 16 h, washed twice (5 min each)
in 2X SSC, 0.1% SDS at 25 1C, and washed twice
(15 min, each) in 0.5x SSC, 0.1% SDS at 65 1C. The
signals were detected with the anti-DIG alkaline
phosphatase antibody, CSPD
s
(Boehringer Mannheim,
Germany), and the substrate on an X-ray lm with a
15 min exposure. One positive plaque was isolated in
phage buffer by the plate lysate method followed by
lambda DNA extraction (Sambrook et al., 1989). Insert
of the positive clone was subcloned into the unique
EcoRI site of pBluescript KS(+).
2.7. DNA sequencing
The sequencing reaction was based on the dideox-
ynucleotide chain termination methods of Sanger et al.
(1977) using [
35
S]dATP and the sequenase version 2.0
polymerase kit (USB Inc., Ohio, USA). Template-
specic and universal primers derived from pBluesceipt
were used in the sequence reactions. Subclones for
sequencing were prepared by ligating suitable restriction
fragments into pBluescript, followed by transformation
into JM 109 cells.
2.8. Analysis of sequence data
The EMBL DataBank was searched with BLAST.
Editing and analysis of the DNA sequence data were
performed with DNASTAR software (DNASTAR Inc.,
Wisconsin). In particular, we used the program MEGA-
LIGN to generate pairwise alignment.
2.9. Northern blot
Total RNA from whole bodies (10 mg) were denatured
and subjected to electrophoresis in a 1.2% agarose gel
containing 2.2 M formaldehyde. Following electrophor-
esis, gels were rinsed in 10 SSC and transferred to
nitrocellulose (Schleicher & Schuell) in 10 SSC. blots
were prehybridized with 1.5 SSPE, 7% SDS, 10%
PEG, 0.1 mg/ml sonicated denatured salmon sperm
DNA and 0.25 mg/ml BSA for 4 h at 65 1C. Hybridiza-
tion was performed for 18 h at 65 1C in the prehybridi-
zation buffer with 5 10
5
cpm/ml of
32
P-labeled probes
(530 bp), prepared by random priming (Feinberg and
Volgelstein, 1983). The lter was then washed twice with
1 SSC, 0.1% SDS at 65 1C for 15 min, and subse-
quently two more times for 15 min with 0.1 SSC,
0.1% SDS at 65 1C before exposure to X-ray lm at
70 1C.
2.10. Peptide synthesis and preparation of antiserum
A peptide sequence corresponding to residues
159172 from the 23 kDa lipocalin was selected and
synthesized with an additional cysteine at its N-terminus
on the pioneer
TM
Peptide Synthesizer (Applied Biosys-
tems). The sequence of this peptide, named H159, is
NH2-CQNAYAVLDKFKISR-COOH. The peptide
was puried by Xtera
TM
RPC18 column loaded on
HPLC (Waters) and its purity was assessed by MALDI-
TOF mass spectrometry, Voyger
TM
DE-STR (Applied
Biosystems). Covalent attachment of peptide to keyhole
limpet hemocyanin (KLH) (Calbiochem, Darmstadt,
Germany) was achieved with m-malemidobenzoyl-N-
hydrosuccinimide ester (MBS) (Sigma, MI, USA) that
couples the N-terminal cysteine of H159 peptide to
amino groups on KLH. The H159-KLH conjugate was
prepped by Superose
TM
6 prepgrade (Amersham
Bioscience, Buckinghamshire, UK) and lyophilized.
This peptide synthesis and KLH conjugation procedure
(Kitagawa and Aikawa, 1976; Sabirov et al., 1998) was
conducted at TaKaRa Korea Biomedical Inc. (Seoul,
Korea).
All 500 ml of H159-KLH in PBS (400 mg/ml) was
mixed with an equal volume of Freunds complete
adjuvant (Sigma, MI, US) and injected into a rabbit
subcutaneously. After two weeks, the same amount of
conjugate in PBS was mixed with Freunds incomplete
adjuvant (Sigma, MI, US) and used for injection. Two
weeks after the second injection, 200 mg of H159-KLH
conjugate in PBS was injected into the rabbit for
boosting antibody production. Blood was collected 3
days after the last injection and serum was prepared.
2.11. SDS-PAGE and Western blot
SDS-PAGE was performed according to the method
of Laemmli (1970) using a 12.5% separating slab gel. All
samples (30 mg) were heated at 90 1C for 9 min in the
presence of 2% of SDS and 5% b-mercaptoethanol.
Following SDS-PAGE, proteins in the gel were
electrotransferred to nitrocellulose (0.45 mm, Bio-Rad,
CA, USA) according to the procedure of Towbin et al.
(1979). The blots were blocked in 20 mM Tris-HCl, pH
7.6, 137 mM NaCl, and 0.2% Tween-20 (buffer A)
containing 5% nonfat dry milk, and then incubated with
antisera against 23 kDa lipocalin at 1:1000 dilution
in buffer A. After washing in buffer A, the blots
ARTICLE IN PRESS
were incubated with horseradish-peroxidase-conjugated
goat anti-rabbit IgG (1:3000) in buffer A for 1 h.
Immunoreactivity was determined using the ECL
chemiluminescence reaction (Amersham, Bioscience,
Buckinghamshire, UK).
3. Results
3.1. Sequence of 23 kDa lipocalin gene and its deduced
primary structure
Fig. 1 shows the cDNA sequence of 23 kDa lipocalin
from H. cunea. In a study to identify H. cunea yolk
protein a PCR product of approximately 530 bp was
isolated and sequenced. Blast search indicated that this
sequence was most closely related to lipocalins. This
product was used as a probe to screen the full-length
cDNA of the 23 kDa lipocalin from cDNA libraries of
pupal and adult whole body. The 23 kDa lipocalin
cDNA presents an open reading frame of 582 bp,
comprising the 5
0
and 3
0
untranslated sequences of 32
and 734 bp, respectively (Fig. 1). The presence of a
purine (A) at position 3 in relation to ATG is in
agreement with the strongest consensus sequnce ob-
served at the translation initiation site (Kozak, 1984).
This open reading frame encodes a 194 residue protein
with a calculated molecular mass of 23 kDa. The N-
terminal stretch of 15 amino acids is characteristic of
signal sequences of eukaryotic secreted proteins (Dyrlov
et al., 2004) and one polyadenylation signal (AATAAA)
was found in the 3
0
-untranslated region.
This protein does not appear to be glycosylated with
either N- or O-linked oligosaccharides, but phosphory-
lated at 16 residues. In the grasshopper Lazarillo,
enzymatic deglycosylation showed that 40% of the
ARTICLE IN PRESS
Fig. 1. Nucleic acid and deduced amino acid sequences of 23 kDa lipocalin cDNA from Hyphantria cunea. Potential cleavage site for N-terminal
leader sequence is indicated by arrow. Conserved cysteine residues are circled and the putative phosphorylation sites are boxed. The putative
polyadenylation signal is underlined.
Lazarillo molecular mass is due to the presence of N-
linked oligosaccharides (Sa nchez et al., 2000). Interest-
ingly, the treatment of enzyme phosphatidylinositol-
specic phospholipase C demonstrated that Lazarillo is
attached to the extracellular sides of the plasma
membrane through a glycosyl-phosphatidylinositol tail
which plays a role as the receptor for a midline
morphogen (Sa nchez et al., 2000). Insecticyanins, which
include Pieris BBP and Maduca insecticyanin, are
characterized by low molecular weight and absence of
glycosylation. Although several lipocalins are N-linked
glycosylated (Flower, 1995), the functional signicance
of these carbohydrates remains obscure (Sa nchez et al.,
2000). Likewise, there is no report on the physiological
role of phosphorylation found in several lipocalins
Gallerin, Bombyrin, and grasshopper Lazarillo.
In the 23 kDa lipocalin, four cysteines are well
conserved which are reported to form two disulde
bridges in another lipocalin (Flower et al., 1993;
Ganfornina et al., 2000).
3.2. Sequence comparisons among lipocalin family
To retrieve protein sequences similar to that of
23 kDa lipocalin, relevant databases were searched using
the BLAST (Altschul et al., 1990) and FASTA
programs. When the protein sequence data are com-
pared to those of several lipocalins reported for other
insects and human, the level of similarity of 23 kDa
lipocalin with these proteins varied from 15.3% to
22.2% (Table 1). These values are commonly found
within the members of this family and rarely exceed
20% (Flower, 1995; Flower et al., 1993). Though there is
low overall similarity in the alignment of lipocalins,
sequence conservation is much greater in SCR1
(31.348.3% identity), SCR2 (2033.3% identity), and
SCR3 (2040%) (data not shown). The crystal structure
of several lipocalins showed eight antiparallel b-barrels
and a C-terminal a-chain in which SCRs are well
maintained to adopt a highly characteristic conforma-
tion. (Flower et al., 1993). The 23 kDa lipocalin has the
typical G-X-W motif in SCR1 and a conserved TDY
triplet in SCR2. In SCR3, arginine (R) is conserved in
H. cunea as in most lipocalins (Fig. 2).
Though their low and partial sequence similarity
makes an overall alignment of these protein families
very unreliable, lipocalins have been aligned using their
structural similarity as an important criterion (Flower et
al., 1993). As Rost reports (1997), only 34% of amino
acids appear to be crucial for protein structure and
function. When protein sequence fold into a unique
structure, very different sequences can fold into similar
structures during protein evolution (Rost, 1997). The
23 kDa lipocalin also fullls the criteria for being a
lipocalin: it contains 194 amino acids, its closest relatives
are Lazarillo, apolipoprotein D, and Bombyrin, dis-
playing the complete preservation of key residues in
lipocalin.
3.3. Stage- and tissue-specic expression of 23 kDa
lipocalin
Changes in the amount of the 23 kDa lipocalin in the
whole body mRNA were studied during the develop-
ment of H. cunea by Northern blot. Hybridization of
23 kDa lipocalin cDNA to total whole body RNA from
different stages showed the presence of a single 1.3 kb
mRNA (Fig. 3). The 23 kDa lipocalin transcript
appeared in small amounts in 4-day-old pupae, and
then abruptly increased in 6-day-old pupae. None was
present during the larval, early or late pupal, and adult
stages. It is noteworthy that 23 kDa lipocalin transcrip-
tion in whole body is strictly restricted in the mid-pupal
stage (4- and 6-day-old pupae).
In an immunoblot analysis, tissue distribution of
23 kDa lipocalin differs from those of BBP and INS
where the hemolymph and epidermis are major sites for
their accumulation (Chinzei et al., 1990; Riddiford et al.,
1990; Saito, 1998). The 23 kDa lipocalin was found at a
trace level in the brain and subesophageal ganglion
during early pupal stages, and its level increased to the
mid-pupal stages (Fig. 4). In brain, the level of 23 kDa
lipocalin remained constant until 10-day-old post-
pupation. The level of 23 kDa lipocalin in subesophageal
ganglion, on the other hand, decreased at the end of the
pupal stage. The 23 kDa lipocalin in fat body and
epidermis appeared only from 5-day-old pupae to 10-
day-old pupae and it was not detected before the pupal
stage or after adult eclosion, showing a pupal specic
appearance. Most biliverdin binding proteins are
synthesized in the larval stage and few are observed
in the pupal and adult stages (Haunerland and
Bowers, 1986; Riddiford et al., 1990; Saito, 1998). The
discordance of developmental pattern for 23 kDa
lipocalin between Northern and Western blots (Figs. 3
and 4) could be explained. Even though the transcript
ARTICLE IN PRESS
Table 1
Sequence homology in members of the lipocalin superfamily
1 2 3 4 5 6 7
22.2 20.6 20.1 18 18 15.3
32.8 20.9 21.2 22.2 20.6
24.3 24.9 20.6 19.6
68.2 47.6 43.9
48.7 42.3
36
Identities were determined by pairwise alignment using MEGALIGN.
The sequence sources are as in Fig. 2.
1, Hc 23 KDa Lc; 2, Dm N-Laz; 3, Hs ApoD; 4, Bm Bombyrin; 5, Gm
Gallerin; 6, Pb BBP; 7, Ms INSa.
of 23 kDa lipocalin was present in brain or subesopha-
geal ganglion at an early pupal stage, RNA from
both tissues only provides a small proportion of
whole body RNA. That is why its transcript is not
found in the 0- or 2-day-old pupal whole body,
though its expression product was found in both tissues
(Figs. 3 and 4). The 23 kDa lipocalin produced during
mid-pupal stage should persist to the end of the pupal
stage (Fig. 4).
H. cunea larvae appear yellowish-white and has
yellowish-brown hemolymph. Besides coloration, BBP
and INS are speculated to function as carrier proteins of
physiologically important lipophilic compounds which
ARTICLE IN PRESS
Fig. 2. Multiple alignments of 23 kDa lipocalin amino acids with several members of the lipocalin family. Conserved cysteine residues among the
lipocalin family are indicated by closed triangle. The structurally conserved regions (SCR) dening the lipocalin family are indicated above. Typical
conserved residues in three SCRs were marked by an asterisk. The sequences were aligned with CLUSTAL W. The sequence sources are as follows,
Hc 23 kDa Lc, 23 kDa lipocalin from Hyphantria cunea; Dm N-Lazarillo, neural lazarillo from Drosophila melanogaster (AAF85707); Hs ApoD,
apolipoprotein D from Homo sapiens (AAB59517); Bm Bombyrin, Bombyrin from Bombyx mori (BAB47155); Gm Gallerin, Gallerin from Galleria
mellonella (AAA85089); Pb BBP, bilin binding protein from Pieris brassicae (CAA54063); Ms INSa, insecticyanin a from Manduca sexta
(CAA45969).
Fig. 3. Developmental changes of 23 kDa lipocalin mRNA in the
whole body from Hyphantria cunea. 10 mg of total RNA was applied to
each lane and subjected to Northern blot analysis. 3L, 5L, and 7L,
third, fth, and seventh instar larvae; PP, prepupae; P0P10, pupae at
days 010; Ad, adult. The lower panel shows ribosomal RNA (rRNA)
on the same lter previous to hybridization as a control for the
amounts of RNA loaded.
are poorly soluble in the hemolymph (Chinzei et al.,
1990). The possibility of physiological role in coloration
seems to be weak. We also speculate that a physiological
function other than coloration for 23 kDa lipocalin
might exist.
3.4. Possible physiological function of 23 kDa lipocalin in
brain
We observed that brain size distinctly increases during
the mid-pupal stage (Fig. 5). The 23 kDa lipocalin
accumulates during mid-pupal stage (Fig. 4) when the
enlargement of brain occurs, suggesting that this
lipocalin could have some roles in the brain develop-
ment. But the exact function of 23 kDa lipocalin in brain
and subesophageal ganglion is not elucidated by the
present study. The subesophageal ganglion, whose
activity is under the control of brain, releases diapause
hormone which acts on developing ovaries in the mid-
pupal stage and in Bombyx mori, induces embryonic
diapause (Yamashita, 1996). We attempted to nd a
relationship between 23 kDa lipocalin and diapause by
Western blots to investigate quantitative changes of
lipocalin in the brain and subesophageal ganglion of H.
cunea under natural diapause (data not shown). No
signicant quantitative differences were found in 23 kDa
lipocalin from both tissues. Therefore we tested other
possible functions of 23 kDa lipocalin in H. cunea.
To elucidate the potential involvement of 23 kDa
lipocalin in wound healing in brain, we minced brains
dissected from 5-day-old pupae and incubated them for
12 and 24 h, respectively. The extracted protein from
brain was analyzed by Western blot using antibodies
against 23 kDa lipocalin, apolipophorin- (Kim et al.,
2004), storage protein-1 (Cheon et al., 2002), ferritin
(Kim et al., 1996), and actin (Sigma, MI, US),
respectively (Fig. 6). Relatively small amounts of
23 kDa lipocalin accumulated in the pupal brain, but it
was highly reactive to cutting injury, inducing an
increase of 23 kDa lipocalin up to 24 h post-injury
(Fig. 6). In contrast, a high concentration of apolipo-
phorin-III is present in pupal brain at normal condition.
ARTICLE IN PRESS
Fig. 4. Accumulation of 23 kDa lipocalin in several organs during
development. Proteins were fractionated by SDS-PAGE (A) and a
Western blot (B) was performed using antibody against 23 kDa
lipocalin. Stages are as in Fig. 3. Br, brain; Sg, subesophageal
ganglion; Fb, fat body; Ep, epidermis; He, hemolymph.
Brain Size
1.4
1.2
1
0.8
0.6
0.4
0.2
0
m
i
l
l
i
m
e
t
e
r

(
m
m
)
7L P1 P3 P5 Ad
Developmental stage
*
*
Fig. 5. Changes of brain size during development. Shaded and open
bars indicate width and height of brain, respectively. Stages are as in
Fig. 3. The results shown are the average of 30 female animals with
standard deviation indicated by T-bars. Asterisks denote a signicant
differences, 0.01oPp0:05**.
The increased accumulation of apolipophorin-III in
brain tissue is presumably due to cutting injury.
Apolipophorin-III is abundantly expressed in the
cytoplasm of dying muscles, interneurons, and motor
neurons at the time of cell death in Manduca sexta (Sun
et al., 1995). The accumulation of 23 kDa lipocalin
against injury is unexpected. The common structure of
the lipocalin is eight b-strands of the barrel enclosing a
ligand binding site (Flower et al., 2000), while apolipo-
phorin-III is an exchangeable insect apolipoprotein
consisting of ve amphipathic a-helices and this protein
is able to open reversibly on association with hydro-
phoic surfaces (Breiter et al., 1991). Therefore, the
structural similarity between 23 kDa lipocalin and
apolipophorin-III seems to be distant, but both proteins
commonly have afnity to hydrophobic molecules.
Since dying cells could potentially contain high con-
centrations of hydrophobic residue liberated by degra-
dative processes, one could hypothesize that might act
to inhibit precipitation of constituents intended for
export from the cell (Sun et al., 1995). Ferritin showing
three subunits (26, 31, and 32 kDa) was detected in brain
but showed no ucturation of amount between control
and injury groups (Fig. 6).
The 23 kDa lipocalin shows homology to those of
apoD from human (20.6%), Lazarillo from Drosophila
(22.2%), and Bombyrin from Bombyx (20.1%) (Table
1). Lazarillo is expressed during grasshopper embry-
ogenesis on the surface of a subset of central nervous
system neurons, and perturbing the function of Lazarillo
specically affects the pathway decisions made by a
subset of developing neurons (Sa nchez et al., 2000).
In Hyphantria embryo, we could not nd any
accumulation of 23 kDa lipocalin by Western and
Northern blots (data not shown). It majorly accumu-
lates in nerve tissues during the middle pupal stages
parallel with the growth of brain. Spreyer et al. (1990)
and Boyles et al. (1990) both reported the increased
apolipoprotein D secretion in regenerating nerve tissue.
Spreyer et al. (1990) reported a 40-fold accumulation of
apolipoprotein D in the extracellular space of regener-
ating crushing rat neurons and Boyles et al. (1990)
reported a 500-fold accumulation of apolipoprotein D
expression in the regenerating peripheral nerve of rat,
rabbits, and monkeys. In comparison, 23 kDa lipocalin
shows about 5-fold increase in amount against injury
(data not shown). While this increase is less substantial
relative to apolipoprotein in vertebrates, it suggests that
23 kDa lipocalin may have an important role to play in
the repair process of brain. In human and rabbit spleen,
apolipoprotein D could be a scavenger for heme-related
toxic molecules (Se guin et al., 1995). The same protein
could have a role in either growing or degradative cells
in response to different conditions. Future research will
be needed to verify the function of pupae-specic
23 kDa lipocalin in nerve tissue of H. cunea.
Acknowledgements
This research was supported by a grant from Korea
Science Engineering Foundation (ABRL Program, R14-
2002-056-01001-0) and Brain Korea 21 Project of
Gyeongsang National University.
References
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990.
Basic local alignment search tool. J. Mol. Biol. 215, 403410.
Boyles, J.K., Notterpek, L.M., Anderson, L.J., 1990. Accumulation of
apolipoproteins in the regenerating and remyelinating mammalian
peripheral nerve. Identication of apolipoprotein D, apolipopro-
tein A-IV, apolipoprotein E, and apolipoprotein A-I. J. Biol.
Chem. 265, 1780517815.
Breiter, D.R., Kanost, M.R., Benning, M.M., Wesmberg, G., Law,
J.H., Wells, M.A., 1991. Molecular structure of an apolipoprotein
determined at 2.5 A
resolution. Biochemistry 30, 603608.

Cheon, H.M., Hwang, S.J., Kim, H.J., Jin, B.R., Chae, K.S., Yun,
C.Y., Seo, S.J., 2002. Two juvenile hormone suppressible storage
proteins may play different roles in Hyphantria cunea Drury. Arch.
Insect Biochem. Physiol. 50, 157172.
Cheon, H.M., Kim, H.J., Yun, C.Y., Lee, H.J., Lee, I.H., Shirk, P.D.,
Seo, S.J., 2003. Fat body expressed yolk protein genes in
Hyphantria cunea are related to the YP4 follicular epithelium yolk
protein subunit gene of pyralid moths. Insect Mol. Biol. 12,
383392.
ARTICLE IN PRESS
SP1
23 KDa Lc
ApoLpIII
Actin
Ferritin
0 12 24 (h)
Fig. 6. Western blot of brain protein from 5-day-old pupae with or
without injury. Twenty brains for each group were incubated for 12
and 24 h after injury. The proteins extracted from brains were
separated by SDS-PAGE and Western blot was performed using ve
different antibodies. Lane 1, not injured control; Lane 2 and 3, 12 and
24 h post-injury. Lc, lipocalin; ApoLp III, apolipophorin III; SP1,
storage protein 1 from Hyphantria cunea.
Chino, H., Abe, U., Takahashi, K., 1983. Purication and character-
ization of a biliverdin-binding cyanoprotein from the locust
haemolymph. Biochem. Biophys. Acta. 48, 109115.
Chinzei, Y., Haruna, T., Miura, K., Numata, H., Nakayama, S., 1990.
Purication and characterization of biliverdin-associated cyano-
protein from eggs and hemolymph of the bean bug, Riptortus
clavatus. Insect Biochem 20, 545555.
Dyrlov, B.J., Nielsen, H., Von Heijne, 2004. Improved prediction of
signal peptides: SignalP 3.0. J. Mol. Biol. 340, 783795.
Feinberg, A.P., Volgelstein, B., 1983. A technique for radiolabeling
DNA restriction endonuclease fragments to high specic activity.
Anal. Biochem. 132, 613.
Filippov, V.A., Filippova, M.A., Kodrik, D., Sehnal, F., 1995. Two
lipocalin-like peptides of insect brain. In: Suzuki, A., Kataoka, H.,
Matsumoto, S. (Eds.), Molecular Mechanisms of Insect Metamor-
phosis and Diapause. Industrial Publishing & consulting, Inc,
Tokyo, pp. 3543.
Flower, D.R., 1995. Multiple molecular recognition properties of the
lipocalin protein family. J. Mol. Recognit. 8, 185195.
Flower, D.R., 1996. The lipocalin protein family: structure and
function. Biochem. J. 318, 114.
Flower, D.R., North, A.C.T., Arkwood, T.K., 1993. Structure and
sequence relationships in the lipocalins and related proteins.
Protein Sci. 2, 753761.
Flower, D.R., North, A.C.T., Sansom, C.E., 2000. The lipocalin
protein family; structural and sequence overview. Biochim.
Biophys. Acta. 1482, 924.
Ganfornina, M.D., Sa nchez, D., Bastiani, M.J., 1995. Lazarillo, a new
GPI-linked surface lipocalin, is restricted to a subset of neurons in
the grasshopper embryo. Development 121, 123134.
Ganfornina, M.D., Gutie rrez, G., Bastiani, M., Sa nchez, D., 2000. A
phylogenetic analysis of the lipocalin protein family. Mol. Biol.
Evol. 17, 114126.
Goodman, W., Adams, G.B., Trost, J.T., 1985. Purication and
characterization of a biliverdin-associated protein from the
hemolymph of Manduca sexta. Biochemistry 24, 11681175.
Haunerland, N.H., Bowers, W.S., 1986. A larval specic lipoprotein:
purication and characterization of a blue chromoprotein
from Heliothis zea. Biochem. Biophys. Res. Commun. 134,
580586.
Huber, R., Schneider, M., Mayr, I.A., Messerschmidt, A., Flugrath, J.,
Kayser, H., 1987a. Crystallization, crystal structure analysis and
preliminary molecular model of the bilin binding protein from the
insect, Pieris brassicae. J. Mol. Biol. 195, 423434.
Huber, R., Schneider, M., Muller, I., Deutzann, R., Suter, F., Zuber,
H., Falk, H., Kayser, H., 1987b. Molecular structure of the bilin
binding protein (BBP) from Pieris brassicae after renement at
2.0 A
resolution. J. Mol. Biol. 198, 499513.

Ito, T., Tanaka, M., 1960. Rearing of the silkworm on an articial diet
and the segregation of pentamolters. J. Seric. Sci. Japan 29,
191196.
Kim, R.A., Lee, S.G., Yun, C.Y., 1996. Purication of ferritin of larval
hemolymph from fall webworm, Hyphantria cunea. Korean J.
Entomol. 26, 135142.
Kim, H.J., Je, H.J., Park, S.Y., Lee, I.H., Jin, B.R., Yun, H.K., Yun,
C.Y., Han, Y.S., Kang, Y.J., Seo, S.J., 2004. Immune activation of
apolipophorin- and its distribution in hemocyte from Hyphantria
cunea. Insect Biochem Mol. Biol. 34, 10111023.
Kitagawa, T., Aikawa, T., 1976. Enzyme coupled immunoassay of
insulin using a novel coupling reagent. J. Biochem. 79, 233236.
Kozak, M., 1984. Compilation and analysis of sequences upstream
from the translational start site in eukaryotic mRNAs. Nucleic
Acids Res. 12, 857872.
Laemmli, U.K., 1970. Cleavage of structural proteins during the
assembly the head of bacteriophage T4. Nature 227, 680685.
Li, W.C., Riddiford, L.M., 1994. The two duplicated insecticyanin
genes, ins-a and ins-b are differentially expressed in the tobacco
hornworm, Manduca sexta. Nucleic acids Res. 22, 29452950.
Riddiford, L.M., Palli, S.R., Hiruma, K., Li, W., Green, J., Hice,
R.H., Wolfgang, W.J., Webb, B.A., 1990. Developmental expres-
sion, synthesis, and secretion of insecticyanin by the epidermis of
the tobacco hornworm, Manduca sexta. Arch. Insect Biochem.
Physiol. 14, 171178.
Riley, C.T., Barbeau, B.K., Keim, P.S., Ke zdy, F.J., Heinrikson, R.L.,
Law, J.H., 1984. The covalent protein structure of insecticyanin, a
blue biliprotein from the hemolymph of the tobacco hornworm,
Manduca sexta. L. J. Biol. Chem. 259, 1315913165.
Rost, B., 1997. Protein structures sustain evolutionary drift. Fold.
Design 2, 1924.
Sabirov, A.N., Kim, Y.D., Kim, H.J., Samukov, V.V., 1998. FMOC-
and NSC-Groups as a base labile N(a)-amino protection: a
comparative study in the automated SPPS. Protein Peptide lett.
5, 5762.
Saito, H., 1998. Purication and characterization of two insecticyanin-
type proteins from the larval hemolymph of the Eri-silkworm,
Samia cynthia ricini. Biochem. Biophys. Acta. 1380, 141150.
Sakai, M., Wu, C., Suzuki, K., 2001. Nucleotide and deduced amino
acid sequences of a cDNA encoding a lipocalin protein in the
central nervous system of Bombyx mori. J. Insect Biochem. Sericol.
70, 105111.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.
Sa nchez, D., Ganfornina, M.D., Bastiani, M.J., 1995. Developmental
expression of the lipocalin Lazarillo and its role in axonal
pathnding in the grasshopper embryo. Development 121,
135147.
Sa nchez, D., Ganfornina, M.D., Torres-Schumann, S., Speese, S.D.,
Lora, J.M., Bastiani, M.J., 2000. Characterization of two novel
lipocalins expressed in the Drosophila embryonic nervous system. J.
Dev. Biol. 44, 349359.
Sanger, F., Nicklen, S., Coulson, A.R., 1977. DNA sequencing with
chain-terminating inhibitors. Proc. Nat. Acad. Sci. USA 74,
54635467.
Se guin, D., Desforges, M., Rassart, E., 1995. Molecular characteriza-
tion and differential mRNA tissue distribution of mouse apolipo-
protein D. Mol. Brain Res. 30, 242250.
Spreyer, P., Schaal, H., Kuhn, G., Rothe, T., Unterbeck, A., Olek, K.,
Mller, H.W., 1990. Regeneration-associated high level expression
of apolipoprotein D mRNA in endoneurial broblasts of
peripheral nerve. The EMBO J. 9, 24782484.
Sun, D., Ziegler, R., Milligan, C.E., Fahrbach, S., Schwartz, L.M.,
1995. Apolipophorin is dramatically upregulated during the
programmed death of insect skeletal muscle and neurons. J.
Neurobiol. 26, 119129.
Suter, F., Kayser, H., Zuber, H., 1988. The complete amino acid
sequence of the bilin-binding protein from Pieris brassicae and its
similarity to a family of serum transport proteins like the retinol-
binding proteins. Biol. Chem Happe-Seyle 369, 497505.
Towbin, H., Staehelin, T., Gordon, J., 1979. Electrophoretic transfer
of prteins from polyacrylamide gels to nitrocellulose sheets:
Procedure and some applications. Proc. Nat. Acad. Sci. USA 76,
43504354.
Yamashita, O., 1996. Diapause hormone of the silkworm, Bombyx
mori; structure, gene expression and function. J. Insect Physiol. 42,
669679.
Yoshiga, T., Tojo, S., 1995. Purication and characterization of four
biliverdin-binding proteins from larval hemolymph of the common
cutworm, Spodoptera litura. Insect Biochem. Mol. Biol. 25,
575581.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
The transcriptome of the salivary glands of the female western
black-legged tick Ixodes pacicus (Acari: Ixodidae)
Ivo M.B. Francischetti
a,
, Van My Pham
a
, Ben J. Mans
a
, John F. Andersen
a
,
Thomas N. Mather
b
, Robert S. Lane
c
, Jose M.C. Ribeiro
a
a
Section of Vector Biology, Laboratory of Malaria and Vector Research, National Institute of Allergy and Infectious Diseases,
National Institutes of Health, Rockville, MD 20892-8132, USA
b
Center for Vector-borne Disease, University of Rhode Island, Kingston, RI 02881-0804, USA
c
Department of Environmental Science, Policy and Management, University of California, Berkeley, CA 94720, USA
Received 14 April 2005; received in revised form 19 May 2005; accepted 20 May 2005
Abstract
Sequencing of an Ixodes pacicus salivary gland cDNA library yielded 1068 sequences with an average undetermined nucleotide
of 1.9% and an average length of 487 base pairs. Assembly of the expressed sequence tags yielded 557 contigs, 138 of which appear
to code for secreted peptides or proteins based on translation of a putative signal peptide. Based on the BLASTX similarity of these
contigs to 66 matches of Ixodes scapularis peptide sequences, only 58% sequence identity was found, indicating a rapid divergence of
salivary proteins as observed previously for mosquito and triatomine bug salivary proteins. Here we report 106 mostly full-length
sequences that clustered in 16 different families: Basic-tail proteins rich in lysine in the carboxy-terminal, Kunitz-containing proteins
(monolaris, ixolaris and penthalaris families), proline-rich peptides, 5-, 9.4- and 18.7-kDa proteins of unknown functions, in
addition to metalloproteases (class PIII-like) similar to reprolysins. We also have found a family of disintegrins, named ixodegrins
that display homology to variabilin, a GPIIb/IIIa antagonist from the tick Dermacentor variabilis. In addition, we describe peptides
(here named ixostatins) that display remarkable similarities to the cysteine-rich domain of ADAMST-4 (aggrecanase). Many
molecules were assigned in the lipocalin family (histamine-binding proteins); others appear to be involved in oxidant metabolism,
and still others were similar to ixodid proteins such as the anticomplement ISAC. We also identied for the rst time a neuropeptide-
like protein (nlp-31) with GGY repeats that may have antimicrobial activity. In addition, 16 novel proteins without signicant
similarities to other tick proteins and 37 housekeeping proteins that may be useful for phylogenetic studies are described. Some of
these proteins may be useful for studying vascular biology or the immune system, for vaccine development, or as immunoreagents to
detect prior exposure to ticks. Electronic version of the manuscript can be found at http://www.ncbi.nlm.nih.gov/projects/omes/.
Published by Elsevier Ltd.
Keywords: Ixodes pacicus; Sialome; Tick; Blood-feeding; Kunitz inhibitor; Lyme disease; Vascular biology; Ixolaris; Vector biology; Transcriptome;
Proteome
1. Introduction
Lyme disease is the most prevalent vector-borne
disease in the US and is transmitted by the tick vectors
Ixodes scapularis and I. pacicus in eastern and western
North America, respectively (Barbour, 1998). Humans
usually acquire Lyme disease when an infected nymphal-
stage Ixodes sp. tick attaches and transmits the
spirochete Borrelia burgdorferi (Burgdorfer et al.,
1985). I. scapularis and I. pacicus transmit other
zoonotic agents besides the Lyme disease spirochete,
such as Anaplasma phagocytophilum (both species) or
Babesia microti (I. scapularis only) (Barbour, 1998).
ARTICLE IN PRESS
0965-1748/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.ibmb.2005.05.007

fax: +1 301 480 2571.
E-mail address: ifrancischetti@niaid.nih.gov
(I.M.B. Francischetti).
Transmission is facilitated by tick saliva that operates
not only as a carrier for Borrelia sp. but also contains a
large repertoire of molecules that counteract the host
response to injury (Ribeiro and Francischetti, 2003),
allowing ticks to feed for days (Sonenshine, 1985).
Accordingly, many biologic activities have been de-
scribed in tick saliva, including molecules that impair
platelet aggregation or neutrophil function (Ribeiro
et al., 1985) in addition to coagulation inhibitors such as
ixolaris and penthalaris that block Factor VIIa/tissue
factor complex (Francischetti et al., 2002a, 2004a) and
SALP 14, which targets Factor Xa (Narasimhan et al.,
2002). Enzymes such as a kininase that degrades
bradykinin (Ribeiro and Mather, 1998), an apyrase that
destroys ADP (Ribeiro et al., 1985), and a metallopro-
tease with brin(ogen)olytic activity (Francischetti et al.,
2003) also have been reported. Tick saliva is also rich in
small molecules such as prostacyclin, a potent inhibitor
of platelet activation and strong inducer of vasodilation
(Ribeiro et al., 1988).
As for the immune system, an inhibitor of the
alternative complement pathway exists in ixodid tick
saliva (Valenzuela et al., 2000). Immunomodulators
affecting NK cell function (Kopecky and Kuthejlova,
1998)in addition to inhibitors of the proliferation of T
lymphocytes and an IL-2 binding activityalso are
present in this secretion (Ramachandra and Wikel, 1992;
Gillespie et al., 2001). Finally, saliva is important in
transmission of tick-borne pathogens, as it may enhance
pathogen transmission (for a review, see Wikel, 1999).
The pace of discovery of tick salivary proteins has
been greatly increased by novel molecular biology
techniques and bioinformatics analysis (Ribeiro and
Francischetti, 2003). Our goal here has been to further
study the complexity of I. pacicus salivary glands. We
report the full-length clone of 87 novel sequences and
discuss their potential role in modulating host inam-
matory and immune responses.
2.1. Reagents
All water used was of 18 MO quality and was
produced using a MilliQ apparatus (Millipore, Bedford,
MA, USA). Organic compounds were obtained from
Sigma (St. Louis, MO, USA) or as stated otherwise.
2.2. I. pacicus ticks
2.2.1. Salivary gland cDNA library construction and
sequencing
Ticks were collected in northern California by
dragging low vegetation with a tick-drag. Salivary
glands were excised and kept at 80 1C until use. The
mRNA from two pairs of I. pacicus salivary glands was
obtained using a Micro-Fast Track mRNA isolation kit
(Invitrogen, San Diego, CA, USA) according to the
manufacturers instructions. The PCR-based cDNA
library was made following the instructions for the
SMART cDNA library construction kit (Clontech,
Palo Alto, CA, USA) as described in detail in the
supplemental data in Francischetti et al. (2004b).
Cycle sequencing reactions using the DTCS labeling
kit (Beckman Coulter, Fullerton, CA, USA) were
performed as reported (Francischetti et al., 2004b)
and can be found as supplemental data at http://
www.ncbi.nlm.nih.gov/projects/omes/ in the section
Poisonous Animals.
2.2.2. CDNA sequence clustering and bioinformatics
Other procedures were as reported in detail in the
supplemental data described in Francischetti et al.
(2004b) and can be found as supplemental data at
http://www.ncbi.nlm.nih.gov/projects/omes/ in the sec-
tion Poisonous Animals.
2.2.3. Structural bioinformatics and molecular modeling
Molecular model of the histamine-binding protein-
like lipocalin gi 51011604 superimposed with the crystal
structure of Rhipicephalus appendicuatus histamine-
binding protein. The 3D-PSSM web server V2.6.0,
found at http://www.sbg.bio.ic.ac.uk/ server was used
to generate a model of gi 51011604 based on sequence
alignment using PSI Blast, secondary structure predic-
tion and search of a fold database of known structures
(Kelley et al., 2000).
2.2.4. Electronic version of the manuscript
The electronic version of the manuscript containing
gures and table with hyperlinks can be found at http://
www.ncbi.nlm.nih.gov/projects/omes/, in the section
Salivary transcriptomes (sialome) of vector arthropods
(I. pacicus).
3. Results and discussion
I. scapularis and I. pacicus are the respective vectors
for B. burgdorferi in the eastern and western US (Fig. 1).
After attachment to the host, infected ticks transmit B.
burgdoferi after 12 days of blood-feeding (Barbour,
1998) via saliva, a secretion that contains a cocktail of
bioactive molecules (Ribeiro and Francischetti, 2003).
Actually, the identication of the transcripts and
proteins present in the salivary gland of ticks such as
I. scapularis (Valenzuela et al., 2002), Boophilus micro-
plus (Santos et al., 2004), and Rhipicephalus appendicu-
latus (Nene et al., 2004) have been identied recently.
Here we identied secretory genes from the salivary
gland of I. pacicus by constructing a unidirectional
ARTICLE IN PRESS
I.M.B. Francischetti et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11421161 1143
PCR-based cDNA library (see Section 2). Next, 735
cDNA were randomly sequenced followed by bioinfor-
matics analysis that included: (i) clustering at high
stringency levels, (ii) BLAST search against the non-
redundant and protein motifs databases, and (iii)
submission of the translated sequences to the Signal P
server (see Materials and methods). This initial ap-
proach allowed us to obtain a ngerprint of the protein
families or clusters present in this particular salivary
gland. Several sequences were then selected based on
novelty or the protein family it assigns for and extension
of their corresponding cDNA were performed until the
poly A was reached. Among these clusters, 87 novel full-
length cDNA coding proteins or peptides were obtained,
most of which appear to be secreted in the saliva.
Our results are presented in Table 1, which describes
the sequence size, the presence of a putative signal
peptide, the molecular weight of the mature peptide, the
isoelectric point, and other parameters (each accession
numbers and sequence information is hyperlinked).
Fifteen large protein families of putative secreted
proteins were found. Some sequences appeared to code
for housekeeping proteins, whereas others without
database hits but containing an open-reading frame
with or without signal peptide were considered novel or
unknown-function proteins. Considering the diverse
roles of putative secreted proteins in blood feeding, a
brief description for each protein family is presented
below.
3.1. Group 1: Basic-tail proteins (BTP)
This family of proteins is highly represented in the
salivary glands of both I. pacicus and I. scapularis ticks.
Fig. 2A shows the alignments of the BTP of these ticks
where a highly conserved signal peptide indicates their
common origin from an ancestral gene. Fig. 2A also
shows that the pattern of these sequences contain six
cysteines (XnCX14CX3CX18CX9CX4CXn) followed
by a basic tail with high content of lysines (Lys, K).
ARTICLE IN PRESS
Fig. 1. Established and reported distribution of the Lyme disease vectors I. scapularis (I. dammini) and I. pacicus by county, United States,
19071996. Distribution was reported by the Centers for Disease Control and Prevention and can be found at http://www.cdc.gov/ncidod/dvbid/
lyme/tickmap.htm.
A
R
T
I
C
L
E
I
N
P
R
E
S
S
Table 1
NCBI Id
number and link
Seq. size SigP result Cleavage
position
Mature MW pI Best match to non-redundant NCBI protein
database
E-value Identication
Probably secreted proteins
Group 1sequences with basic tail similar to Group I peptides of I. scapularisPutative anti-hemostatic
51011436 120 SIG 2122 11.042 8.53 14 kDa salivary gland protein [Ixodes 1E059 Group 1 basic tail salivary peptide
51011428 120 SIG 2122 11.028 8.53 Putative secreted protein [Ixodes sca 9E059 Group 1 basic tail salivary peptide
51011462 125 SIG 2122 11.542 9.04 Putative secreted salivary protein [I 1E062 Group 1 basic tail salivary peptide
51011546 138 SIG 2223 13 9.46 Putative secreted salivary protein [I 2E058 Group 1 basic tail salivary peptide
51011492 115 SIG 2122 10.613 8.92 Salivary secreted protein [Ixodes sca 4E055 Group 1 basic tail salivary peptide
51011432 118 SIG 2122 10.678 4.58 Salivary secreted protein [Ixodes sca 3E007 Group 1 basic tail salivary peptide
Group 2Similar to Group 1 proteins from I. scapularis and I. pacicus, without basic tailPutative anti-hemostatic
51011488 91 SIG 2021 7.855 9.22 Putative 8.4 kDa secreted protein [Ix 6E043
51011490 90 SIG 1819 8.011 9.21 Putative 8.4 kDa secreted protein [Ix 3E044
Group 3Sequences with Kunitz domains
Monolaris family (1-Kunitz domain)putative anti-coagulant
51011560 85 SIG 2122 6.953 4.67 Putative secreted protein [Ixodes sca 4E029 Group 2 of Ixodes scapularissingle Kunitz
proteins
Ixolaris family (2-Kunitz domains)anti-coagulant
51011542 167 SIG 3132 14.94 5.03 Ixolaris [Ixodes scapularis] 205 3e052 3E052 Ixolaris homologue
51011606 186 SIG 2021 19.074 9.2 CG33103-PB [Drosophila melanogaster 1E009 Kunitz and thrombospondin similarity
51011414 142 SIG 2021 13.999 9.08 Tissue factor pathway inhibitor 2 [ 4E022 similarity to tissue factor pathway inhibitor 2
51011412 167 SIG 2122 16.453 7.44 Putative secreted protein [Ixodes sca 2E014 Kunitz protease inhibitor domain
Penthalaris (5-Kunitz domains)anti-coagulant
Group 4of related sequences rich in proline found in other Ixodidaeunknown function
51011452 65 SIG 1920 4.406 8.96 Putative secreted protein [Ixodes sca 3E023 Group 3 collagen-like salivary secreted peptides
Group 5Sequences similar to I. scapularis 18.7 kDa proteinunknown function
51011468 189 SIG 2021 18.945 4.92 Putative 18.7 kDa secreted protein [I 4E042 Putative 18.9 kDa secreted protein
51011470 189 SIG 2021 19.004 4.95 Putative 18.7 kDa secreted protein [I 4E040 Putative 19 kDa secreted protein [Ixodes
pacicus]
I
.
M
.
B
.
F
r
a
n
c
i
s
c
h
e
t
t
i
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
1
4
2
1
1
6
1
1
1
4
5
A
R
T
I
C
L
E
I
N
P
R
E
S
S
Table 1 (continued )
NCBI Id
number and link
position
database
51011570 189 SIG 2021 19.028 4.92 Putative 18.7 kDa secreted protein [I 3E043 Putative 19 kDa secreted protein [Ixodes
pacicus]
Group 6Sequences similar to I. scapularis 5.3 kDa peptideunknown function
51011582 61 SIG 1819 4.283 9.38 Putative 5.3 kDa secreted protein [Ix 0.015 Similar to I. scapularis 5.3 kDa peptide
51011562 79 SIG 2223 6.314 9.57 ENSANGP00000017973 [Anopheles gambi 0.34 Similar to I. Scapularis 5.3 kDa peptide
Group 7Sequences similar to I. scapularis 9.4 kDa peptideunknown function
51011580 103 SIG 2021 9.57 8.23 Putative 9.4 kDa secreted protein [Ix 3E019 Similar to I. scapularis 9.4 kDa peptide
51011396 101 SIG 2021 9.416 8.46 Putative 9.4 kDa secreted protein [Ix 6E035 Similar to I. scapularis 9.4 kDa peptide
51011434 79 SIG 1819 6.779 5.19 Putative 7 kDa secreted protein [Ixod 2E034 Similar to I. scapularis 9.4 kDa peptide
Group 8Metalloprotease familyputative anti-hemostatic
51011584 344 CYT Truncated secreted metalloprotease [I 0.0 Truncated secreted metalloprotease
Group 9Ixodegrin family: disintegrinsputative anti-hemostatic
51011476 60 SIG 2122 4.173 6.47 HL01481p [Drosophila melanogaster] 33 1.8 1.8 Cys-rich
Group 10Ixostatin family: short-coding cysteine-rich peptides similarly found in ADAMST-4 (Thrombospondins)putative anti-hemostatic
51011550 115 SIG 1819 10.688 5.6 Thrombospondin [Ixodes scapularis] 80 1e014 1E014 Kunitz protease inhibitor domain
Group 11Histamine binding proteinsputative anti-hemostatic
51011424 313 SIG 1718 33.149 5.08 Putative secreted histamine binding p 1E101 Putative secreted histamine binding protein
51011558 191 SIG 1819 19.53 4.49 Putative secreted histamine binding p 5E030 Putative secreted histamine binding protein
51011480 196 SIG 1819 19.708 5.66 Putative 22.7 kDa secreted protein [I 1E093 Putative secreted histamine binding protein
51011422 265 CYT Putative secreted histamine binding p 1E142 Truncated histamine binding protein
51011604 244 SIG 2021 25.901 8.85 Putative protein [Ixodes scapularis] 331 7e090 7E090 Putative secreted histamine binding protein
51011586 215 SIG 2021 22.52 8.71 Putative protein [Ixodes scapularis] 333 2e090 2E090 Putative secreted histamine binding protein
51011498 160 CYT Histamine binding protein [Ixodes sca 4E017 Truncated histamine binding protein
51011532 210 SIG 1617 22.767 9.35 Serotonin and histamine binding prote 2E005 Putative secreted histamine binding protein
51011410 214 SIG 2728 21.262 6.12 Putative 22.5 kDa secreted protein [I 0.79 Putative secreted histamine binding protein
51011416 193 SIG 19-20 20.018 6.16 Putative secreted protein [Ixodes sca 4E027 Putative secreted histamine binding protein
Group 12Neuropeptide-like protein with GGY repeatsputative anti-microbial
51011444 73 SIG 2324 4.758 9.63 Neuropeptide-like protein (nlp-31) 6E017 Reference
51011404 78 SIG 2324 5.416 9.7 Neuropeptide-like protein (nlp-31) 6E020 Reference
Group 13Oxidant metabolism
51011420 116 CYT Plasma glutathione Peroxidase [Homo sa 5E027 Truncated glutathione peroxidase
51011588 189 SIG 2324 17.439 7.15 Mn superoxide dismutase [Melopsittacu 8E051 Mn superoxide dismutase
Group 14Similar to other Ixodid proteins
51011484 119 SIG 1617 11.293 9.01 15 kDa salivary gland protein [Ixodes 5E026 Salp15 family
51011390 124 SIG 2021 11.314 4.66 15 kDa salivary gland protein [Ixodes 0.001 Salp15 family
51011590 154 SIG 1819 14.408 5.14 Salivary gland 16 kDa protein [Ixodes 1E031 Salp15 family
51011394 123 SIG 1819 11.19 6.26 Salivary gland 16 kDa protein [Ixodes 4E010 Domain 8 of human ADAMS
51011564 108 SIG 2425 9.089 8 16 kDa salivary gland protein A [Ixod 4E004 Some similarity with factor VII
51011592 178 SIG 2223 17.328 4.38 20 kDa salivary gland protein [Ixodes 7E065 Anti-complement, ISAC
51011398 119 SIG 2627 9.778 7.76 Is3 [Ixodes scapularis] 36 0.13 0.13 Similar to is3 protein
I
.
M
.
B
.
F
r
a
n
c
i
s
c
h
e
t
t
i
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
1
4
2
1
1
6
1
1
1
4
6
A
R
T
I
C
L
E
I
N
P
R
E
S
S
51011402 240 SIG 1819 21.295 9.03 Hypothetical protein [Ixodes ricinus] 267
2e070
2E070 Possible cuticle or salivary duct protein
Group 15Novelunknown
51011472 87 SIG 2324 6.298 4.13 Insecticidal neurotoxin Tx4(5-5 0.75 Cys-rich, weak similarity to neurotoxin
51011482 120 SIG 2021 11.203 4.36 Nematocyst outer wall antigen precurs 0.077
51011392 163 SIG 3031 14.98 8.58 Probable K5 antigen synthesis [Vibrio 3.4
51011602 117 SIG 1920 11.113 9.59 AGR_C_2052p [Agrobacterium tumefaci 0.13
51011598 78 SIG 2324 5.358 4.9 OSJNBa0086B14.5 [Oryza sativa (japon 0.35 polyGly tailglue?
51011474 45 SIG 1920 2.496 4.84 COG3451: Type IV secretory pathwa 2.5 HEAHEAHEA protein
51011540 59 SIG 2223 4.117 7.92
51011566 122 BL 1920 11.717 9.83 Predicted protein [Ustilago maydis 521] 31 4.2 4.2 Unknown, possible Dopachrome tautomerase
precursor
51011384 124 SIG 2324 10.597 5.28 Keratin associated protein 18-7; ke 6E005 Very cys-richglue?
51011400 168 SIG 2122 15.407 6.54 Putative protein (4I100) [Caenorhab 0.003
51011594 47 SIG 3738 0.976 11 Hypothetical protein Tb927.2.4250 [ 1.5
51011478 99 SIG 1920 8.723 9.06 Hypothetical protein MGC63561 [Dani 4E025 Unknown conserved protein
51011426 54 SIG 2021 4.02 7.01 Hypothetical protein CBG15127 [Caeno 0.099
51011406 225 SIG 2324 22.54 4.92 Cytotoxin [Bacteriophage phi CTX] 4 6E004
51011600 205 SIG 1718 21.659 8.65 Similar to tenascin-N [Rattus norve] 5E020
51011418 196 SIG 2627 18.516 8.71 CG17035-PA [Drosophila melanogaster] 2E025 Group XIV secreted phospholipase A2
Group 16Probably housekeeping proteins
Other possible housekeeping proteins
51011496 74 CYT CG32446-PA [Drosophila melanogaster] 3E014 Copper transport protein
51011504 174 CYT CG3595-PA [Drosophila melanogaster] 5E080 Myosin regulatory light chain
51011578 147 CYT CG7013-PA [Drosophila melanogaster] 7E047 ARMET-like protein precursor, truncated
51011442 93 CYT CG7630-PA [Drosophila melanogaster] 0.030 Unknown
51011554 101 CYT Heat shock protein 10 [Gallus gallu 8E029 Heat shock protein 10
51011552 80 CYT Hypothetical protein Magn027998 [ 0.34 Unknown
51011386 265 CYT Isopentenyl-diphosphate delta-is 3E063 Isopentenyl-diphosphate delta-isomerase
51011506 232 CYT Similar to Shwachman-Bodian-Diamond 9E084 ShwachmanBodianDiamond syndrome
homolog
51011446 66 CYT Unnamed protein product [Tetraodon n 1.8
51011524 66 CYT Unnamed protein product [Tetraodon n 1.4
Kunitz-containing intracellular proteins
51011388 179 CYT Putative secreted protein [Ixodes sca 4E072 Truncated peptide with Kunitz protease
inhibitor domain
51011382 205 CYT Putative secreted protein [Ixodes sca 9E028 Truncated peptide with Kunitz protease
inhibitor domain
Ribosomal proteins
51011528 165 CYT CG3195-PA [Drosophila melanogaster] 1E068 Ribosomal protein
51011568 123 CYT Ribosomal protein L35 [Mus musculus 7E046 Ribosomal protein
51011502 100 CYT 60S Ribosomal protein L37 4gnl| 1E037 Ribosomal protein
51011508 105 CYT Ribosomal protein L44 [Chlamys farreri] 194
3e049
3E049 Ribosomal protein
51011522 268 CYT Ribosomal protein L6 [Gallus gallus 7E057 Ribosomal protein
51011548 90 CYT Ribosomal protein L7a [Argopecten irr 8E035 Ribosomal protein
51011530 112 CYT Ribosomal protein L30 [Argopecten irr 1E052 Ribosomal protein
51011534 269 CYT Unknown (protein for MGC:73183); wu 1E113 Ribosomal protein
51011526 151 CYT Ribosomal protein S14; wu:fa92e08 [ 8E072 Ribosomal protein
51011510 149 CYT Ribosomal protein S15 [Argopecten irr 2E068 Ribosomal protein
I
.
M
.
B
.
F
r
a
n
c
i
s
c
h
e
t
t
i
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
1
4
2
1
1
6
1
1
1
4
7
A
R
T
I
C
L
E
I
N
P
R
E
S
S
Table 1 (continued )
NCBI Id
number and link
position
database
51011516 25 CYT Ribosomal protein L41 [Mus musculus] 2E006 Ribosomal protein
51011518 81 CYT Ribosomal protein S4 [Argopecten irra 2E032 Ribosomal protein
51011520 114 BL Ribosomal protein, large P2 [Mus mu 3E037 Ribosomal protein
51011536 133 CYT FinkelBiskisReilly murine sarcoma 3E036 Ribosomal protein
51011494 209 CYT 40S ribosomal protein S5 [Dermacentor 1E110 Ribosomal protein
Oxidant metabolism
51011512 230 CYT Putative glutathione S-transferase [D 1E036 Glutathione S-transferase
51011500 220 CYT Glutathione S-transferase [Boophilus 1E080 Glutathione S-transferase
Vacuolar sorting protein?
51011408 222 CYT Neuroendocrine differentiation factor 1E073 Vacuolar sorting protein VPS24
Energy metabolism
51011608 109 CYT Cytochrome c 4gnl|BL_ORD_ID|145934 8E050 Cytochrome c
51011610 153 CYT Cytochrome c oxidase subunit Va [Rhyz 5E050 Cytochrome c oxidase subunit Va
51011612 73 CYT Cytochrome oxidase subunit VIIc [Mac 5E013 Cytochrome oxidase subunit VIIc
51011614 152 CYT ATP synthase c-subunit [Dermacentor v 1E065 ATP synthase c-subunit
51011616 134 CYT CG2140-PB [Drosophila melanogaster] 6E036 Cytochrome b5
51011618 69 CYT ENSANGP00000013087 [Anopheles gambi] 0.002 Cytochrome c oxidase polypeptide VIII
51011620 182 CYT CG9350-PA [Drosophila melanogaster] 2E010 Probable NADH-ubiquinone oxidoreductase
subunit
I
.
M
.
B
.
F
r
a
n
c
i
s
c
h
e
t
t
i
e
t
a
l
.
/
I
n
s
e
c
t
B
i
o
c
h
e
m
i
s
t
r
y
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
3
5
(
2
0
0
5
)
1
1
4
2
1
1
6
1
1
1
4
8
On the other hand, some proteins from this family, such
as gi 22652868 and gi 22164158, display a negatively
charged tail composed of six glutamic acid (Glu, E)
anionic residues (Fig. 2A). The evolutionary relation-
ships of BTP were inferred by constructing the
phylogenetic tree using the NJ algorithm; a cladogram
is shown in Fig. 2B. Of interest, these proteins share
sequence similarities to exogenous anticoagulants such
as SALP 14 (gi 15428308) from I. scapularis. SALP 14 is
a FXa inhibitor that appears to interact with the
catalytic domain of FXa and with the so-called exosite
(Narasimhan et al., 2002). Exositesregions far from
the catalytic site and known to determine specicity and
afnity of blood coagulation factors toward sub-
stratesalso are critical for the assembly of the
prothrombinase, a multimolecular complex that leads
to thrombin generation (Krishnaswamy, 2005). Target-
ing these domains appears to be an effective strategy
evolved by blood-feeding arthropods to effectively
impair blood coagulation. In fact, we recently reported
that ixolaris, a FX(a) scaffold-dependent inhibitor
of Factor VIIa/tissue factor complex, specically
recognizes the FXa heparin-binding exosite (Monteiro
et al., 2005).
ARTICLE IN PRESS
(A)
gi|22652868| MGLTEIMLVL-VSLAFVATAAAHDCQNGTRPASEEKREGCDYYCWNTETKSWDKFFFGNGERCFYNNGDEGLCQNGECHLTTDSGVPNDTDAKIEETEEELEA-----------------------------------
gi|22164158| MGLTEIMLVL-VSLAFVATAAAHDCQNGTRPASEEKREGYDYYCWNTETKSWDKFFFGNGERCFYNNGDEGLCQNGECHLTTDSGVPNDTDAKIEETEEELEA-----------------------------------
gi|22164164| MGLTGTTLVL-VSLVFFGSAAAHNCKNGTRPASEENREGCDYYCWNDGTNSWDQFFFGNGEICFYNSGEKGICQNGECHLTNNSGGPNETDENTPATTEKPKQKKKKTKKPKKPKRKSKKDQ----------------
gi|22164176| MGLTGTTLVL-VSLAFFGSAAAHNCKNGTRPASEENREGCDYYCWNDGTNSWDQFFFGNGEICFYNSGEKGICQNGECHLTNNSGGPNETDDNTPAPTEKPKQKKKKPKKPKKPKRKSKKDH----------------
gi|15428308| MGLTGTMLVL-VSLAFFGSAAAHNCQNGTRPASEQDREGCDYYCWNAETKSWDQFFFGNGEKCFYNSGDHGTCQNGECHLTNNSGGPNETDDYTPAPTEKPKQKKKKTKKTKKPKRKSKKDQEKNL------------
gi|22164190| MGLTGTTLML-VCVAFFGTAAAHNCKNGTRPASEENREGCDYYCWNEVTNSWDQFFFGNGERCFYNTGENGKCQNGECHLTTNSDGPNETDDNTPPPTEKPK------------------------------------
gi|22164156| MGLTGTTLVL-VCVAFFGSAAAHNCQNGTRPASEENREGCDYYCWNEVTNSWDQFFFGNGERCFYNTGENGKCQNGECHLTTNSDGPNETDDNTPPPTEKPKQKKKKPKKPKKPKRKSKKDQ----------------
gi|22164184| MGLTGTTLVL-VCVAFFGTAAAHNCKNGTRPASEENREGCDYYCWNEVTNSWDQFFFGNGERCFYNTGENGKCQNGECHLTTNSGGPDDTDDNTPPPTEKPKQKKKKPKKPKKPKRKSKKDQ----------------
gi|22164186| MGLTGTTLVL-VCVAFFGSAAAHNCQNGTRPASEKNREGCDYYCWNAETKSWDQFFFGDGERCFYNTGENGTCRNGECHLTTSSGGPNETDDNTPPPTEKPKQKKKKPKKTKKPKRKSRKDQ----------------
gi|22164174| MGLTGTTLVL-VSLAFFGSAAAHNCQNGTRPTSEKNREGCDFYCWNADTNLWDKFFFGNGEKCFYNTGEKGTCLNGECHLTTSSGGPDDTGDNTPPPTEKPKQKKKKPKKTKKPKRKSKKDQKENF------------
gi|22164194| MGLTGTALVL-VSLAFFGSAAAHNCQNGTRPASEENREGCDYYCWNSETQSWDQYFFGDGERCFYNSGDRGICQNGECHLTTSSGGPDDTDENTPPPTEKPKQKKKKPKKTKEPKRKSKKD-----------------
IP_5_100_90_1_CLU MGLTGATLVL-VSLAFFGSAAAHNCKNGTRPASEENREGCDFYCWSTDTNSWEIFFFGNGEECFYNNGDRGTCQDGACHLTTHSGGPNETDDYTPAPTEKPKQKKKKPKKTKKPKRNSKKD-----------------
IP_clu3 MGLTGATLVQGVSLAFFGSAAAHNCKNGTRPASEENREGCDFYCWSTDTNSWEIFFFGNGEECFYNNGDRGTCQDGACHLTTHSGGPNETDDYTPAPTEKPEQKKKKPKKTKKPKRNTKKKKKKKKKKKNFLGPPGPH
IP_5_100_90_1_CLU1 MGLTGATLVL-VSLAFFGSAAAHNCQNGTRPASEENREGCDFYCWNTDTNSWDIFFFGNGEKCFYNNGDRGTCQDGACHLTTHSGGPNETDDYTPAPTEKPKQKKKKPKKTKKPKRKSKKD-----------------
IP_5_100_90_2A_CLU MGLTGITLVL-VSLAFFGSAAAHNCQNGTRPASEENREGCDFYCWNAGTNSWDIFFFGNGEKCFYNNGDRGTCQDGACHLTTHSGGPNETDDYTPAPTEKPKQKKKKPKKTKKPKRKSKKDKEG-----NF-------
IP_5_100_90_2_CLU MGLTGATLVL-VSLAFFGSAAAHNCKNGTRPASEETREGCDFYCWNTDTSSWDIFFFGNGEKCFYNNGDRGTCRDGACHLTTLSGGPNETDDYTSAPTEKPKQKKKKLKKTKKPKRKSKKD-----------------
gi|45593710| MGLTGTTLVL-ASLAFFGSAAAHNCKNGTRPASEEKREGCDFYCWNSDTSRCDQFFFRDGETCFYNNGDRGSCQNGECHLNTNSGVPTHNDDYTPSPTEKPKQKKKKPKKTKKPKRQSNKD-----------------
IP-7-60-92-2-CLU MGLTGITLVL-VSFAFFGSVAAHNCQNGTRPTSEQNREGCDYYCWNTDTKSWDQFFFGNGERCFYSNGDTGVCTNGECHLNTESSVPTETDVETPAPTKKPKQKKKKQKKTKKPKR----------------------
IP_5_100_90_25_CLU MGLTGITLVL-VSFAFFGSVAAHNCQNGTRPASEQNREGCDYYCWNTDTKSWDQFFFGNGERCFYSNGDTGVCTNGECHLNTESGVPTETDVETPAPTKKPKHKKKKQKKSKKPKG----------------------
IP_5_100_90_6_CLU MGLTGTTLVL-VSFAFFGSVAAHNCQNGTRPASEENREGCDYYCWNTDTRSWEQFFFGNGERCFYNTGEKGECKNGECHLTTESGVPTDTDVDTPAPTKKPKQKKKKQKKTKKPKR----------------------
gi|22164182| MEFTGITLVL-VSLTFFGSAAAETCRNGTRPGSQTQREGCDYYCWNSQTSSWDKYFFGDNEPCFYNTGLRGTCQNGECHLTSEGGVPTDPNQYPSEPTEKPKKNKKKSKKTKKPKKTKKPKDN---------------
gi|22164160| MEFTGITLLL-VSLAFFGSAAAETCRNGTRPASQTQREGCDYYCWNLQTSSWDKYFFGDNEPCFYNTGLRGTCQNGECHLTSEGGVPTDPNQYPSEPTEKPKKNKKKSKKTKKPKKTKKPKDN---------------
gi|22164178| MEFTGITLVL-VSVAFFGSAAAETCRNGTRPASQTDREGCDYYCWNTLTSSWDKYFFGDEEPCFYNTGLRGTCKNGGCHLTSEGNVPTDPDQYPSEPTEKPKKSKKKSKKTKKPKKTKKPKDN---------------
gi|22164180| MEFTGITLVL-VSVAFFGSAAAETCRNGTRPASQTDREGCDYYCWNTLTSSWDKYFFGDEEPCFYNTGLRGTCKNGECHLTSEGGVPTDPNQYPSEPTEKPKKNKKKSKKTKKPKKSKKPKDN---------------
gi|22164168| MELTGITLVL-VSLALFGSAAAETCRNGTRPASQTDREGCDYYCWNTLTSSWDKYFFGDEEPCFYNTGLKGTCKNGECHLTSEGGVPTDPHQYPSEPTEKPKKNKKKSKKTKKPKKSKKPKNN---------------
____________________
Poly K (lysine) tail
(B)
0.1
gi|22652868|
gi|22164158|
gi|22164182|
gi|22164160|
gi|22164168|
gi|22164178|
gi|22164180|
IP 5 100 90 6 CLU
IP-7-60-92-2-CLU
IP 5 100 90 25 CLU
gi|45593710|
IP 5 100 90 2A CLU
P 5 100 90 2 CLU
IP 5 100 90 1 CLU1
IP 5 100 90 1 CLU
IP clu3
gi|15428308|
gi|22164164|
gi|22164176|
gi|22164174|
gi|22164186|
gi|22164156|
gi|22164190|
gi|22164184|
gi|22164194|
Fig. 2. Group 1: Basic tail proteins (BTP). (A) Alignments of peptides from I. pacicus (Table 1) and I. scapularis BTP deduced from cDNA libraries.
Conserved amino acid residues are shown in black background. Lysine residues (K) are shown in bold (Poly K, lysine tail). (B) The bar represents the
degree of divergence among sequences.
The fact that BTP and SALP 14 contain a poly-Lys
tail adds an additional layer of anticoagulation, as it
directs the inhibitor to negatively charged membranes
(e.g., activated platelets) critical for productive blood
coagulation complex assembly (Broze, 1995). As a
result, the effective concentration of the inhibitor is
increased at sites that are predominantly pro-coagulant.
Also, we speculate that FXawhich is usually protected
from physiologic inhibitors (e.g., TFPI, heparin/ATIII)
when the prothrombinase is fully assembled (Mast and
Broze, 1996; Rezaie, 2001)would be more susceptible
to these bifunctional molecules. Demonstration that
proteins rich in positively charged residues effectively
block the coagulation cascade comes from studies
performed with a recombinant Rhodnius prolixus
salivary lipocalin (nitrophorin-7, NP-7). NP-7 contains
a cluster of positively charged residues in the N-terminus
and specically binds to anionic phospholipids, prevent-
ing thrombin formation by the prothrombinase (Ander-
sen et al., 2004). Finally, a bifunctional fusion protein
containing Kunitz and annexin domains was shown
recently to inhibit the initiation of blood coagulation
(Chen et al., 2005).
3.2. Group 2: Similar to Group 1, but without the basic
tail
These sequences contain a cysteine pattern identical to
Group 1 peptides except that, remarkably, the poly K
tail is missing. Many other amino acids also are not
conserved. Sequence alignment between the Group 1
peptides (containing poly K and poly E) and the
peptides similar to Group 1 is shown in Fig. 3A. Fig.
3B shows that these proteins come from a common
ancestor that appears to have evolved to display
different functions. The function of the peptides of
Group 2 deserves further investigation.
3.3. Group 3: Kunitz-containing proteins
Kunitz domains are about 60 residues and contain six
specically spaced cysteines (XnCX8CX15CX7CX12
CX3CXn) that form disulphide bonds typically repre-
sented by bovine pancreatic trypsin inhibitor (BPTI). In
most cases, they are reversible inhibitors of serine
proteases that bind the active site (Laskowski and Kato,
1980); however, Kunitz inhibitors such as the dendro-
toxins from Dendroaspis angusticeps snake venom block
K+ channel but display negligible protease inhibitory
properties (Harvey, 2001). Kunitz-containing proteins
also interact with protease exosites (Monteiro et al.,
2005) or platelets (Mans et al., 2002a). Of note,
sequencing the I. pacicus cDNA library yields a
number of proteins containing Kunitz-like domains.
The alignment of BPTI, snake venom, and I.
scapularis and I. pacicus single Kunitz-like proteins is
shown in Fig. 4A. Some I. pacicus proteins contain
one-Kunitz-like domain, here named the Monolaris-1
family (or similar to 6.58.4-kDa proteins from I.
scapularis). These molecules display the following
cysteine pattern: XnCX8CX18CX5CX12CX3CXn.
Other single-Kunitz sequences present in I. scapularis
belong to the Monolaris-2 family (or similar to 7.98.7-
kDa proteins from I. scapularis) and display the
sequence pattern XnCX8CX15CX8CX11CX3CXn. We
could not, however, nd members of the Monolaris-2
family sequences in our I. pacicus cDNA library. Fig.
4A also shows that the well-known tick anticoagulant
peptide from the soft tick Ornithodoros moubata (Wax-
man et al., 1990) has Kunitz-like folding with the
sequence pattern XnCX9CX17CX5CX15CX3CXn. At
present, the functions of Monolaris-1 and -2 are
unknown, but they may target specic proteases. The
phylogenetic tree shown in Fig. 4B suggests that snake
venom peptides containing Kunitz domains (non-
neurotoxic or neurotoxic) and the tick families of
Monolaris and basic tail peptides have diverged into
two different main groups from a commom ancestor,
suggesting that these proteins have evolved to perform
different functions.
Additionally, cDNAs were sequenced coding for
proteins containing two- or ve-Kunitz domains. These
ARTICLE IN PRESS
Fig. 3. Group 2: Similar to Group 1, without basic tail. (A) Alignment
of Group 2 peptides (Table 1). Conserved amino acid residues are
shown in gray background. (B) The unrooted cladogram of all
sequences. The bar represents the degree of divergence among
sequences.
proteins share sequence similarity to ixolaris (Fran-
cischetti et al., 2002b) and penthalaris (Francischetti
et al., 2004a), two I. scapularis TFPI salivary proteins that
prevent initiation of blood coagulation through specic
inhibition of the Factor VIIa/tissue factor complex. It is
possible that these proteins block other proteases (Ruf,
2004) or affect angiogenesis (Hembrough et al., 2004).
Fig. 4C depicts the predicted secondary folding of I.
scapularis and I. pacicus Kunitz-like-containing pro-
teins based on the crystal structure determined for BPTI
(Huber et al., 1974).
3.4. Group 4: Proline-rich proteins
Group 3 cDNA sequences code for short peptides of
mature molecular mass ranging from 3.5 to 4.8 kDa of
both basic and acidic nature (Table 1). Alignments and
cladograms (presented in Fig. 5A and B, respectively),
show that all sequences are relatively glycine and proline
rich in both I. pacicus and I. scapularis salivary glands.
Some sequences display weak matches to proteins
annotated as collagen in the NR database; these possess
two conserved cysteine residues in the mature peptide
and remarkable conservation of the secretory signal
peptide (Fig. 5). Most amino acids of the predicted
signal secretory peptide are conserved, versus few on the
mature peptide, suggesting functional diversity. The
possible function of these peptides remains to be
characterized, but taking into account its similarity to
collagen, it may somehow affect vascular biology
through inhibition of cellcell, cellmatrix, or cellli-
gand interactions. These peptides may also function
as adhesive molecules to cement the tick into their
hosts skin.
ARTICLE IN PRESS
Fig. 4. Group 3: Kunitz-containing proteins. (A) Alignment of Group 3 peptides (Table 1) with single Kunitz-containing protein from snake venoms.
Conserved amino acid residues are shown in gray background. (B) The phylogram was constructed using protein from snake venom single-kunitz
(neurotoxic or non-neurotoxic from Elapidae and Viperidae families) and tick salivary gland, plus BPTI (all accession numbers are depicted). The bar
represents the degree of divergence among sequences. (C) Predicted secondary folding of Kunitz-containing proteins from Ixodidae sp. based on
BPTI folding (Huber et al., 1974).
3.5. Group 5: Similar to I. scapularis 18.7-kDa protein
This group of proteins (Table 1) is similar to
orthologs described in I. scapularis and code for an
acidic putative protein of unknown function. Only low
e-values have been found when compared with proteins
in the NR database including coagulation factor X (gi
9837158, e-value 0.069), venom metalloprotease acur-
hagin precursor (gi 4689408; e-value 3.8), and pro-
protein convertase subtilin (gi 51771463, e-value 8.4).
Accordingly, this family of proteins may have evolved
from a protease precursor; however, any functional
assignment will be possible only after testing the
recombinant protein in screening assays.
3.6. Groups 6 and 7: Similar to I. scapularis 5- and 9.4-
kDa protein
Groups 6 and 7 code for basic proteins of 5 and
9.4 kDa that also are present in I. scapularis. No protein
motif was identied for either protein; accordingly, the
function of these proteins is not evident.
3.7. Group 8: Metalloprotease
These enzymes are capable of hydrolyzing various
components of the extracellular matrix including bri-
nogen and bronectin and reportedly affect endothelial
cells, leading to apoptosis. These enzymes are organized
ARTICLE IN PRESS
(A)
gi|22164234| MKATLVAICFLATVAFSMGESWGSSTPCPGAPGEPCQNGQPSQPPA---------PGSGPNVHPPENTSPPGRK--------------------------
gi|22164232| MKATLVAICFLAAVAFSMGESWGSSTPCPGAPGEPCQNGNPSKPPA---------PGSGPNVHPPRNTSPPGRK--------------------------
gi|22164236| MKATFMAICFLAAVAFSMGESWGSSTPCPGAPGEPCQNGNPSQPPA---------PGSGPNVRPPQNTSPPGRK--------------------------
gi|22164240| MKATFMVICFLAAVAFSMGESWGSSTPCPGAPGEPCQNGQPSQPPA---------PGSGPNVHPPQSTSPPGRK--------------------------
gi|22164230| MKATLIAICFLAAVAFSMGESWGSSTPCPGAPGEPCGNGNSPGGPS---------NPPGPSQPGMRDPSPPGRK--------------------------
gi|22164226| MKATLIAICFLAAVAFSMGESWGSSTPCPGAPGEPCGNGNSPGGPS---------NSPGPSQPGTRDPSPPGRK--------------------------
gi|22164224| MKATLLAICFLAAVTLTMGESHGSTTPCPGAPGQDCHPGNGPQGPS---------GPSGPQQPGTRGPNPPGTK--------------------------
gi|22164228| MKATLLAICLLAAVTLTMGESHGSTTPCPGAPGEACNPGNGPQGPS---------NSPGLRQPGTSDPSPPGTK--------------------------
gi|22164242| MKATLLAICFLAAVTLTMGESHGSTTPCPGAPGQPCNPLQGSQGPS---------NSPGPNQPGTRGSYAPGNSRK------------------------
gi|22164244| MKATLLAICFLAAVTLTMGESHGSTTPCPGAPGQPCNPLQGSQGPS---------NSPRPNQPGTRGSYAPGNSRK------------------------
gi|22164222| MKATLIAICFLAAVTFSMGETHGSSTPCPTAPGEDCNPGSRLQGPS---------NPSGPNQP-------------------------------------
gi|22164238| MKATLIAICFLAAVTFSMGETHGSSTPCPGAPGEDCNPGSRLQGPS---------NPSGPNHPGTRDTSPPGRK--------------------------
IP_7_60_92_16_CLUB MKATLIAICFLAAVTFSMGESIGSSTPCPNPPGQPCGPGQGPQGP-----------PSGPSHQPGKSPNAPGSSSKVSSRWLPPETCDGG----------
IP_5_100_90_27_CLU MKATLIAICFLAAVTFSMGESIGSSTPCPNPPG-----------P-----------PSGPIHQPGRSPHAPGSSSK------------------------
IP_5_100_90_30_CLU MKATLIAICFLAAVTFSMGESIGSSTPCPNPPGQPCGPGQGPQGP-----------PSGPSHQPGKSPSAPGSSTK------------------------
IP_clu28 MKATLIAICFLAAVTFSMGDTYGSSTPCPNAPGTPCGPGQGPQGP-----------PSAPSHQPDRSPNAPGGSSK------------------------
IP_clu28A MKATLIAICFLAAVTFSMGDTYGSSTPCPNAPGTPCGPGQGPQGP-----------PSAPSHQPDRSPNAPGGSSKAVGGCLQRRAMADRILPRCSKFSE
IP_5_100_90_29_CLU2 MKATLIAICFLAAVAFSMGESWGSSTPCPNPPGQPCGPGQPPQGPS---------QNPGPSPQPPRDTSAPGRK--------------------------
IP_5_100_90_32_CLU MKATLIAICFLAGVAFSMGESWGKLPPCPNPPGQPCGPGQPPQGPS---------QNPGPSPQPPRDTSAPGRK--------------------------
IP_5_100_90_29_CLU4 MKATLIAICFLAGVTFSMGETWGKPPPCSSPPGQPCPPGEGPQGPSSSPRPYPQPEGPSPSPQPPRDQSPPGTR--------------------------
IP_5_100_90_29_CLU5 MKATLIAICFLAGVTFSMGETWGKPPPCSSPPGKPCPPGEGPQGPSSSPRPYPQPEGPSPSPQPPRDQSPPGTR--------------------------
IP-7-60-92-15-CLU MKATLIAICFLAGVAFSMGETYGKPPPCSSPPGKPCPPGEGPQGPSSSPRPYPQPQGPAPSPQPPKDQSPPGTR--------------------------
IP_5_100_90_29_CLU3 MKATLIAICFLAGVAFSMGETYGQPPPCTSPPGEPCPPGEGPQGPPSSPRPYPRPHGPAPSPQPPKDQSPPGTR--------------------------
IP_5_100_90_29_CLU6 MKSTLIAICFLAAVTFTMGETWGNPPPCSSPPGQPCPPGEGPQGPSSSPRPYPQPQGPAPNPQPPRDQSPPGTRT-------------------------
**:*::.**:** *:::**:: *. .**. .** * .
(B)
0.1
gi|22164230|
IP clu28
IP 5 100 90 30 CLU
IP 5 100 90 27 CLU
IP 5 100 90 32 CLU
IP 5 100 90 29 CLU6
IP 5 100 90 29 CLU4
IP-7-60-92-15-CLU
IP 5 100 90 29 CLU3
gi|22164224|
gi|22164242|
gi|22164228|
gi|22164244|
gi|22164222|
gi|22164238|
gi|22164234|
gi|22164236|
gi|22164240|
Fig. 5. Group 4: Proline-rich peptides. (A) Alignment of Group 4 peptides (Table 1). Signal peptide is shown in gray background, and conserved
amino acid residues are shown in black background. (B) The unrooted cladogram of all sequences. The bar represents the degree of divergence
among sequences.
into four classes, PI through PIV, according to size and
domain composition (Bjarnason and Fox, 1995).
Our library contains a truncated cDNA that codes for
a mature metalloprotease similar to one described in I.
scapularis (gi 31322779) (Francischetti et al., 2003) and
I. ricinus (gi 5911708). The alignment of the mature
metalloproteases from I. pacicus, I. scapularis, and I.
ricinus, where the zinc-binding motif HExxHxxGxxH
common to these enzymes was identied, is shown in
Fig. 6A. Fig. 6B compares the PIII class of metallopro-
teases from snake venom and the I. pacicus, I.
scapularis and I. ricinus enzymes. It is clear that enzymes
from both genera have pre-, pro-, metalloprotease,
disintegrin-like, and cysteine-rich-like domains; how-
ever, the Ixodidae disintegrin-like and cysteine-rich like
domains are signicantly shorter in the number of
amino acid residues when compared with the corre-
sponding domains of metalloproteases from the repro-
lysin family (Bjarnason and Fox, 1995). We suggest that
this pattern of cysteines confer a different specicity for
these enzymes. This family of proteins also appears to
account for the a-brinogenase and brinolytic activity
recently reported for I. scapularis saliva (Francischetti
et al., 2003). Degradation of brinogen and brin are
associated with inhibition of platelet aggregation and
clot formation. Metalloproteases also may interact with
endothelial cell integrins, leading to apoptosis and
inhibition of angiogenesis (Francischetti et al, 2005).
3.8. Group 9: GPIIB/IIIa antagonists from the short
neurotoxin family
Inhibitors of platelet aggregation that targets the
brinogen receptor (GPIIbIIIa, integrin aIIbb3) have
been described in the hard tick Dermacentor variabilis
(variabilin) and the soft ticks, Ornithodoros moubata
(disagregin) and O. savignyi (savignygrin) (Karczewski
et al., 1994; Wang et al., 1996; Mans et al., 2002a).
Savignygrin belongs to the Kunitz-BPTI family and
presents the integrin RGD-recognition motif on the
substrate binding loop of the Kunitz fold (Mans et al.,
2002a). In contrast, variabilin possesses an RGD-motif
in its C-terminal region that is not anked by cysteines
(Wang et al., 1996). A search for possible GPIIb/IIIa
antagonists with RGD-motifs and anking cysteines,
termed the Ixodegrins, identied one candidate in I.
pacicus and several homologs in I. scapularis (Table 1).
It is clear that the Ixodegrins are related to variabilin,
but do possess anking cysteines. Variabilin probably
possesses a anking disulphide motif too, but was
missed due to the technical difculties in identifying
cysteines correctly during N-terminal sequencing. Data-
base searches using SAM-T99 (Karplus et al., 1998), a
program that utilizes hidden Markov models to nd
remote homologous sequences, identied dendroaspin
as the highest hit. Dendroaspin, also known as mambin,
is part of the short neurotoxin family found in elapid
snakes (McDowell et al., 1992; Williams et al., 1993;
Sutcliffe et al., 1994). Strikingly, the RGD-active site
loop (loop3) is conserved between snake and tick
integrin antagonists (Fig. 7A). This includes the anking
cysteines involved in a disulphide bond that constricts
the RGD-loop conformation and the anking prolines
that was shown to be important for presentation of the
RGD sequence (Lu et al., 2001). The tick inhibitors
maintain loops 2 and 3 of the short neurotoxin fold, but
do not possess the N-terminal loop 1 and the C-terminal
extension (Fig. 7A). This makes them the shortest
members of the short neurotoxin family described to
date, with only 39 amino acids forming the core active
fold. Phylogenetic analysis of the neurotoxin family
indicates that dendroaspin and tick inhibitors group
within one clade to the exclusion of the other short
neurotoxins (Fig. 7B). This suggests either an extreme
form of convergent evolution, where ticks and elapid
snakes used the same protein fold to evolve the same
function or raises the possibility that ticks or snakes
acquired the ancestral protein via a horizontal gene
transfer event or that there is a true evolutionary
relationship between the ixodegrins and short neurotox-
ins. The fact that orthologs of the Ixodegrins are present
in both Ixodes (prostriate) and Dermacentor (metastri-
ate) ticks, suggests that this inhibitor was present in the
last common ancestor of hard ticks. Snakes evolved
most of their venom properties approximately 6080
million years ago (Fry, 2005), whereas most hard tick
genera diverged at least 110 million years ago or earlier
(Klompen et al., 1996). If tick and snake proteins are
related, then the ancestral gene may have a platelet
antagonist function and the neurotoxic properties (and
the rest of the short neurotoxin foldloop1 and the C-
terminal extension) evolved later. In contrast, soft ticks
in the genus Ornithodoros evolved integrin antagonists
from the BPTI-fold which suggests that hard and soft
ticks evolved different strategies to obtain a blood meal
(Mans et al., 2002b; Mans and Neitz, 2004). Accord-
ingly, ixodegrin may affect platelet or neutrophil
integrin function or neutrophil function.
3.9. Group 10: Ixostatin family, or short-coding cysteine-
rich peptides (thrombospondin)
The two sequences in Group 11 match a sequence
deposited in the NR database from I. scapularis;
alignments are shown in Fig. 8A. These sequences have
been annotated thrombospondin (gi 15428290), but
thrombospondin motifs are lacking. On the contrary,
these short coding region cysteine-rich peptideshere
named ixostatinsare remarkably similar to the cy-
steine-rich domain of ADAMTS (Fig. 8B). Of note,
ADAMTS-4 (a disintegrin and metalloproteinase with
thrombospondin motifs), also known as aggrecanase,
ARTICLE IN PRESS
ARTICLE IN PRESS
Fig. 6. Group 8: Metalloproteases. (A) Alignment of metalloproteases from I. pacicus (Ip) (Table 1), I. scapularis (Is), and I. ricinus (Ir). The
characters in bold represent the conserved Zn binding motif present in the catalytic domain. Asterisks, colons, and stops below the sequences indicate
identity, high conservation, and conservation of the amino acids, respectively. (B) Diagram comparing the protein motifs (pre, pro, catalytic,
disintegrin-like, and cysteine rich-like domains) of class III metalloproteases from snake venoms and tick class III-like metalloproteases.
ARTICLE IN PRESS
Fig. 7. Group 9: Ixodegrin: disintegrins. (A) Alignment of the ixodegrins from I. pacicus (Ixodegrin_Ip), I. scapularis (Ixodegrin_Sc1/2/3), variabilin
and dendroaspin. Shadowed in gray are conserved cysteine regions and the RGD motif. Also indicated are the loops and disulphide bond pattern of
the short neurotoxin fold and the inferred disulphide bond patterns of the ixodegrins. (B) A neighbor-joining tree of the short neurotoxin family.
Indicated are different functional clades found for the family. Snake proteins are referred to by their SwissProt name. Black circles indicate
condence levels 470% from 10 000 bootstraps.
are enzymes involved in cartilage cleavage (Flannery
et al., 2002). The role of the cysteine-rich domain of
ADAMTS proteases is unknown, but it is postulated to
interact with integrins and/or other attachment motifs
of cells and matrix proteins (Porter et al., 2005).
Accordingly, the ixostatin family of peptides could be
involved in disruption of platelet aggregation or
neutrophil function, cellmatrix interactions, or inhibi-
tion of angiogenesis (Porter et al., 2005). The protein
modules of ixostatin and of ADAMST-4 are compared
in Fig. 8C.
3.10. Group 11: Histamine-binding proteins (lipocalins)
Group 11 contains sequences with similarities to
histamine-binding proteins discovered in the saliva of
R. appendiculatus ticks (Paesen et al., 1999). The
alignments of these sequences (Fig. 9A) reveal that they
do not display a highly conserved signal peptide which
suggest that they may not share a common ancestor. In
addition, the mature proteins contain few consensus
sequences indicating that they may have diverged to
perform distinct functions (Fig. 9B). This contention is
also supported by the cladogram presented in Fig. 9B. It
is likely that these proteins function by binding small
ligands such as histamine, serotonine, and adrenaline
(Andersen et al., 2005). Fig. 9C shows a predicted 3-D
model for sequence gi 51011604 that has an e-value of
768 for HBP from R. appendiculatus. The gure shows
amino acid side chains of the histamine-binding protein
from R. appendiculatus (red) surrounding the bound
histamine ligand with the corresponding residues for gi
51011604 shown in cyan. In the histamine-binding
protein, the imidazole ring of the ligand is stabilized
by surrounding aromatic residues, while in the
I. scapularis protein the binding pocket remains hydro-
phobic and fewer aromatic residues are present,
suggesting different ligand specicity. Polar residues
(Tyr 36 and Glu 135) forming electrostatic interactions
with the aliphatic amino group of histamine in the
ARTICLE IN PRESS
Fig. 8. Group 10: Ixostatin: short coding cysteine-rich peptides. (A) Alignment of Group 9 peptides from I. pacicus (Table 1) and I. scapularis.
Conserved amino acid residues are shown in black background. (B) Alignment between ixostatin and the cysteine-rich domain of ADAMST-4
(aggrecanase). (C) Diagram comparing the protein motifs (pre, pro, catalytic, disintegrin-like, cysteine-rich-like, and spacer domains) of ADAMST-4
(Flannery et al., 2002) and ixostatin.
histamine-binding protein are conserved in gi 51011604
suggesting the possibility of a similar role in this protein.
3.11. Group 12: Neuropeptide-like protein (npl-31) with
GGY repeat
A cDNA coding for a protein that shows remarkable
sequence homology to a neuropeptide-like protein (npl-
21) described in Caenorhabditis elegans (Nathoo et al.,
2001). This family of peptides displays a potent
antimicrobial activity toward Drechmeria coniospora,
Neurospora crassa, and Aspergillus fumigatus (Couillault
et al., 2004). Identication of these peptides in ticks
reinforces the notion that saliva contains a cocktail of
antimicrobial peptides. These peptides may prevent
growth of yeast and bacteria that, per se, can elicit an
inammatory/immune response that may be detrimental
to the feeding behavior of the attached ticks. Expression
of these molecules is particularly important vis-a` -vis the
remarkably immunosupressive property of the saliva
(Wikel, 1999) that helps ticks to feed for days but
otherwise creates an appropriate environment for
pathogen overgrowth. The sequence alignments for C.
elegans npl-21 and I. pacicus npl-21-like proteins are
presented in Fig. 10A and the cladogram in Fig. 10B.
This is the rst time that this family of antimicrobial
peptides has been identied in the salivary gland of a
blood-sucking arthropod.
3.12. Group 13: Oxidant metabolism
Proteins with similarity to glutathione peroxidase and
a putative secreted superoxide dismutase were found
(Table 1). These sequences categorize the prominent
ARTICLE IN PRESS
Fig. 9. Group 11: Histamine-binding proteins (lipocalins). (A) Alignment of Group 10 peptides from I. pacicus (Table 1). Conserved amino acid
residues are shown in black background. (B) The unrooted cladogram of all sequences. The bar represents the degree of divergence among sequences.
(C) The gure shows amino acid side chains of the histamine-binding protein from R. appendicuatus (red) surrounding the bound histamine ligand.
The corresponding residues for gi 51011604 are shown in cyan. In the histamine-binding protein, the imidazole ring of the ligand is stabilized by
surrounding aromatic residues. In the I. scapularis protein the binding pocket remains hydrophobic, fewer aromatic residues are present, suggesting a
different ligand specicity.
salivary gland proteins in I. pacicus and demonstrate
the presence of a potent antioxidant in tick saliva. Of
interest, cluster F12_IPL_P23 has sequence similarity to
SALP 25, a protein that catalyzes the reduction of
hydrogen peroxide in the presence of reduced glu-
tathione and glutathione reductase (Das et al., 2001).
The functions of these proteins are likely related to
maintenance of the physiologic redox of cellular
intracellular milieu or to modulation of the extracellular
levels of pro-oxidants often associated with inamma-
tory events.
3.13. Group 14: Similar to other ixodid proteins
A number of sequences show sequence homology to
proteins from Ixodidae described before. We have found
sequences similar to SALP 15, a immunodominat
protein in I. scapularis (Das et al., 2001), and to ISAC,
the anti-complement from I. scapularis (Valenzuela et
al., 2002). We also have found sequences similar to
domain 8 of human ADAMS and Factor VII.
3.14. Group 15: Novel, unknown
Some sequences containing a signal peptide and a
stop codon and with a clear open reading frame were
without database hits and were characterized as
unknown-function proteins. Assignment of function
for these proteins will only be possible after expressing
and screening for testable biologic activities.
3.15. Group 16: Housekeeping cDNA
Thirty-seven sequences with homology to housekeep-
ing protein are given in Table 1. They assign to
ribosomal, glutathione S-transferase, vacuolar assorting
proteins, cytochrome, ATP synthase subunit, and
NADH-ubiquinone oxidoreductase, among other mole-
cules. In addition, housekeeping proteins may be useful
in phylogenetic studies (Black and Piesman, 1994).
3.16. I. pacicus salivary gland protein diversity:
modulators of vascular biology and candidates for an anti-
saliva experimental vaccine
We describe the set of cDNA present in the salivary
glands of I. pacicus salivary gland. Our library contains
a remarkably large degree of redundancy, as shown by
the many related mRNAs. It appears that the long
evolutionary history of ticks may be responsible for the
complexity of transcripts reported here. Also, many
protein families we have identied were found pre-
viously in I. scapularis salivary glands (Valenzuela et al.,
2002) which conrms the diverse nature of these
secretions compared with the salivary composition of
fast feeders such as sand ies (Charlab et al., 1999) and
mosquitoes (Francischetti et al., 2002a). This variability
ARTICLE IN PRESS
Fig. 10. Group 12: Neuropeptide-like (npl-31) peptides. (A) Alignment of Group 11 peptides from I. pacicus (Table 1) and I. scapularis. Conserved
amino acid residues are shown in gray background. (B) The unrooted cladogram of all sequences. The bar represents the degree of divergence among
sequences.
in the tick salivary gland is consistent with the high
polymorphism of salivary proteins among individual
ticks analyzed by SDS-PAGE (Wang et al., 1999). Also,
the diversity across and within species could reect the
range of host species and the need to have modulators of
specic pathways that differ in distinct host species. The
adaptive role of this diversity appears to be explained at
least in part by a gene-duplication phenomenon. This
contention is supported by the diversity of sequences
containing Kunitz-like domains in addition to a weak
similarity observed among members of the lipocalin
family of proteins reported here. It may be that these
inhibitors have evolved to inhibit different proteases or
to bind to different ligands (Andersen et al., 2005). It is
also plausible that gene duplication may help ixodid
ticks to evade the immune system. If so, this may help to
explain why hard ticks can remain attached to many
hosts for days without apparent detrimental effects
(Ribeiro and Francischetti, 2003). Finally, the possible
closer association of I. persulcatus with I. pacicus
makes the former an interesting species for future
salivary gland transcriptome analysis and phylogenetic
studies.
The functions of many tick sequences described in this
paper are unknown. Cloning and expressing select
cDNAs will help in the identication of molecule
specicity and to nd potential targets for gene silencing
(Sanchez-Vargas et al., 2004), and accordingly, our
understanding of how ticks successfully feed on blood.
It also may provide tools to understand vascular biology
ARTICLE IN PRESS
Fig. 11. Negative modulators of vascular biology are present in I. pacicus and I. scaularis saliva. Vascular injury is accompanied by vasoconstriction
and activation of the extrinsic and intrinsic pathways of blood coagulation (Broze, 1995). Vasoconstriction is mediated by molecules such as
serotonine that may be removed by salivary protein with a lipocalin folding (Andersen et al., 2005). The extrinsic pathway is initiated by tissue factor/
factor VIIa complex and effectively blocked by ixolaris (Francischetti et al., 2002b) and penthalaris (Francischetti et al., 2004a). FXa generated by
the intrinsic or extrinsic Xnase may be inhibited by Group 1 peptides containing a basic tail that may prevent productive prothrombinase complex
assemble (Rezaie, 2000; Narasimhan et al., 2002; Andersen et al., 2004, Monteiro et al., 2005). Platelet, neutrophil, and endothelial cell function may
be affected by Ixodegrins (disintegrins) or Ixostatins (short-coding cysteine-rich peptides). Metalloproteases appear to cleave brinogen and brin,
therefore inhibiting platelet aggregation and clot formation (Francischetti et al., 2003). Metalloproteases also may affect endothelial cell function and
angiogenesis (Francischetti et al., 2005). The intrinsic pathway that is activated by contact leads to bradykinin formation, a peptide that increases
vascular permeability and induces pain. Bradykinin is degraded by a salivary kinininase, thus preventing its pro-inammatory effects (Ribeiro and
Francischetti, 2003).
and the immune system. A diagram with the putative
targets of salivary proteins and how they may affect
vascular biology is shown in Fig. 11. Finally, dening
the most abundant antigens or those that may effec-
tively help ticks to feed or transmit Borrelia could
be an effective approach to develop a protective
vaccine directed toward tick salivary molecules (Lane
et al., 1999).
Acknowledgements
We thank Drs. Thomas E. Wellems, Robert W.
Gwadz, and Thomas J. Kindt for encouragement and
support. We are thankful to Brenda Rae Marshal for
editorial assistance.
References
Andersen, J.F., Gudderra, N.P., Francischetti, I.M., Valenzuela, J.G.,
Ribeiro, J.M., 2004. Recognition of anionic phospholipid mem-
branes by an antihemostatic protein from a blood-feeding insect.
Biochemistry 43, 69876994.
Andersen, J.F., Gudderra, N.P., Francischetti, I.M., Ribeiro, J.M.,
2005. The role of salivary lipocalins in blood feeding by Rhodnius
prolixus. Arch. Insect Biochem. Physiol. 58, 97105.
Barbour, A.G., 1998. Fall and rise of Lyme disease and other Ixodes
tick-borne infections in North America and Europe. Br. Med. Bull.
54, 647658.
Bjarnason, J.B., Fox, J.W., 1995. Snake venom metalloendopepti-
dases: reprolysins. Methods Enzymol. 248, 345368.
Black IV, W.C., Piesman, J., 1994. Phylogeny of hard- and soft-tick
taxa (Acari: Ixodida) based on mitochondrial 16S rDNA
sequences. Proc. Natl. Acad. Sci. USA 91, 1003410038.
Broze Jr., G.J., 1995. Tissue factor pathway inhibitor and the revised
theory of coagulation. Annu. Rev. Med. 46, 103112.
Burgdorfer, W., Lane, R.S., Barbour, A.G., Gresbrink, R.A.,
Anderson, J.R., 1985. The western black-legged tick, Ixodes
pacicus: a vector of Borrelia burgdorferi. Am. J. Trop. Med.
Hyg. 34, 925930.
Charlab, R., Valenzuela, J.G., Rowton, E.D., Ribeiro, J.M., 1999.
Toward an understanding of the biochemical and pharmacological
complexity of the saliva of a hematophagous sand y Lutzomyia
longipalpis. Proc. Natl. Acad. Sci. USA 96, 1515515160.
Chen, H.H., Vicente, C.P., He, L., Tollefsen, D.M., Wun, T.C., 2005.
Fusion proteins comprising annexin V and Kunitz protease
inhibitors are highly potent thrombogenic site-directed antic-
oagulants. Blood 105 (10), 39023909.
Couillault, C., Pujol, N., Reboul, J., Sabatier, L., Guichou, J.F.,
Kohara, Y., Ewbank, J.J., 2004. TLR-independent control of
innate immunity in Caenorhabditis elegans by the TIR domain
adaptor protein TIR-1, an ortholog of human SARM. Nat.
Immunol. 5, 488494.
Das, S., Banerjee, G., dePonte, K., Marcantonio, N., Kantor, F.S.,
Fikrig, E., 2001. SALP 25D, an Ixodes scapularis antioxidant, is 1
of 14 immunodominant antigens in engorged tick salivary glands.
J. Infect. Dis. 184, 10561064.
Flannery, C.R., Zeng, W., Corcoran, C., Collins-Racie, L.A.,
Chockalingam, P.S., Hebert, T., Mackie, S.A., McDonagh, T.,
Crawford, T.K., Tomkinson, K.N., laVallie, E.R., Morris, E.A.,
2002. Autocatalytic cleavage of ADAMTS-4 (aggrecanase-1)
reveals multiple glycosaminoglycan-binding sites. J. Biol. Chem.
277, 4277542780.
Francischetti, I.M., Valenzuela, J.G., Pham, V.M., Gareld, M.K.,
Ribeiro, J.M.C., 2002a. Toward a catalog for the transcripts and
proteins (sialome) from the salivary gland of the malaria vector
Anopheles gambiae. J. Exp. Biol. 205, 24292451.
Francischetti, I.M., Valenzuela, J.G., Andersen, J.F., Mather, T.N.,
Ribeiro, J.M.C., 2002b. Ixolaris, a novel recombinant tissue factor
pathway inhibitor (TFPI) from the salivary gland of the tick,
Ixodes scapularis: identication of factor X and factor Xa as
scaffolds for the inhibition of factor VIIa/tissue factor complex.
Blood 99, 36023612.
Francischetti, I.M., Mather, T.N., Ribeiro, J.M., 2003. Cloning of a
salivary gland metalloprotease and characterization of gelatinase
and brin(ogen)lytic activities in the saliva of the Lyme disease tick
vector Ixodes scapularis. Biochem. Biophys. Res. Commun. 305,
869875.
Francischetti, I.M., Mather, T.N., Ribeiro, J.M.C., 2004a. Penthalaris,
a novel recombinant ve-Kunitz tissue factor pathway inhibitor
(TFPI) from the salivary gland of the tick vector of Lyme disease,
Ixodes scapularis. Thromb. Haemostasis 91, 886898.
Francischetti, I.M., My-Pham, V., Harrison, J., Gareld, M.K.,
Ribeiro, J.M.C., 2004b. Bitis gabonica (Gaboon viper) snake
venom gland: toward a catalog for the full-length transcripts
(cDNA) and proteins. Gene 337, 5569.
Francischetti, I.M., Mather, T.N., Ribeiro, J.M.C., 2005. Tick saliva is
a potent inhibitor of endothelial cell proliferation and angiogenesis.
Thromb. Haemost. 94 (1), 167174.
Fry, B.G., 2005. From genome to venome: molecular origin and
evolution of the snake venom proteome inferred from phylogenetic
analysis of toxin sequences and related body proteins. Genome
Res. 15, 403420.
Gillespie, R.D., Dolan, M.C., Piesman, J., Titus, R.G., 2001.
Identication of an IL-2 binding protein in the saliva of the Lyme
disease vector tick, Ixodes scapularis. J. Immunol. 166, 43194326.
Harvey, A.H., 2001. Twenty years of dendrotoxins. Toxicon 39, 1526.
Hembrough, T.A., Ruiz, J.F., Swerdlow, B.M., Swartz, G.M.,
Hammers, H.J., Zhang, L., Plum, S.M., Williams, M.S., Strick-
land, D.K., Pribluda, V.S., 2004. Identication and characteriza-
tion of a very low density lipoprotein receptor-binding peptide
from tissue factor pathway inhibitor that has antitumor and
antiangiogenic activity. Blood 103, 33743380.
Huber, R., Kukla, D., Bode, W., Schwager, P., Bartels, K.,
Deisenhofer, J., Steigemann, W., 1974. Structure of the complex
formed by bovine trypsin and bovine pancreatic trypsin inhibitor.
II. Crystallographic renement at 1.9 A
resolution. J. Mol. Biol. 89,

73101.
Karczewski, J., Endris, R., Connolly, T.M., 1994. Disagregin is a
brinogen receptor antagonist lacking the ArgGlyAsp sequence
from the tick, Ornithodoros moubata. J. Biol. Chem. 269,
67026708.
Karplus, K., Barrett, C., Hughey, R., 1998. Hidden Markov models
for detecting remote protein homologies. Bioinformatics 14,
846856.
Kelley, L.A., MacCallum, R.M., Stemberg, M.J., 2000. Enhanced
genome annotation using structural proles in the program 3D-
PSSM. J. Mol. Biol. 299, 499520.
Klompen, J.S., Black IV, W.C., Keirans, J.E., Oliver Jr., J.H., 1996.
Evolution of ticks. Annu. Rev. Entomol. 41, 141161.
Kopecky, J., Kuthejlova, M., 1998. Suppressive effect of Ixodes ricinus
salivary gland extract on mechanisms of natural immunity in vitro.
Parasite Immunol. 20, 169174.
Krishnaswamy, S., 2005. Exosite-driven substrate specicity and
function in coagulation. J. Thromb. Haemostasis 3, 5467.
Lane, R.S., Moss, R.B., Hsu, Y.P., Wei, T., Mesirow, M.L., Kuo,
M.M., 1999. Anti-arthropod saliva antibodies among residents of a
ARTICLE IN PRESS
community at high risk for Lyme disease in California. Am. J.
Trop. Med. Hyg. 61, 850859.
Laskowski Jr., M., Kato, I., 1980. Protein inhibitors of proteinases.
Annu. Rev. Biochem. 49, 593626.
Lu, X., Sun, Y., Shang, D., Wattam, B., Egglezou, S., Hughes, T.,
Hyde, E., Scully, M., Kakkar, V., 2001. Evaluation of the role of
proline residues anking the RGD motif of dendroaspin, an
inhibitior of platelet aggregation and cell adhesion. Biochem.
J. 355, 633638.
Mans, B.J., Neitz, A.W., 2004. Adaptation of ticks to a blood-feeding
environment: evolution from a functional perspective. Insect
Mans, B.J., Louw, A.I., Neitz, A.W., 2002a. Savignygrin, a platelet
aggregation inhibitor from the soft tick Ornithodoros savignyi,
presents the RGD integrin recognition motif on the Kunitz-BPTI
fold. J. Biol. Chem. 277, 2137121378.
Mans, B.J., Louw, A.I., Neitz, A.W., 2002b. Evolution of hemato-
phagy in ticks: common origins for blood coagulation and platelet
aggregation inhibitors from soft ticks of the genus Ornithodoros.
Mol. Biol. Evol. 19, 16952705.
Mast, A.E., Broze Jr., G.J., 1996. Physiological concentrations of
tissue factor pathway inhibitor do not inhibit prothrombinase.
Blood 87, 18451850.
McDowell, R.S., Dennis, M.S., Louie, A., Shuster, M., Mulkerrin,
M.G., Lazarus, R.A., 1992. Mambin, a potent glycoprotein IIb-
IIIa antagonist and platelet aggregation inhibitor structurally
related to the short neurotoxins. Biochemistry 31, 47664772.
Monteiro, R.Q., Rezaie, A.R., Ribeiro, J.M., Francischetti, I.M.,
2005. Ixolaris: a factor Xa heparin-binding exosite inhibitor.
Biochem. J. 387, 871877.
Narasimhan, S., Koski, R.A., Beaulieu, B., Anderson, J.F., Rama-
moorthi, N., Kantor, F., Cappello, M., Fikrig, E., 2002. A novel
family of anticoagulants from the saliva of Ixodes scapularis. Insect
Mol. Biol. 11, 641650.
Nathoo, A.N., Moeller, R.A., Westlund, B.A., Hart, A.C., 2001.
Identication of neuropeptide like protein gene families in
Caenorhabditis elegans and other species. Proc. Natl. Acad. Sci.
USA 98, 1400014005.
Nene, V., Lee, D., Kanga, S., Skilton, R., Shah, T., de Villiers, E.,
Mwaura, S., Taylor, D., Quackenbush, J., Bishop, R., 2004. Genes
transcribed in the salivary glands of female Rhipicephalus
appendiculatus ticks infected with Theileria parva. Insect Biochem.
Mol. Biol. 34, 111711128.
Paesen, G.C., Adams, P.L., Harlos, K., Nuttall, P.A., Stuart, D.I.,
1999. Tick histamine-binding proteins: isolation, cloning, and
three-dimensional structure. Mol. Cell 3, 661671.
Porter, S., Clark, I.M., Kevorkian, L., Edwards, D.R., 2005. The
ADAMTS metalloproteinases. Biochem. J. 386, 1527.
Ramachandra, R.N., Wikel, S.K., 1992. Modulation of host-immune
responses by ticks (Acari: Ixodidae): effect of salivary gland
extracts on host macrophages and lymphocyte cytokine e produc-
tion. J. Med. Entomol. 29, 818826.
Rezaie, A.R., 2000. Heparin-binding exosite of factor Xa. Trends
Cardiovasc. Med. 10, 333338.
Rezaie, A.R., 2001. Prothrombin protects factor Xa in the prothrom-
binase complex from inhibition by the heparin-antithrombin
complex. Blood 97, 23082313.
Ribeiro, J.M., Francischetti, I.M., 2003. Role of arthropod saliva in
blood feeding: sialome and post-sialome perspectives. Annu. Rev.
Entomol. 48, 7388.
Ribeiro, J.M., Mather, T.N., 1998. Ixodes scapularis: salivary kininase
activity is a metallo dipeptidyl carboxypeptidase. Exp. Parasitol.
89, 213221.
Ribeiro, J.M., Makoul, G.T., Levine, J., Robinson, D.R., Spielman,
A., 1985. Antihemostatic, antiinammatory, and immunosuppres-
sive properties of the saliva of a tick, Ixodes dammini. J. Exp. Med.
161, 332344.
Ribeiro, J.M., Makoul, G.T., Robinson, D.R., 1988. Ixodes dammini:
evidence for salivary prostacyclin secretion. J. Parasitol. 74,
10681069.
Ruf, W., 2004. Protease-activated receptor signaling in the regulation
of inammation. Crit. Care Med. 32, S287S292.
Sanchez-Vargas, I., Travanty, E.A., Keene, K.M., Franz, A.W., Beaty,
B.J., Blair, C.D., Olson, K.E., 2004. RNA interference, arthropod-
borne viruses, and mosquitoes. Virus Res. 102, 6574.
Santos, I.K., Valenzuela, J.G., Ribeiro, J.M., de Castro, M., Costa,
J.N., Costa, A.M., da Silva, E.R., Neto, O.B., Rocha, C., Daffre,
S., Ferreira, B.R., da Silva, J.S., Szabo, M.P., Bechara, G.H., 2004.
Gene discovery in Boophilus microplus, the cattle tick: the
transcriptomes of ovaries, salivary glands, and hemocytes. Annu.
N. Y. Acad. Sci. 1026, 242246.
Sonenshine, D.E., 1985. Pheromones and other semiochemicals of the
acari. Annu. Rev. Entomol. 30, 128.
Sutcliffe, M.J., Jaseja, M., Hyde, E.I., Lu, X., Williams, J.A., 1994.
Three-dimensional structure of the RGD-containing neurotoxin
homologue dendroaspin. Nat. Struct. Biol. 1, 802807.
Valenzuela, J.G., Charlab, R., Mather, T.N., Ribeiro, J.M., 2000.
Purication, cloning, and expression of a novel salivary antic-
omplement protein from the tick, Ixodes scapularis. J. Biol. Chem.
275, 1871718723.
Valenzuela, J.G., Francischetti, I.M., Pham, V.M., Gareld, M.K.,
Mather, T.N., Ribeiro, J.M., 2002. Exploring the sialome of the
tick Ixodes scapularis. J. Exp. Biol. 205, 28432864.
Wang, X., Coons, L.B., Taylor, D.B., Stevens Jr., S.E., Gartner, T.K.,
1996. Variabilin, a novel RGD-containing antagonist of glycopro-
tein IIb-IIIa and platelet aggregation inhibitor from the hard tick
Dermacentor variabilis. J. Biol. Chem. 271, 1778517790.
Wang, H., Kaufman, W.R., Nuttall, P.A., 1999. Molecular indivi-
duality: polymorphism of salivary gland proteins in three species of
Ixodid tick. Exp. Appl. Acarol. 23, 969975.
Waxman, L., Smith, D.E., Arcuri, K.E., Vlasuk, G.P., 1990. Tick
anticoagulant peptide (TAP) is a novel inhibitor of blood
coagulation factor Xa. Science 248, 593596.
Wikel, S.K., 1999. Tick modulation of host immunity: an important
factor in pathogen transmission. Int. J. Parasitol. 29, 851859.
Williams, J.A., Lu, X., Rahman, S., Keating, C., Kakkar, V., 1993.
Dendroaspin: a potent integrin receptor inhibitor from the venoms of
Dendroaspis viridis and D. jamesonii. Biochem. Soc. Trans. 21, 73S.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Development and characterization of a double subgenomic
chikungunya virus infectious clone to express heterologous genes in
Aedes aegypti mosqutioes
Dana L. Vanlandingham
a
, Konstantin Tsetsarkin
a
, Chao Hong
a
, Kimberly Klingler
a
,
Kate L. McElroy
a
, Michael J. Lehane
b
, Stephen Higgs
a,
a
Department of Pathology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX, USA
b
Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK
Abstract
Three full-length infectious cDNA clones based on the alphavirus chikungunya (CHIKV) were developed and characterized in
vitro and in vivo. The full-length clone retained the viral phenotypes of CHIKV in both cell culture and in mosquitoes and should be
a valuable tool for the study of virus interactions in an epidemiologically signicant natural vector, Aedes aegypti. Two additional
infectious clones were constructed that express green uorescent protein (EGFP) in the midgut, salivary glands, and nervous tissue
of Aedes aegypti mosquitoes following oral infection. The two constructs differed in the placement of the subgenomic promoter and
the gene encoding EGFP. Viruses derived from the pCHIKic EGFP constructs (5
0
CHIKV EGFP and 3
0
CHIKV EGFP) expressed
EGFP in 100% of the Ae. aegypti mosquitoes tested on days 7 and 14 post infection (p.i.). The 5
0
CHIKV EGFP disseminated to
90% of the salivary glands and nervous tissue by day 14 p.i. Dissemination rates of this new viral vector exceeds those of previous
systems, thus expanding the repertoire and potential for gene expression studies on this important vector species.
Keywords: Mosquitoes; Aedes aegypti; Chikungunya virus; Green uorescent protein; Gene expression
1. Introduction
Since the 1980s there has been an effort to develop
genetically engineered mosquitoes to reduce the impact
to humans of various arthropod-borne pathogens,
notably Plasmodium spp., and arboviruses such as
dengue virus. The basis for this approach is that the
capacity of vectors to be infected with, or to transmit,
specic pathogens can be altered by genetic manipula-
tion of the mosquito vector. It is assumed that pathogen
transmission in the eld can be reduced by replacing
existing competent vector populations with mosquitoes
that have been genetically modied to be less competent.
Technology to genetically engineer mosquitoes has
improved dramatically following the discovery and
development of transformation systems using transpo-
sable elements such as Hermes (Warren et al., 1994),
Minos (Franz and Savakis, 1991), piggyBac (Cary et al.,
1989), and Mariner (Jacobson and Hartl, 1985). Using
these tools, several species of mosquitoes have been
transformed, including Aedes aegypti, Anopheles ste-
phensi, An. gambiae, and Culex spp (Jasinskiene et al.,
2000, 1998; Coates et al., 1998; Pinkerton et al., 2000;
Grossman et al., 2001; Perera et al., 2002). Despite these
successes, the methodology and requirements for gen-
erating transgenic lines and the maintenance and
expansion of them, is labor intensive and requires
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.008

fax: +1 409 747 2437.
E-mail address: sthiggs@utmb.edu (S. Higgs).
considerable expertise and resources. For these reasons,
techniques that can rapidly evaluate gene activity prior
to germ-line integration are essential to ensure that
efforts are focused on strategies with the greatest chance
of success.
A full length cDNA clone of the alphavirus Sindbis
(dsSIN), pTE/3
0
2J, was developed as a transient
expression system for heterologous RNAs and proteins
and has proved to be an efcient expression system in
cell culture and mosquitoes (Hahn et al., 1992; Higgs
et al., 1995). A number of gene sequences have since
been delivered to mosquitoes using this, and other
dsSIN expression systems including: antisense con-
structs to interfere with viral replication (Powers et al.,
1996; Olson et al., 1996; Higgs et al., 1998; Adelman
et al., 2001), to silence genes such as luciferase (Johnson
et al., 1999), and mosquito genes such as prophenolox-
idase (Shiao et al., 2001; Tamang et al., 2004); genes to
over-express toxin genes (Higgs et al., 1995), single chain
antibodies, (James et al., 1999), and to knockout of
endogenous genes using RNAi (Attardo et al., 2003).
The dsSIN pTE/3
0
2J system has also been used to infect
larval arthropods by feeding infected cells that expressed
green uorescent protein (GFP) or defensin genes
(Higgs et al., 1999; Cheng et al., 2001). In addition to
the GFP reporter system, other reporters have been used
such as chloramphenicol acetyltransferase (Olson et al.,
1994).
Despite the successful employment of the dsSIN
expression system, a shortcoming of the original system
was the relatively low efciency with which it infects and
disseminates from the midgut following oral infection.
Although this problem has been addressed using various
strategies (Seabaugh et al., 1998; Olson et al., 2000;
Higgs et al., 1997; Pierro et al., 2003; Foy et al., 2004),
the oral infectivity and dissemination rates of dsSIN
expression systems are frequently too low for use with
genes which are difcult to characterize in Ae. aegypti.
The SINV strain MRE 16 (Oriental/Australian geno-
type), has been found to be more orally infectious to Ae.
aegypti than prototype SINV strain AR339 (Bowen,
1987). Several expression systems based on the MRE 16
strain have been developed. ME2/5
0
2J/GFP, contains
the E2 gene of MRE 16 in the dsSIN cDNA clone pTE/
5
0
2J backbone and expresses the reporter gene EGFP in
63% of the midguts with dissemination of virus and
expression of EGFP in 67% of the mosquitoes analyzed
following oral infection of Ae. aegypti (Pierro et al.,
2003). A full length infectious clone of MRE 16,
5
0
dsMRE 16ic-EGFP, increased the infection rate to
93% of the Ae. aegypti posterior midguts and 68% of
the whole mosquitoes (Foy et al., 2004).
For practical reasons, the SINV-based expression
systems have been progressively modied to infect and
disseminate in Ae. aegypti. However, SINV is primarily
transmitted by Culex mosquitoes in nature (Karabatsos,
1985) and does not normally infect Ae. aegypti. The
MRE 16 strain was one of three isolates from Cx.
tritaeniorhynchus collected in Sarawak Malaysia, which
replicated in AP-61 (Ae. pseudoscutellaris) cell culture
(Pudney et al., 1979; C.J. Leake, personal communica-
tion). Under experimental conditions it was found to
replicate in Ae. aegypti (Bowen, 1987), hence its
development for gene expression studies. Although
MRE 16 derived viruses replicate in this species, it is
possible that because the virus does not naturally infect
Ae. aegypti, the infection process and tissue tropisms
might not be optimal. In this paper we describe the
development and characterization of an expression
system based on CHIKV, a virus that naturally infects
Ae. aegypti.
CHIKV (strain 37997) was recently characterized in
Ae. aegypti mosquitoes and was found to have
consistently high infection and dissemination rates
(Vanlandingham et al., 2005). These data indicate that
CHIKV (37997) has the potential to express hetero-
logous genes, following oral infection of Ae. aegypti, at a
consistently higher rate than most SINV-based expres-
sion systems. CHIKV and SINV are arthropod-borne
viruses in the family Togaviridae. These viruses consist
of a positive sense, linear, ssRNA genome which is
11.7 Kb in size. The 5
0
terminus of the virus is capped
and the 3
0
terminus is polyadenylated. The nonstructural
proteins (nsP14) are encoded at the 5
0
end of the
genome followed by the structural proteins which are
encoded from a subgenomic promoter at the 3
0
end. The
structural proteins consist of a capsid, two envelope
glycoproteins (E1 and E2), and two small peptides E3
and 6 K (Strauss and Strauss, 1994).
Here we describe the development and characteriza-
tion of three novel expression systems based on CHIKV
(37997) which are orally infectious in Ae. aegypti and
have high infection and dissemination rates. The ability
of the 5
0
CHIK EGFP virus to express a heterologous
gene in 100% of mosquitoes midguts and to disseminate
to 90% of the salivary glands and head tissues, following
oral infection, will enable the biological characterization
of endogenous genes. The expression of EGFP from an
epidemiologically important virus in Ae. aegypti is a
signicant improvement over the SINV system for
the study of virusvector relationships with Ae. aegypti
mosquitoes.
2.1. Viruses
The 37997 strain of CHIKV was obtained from
the World Reference Center for Arboviruses at the
University of Texas Medical Branch, Galveston, TX.
CHIKV was originally isolated from Ae. furcifer
ARTICLE IN PRESS
D.L. Vanlandingham et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11621170 1163
mosquitoes from Kadougou, Senegal in 1983 and was
passed once in Ae. pseudoscutellaris (AP-61) cells and
twice in Vero (green monkey kidney) cells. Stock virus
was produced following a single passage in Vero cells,
grown at 37 1C in Leibovitz L-15 media with 10% fetal
bovine serum (FBS), 100U penicillin, and 100 mg/mL
streptomycin. Virus was harvested when cells showed
75% cytopathic effect (CPE) and aliquoted and stored
at 80 1C for use in all experiments.
2.2. RNA extraction
RNA was generated using C6/36 cells that were
inoculated with the CHIKV (37997). Virus was har-
vested from cell culture supernatant using the QIAamp
Viral RNA Mini kit (Qiagen, Valencia, CA) following
the manufacturers protocol. RNA was stored at 80 1C
for later use.
2.3. Reverse transcription and sequencing
CHIKV (37997) RNA was reverse transcribed to
produce cDNA using random hexanucleotide primers
(Promega, Madison, WI) and Superscript II (Invitrogen
Life Technologies) following manufactures instructions.
cDNA was amplied with Taq DNA polymerase (New
England BioLabs, Beverly, MA) with 35 cycles at 94 1C,
20 s; 55 1C, 20 s; 72 1C, 2 min; nal extension at 70 1C for
5 min. Amplied PCR products were analyzed by
electrophoreses on 1% agarose gel and gel-puried
using the QIAquick Gel Extraction Kit (Qiagen). The
puried PCR products were used for direct sequencing.
Sequencing of the 5
0
and 3
0
ends. The 3
0
terminal
sequence was determined using the 3
0
RACE method
(Frohman, 1994). The 5
0
terminal sequence was deter-
mined using the 5
0
RACE kit (Ambion, Austin, Texas)
following the manufacturers instruction.
2.4. Construction of infectious clones
Three plasmids, pCHIKic, 5
0
pCHIKic EGFP, and 3
0
pCHIKic EGFP, were prepared by standard PCR-based
cloning methods. CHIKV DNA fragments were sub-
stituted into an alphavirus onyong nyong/pBluescript II
SK(+) infectious clone (p5
0
dsONNic-Foy) which was
kindly provided by Ken E. Olson and Brian Foy (Brault
et al., 2004). This clone was modied by substituting
the T7 promoter with an SP6 promoter and the removal
of restriction sites. The PCR amplied fragments of
CHIKV (37997) were produced using high-delity PFU
polymerase (Stratagene, La Jolla, CA). The fragments
were ligated either singly or in tandem with T4 DNA
ligase (Stratagene) and transformed into XL10-Gold
cells (Stratagene). All plasmids were extracted using
QIAprep Spin Miniprep Kit (Qiagen). The construction
of the 5
0
pCHIKic EGFP is illustrated in Fig. 1. The 3
0
pCHIKic EGFP and the pCHIKic clones were con-
structed by similar methods. The 5
0
and 3
0
pCHIKic
EGFP plasmids have the capacity to accept an insert of
at least 724bp in length using restriction sites AscI, PacI
or EcoRI. Additional information, such as the sequences
of primers and maps of the clones are available from the
authors upon request.
2.5. In vitro transcription of the pCHIKic clones
Infectious virus from the pCHIKic clones (CHIKV, 5
0
CHIKV EGFP, and 3
0
CHIKV EGFP) were produced
by linearization with NotI which was in vitro tran-
scribed from SP6 promoter using the mMESSAGE
mMACHINE kit (Ambion) following manufactures
instructions. RNA was electroporated into BHK-21 S
cells as previously described (Higgs et al., 1997). Cell
culture supernatant containing virus was harvested,
alloquoted and stored at 80 1C when the cells showed
75% CPE.
2.6. In vitro growth kinetics of viruses
One vertebrate-derived, Vero, and two mosquito-
derived, C6/36 (Ae. albopictus) and MOS-55 (An.
gambiae), cell lines were used for these studies. All cells
were maintained in L-15 medium with 10% FBS, 100 U/
mL penicillin, and 100 mg/mL streptomycin. Vertebrate
and mosquito cells were maintained at 37 1C and 28 1C,
respectively. CHIKV (37997) and infectious virus from
pCHIKic (CHIKV) were grown on conuent cell
monolayers, 25 cm
2
asks were infected with a standard
1 mL inoculum by rocking at room temperature for 1 h.
The inoculum was then removed and after three washes
with 5 mL L-15, 5.5 mL of medium was added per ask.
A sample of 0.5 mL was removed immediately. Addi-
tional 0.5 mL samples were collected at 24 h intervals
and replaced with 0.5 mL of fresh medium. Samples
ARTICLE IN PRESS
Fig. 1. Construction of 5
0
pCHIKic EGFP.
were stored at 80 1C until titrated. Data represents
virus production for a standardized monolayer area
(25 cm
2
). Due to a difference in the size of individual
cells, the multiplicity of infection varied for the different
cell lines. Expression of EGFP was assessed following
infection of 5
0
CHIKV EGFP and 3
0
CHIKV EGFP in
Vero and C6/36 cells using protocols above. Viruses
were compared at 48 h p.i. for the amount of EGFP
expression.
2.7. Mosquitoes
The white-eyed Higgs variant of the Rexville D strain
of Ae. aegypti were reared at 27 1C and 80% relative
humidity under a 16 h light: 8 h dark photoperiod, as
previously described (Wendell et al., 2000; Miller and
Mitchell, 1991). Adults were supplied with a cotton wool
pad soaked in a 10% sucrose solution ad libitum and fed
on anaesthetized hamsters once per week for egg
production.
2.8. Virus infections of mosquitoes
Four-day old adult female Ae. aegypti mosquitoes
were fed a blood meal containing one of the four viruses
to be analyzed. Fresh virus was grown from stock and
harvested from Vero cells when 75% of the cells showed
CPE. The viral supernatant was mixed with an equal
volume of debrinated sheep blood (Colorado Serum
Company, Denver, CO). As a phagostimulant, adeno-
sine triphosphate at a nal concentration of 2 mM, was
added to the blood meal. Mosquitoes were fed using an
isolation glove box located in a Biosafety Level 3
insectary. Infectious blood was heated to 37 1C and
placed in a Hemotek feeding apparatus (Discovery
Workshops, Accrington, Lancashire, United Kingdom)
and mosquitoes were allowed to feed for 1 h (Cosgrove
et al., 1994). Fully engorged females were separated
from unfed females and were placed into new cartons.
Three to eight mosquitoes were removed for titration on
days 0, 1, 2, 3, 7, and 14 p.i. and were stored at 80 1C.
Day 0 samples, collected immediately after feeding, were
used to determine the titer of virus imbibed and to
evaluate continuity between experiments.
2.9. Titrations
Viral samples harvested from cell culture and
mosquitoes were quantied as tissue culture infectious
dose 50 endpoint titers (log
10
TCID
50
/mL) using a
standardized procedure (Higgs et al., 1997). Briey,
100 mL samples of cell culture supernatant/mosquito
triturate were pipetted into wells of the rst column of a
96-well plate, serially diluted in a 10-fold series, seeded
with Vero cells and incubated at 37 1C for 7 days. Prior
to titration, each mosquito was triturated in 1mL of L-
15 medium and ltered through a 0.22 mM syringe lter
(Millipore, Carrigwohill, Cork, Ireland).
2.10. Immunouorescence Assay (IFA) and EGFP
analysis
Midguts and salivary glands were dissected from 7
and 14-day p.i. mosquitoes for analysis to determine
dissemination rates. The mosquitoes were dissected on
glass microscope slides in phosphate buffered saline. For
IFA, salivary glands were air dried, xed in cold acetone
for 10 min and stained using a cross-reactive mouse
hyperimmune ascitic uid raised against chikunguna
virus as the primary antibody and amplifying the signal
using indirect IFA protocols previously described
(Gould et al., 1985a, b; Higgs et al., 1997). For analysis
of EGFP expression, midguts and salivary glands were
dissected directly into glycerol-saline and immediately
examined for EGFP expression under an Olympus IX-
70 epiuorescence microscope. Differences in the infec-
tion and dissemination rates based on IFA or EGFP
analysis were tested for signicance using Fishers Exact
Test, SPSS version 11.5 (SPSS Inc. Chicago, IL).
3. Results
CHIKV (37997) and CHIKV derived from pCHIKic
in the vertebrate cell line, Vero and two invertebrate cell
lines, C6/36 (Ae. albopictus) and MOS-55 (An. gambiae),
displayed similar in vitro growth characteristics (Fig. 2).
The peak titer of both CHIKV (37997) and CHIKV in
Vero and C6/36 cells was reached at day 2 p.i. The titers
decreased at similar rates from day 2 p.i. to day 6 p.i.
(Fig. 2). 5
0
CHIKV EGFP and 3CHIKV EGFP were
compared in Vero and C6/36 cells to assess the levels of
EGFP expression in cell culture. The 3
0
clone expressed
EGFP at a markedly higher intensity than the 5
0
clone in
both cell types examined.
In vivo experiments were conducted in Ae. aegypti
mosquitoes to compare the CHIKV (37997) and the
three viruses derived from infectious clones. The blood
meal titers for the CHIKV (37997) and CHIKV were
identical, 7.95 log
10
TCID
50
/mL, and the percent of
infected mosquitoes and titers of virus in the mosquitoes
were similar by whole body titrations of mosquitoes at
six time points p.i. (Table 1). The two clones that
expressed EGFP had slightly lower blood meal titers
when compared to the CHIKV. Both the 5
0
and the
3
0
CHIKV EGFP had a blood meal titer of 7.52 log
10
TCID
50
/mL. Although the blood meal titers were
slightly different between the viruses with or without
EGFP, all of the viruses infected 100% of the
mosquitoes on day 14 p.i. (Table 1).
IFA and EGFP were used to determine the percent of
mosquitoes infected on days 7 and 14 p.i. (Table 2). IFA
ARTICLE IN PRESS
and EGFP data were compared using dissected midguts
to determine infection rates and dissected salivary
glands to determine dissemination rates. Expression of
EGFP in the eyes of day 14 p.i. Ae. aegypti indicated
infected nervous tissue, virus was not observed in other
tissues. Two experiments were completed to compare
virus derived from the 5
0
and 3
0
pCHIKic EGFP
constructs (Table 2). In both experiments, the infection
rates were 100% on days 7 and 14 p.i. for both viruses
(Table 2). The dissemination rates were similar based on
antigen detection by IFA and by EGFP visualization for
the 5
0
CHIKV EGFP with 100% and 90% dissemina-
tion by IFA and EGFP, respectively (Table 2). The 3
0
CHIKV EGFP dissemination rates on day 14 p.i. were
different for the two experiments. The percent of
mosquitoes with disseminated infections by IFA were
100% on day 14 p.i.; where as, the percent of
mosquitoes with disseminated infections by EGFP were
70% (Table 2).
4. Discussion
Previous experiments which compared the growth of
CHIKV (37997) on Vero, C6/36, and Mos-55 cells had
shown the CHIKV (37997) was able to replicate in Vero
and C6/36 cell lines but was unable to replicate in the
Moss-55 cell line (Vanlandingham et al., 2005). As
expected, neither CHIKV (37997) nor the virus from the
infectious clone, that did not contain EGFP (CHIKV),
grew in the MOS-55 cells (Fig. 2) indicating that the in
vitro phenotype of the CHIKV produced from the
infectious clone has been retained and is similar to that
of the parental virus. The infection rates of these two
viruses were also found to be retained in vivo following
oral infection of Ae. aegypti mosquitoes (Table 1). The
CHIKV 37997 infected 100% of the mosquitoes
examined at all time points p.i. These results are similar
to the virus derived from the infectious clone which
infected 100% of the mosquitoes on all time points p.i.
ARTICLE IN PRESS
CHIKic
0
2
4
6
8
10
0 1 2 3 4 5 6
Days post-infection
T
i
t
e
r

(
L
o
g
1
0

T
C
I
D
5
0
/
m
L
)
CHIK - 37997
0
2
4
6
8
10
0 1 2 3 4 5 6
Days post-infection
T
i
t
e
r

(
L
o
g
1
0

T
C
I
D
5
0
/
m
L
)
Vero
C6/ 36
MOS-55
Vero
C6/ 36
MOS-55
Fig. 2. In vitro growth of CHIKV strain 37997 and virus produced from pCHIKic in Vero, C6/36, and MOS-55 cell lines.
Table 1
Infection rates of CHIKV (37997) and virus derived from pCHIKic
(CHIKV), 5
0
pCHIKic EGFP (5
0
CHIKV EGFP), and 3
0
pCHIKic
EGFP (3
0
CHIKV EGFP) in Ae. aegypti
Virus titer
a
Day p.i. Titer7S.D. Infected/Total (%)
CHIKV 0 6.770.3 3/3 (100)
37997 1 5.970.8 8/8 (100)
2 4.570.0 5/5 (100)
3 4.870.2 5/5 (100)
7 6.870.6 8/8 (100)
14 5.070.4 7/7 (100)
CHIKV 0 6.570.0 3/3 (100)
1 6.270.5 8/8 (100)
2 5.970.5 8/8 (100)
3 4.570.9 6/8 (75)
7 7.072.2 8/8 (100)
14 5.070.4 7/7 (100)
5
0
CHIKV 0 6.470.5 3/3 (100)
EGFP 1 5.070.4 5/8 (63)
2 2.471.2 4/8 (50)
3 2.770.1 3/6 (50)
7 5.670.87 6/6 (100)
14 4.270.26 8/8 (100)
3
0
CHIKV 0 6.170.3 3/3 (100)
EGFP 1 4.870.9 8/8 (100)
2 4.470.6 8/8 (100)
3 5.170.9 8/8 (100)
7 5.071.3 8/8 (100)
14 5.270.7 8/8 (100)
a
Virus titers for blood meals: CHIKV 379977.95 log
10
TCID
50
/mL,
virus derived from pCHIKic7.95 log
10
TCID
50
/mL, 5
0
CHIKV
EGFP7.52 log
10
TCID
50
/mL, 3
0
CHIKV EGFP7.52 log
10
T-
CID
50
/mL.
except day 3 p.i. The average whole body mosquito titer
of both viruses on day 14 p.i. was 5.0 log
10
TCID
50
/mL.
These results indicate that this infectious clone will be
useful for the study of CHIKV in Ae. aegypti.
The two viruses that express the reporter gene, EGFP,
were compared in Ae. aegypti mosquitoes. These viruses
differ in the placement of the EGFP sequence within the
viral genome. Previous studies have indicated that the
placement of the reporter gene at either the 5
0
or the 3
0
position within various alphaviruses produce differences
in expression levels of the reporter gene and in the
stability of the construct (Higgs et al., 1995). In the 5
0
pCHIKic EGFP, the EGFP was placed downstream of
the nonstructural genes and a RNA subgenomic
promoter. The EGFP is followed by an additional
internal RNA subgenomic promoter sequence and the
viral structural genes (Fig. 1). The 5
0
position has been
shown to be more stable than the 3
0
position in two
SINV expression systems (ME2 5
0
2J/GFP and TE/5
0
2J/
GFP) following several passages in cell culture. Pierro
et al (2003) found that genes encoding GFP placed at
the 5
0
position expressed GFP in more than 90% of the
cells following ve passages in vero cells. In the 3
0
pCHIKic EGFP construct, the EGFP is expressed from
an additional RNA subgenomic promoter which is
located at the extreme 3
0
end of the structural genes of
the virus. Studies using various SIN expression systems
have indicated that the 3
0
construction is unstable after
multiple passages in cell culture. This instability is
characterized by the ability to detect viral antigen in the
absence of GFP expression (Higgs et al., 1999; Pierro
et al., 2003).
Ae. aegypti mosquitoes infected with either the 5
0
or 3
0
CHIKV EGFP were analyzed by IFA and EGFP
expression in the midguts and salivary glands on days
7 and 14 p.i. Nervous tissue was also examined on day
14 p.i. by analysis of EGFP expression in the eyes
(Fig. 3). These tissues and time points were selected
based on previous experiments with ONNV and
CHIKV (Vanlandingham et al., 2005). The tissue
tropisms of CHIKV EGFP differ from those observed
for SINV at similar time points (Foy et al., 2004; Pierro
et al., 2003; Raymes-Keller et al., 1995) being less focal
in the midgut at early time points and more intense in
infected tissues at late time points. EGFP was expressed
in a higher percent of the salivary glands on day 14 p.i.
for 5
0
CHIKV EGFP when compared to 3
0
CHIKV
EGFP (Table 2). The intensity of EGFP expression
was greater for 3
0
CHIKV EGFP on day 14 p.i. (Fig. 3).
The 3
0
CHIKV EGFP disseminated in 100% of the
mosquito salivary glands examined by IFA and 70% of
the mosquito salivary glands and eyes when examined
by EGFP expression (Table 2). The nding that virus
disseminates at a higher level than the expression of
EGFP has been demonstrated for other alphavirus
expression systems which used the 3
0
construction
(Olson et al., 2000).
The high level of infection and efcient dissemina-
tion of these three CHIKV infectious clones will enable
studies of virus tropisms in an epidemiologi-
cally important vector with a naturally infectious virus.
The 5
0
and 3
0
CHIKV constructs will provide additional
tools for gene expression and knockout studies, for
example RNAi, in Ae. aegypti mosquitoes. Further-
more, by increasing the repertoire of alphaviral in-
fectious clones, we have the capacity to produce
chimeric viruses with specic gene or amino acid
substitutions that will help in the identication of the
molecular determinants of the viral infection process in
mosquitoes.
ARTICLE IN PRESS
Table 2
Infection and dissemination rates based on antigen detection by IFA and by visualization of EGFP for the CHIKV 37997, and viruses derived from
the pCHIKic, 5
0
pCHIKic EGFP and 3
0
pCHIKic EGFP in Ae. aegypti mosquitoes
Virus strain Day p.i. Infected (IFA)
a
positive/total (%)
Disseminated (IFA)
a
positive/total (%)
Infected (EGFP)
b
positive/total (%)
Disseminated (EGFP)
b
positive/total (%)
CHIKV 7 10/10 (100) 9/10 (100) na
c
na
37997 14 10/10 (100) 10/10 (100) na na
CHIKV 7 9/10 (90) 9/10 (90) na na
14 7/10 (70) 5/10 (50) na na
5
0
CHIKV 7 10/10 (100) 8/10 (80) 10/10 (100) 7/10 (70)
EGFP 14 10/10 (100) 10/10 (100) 10/10 (100) 9/10 (90)
3
0
CHIKV 7 5/5 (100) 1/5 (20) 10/10 (100) 4/10 (40)
EGFP 14 5/5 (100) 5/5 (100) 10/10 (100) 7/10 (70)
a
Virus titers of blood meals, analyzed by IFA: CHIKV 379977.95 log
10
TCID
50
/mL, viruses produced from pCHIKic7.95 log
10
TCID
50
/mL, 5
0
pCHIKic EGFP7.52 log
10
TCID
50
/mL, 3
0
10
TCID
50
/mL.
b
Virus titers of blood meal titers, analyzed by EGFP: 5
0
pCHIKic EGFP7.95 nd
4
, 3
0
10
TCID
50
/mL.
c
na not applicable.
ARTICLE IN PRESS
Fig. 3. EGFP expression on days 7 and 14 p.i. with virus derived from 5
0
and 3
0
pCHIKic EGFP in midguts, salivary glands, and eyes. (A) 3
0
CHIKV
EGFP day 7 p.i. midgut; (B) 3
0
CHIKV EGFP day 7 p.i. salivary gland; (C) 3
0
CHIKV EGFP day 14 p.i. midgut; (D) 3
0
CHIKV EGFP day 14 p.i.
salivary gland; (E) 3
0
CHIKV EGFP day 14 p.i. eyes; (F) 5
0
CHIKV EGFP day 7 p.i. midgut; (G) 5
0
CHIKV EGFP day 7 p.i. salivary gland; (H)
5
0
CHIKV EGFP day 14 p.i. midgut; (I) 5
0
CHIKV EGFP day 14 p.i. salivary gland; (J) 5
0
CHIKV EGFP day 14 p.i. eyes.
Acknowledgments
The authors would like to thank Jing Huang for
rearing and dissection of the Ae. aegypti used for these
experiments. We would also like to thank Dr. R. Tesh
for supplying the virus used to make the infectious
clones. This work was supported by: a grant from the
National Institutes of Health, Genetic Determinants of
Epidemic Alphavirus Transmission (Grant Number:
5RO1AI47877-3); DARPA, cooperative agreement
NOO178-02-2-9002 with the Chemical Biological Radi-
ological Defense Division (B50) of the Naval Surface
Warfare center, Dahlgren Division; Kate L. McElroy
was supported by a CDC Fellowship Training Program
in Vector-Borne Infectious Diseases T01/CCT622892.
References
Adelman, Z.N., Blair, C.D., Carlson, J.O., Beaty, B.J., Olson, K.E.,
2001. Sindbis virus-induced silencing of dengue viruses in
mosquitoes. Insect Mol. Biol. 10, 265273.
Attardo, G.M., Higgs, S., Klingler, K.A., Vanlandingham, D.L.,
Raikhel, A.S., 2003. RNA interference-mediated knockdown of a
GATA factor reveals a link to anautogeny in the mosquito Aedes
aegypti. Proc. Natl. Acad. Sci. USA 100, 1337413379.
Bowen, M.D., 1987. Transovarial transmission of arboviruses by their
vectors: experimental transovarial transmission of Sindbis virus
(genus Alphavirus) by Aedes aegypti. Thesis. London School of
Hygiene and Tropical Medicine.
Brault, A.C., Foy, B.D., Myles, K.M., Kelly, C.L., Higgs, S., Weaver,
S.C., Olson, K.E., Miller, B.R., Powers, A.M., 2004. Infection
patterns of onyong nyong virus in the malaria-transmitting
mosquito, Anopheles gambiae. Insect Mol. Biol. 13, 625635.
Cary, L.C., Goebel, M., Corsaro, B.G., Wang, H.G., Rosen, E.,
Fraser, M.J., 1989. Transposon mutagenesis of baculoviruses:
analysis of Trichoplusia ni transposon IFP2 insertions within the
FP-locus of nuclear polyhedrosis viruses. Virology 172, 156169.
Cheng, L.L., Bartholomay, L.C., Olson, K.E., Lowenberger, C.,
Vizioli, J., Higgs, S., Beaty, B.J., Christensen, B.M., 2001.
Characterization of an endogenous gene expressed in Aedes aegypti
using an orally infectious recombinant Sindbis virus. J. Insect Sci.
1, 1024.
Coates, C.J., Jasinskiene, N., Miyashiro, L., James, A.A., 1998.
Mariner transposition and transformation of the yellow fever
mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. USA 95,
37483751.
Cosgrove, J.B., Wood, R.J., Petric, D., Evans, D.T., Abbott, R.H.,
1994. A convenient mosquito membrane feeding system. J. Am.
Mosq. Control Assoc. 10, 434436.
Foy, B.D., Myles, K.M., Pierro, D.J., Sanchez-Vargas, I., Uhlirova,
M., Jindra, M., Beaty, B.J., Olson, K.E., 2004. Development of a
new Sindbis virus transducing system and its characterization in
three Culicine mosquitoes and two Lepidopteran species. Insect
Mol. Biol. 13, 89100.
Franz, G., Savakis, C., 1991. Minos, a new transposable element from
Drosophila hydei, is a member of the Tc1-like family of
transposons. Nucleic Acids Res. 19, 6646.
Frohman, M.A., 1994. On beyond classic RACE (rapid amplication
of cDNA ends). PCR Methods Appl. 4, S40S58.
Gould, E.A., Buckley, A., Cammack, N., 1985a. Use of the biotin-
streptavidin interaction to improve avivirus detection by im-
munouorescence and ELISA tests. J. Virol. Methods 11, 4148.
Gould, E.A., Buckley, A., Cammack, N., Barrett, A.D., Clegg, J.C.,
Ishak, R., Varma, M.G., 1985b. Examination of the immunological
relationships between aviviruses using yellow fever virus mono-
clonal antibodies. J. Gen. Virol. 66, 13691382.
Grossman, G.L., Rafferty, C.S., Clayton, J.R., Stevens, T.K.,
Mukabayire, O., Benedict, M.Q., 2001. Germline transformation
of the malaria vector, Anopheles gambiae, with the piggyBac
transposable element. Insect Mol. Biol. 10, 597604.
Hahn, C.S., Hahn, Y.S., Braciale, T.J., Rice, C.M., 1992. Infectious
Sindbis virus transient expression vectors for studying antigen
processing and presentation. Proc. Natl. Acad. Sci. USA 89,
26792683.
Higgs, S., Olson, K.E., Kamrud, K.I., Powers, A.M., Beaty, B.J., 1997.
Viral expression systems and viral infections in insects. In:
Crampton, J.M., Beard, C.B., Louis, C. (Eds.), The Molecular
Biology of Disease Vectors: A Methods Manual. Chapman & Hall,
UK, pp. 457483.
Higgs, S., Olson, K.E., Klimowski, L., Powers, A.M., Carlson, J.O.,
Possee, R.D., Beaty, B.J., 1995. Mosquito sensitivity to a scorpion
neurotoxin expressed using an infectious Sindbis virus vector.
Insect Mol. Biol. 4, 97103.
Higgs, S., Oray, C.T., Myles, K., Olson, K.E., Beaty, B.J., 1999.
Infecting larval arthropods with a chimeric, double subgenomic
Sindbis virus vector to express genes of interest. Biotechniques 27,
908911.
Higgs, S., Rayner, J.O., Olson, K.E., Davis, B.S., Beaty, B.J., Blair,
C.D., 1998. Engineered resistance in Aedes aegypti to a West
African and a South American strain of yellow fever virus. Am. J.
Trop. Med. Hyg. 58, 663670.
Jacobson, J.W., Hartl, D.L., 1985. Coupled instability of two X-linked
genes in Drosophila mauritiana: germinal and somatic mutability.
Genetics 111, 5765.
James, A.A., Beerntsen, B.T., Capurro, M.L., Coates, C.J., Coleman,
J., Jasinskiene, N., Krettli, A.U., 1999. Controlling malaria
transmission with genetically engineered, Plasmodium-resistant
mosquitoes: milestones in a model system. Parassitologia 41,
461471.
Jasinskiene, N., Coates, C.J., Benedict, M.Q., Cornel, A.J., Rafferty,
C.S., James, A.A., Collins, F.H., 1998. Stable transformation
of the yellow fever mosquito, Aedes aegypti, with the Hermes
element from the housey. Proc. Natl. Acad. Sci. USA 95,
37433747.
Jasinskiene, N., Coates, C.J., James, A.A., 2000. Structure of hermes
integrations in the germline of the yellow fever mosquito, Aedes
aegypti. Insect Mol. Biol. 9, 1118.
Johnson, B.W., Olson, K.E., Len-Miura, T., Rayms-Keller, A.,
Carlson, J.O., Coates, C.J., Jasinskiene, N., James, A.A., Beaty,
B.J., Higgs, S., 1999. Inhibition of luciferase expression in
transgenic Aedes aegypti mosquitoes by Sindbis virus expression
of antisense luciferase RNA. Proc. Natl. Acad. Sci. USA 96,
1339913403.
Karabatsos, N., 1985. International Catalogue of Arboviruses, third
Ed. American Society of Tropical Medicine and Hygiene, San
Antonio, TX.
Miller, B.R., Mitchell, C.J., 1991. Genetic selection of a avivirus-
refractory strain of the yellow fever mosquito Aedes aegypti. Am. J.
Trop. Med. Hyg. 45, 399407.
Olson, K.E., Higgs, S., Gaines, P.J., Powers, A.M., Davis, B.S.,
Kamrud, K.I., Carlson, J.O., Blair, C.D., Beaty, B.J., 1996.
Genetically engineered resistance to dengue-2 virus transmission
in mosquitoes. Science 272, 884886.
Olson, K.E., Higgs, S., Hahn, C.S., Rice, C.M., Carlson, J.O., Beaty,
B.J., 1994. The expression of chloramphenicol acetyltransferase in
Aedes albopictus (C6/36) cells and Aedes triseriatus mosquitoes
using a double subgenomic recombinant Sindbis virus. Insect
ARTICLE IN PRESS
Olson, K.E., Myles, K.M., Seabaugh, R.C., Higgs, S., Carlson, J.O.,
Beaty, B.J., 2000. Development of a Sindbis virus expression
system that efciently expresses green uorescent protein in
midguts of Aedes aegypti following per os infection. Insect Mol.
Biol. 9, 5765.
Perera, O.P., Harrell, II, R.A., Handler, A.M., 2002. Germ-line
transformation of the South American malaria vector, Anopheles
albimanus, with a piggyBac/EGFP transposon vector is routine and
highly efcient. Insect Mol. Biol. 11, 291297.
Pierro, D.J., Myles, K.M., Foy, B.D., Beaty, B.J., Olson, K.E., 2003.
Development of an orally infectious Sindbis virus transducing
system that efciently disseminates and expresses green uorescent
protein in Aedes aegypti. Insect Mol. Biol. 12, 107116.
Pinkerton, A.C., Michel, K., OBrochta, D.A., Atkinson, P.W., 2000.
Green uorescent protein as a genetic marker in transgenic Aedes
Powers, A.M., Kamrud, K.I., Olson, K.E., Higgs, S., Carlson, J.O.,
Beaty, B.J., 1996. Molecularly engineered resistance to California
serogroup virus replication in mosquito cells and mosquitoes. Proc.
Natl. Acad. Sci. USA 93, 41874191.
Pudney, M., Leake, C.J., Varma, M.G.R., 1979. Replication of
arboviruses in arthropod in vitro systems. In: Edouard Kurstak, X.
(Ed.), Arctic and Tropical Arboviruses. Academic Press, New
York, pp. 245262.
Raymes-Keller, A., Powers, A.M., Higgs, S., Olson, K.E., Kamrud,
K.I., Carlson, J.O., Beaty, B.J., 1995. Replication and expression of
a recombinant Sindbis virus in mosquitoes. Insect Mol. Biol. 4,
245251.
Seabaugh, R.C., Olson, K.E., Higgs, S., Carlson, J.O., Beaty, B.J.,
1998. Development of a chimeric Sindbis virus with enhanced per
Os infection of Aedes aegypti. Virology 243, 99112.
Shiao, S.H., Higgs, S., Adelman, Z., Christensen, B.M., Liu, S.H.,
Chen, C.C., 2001. Effect of prophenoloxidase expression knockout
on the melanization of microlariae in the mosquito Armigeres
subalbatus. Insect Mol. Biol. 10, 315321.
Strauss, J.H., Strauss, E.G., 1994. The alphaviruses: gene
expression, replication, and evolution. Microbiol. Rev. 58,
491562.
Tamang, D., Tseng, S.M., Huang, C.Y., Tsao, I.Y., Chou, S.Z., Higgs,
S., Christensen, B.M., Chen, C.C., 2004. The use of a double
subgenomic Sindbis virus expression system to study mosquito
gene function: effects of antisense nucleotide number and duration
of viral infection on gene silencing efciency. Insect Mol. Biol. 13,
595602.
Vanlandingham, D.L., Hong, C., Klingler, K., Tsetsarkin, K.,
McElroy, K.L., Powers, A.M., Lehane, M.J., Higgs, S., 2005.
Differential infectivities of onyong nyong and chikungunya virus
isolates in Anopheles gambiae and Aedes aegypti mosquitoes. Am.
J. Trop. Med. Hyg. 72, 616621.
Warren, W.D., Atkinson, P.W., OBrochta, D.A., 1994. The Hermes
transposable element from the house y, Musca domestica, is a
short inverted repeat-type element of the hobo, Ac, and Tam3
(hAT) element family. Genet. Res. 64, 8797.
Wendell, M.D., Wilson, T.G., Higgs, S., Black, W.C., 2000. Chemical
and gamma-ray mutagenesis of the white gene in Aedes aegypti.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Molecular cloning and analysis of a novel teratocyte-specic
carboxylesterase from the parasitic wasp, Dinocampus coccinellae
Ravikumar Gopalapillai, Keiko Kadono-Okuda, Takashi Okuda
National Institute of Agrobiological Sciences, Owashi 1-2, Tsukuba, Ibaraki 305 8634, Japan
Abstract
Teratocytes derived from the embryonic membrane (serosa) of parasitoids are released into the host hemocoel when the parasitoid
eggs hatch, where they perform several functions during the post-embryonic stage. A full-length cDNA encoding a putative
carboxylesterase was isolated from the teratocytes of Dinocampus coccinellae and was designated as teratocyte-specic
carboxylesterase (TSC). It contained an open reading frame of 2571 bp coding for a protein of 857 amino acids with a calculated
molecular mass of 89 kDa. The deduced amino acid sequence had many structural features that are highly conserved among serine
hydrolases including Ser, Glu and His as a catalytic triad, carboxylesterase type-B (FGGNPNSVTLLGYSAG)/ lipase-serine
(VTLLGYSAGA) active sites, and six N-glycosylation sites. Interestingly, the mRNA encoding the TSC gene was expressed
exclusively in teratocytes but not in the parasitoid larva or in the non-parasitized host. Most notably, the TSC protein was
distinguished by an insertion of 294 amino acids towards the N-terminal region and was anked by carboxylesterase domains.
Furthermore, sequence alignment and homology search revealed these additional amino acids to be unique to TSC and the insertion
contributed signicantly to its molecular mass resulting in a larger protein than other esterases. In addition to sequence analysis, the
possible role of TSC in relation to the host (Coccinella septempunctata) and parasitoid (D. coccinellae) system is discussed.
Keywords: Dinocampus coccinellae; Coccinella septempunctata; Host-parasitoid relationship; Teratocytes; Carboxylesterase
1. Introduction
Parasitoids have evolved a high degree of nutritional,
physiological, and behavioral interactions with their
hosts. For survival in their hosts, parasitoid wasp-
derived components such as venom, polydnavirus and
teratocytes play important roles by manipulating the
hosts physiology (Beckage and Gelman, 2004). Terato-
cytes are unique cells originated from the serosal
membrane of some endoparasitoid embryos which
become dissociated after hatching (Dahlman, 1990;
Lawrence, 1990; Buron and Beckage, 1997). They have
nutritive and immunosuppressive roles in some species
and can also regulate host growth and development in
others (Kitano et al., 1990; Strand and Wong, 1991;
Pennacchio et al., 1992; Dahlman and Vinson, 1993).
Although several studies on teratocytes are available in
literature, only one teratocyte-specic gene has been
characterized at molecular level (Rana et al., 2002;
Dahlman et al., 2003).
The braconid wasp, Dinocampus coccinellae parasi-
tizes several aphidophagous lady beetle species. In a host
beetle Coccinella septempunctata, teratocytes increase in
size and decrease in number during parasitism indicating
that D. coccinellae teratocytes primarily provide nutri-
tion for developing parasitoid larvae in the host
(Kadono-Okuda et al., 1995). The teratocytes synthesize
a great amount of teratocyte-specic protein of 540 kDa,
with a major subunit of 94 kDa, a hexamerin and
is shown to be nutritive in function (Okuda and
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.010

fax: +81 29 838 6110.
E-mail address: oku@affrc.go.jp (T. Okuda).
Kadono-Okuda, 1995). In addition it also exhibited a
strong esterase activity (Kadono-Okuda et al., 1998).
We now report the cloning and sequence analysis of the
major teratocyte-specic gene that showed domain
similarity to members of the carboxylesterase family.
Although tetrocyte-specic carboxylesterase (TSC) ex-
hibits homology to carboxylesterases, the presence of
the 294 amino acids insertion anked by carboxylester-
ase domains makes this protein structurally distinct
from other esterases.
2.1. Insects rearing and teratocytes sampling
Adults of C. septempunctata, which is the main host of
D. coccinellae, were collected from the vicinity of
Tsukuba and kept in the laboratory at 25 1C. After
several days, parasitoid larvae came out from the hosts
and made cocoons. Subsequently the emerged wasps
were fed on 30% sucrose solution until used for
parasitization. Unparasitized coccinellids were reared
under short day (12L: 12D) at 25 1C, which are
diapause-averting conditions (Okuda and Hodek,
1983). Coccinellids were parasitized 7 days after adult
eclosion in a Petri dish under careful observation to
avoid super-parasitism. Under such conditions, para-
sitoid larvae complete pre-pupal development within
about 19 days without undergoing diapause. Parasitized
hosts were submerged in ice-cold Ringers solution
(110 mM NaCl, 1.8 mM KCl, 1.1 mM CaCl
2
, 2.4 mM
NaHCO
3
) and dissected. The teratocytes were distrib-
uted in the host body cavity and dislodged from host
tissues by pipetting ice-cold Ringers solution over the
carcass several times. The teratocytes were then trans-
ferred to a watch dish and rinsed several times with the
Ringers solution to remove the host hemocytes.
Following this they were homogenized in extraction
buffer (0.1 M Tris-HCl (pH 8.3), 0.8 M NaCl, 2 mM
PMSF, 0.2% DOC, and 0.2% Triton X), centrifuged at
13000 rpm in 1.5 ml tubes for 10 min at 4 1C. The
supernatant was collected and used for SDS-PAGE and
the demonstration of lipase activity.
2.2. Protein purication and amino acid sequencing
Teratocytes from day 12 after parasitization were
used for protein sequencing. The 94 kDa band of the
teratocyte protein was cut out from a 2.520% gradient
SDS-PAGE gel (Fig. 1) and electroeluted. After acetone
precipitation, the protein was puried by reverse phase
HPLC (Shimadzu, Japan) with a C8 column (Tosoh,
Japan) using 0.1% TFA and acetonitrile. Puried
protein was dried up and dissolved in CNBr (in 70%
formic acid) and digestion was carried out at room
temperature over night. Resultant fragments were
isolated by reverse phase HPLC using the same
conditions as above and amino acid residues were
determined by a protein sequencer (Applied Biosystem
473 A).
2.2.1. cDNA cloning
Total RNA was extracted from the teratocytes using
Isogen (Nippon Gene, Japan) and was used as template
for rst strand cDNA synthesis employing a 5
0
end
adapter primer (SmartIIoligo, Clontech), an oligo dT
primer (Amersham) and Superscript II reverse tran-
scriptase (Invitrogen). The rst strand cDNA was then
used as template in PCR with degenerate primers
designed from the amino acid sequence of the puried
teratocyte protein. The primer set included the sense
primer, GA(C/T)AA(C/T)CA(A/G)AA(C/T)CC(I)AA
(C/T)CC(I)GT (DNQNNANPV) and the antisense
primer, (A/G)TA(A/G)TG(A/G)TA(A/G)TT(A/G)
TA(I)AC(I)GG(T/C)TG (QPVYNYHY). Three inde-
pendent amplications were carried out to avoid any
nucleotide mutations. The resultant 2.1 kb PCR product
was cloned into pGEMT-Easy vector (Promega) and
nucleotide sequence was determined by an automated
DNA sequencer (ABI 3100).
2.2.2. Rapid Amplication of cDNA Ends (RACE)
First strand cDNA was used as templates for 5
0
and 3
0
RACE PCR using adapter primers as above and the
following gene specic primers: 5
0
RACE Primer:
GTGCTTCAGCATAACGAATACCACGATACGCA
RACE, and 3RACE Primer: GCTATCACGGACTCT
ARTICLE IN PRESS
Fig. 1. Prole of teratocyte protein analyzed by SDS-PAGE on a
2.520% gradient gel. M: Protein markers (Bio-Rad) TC: crude
teratocyte extract of day 12 after parasitization. An arrow indicates a
94 kDa teratocyte-specic polypeptide.
R. Gopalapillai et al. / Insect Biochemistry and Molecular Biology 35 (2005) 11711180 1172
CTCAACTTATGCCGATGC. RACE products were
processed and sequenced as above. Direct sequencing
was also done for conrmation.
2.3. RT PCR
Total RNA was extracted from teratocytes (day 12),
parasitoid larvae (day 12) and fat body of non-
parasitized host beetles. The rst strand cDNA was
provided as template for RT-PCR using the following
primers from both ends of the open reading frame of the
cloned and sequenced cDNA: sense primer- ATGGA-
GATGAAGATTTTGTTGGGATTGTTG; antisense
primer- TTAGTAGTGATGCATATTGTTAATCGT-
CATTTGC.
2.4. Protein expression
The TSC cDNA without its signal sequence was
cloned into pIVEX2.3d plasmid (Roche) between Nco1
and Sac1 sites and in frame with the C-terminal His
6
-
tag. After sequence verication, the constructed plasmid
was used for expression in a cell free protein expression
system (RTS100/500 Escheriachia coli HY kit from
Roche). Chaperones were added to enhance the folding
of the protein according to the Roches application
manual. A negative control reaction was performed
using plasmid without the insert. Purication and
detection of the recombinant TSC were done using
His-Trap HP kit (Amersham), and AntiHis
6
antibody
(Roche) followed by chemiluminescence detection sys-
tem (WesternBreeze, Invitrogen), respectively. TNT
System (Promega) was also used for TSC expression
according to the manufactures protocol. SDS PAGE
and Western blot were done according to standard
protocols.
2.5. Lipase activity of teratocytes
Hemolymph of non-parasitized coccinellids was
collected by reex bleeding, hemocytes removed and
the lipophorin was isolated by potassium bromide (KBr)
density gradient ultracentrifugation (Shapiro et al.,
1984). High-density lipophorin (HDLp, density
1.078 g/ml), formed as a clear yellow band was collected,
desalted and used immediately. Teratocyte extract of
day 12 (protein concentration 5 mg/ml) was used for
lipase activity as described by Kawooya et al. 1991.
Three groups were used: (A) 1 ml of lipophorin plus
500 ml of extraction buffer, (2) 1 ml of lipophorin plus
500 ml teratocyte extract, (C) 1 ml of lipophorin plus
500 ml of extraction buffer containing 210 U of
Pseudomonas Sp lipase (Toyobo, Japan). The reactions
were incubated at 37 1C for 3 h, terminated by keeping
on ice and then subjected to KBr density gradient
ultracentrifugation.
2.6. Other methods
Unless otherwise indicated, all molecular biology
techniques were performed essentially as described in
Sambrook et al. (1989). Protein concentration was
measured using Bio-Rad protein assay kit. Triacylgly-
ceride (TAG) was estimated using Iatroscan TLC
Analyzer (Iatroscan, Japan). Signal sequence was
predicted by the SignalP program (http://www.cbs.
dtu.dk/services/SignalP/). NCBI (www.ncbi.nlm.nih.
gov), PSORT II (http://psort.nibb.ac.jp/), ExPASy
(http://au.expasy.org/prosite/) and PredictProtein
(http://www.embl-heidelberg.de/predictprotein/predict-
protein.html) programs were used for protein and
nucleotide analyses. Phylogenetic analysis was done by
ClustalX (Thompson et al., 1997).
3. Results
Using degenerative primers designed from the amino
acid fragments of the puried teratocyte protein and
subsequent PCR amplications including RACE, we
have obtained a full-length cDNA from the teratocytes
of D. coccinellae (Fig. 2). The 2894 bp cDNA sequence
revealed an open reading frame (ORF) of 2571 bp
encoding a protein of 857 amino acid residues. There
were two start codons at the 5
0
end positions at 122124
and 128130. Based on the Kozak sequence, the rst
ATG was considered to be the translation initiation
codon (Kozak, 1984). The stop codon was succeeded by
a polyadenylation signal (AATAAA) 138 bp down-
stream and a Poly (A)
+
tail further downstream. The
rst 18 amino acids comprised a putative signal peptide
and the next 14 amino acid residues exactly matched the
N-terminal sequence determined by Edman degradation
of the puried teratocyte protein. All other partial
sequences obtained were also almost identical and were
present in the ORF. The ORF predicted a protein with a
molecular mass of 89 kDa and was consistent with the
estimated molecular mass of 94 kDa for the fully
processed native protein (Okuda and Kadono-Okuda,
1995). The predicted isoelectric point of the protein was
6.41. Further it was characterized by the presence of six
N-glycosylation sites, a carboxylesterase type-B signa-
ture 2 motif (EDCLRLNVYT) at positions 101110,
another carboxylesterase type-B serine motif (FGGN
PNSVTLLGYSAG) at positions 479494, as well as a
lipase serine active site (VTLLGYSAGA) at positions
486495, and notably the catalytic triad, Ser492,
Glu629, His752 which may form the charge-relay
system. The sequence Gly Xaa-Ser-Xaa-Gly (Gly490-
Tyr491-Ser492-Ala493-Gly494) is conserved in all es-
terases/lipases (Fig. 3). These data clearly showed
that the inferred amino acid sequence of the cloned
gene had all the features characteristic of an active
ARTICLE IN PRESS
ARTICLE IN PRESS
ACACAGTCGCATTCTCGACGTTGTAAGAAATATACTTGCATTGTCAACTACTGCTTTCGA 60
GAATTCGTTTTTACAATTAACAAAAAAAAAAAATTATTTTTTTTGTTTTGTAAAAAAAAA 120
TATGGAGATGAAGATTTTGTTGGGATTGTTGCTATGCTGTGCTATAGCAATTGCCACAGA 180
M E M K I L L G L L L C C A I A I A T D 20
CAACCAAAATAATGCCAATCCAGTGGTACAGACGTCTGCTGGTACCATCCAAGGTTCATT 240
N Q N N A N P V V Q T S A G T I Q G S L 40
ATGGAAAACACGTTTGGGAAAAACAATTTATGCGTATCGTGGTATTCGTTATGCTGAAGC 300
W K T R L G K T I Y A Y R G I R Y A E A 60
ACCAACCGGTCAAAACCGTTTCAAACAAGCAATTCCAGTTAAACCACACAGTGGAGTTTA 360
P T G Q N R F K Q A I P V K P H S G V Y 80
TGATGCCACCCAAGATGGACCATTGTGCCCACAACCCGTTTCAAACAACAGAATAATCTC 420
D A T Q D G P L C P Q P V S N N R I I S 100
TGAAGATTGTCTCCGTTTGAATGTCTACACCACCTCATCACCACAAAGTTCATCCGGCAG 480
E D C L R L N V Y T T S S P Q S S S G S 120
TCAATCTGGTAGCCAACCTAGTAATCAACCTGGAGCTCAATCTGGCAATCAACAAGGCAA 540
Q S G S Q P S N Q P G A Q S G N Q Q G N 140
TTATTTCCATAGCCAATCCGGAAATCAACCTGGTGGTGGCCAATCTGGAAATCAACCTGG 600
Y F H S Q S G N Q P G G G Q S G N Q P G 160
TGGCCAATGGGGAAATCAACCTAGTGGCCAATGGGGAAATCAACCTGGTAGCCAAACCGG 660
G Q W G N Q P S G Q W G N Q P G S Q T G 180
AAATCAACCTGGTAGCCAATGGGGAAATCAACCTGGTAGCCAATCCGGAAATCAACCTGG 720
N Q P G S Q W G N Q P G S Q S G N Q P G 200
TGCCGGATTTGGAACTCATTCTGGAAGTCAATCCGGAAACCAGCCCGGTAACCAACCTGG 780
A G F G T H S G S Q S G N Q P G N Q P G 220
CAACCAACCCGGCAGCCCATCAACCTATGTATATGGCACATTAGCTGGTACCTCACCCAC 840
N Q P G S P S T Y V Y G T L A G T S P T 240
CGGTAACCAACCTGGCAATCAACCTGGTAGCCAATCCGGTTATCAACCTGGAAACCAATC 900
G N Q P G N Q P G S Q S G Y Q P G N Q S 260
GGGAAGTCAATCTGGCAGTCCATCCTCCCATCCATATTGGTATCATTCTGATAGCCAATC 960
G S Q S G S P S S H P Y W Y H S D S Q S 280
CGGTGGTCAATCTGGTGGTCAAACAGGTGGCCAACCTGGTGGTCAACCCGGAGGCCAACC 1020
G G Q S G G Q T G G Q P G G Q P G G Q P 300
TGGTAGCCAACCTGGTAGCCAATCCGGAAATCAAGCTGGTGGCCAATTCGGTGGCCAAAC 1080
G S Q P G S Q S G N Q A G G Q F G G Q T 320
TGGCGGTCAACCCGGTAGCCATTCTGGTAGCCAATCTGGAAATCAAGCTGGTGGCCAATT 1140
G G Q P G S H S G S Q S G N Q A G G Q F 340
CGGTGGCCAATCTGGTGGTCAACCTGGCAGTCAACCTGGTAGCCATTCTGGTAGCCAATC 1200
G G Q S G G Q P G S Q P G S H S G S Q S 360
TGGAAATCAAGCTGGTGGCCAATTCGGTGGCCAATCTGGTGGTCAACCTGGCAGTCAACC 1260
G N Q A G G Q F G G Q S G G Q P G S Q P 380
CGGTAGCCAATCCGGTTTCCAACCGATCAGTTCATCTGGCAGTCAATCCGGTAATCAATT 1320
G S Q S G F Q P I S S S G S Q S G N Q F 400
CGGAAATAAAGACGTTGTTGTCTTCCTCCATCCTGGCGCATTCTACTCCTACTCTGGAAC 1380
G N K D V V V F L H P G A F Y S Y S G T 420
TGTAACAAATCAAAACCGTTTGCCATCCGACCAAGCTGATCTTGCCAAAAAACAAGCCCA 1740
V T N Q N R L P S D Q A D L A K K Q A Q 540
AATTCTTGGATGCCCCATCGACACATATGACAATATGTTTAACTGCTTATACTCAAAATC 1800
I L G C P I D T Y D N M F N C L Y S K S 560
TTCAAATGATTTCGGACCGGAAAATTTGCTCGACCGCGATATCGTTTTAGTCACAGTTAA 1440
S N D F G P E N L L D R D I V L V T V N 440
CTACCGTTTGGGATCATTAGGATTTTTGAGTGTAGGCGATGCACGTGCACCAGGAAATGC 1500
Y R L G S L G F L S V G D A R A P G N A 460
AGGTCTCAAGGATCAAGTTCAAGCTCTTCGTTGGATCCAACAAAATATTCACAACTTTGG 1560
G L K D Q V Q A L R W I Q Q N I H N F G 480
TGGTAATCCAAATTCAGTTACATTGTTGGGCTACAGCGCTGGTGCATGGAGTGTGTCATT 1620
G N P N S V T L L G Y S A G A W S V S L 500
ACACATCGTATCACCAATGAGCAGAGGTTTATTCCACCGTGCAATTGCCATGAGTGGTGC 1680
H I V S P M S R G L F H R A I A M S G A 520
Fig. 2. Nucleotide and inferred amino acid sequences for teratocyte-specic carboxylesterase of D. coccinellae. The putative signal peptide is shown
in broken lines and the stop codon is indicated by an asterisk. A potential polyadenylation signal is in bold and the polyA tail in italics. The predicted
N-glycosylation sites are shown in grey, carboxylesterase and lipase active sites are indicated in bold in the amino acid sequence. The amino acids
determined by Edman degradation are underlined, whereas, unmatched residues are indicated below at their respective positions. The amino acids
that are unique to TSC are boxed.
carboxylesterase and hence it was termed TSC. How-
ever, TSC was distinguished by the insertion of 294
contiguous amino acids from position 111404 towards
its N-terminal and thus anked by carboxylesterase
domains. Alignment of the complete amino acid
sequence of TSC with other insect carboxylesterases
resulted in a long gap due to this insertion and a further
homology search showed no similarity of this part to
any known esterases, indicating that these inserted
amino acids are unique to TSC.
RT-PCR analysis showed that the carboxylesterase
gene was exclusively expressed in teratocytes (Fig. 4).
No transcript was detected in parasitoid larva or in the
non-parasitized host. Thus, the parasitoid larva that is
basically from the same source as teratocytes did not
seem to transcribe the TSC gene. The TSC cDNA
without its signal sequence was cloned in an expression
vector in frame with the C-terminal His
6
-tag and
expressed in a cell-free system. As shown in Fig. 5, the
expressed protein was of expected size (90.1 kDa) in
which the C-terminal His
6
-tag and other linker amino
acids contributed 1.1 kDa. This was in good agreement
with the molecular mass of the native protein after
glycosylation and other post-translational modications
as mentioned above. Although our data clearly indicate
that the target gene was cloned from the teratocytes, the
expressed protein from both cell free systems did not
yield a functional enzyme (data not shown) indicating
that it requires post-translational processing; especially
glycosylation, and this could not be induced by the
ARTICLE IN PRESS
CATCTGGACCCCAGTCGTTGAACAGAATCTATCATCAGGCAGTAGCAACAACAATGCAGA 1920
I W T P V V E Q N L S S G S S N N N A E 600
AGCATTTATTAGTGCTCAACCCGTTGACATTATCCGTTCAAAACAAGCCAATTTCGTTCC 1980
A F I S A Q P V D I I R S K Q A N F V P 620
TTTAATCACCGGCGTCAACAAAGATGAACTCGGTGGAGTTGTCATAGTCGCTGAGGAACA 2040
L I T G V N K D E L G G V V I V A E E Q 640
AGCTCAAAGTGGAAACAGTTCAATCTATGATGAATTCAACAGCAAATGGGAACAGGTTGC 2100
A Q S G N S S I Y D E F N S K W E Q V A 660
ACCAATAAGCTTTTCTTATGAACGTGACACACCAAGATCATCAAGTATCAGCCGAGACTT 2160
P I S F S Y E R D T P R S S S I S R D L 680
GAAATCATTCTACTTGCGAGATCAACCAGTTAAACAAGGAAGCTATCACGGACTCTCTCA 2220
K S F Y L R D Q P V K Q G S Y H G L S Q 700
ACTTTATGCCGATGCATTGATCATCTTCCAAGGACACCGTTTCGAAAGATTGATGGCTAA 2280
L Y A D A L I I F Q G H R F E R L M A N 720
CTATTCATCTCAACCAGTTTACAACTATCATTATGTCTATCCCGCTTGTGAAAGTTTCGC 2340
Y S S Q P V Y N Y H Y V Y P A C E S F A 740
A E
CAAATGGTCTAATGGATCTCATTTCGGTGTTGTTCATCACGATGAGTTACTTCTCCTCTT 2400
K W S N G S H F G V V H H D E L L L L F 760
CAAAATGAACAAATATCCAAATGTGTGCAACCGAGATGTCAAGACACTCGAACGTCTTAC 2460
K M N K Y P N V C N R D V K T L E R L T 780
K K
TGGAATAATTGCCAACTTTGCCAAAACTGGTGAACCAATTCCACAAAATGACGCTGTCAA 2520
G I I A N F A K T G E P I P Q N D A V N 800
CTACTCCAACGTTAAATGGCAACCGTCAACCCAAAATCATCCACAACATTTGGAAATTGG 2580
Y S N V K W Q P S T Q N H P Q H L E I G 820
CGAGGAATTGTCTATCGTTAATGGACCAGTTTATGAAAACAGAATGAATGAATGGGAAAA 2640
E E L S I V N G P V Y E N R M N E W E K 840
ATTATTCCCATTGTCGACAATGCAAATGACGATTAACAATATGCATCACTACTAAATTCA 2700
L F P L S T M Q M T I N N M H H Y * 857
K K
GATTTTCATATATATCTTTGAGATATATACATATATTATATATATTATTCATGATGGCAT 2760
TTCGTTTTCTTCCTCTTTTTTCTCTTTTTTTTTTTCTCTTCTTATTGGCTAATTTTAAAT 2820
TGTATTTACATGAATAAATATTTTGTTGATTGAATTAAAAAAAACCCGAAAAAAAAAAAA 2880
AAAAAAAAAAAAAA
AGCCGAAGATTATGCCTATTCCCTTCCCAAATTCGCTGAATTCCACGGTGATCCCGTCCT 1860
A E D Y A Y S L P K F A E F H G D P V L 580
Fig. 2. (Continued)
expression systems we used. Since the cloned TSC gene
had both esterase and lipase motifs, and the esterase
activity of this protein was already shown (Kadono-
Okuda et al., 1998), we were interested to know whether
it also possesses lipase activity. To this point, we have
used crude teratocyte extracts, as in the case of the
ARTICLE IN PRESS
Dinocampus ------MEMKILLGLLLCCAIAIATDNQNNANPVVQTSAGTIQGSLWKTRLGKTIYAYRG 54
Myzus ------MKNTCGILLNLFLFIGCFLTCSASNTPKVQVHSGEIAGGFEYTYNGRKIYSFLG 54
Drosophila ----------MNYVGLGLIIVLSCLWLGSNASDTDDPLLVQLPQGKLRGRDNGSYYSYES 50
Nilaparvata ---MAVKWEMVGLAWAALLAFASFSAADNSVPVVHDTASGDLSGKFLTLTPNRTIEAYLG 57
Lucilia MNFNVSLMEKLKWKIKCIENKFLNYRLTTNETVVAETEYGKVKGVKRLTVYDDSYYSFEG 60
Anisopteromalus -----------------------------MERPEVKTLSGQVRGLKQISVEGIGFYAFKG 31
. : . :: .
Dinocampus IRYAEAPTGQNRFKQAIPV-KPHSGVYDATQDGPLCPQPVS-----NNRIISEDCLRLNV 108
Myzus IPYASPPVQNNRFKEPQPV-QPWLGVWNATVPGSACLGIEF--GSGSKIIGQEDCLFLNV 111
Drosophila IPYAEPPTGDLRFEAPEPYKQKWSDIFDATKTPVACLQWDQFTPGANKLVGEEDCLTVSV 110
Nilaparvata IPYAQPPIGSRRFKDPEPF-GKWLGTFNGTKEPTKCLQVNGF-LPGKPVEGSEDCLYLNV 115
Lucilia IPYAQPPVGELRFKAPQRP-TPWDGVRDCCNHKDKSVQVD---FITGKVCGSEDCLYLSV 116
Anisopteromalus IPYAKPPVGELRFKDPVPI-EPWQEVREATEFGPMAAQFD---VISKFSGGSDDCLYINV 87
* **..* . **: . : . .:*** :.*
Dinocampus YT--------------VVVFLHPGAFYSYSGTSNDFGPE-NLLDRDIVLVTVNYRLGSLG 153
Myzus YTPKLPQENSAGDLMNVIVHIHGGGYYFGEG--ILYGPHYLLDNNDFVYVSINYRLGVLG 169
Drosophila YKPKNSKR----NSFPVVAHIHGGAFMFGAA--WQNGHENVMREGKFILVKISYRLGPLG 164
Nilaparvata YTPSRNG-----VGYPVMVFIHGGGFVDGDGTSGFYGPDKLLLTKDIILVTIHYRLGFLG 170
Lucilia YTNNLNPE----TKRPVLVYIHGGGFIIGENHRDMYGPD-YFIKKDVVLINIQYRLGALG 171
Anisopteromalus YTKKINSN----VKQPVMFYIHGGGFIFGSGNDFFYGPD-FLMRKDIVLVTFNYRLGVFG 142
*. *: .:* *.: * . : ..: :.. **** :*
Dinocampus FLSVGDAR--APGNAGLKDQVQALRWIQQNIHNFGGNPNSVTLLGYSAGAWSVSLHIVSP 211
Myzus FASTGDG--VLPGNNGLKDQVAALKWIQQNIVAFGGDPNSVTITGMSAGASSVHNHLISP 227
Drosophila FVSTGDRD--LPGNYGLKDQRLALKWIKQNIASFGGEPQNVLLVGHSAGGASVHLQMLRE 222
Nilaparvata FASLDDG--DFAGNYGLKDQSLALKWVKENIAKFGGDGDKVTVVGESAGAASAHFHILSP 228
Lucilia FLSLNSEDLNVPGNAGLKDQVMALRWIKNNCANFGGNPDNITVFGESAGAASTHYMMLTE 231
Anisopteromalus FLNLEHE--VAPGNQGLKDQVMALKWVRDNIANFGGDSENVTIFGESAGGASVHYLTVSP 200
* . .** ***** **:*:::* ***: :.: : * ***. *. :
Dinocampus MSRGLFHRAIAMSGAVTNQNRLPSDQA-DLAKKQAQILGCPIDTYDNMFNCLYSKSAEDY 270
Myzus MSKGLFNRAIIQSGSAFCHWSTAENVA-QKTKYIANLLGCPTNNSVEIVECLRSRPAKAI 286
Drosophila DFGQLARAAFSFSGNALDPWVIQKGARGRAFELGRNVGCESAEDSTSLKKCLKSKPASEL 282
Nilaparvata QSQGLFQRAILLSGTADCPWAVSTAHQ-NGNLTAKMASLVNCSADTSATELLECLRKVEG 287
Lucilia QTRGLFHRGILMSGNAICPWANTQCQH-RAFTLAKLAGYKGEDNDKDVLEFLMKAKPQDL 290
Anisopteromalus LAKGLFHKAISQSGVFMNPWASVSGEP-RKKAYELCELLG--KKTTDPVEIVKFLRTVDT 257
* . .: ** . . : :
Dinocampus AYSLPKFAEFHGDPVLIWTPVVEQNLSSGSSNNNAEAFISAQPVDIIRSKQANFVPLITG 330
Myzus AKSYLNFMPWRNFPFTPFGPTVEVAG--------YEKFLPDIPEKLVP----HDIPVLIS 334
Drosophila VTAVRKFLIFSYVPFAPFSPVLEPSD-------APDAIITQDPRDVIKSGKFGQVPWAVS 335
Nilaparvata SEFLIHN--EKFQTVWKGYSVPIVIFRPTIESHSGNAFITQESYKSQ-----SKKPMMIG 340
Lucilia IKLEEKVLTLEERTNKVMFPFGPTVEPYQTADCVLPKHPREMVKTAWG----NSIPTMMG 346
Anisopteromalus MKLIEHQGELQIQELQKKCLSAFVPGVDDKSPNPFMPFSREVAVEQA-----AHVPYLIG 312
: * .
Dinocampus VNKDELGGVVIVAEEQAQSGNSSIYDEFNSKWEQVAPISFSYERDTPRSSSISRDLKSFY 390
Myzus IAQDEGLIFSTFLGLENGFNELNNNWNEHLPHILDYNYTISNE---NLRFKTAQDIKEFY 391
Drosophila YVTEDGGYNAALLLKERKSGIVIDDLNERWLELAPYLLFYRDTKTKKDMDDYSRKIKQEY 395
Nilaparvata ATSDEGALVLAILKRDKTRSLESALSEFDKRFTEIMPVEGDFL-DDPDHKERAEKIKTEY 399
Lucilia NTSYEGLFFTSILKQMPMLVKELETCVNFVPSELADAER--------TAPETLEMGAKIK 398
Anisopteromalus YNDREGTLLYKIFENDDFESKNLRFEEFIHPNFAETLKR---------KKISLEDLKRMY 363
: . .
Dinocampus LRDQPVKQGSYHGLSQLYADALIIFQGHRFERLMANYSS-QPVYNYHYVYPACESFAKWS 449
Myzus FGDKPISKETKSNLSKMISDRSFGYGTSKAAQHIAAKNT-APVYFYEFGYSGNYSYVAFF 450
Drosophila IGNQRFDIESYSELQRLFTDILFKNSTQESLDLHRKYGK-SPAYAYVYDNPAEKGIAQVL 454
Nilaparvata FGNSTISNETLPQLTKLYSDTYFLNGIKSTLSRHEGEKY-VYKFGYEGSYSISQLLSGDP 458
Lucilia KAHVTGETPTADNFMDLCSHIYFWFPMHRLLQLRFNHTSGTPVYLYRFDFDSEDLINPYR 458
Anisopteromalus FKNKKISKETTGKFIDLFSDMYFIQGIHQVARVQAERNS-APTYMYQFTYDQGPNFSKGM 422
. . : : : :. : : *
Dinocampus NGSHFG----VVHHDELLLLFKMNKYPNVCNRDVKTLER-------LTGIIANFAKTGEP 498
Myzus DPKSYSRGSSPTHGDETNYVLKVDGFTVYDNEEDRKMIK------TMVNIWATFIKSGVP 504
Drosophila ANRTDYD-FGTVHGDDYFLIFENFVRDVEMRPDEQIISR------NFINMLADFASSDNG 507
Nilaparvata TYRN-----GVCHADDLFYLFPMKPFLGLRVGSETEKD--KEISAKFVDLITNFVIEGNP 511
Lucilia IMRSGRGVKGVSHADELTYFFWNQLAKRMPKESREYKT-----IERMTGIWIQFATTGNP 513
Anisopteromalus FSIDE---PGSTHMDELIYLFSMKFQETLNMEPIDKKSPHFRVMEQMVELWTNFAKYGRP 479
* *: .: : : * .
Dinocampus IPQNDAVNYSNVKWQPSTQNHPQHLEIGEELSIVNGPVYENRMNEWEKLFPLSTMQMTIN 558
Myzus DTENSEIWLP---VSKNPADLFRFTKITQQQTFEAREQSTMAIMNFGVAYHYQNILNLMC 561
Drosophila SLKYGECDFKDSVGSEKFQLLAIYIDAARIGSMWNFRKLHE------------------- 548
Nilaparvata N---SKSEP--SIWTPSSK-DVDFLSISTEGNFEMKKNFPGA------------------ 547
Lucilia YSNEIEGMEN-VSWDPIKKSDEVYKCLNISDELKMIDVPEMDKIKQWESMFEKHRDLF-- 570
Anisopteromalus IPAPTELLP--VHWLPMND-GTVLRYLNIGEELRMEKVLNIEERYDYKLICHREKV---- 532
:
Dinocampus NMHH 562
Myzus QMT- 564
Fig. 3. An alignment using ClustalX program of the inferred amino acid sequence of TSC of D. coccinellae (AY498693) with ve insect
carboxylesterases: M. persicae (P35502); D. melanogaster (A28022); N. lugens (AAG40239); L. cuprina (AAB67728); and A. calandrae (AF064523).
GenBank accession numbers are given in parentheses. Functionally important residues (catalytic triad) are indicated by arrows. Identical amino acid
residues are shown by asterisks. The : and indicate strong and weak group of amino acids that are conserved respectively. Amino acids from
position 111 to 404 were removed from TSC sequence to avoid a long gap in the alignment (see text for details).
esterase assay, together with the lipophorin (Lp) of the
host beetle.
As in many insects the lipophorin (HDLp) of the host
coccenellids is yellow in color due to carotenoids and its
density was 1.078 g/ml. This HDLp formed a clear
yellow band upon KBr density gradient centrifugation.
The density of the Lp increased to 1.14 g/ml when it was
incubated with teratocytes and analyzed after density
gradient centrifugation. This indicates the conversion of
HDLp to very high-density lipophorin (VHDLp)
resulting in the band shift (Fig. 6). Increased density
of Lp may be caused by the hydrolysis of lipids by the
teratocytes. Lp incubated with bacterial lipase also
showed a similar result and the disappearance of the
yellow band was noticed when increased quantities of
lipase or teratocyte extracts were used (data not shown).
We have observed a marked decrease in triacylglycer-
ide (TAG) in the fat body of parasitized host compared
to the TAG in the fat body of non-parasitized host
(Fig. 7B). Further, TAG in the teratocytes showed a
gradual increase from day 7 and reached a maximum on
day 12 after parasitization. After this, the TAG content
either reached a plateau or slightly decreased (Fig. 7A),
with decrease in number of the teratocytes (Kadono-
Okuda et al., 1995).
4. Discussion
Carboxylesterases are a group of serine esterases that
catalyze the hydrolysis of a wide variety of ester
and amide containing endogenous and xenobiotic
compounds (Heyman, 1980). Some of them hydrolyze
palmitoyl CoA, acyl-carnitine, and mono-and diacylgly-
cerols and are thought to be involved in lipid
metabolism (Satoh, 1987) while others are able to
hydrolyze a broad range of substrates (Oakeshott
et al., 1999). Furthermore, carboxylesterases are
glycoproteins characterized by high mannose and
N-glycosylation may play an important role in their
catalytic activity (Kroetz et al., 1993). Previous work
shows that teratocytes of D. coccinellae synthesize a
teratocyte-specic polypeptide (Okuda and Kadono-
Okuda, 1995), a high mannose containing glycoprotein,
produced as a hexamer, and the molecular mass of its
major subunit is 94 kDa (Kadono-Okuda et al., 1998).
This is the most prominent protein in teratocytes
observed when analyzed on a SDS-PAGE with Coo-
massie blue staining indicating its abundance (Okuda
and Kadono-Okuda, 1995). Moreover, it exhibits a
strong esterase activity but is devoid of juvenile
hormone esterase activity (Kadono-Okuda et al.,
ARTICLE IN PRESS
Fig. 5. Western blot analysis of TSC expressed in a cell- free protein
expression system on a 7.5% SDS-PAGE gel. The TSC cDNA was
cloned into pIVEX2.3d plasmid and expressed using RTS100 E. coli
kit (Roche). TSC, TSC protein of molecular mass 90.1 kDa containing
His
6
-tag at C-terminal. C, negative control (plasmid without insert).
Five microlitre of each reaction product was loaded. Protein markers
(Bio-Rad) on the left.
Fig. 4. RT-PCR analysis demonstrating TSC gene expression. RNA
was isolated from teratocytes (TC), parasitoid larva (L) and fat bodies
of non-parasitized host (NPH) reverse transcribed, amplied by PCR
with TSC-specic primers and the amplication products were
analyzed on a 0.8% agarose gel. Molecular size markers (M)
(Invitrogen) are indicated at left.
1998). The deduced amino acid sequence of the
teratocyte cDNA has the most characteristics of an
active carboxylesterase enzyme. These include a cataly-
tic triad, a conserved esterase/lipase motif, two carbox-
ylesterase type-B (involved in lipid metabolism) motifs,
a lipase-serine active site, and six N-glycosylation sites.
Several lines of evidence indicate that we have cloned the
major gene from the teratocytes of D. coccinellae. First,
the cDNA sequence obtained would encode peptides
similar to that of fragments obtained from Edman
degradation of the puried teratocyte protein. Second,
regions of the inferred amino acid sequence of TSC have
esterase motifs and third, the predicted molecular mass
of the protein (89 kDa) is in good agreement with that of
its native protein (94 kDa). This difference in molecular
mass is mainly due to glycosylation of the native protein
and it is further supported by the presence of six
N-glycosylation sites in the TSC sequence. The presence
of a signal peptide and the absence of a consensus
C-terminal endoplasmic reticulum retention signal,
HXEL may indicate that TSC protein is secretory.
Interestingly, TSC has a clear HDEL tetrapeptide
(753756), albeit 101 amino acids away from the
C-terminal. Additionally, a secretory signal TEHT
(Medda and Proia, 1992) found at the C-terminal of
some carboxylesterases is absent in TSC. The PSORT II
program predicts that TSC can be both cytoplasmic and
extracellular in equal probability. Retention of rat liver
carboxylesterase in the endoplasmic reticulum is still
possible with a signal peptide and the absence of a
retention sequence (Takagi et al., 1988). Taken together
these data may thus agree with our previous report
(Kadono-Okuda et al., 1998) that teratocyte protein largely
accumulates in teratocyte cells with a little secretion.
While most insect carboxylesterases are implicated in
insecticide resistance (Field et al., 1988; Karunaratne et
al., 1993; Newcomb et al., 1997), many vertebrate
carboxylesterases are reported to be involved in the
hydrolysis of several lipids (Satoh, 1987). Our data do
not seem to support a detoxication or xenobiotic
degradation by TSC. The host beetles have a defensive
alkaloid Coccinellin which is bitter in taste and indeed
toxic (Tursch et al., 1971). However, the alkaloid does
not contain ester or amide bonds to be digested by
carboxylesterase.
ARTICLE IN PRESS
Fig. 7. Triacylglyceride content (in mg per animal) in teratocytes and
parasitoid larvae (A), fat bodies of parasitized and non-parasitized
host (B) after different days of parasitization. (A) Open circle: total
teratocytes in a host beetle, closed circle: a parasitoid larva (B) Open
circle: fat bodies from a non-parasitized host beetle, closed circle: fat
bodies from a parasitized host beetle, Values are mean7SE of four
animals.
Fig. 6. Hydrolysis of lipids by teratocyte extract. HDLp incubated in
the absence (Control) and in the presence of teratocyte extract (TC).
Subsequent density gradient ultracentrifugation showing Lp band shift
in the right tube (TC) due to the formation of VHDLp from HDLp.
As mentioned above, TSC is marked by the presence
of a lipase serine active site and two carboxylesterase
type-B motifs that are all thought to be involved in lipid
metabolism. TSC may be involved in the break down of
host lipids following lipid accumulation in teratocytes
(Fig. 7) as part of the utilization of fat as an energy
source for the growing parasitoid larva. This is
consistent with a nutritional role of D. coccinellae
teratocytes (Okuda and Kadono-Okuda, 1995). Terato-
cytes serving as a nutrient source for developing
parasitoid larvae are reported in several species such
as Aphidius ervi (Falabella et al., 2000), Microplitis
mediator (Quin et al., 2000), Microctonus aethiopoides
(Barratt and Sutherland, 2001) and Cotesia kariyai
(Nakamatsu et al., 2002) parasitoid systems.
Although the functions of teratocytes are not
completely well understood, there is considerable
published evidence to support their role in parasitoid
development (reviewed by Beckage and Gelman, 2004).
Microplitis croceipes teratocytes produce a 13.9 kDa
protein (monomer, TSP14) that inhibits host protein
synthesis that is linked to larval growth and develop-
ment (Zhang and Dahlman, 1989; Dahlman, 1990).
TSP14 appears to be the only teratocyte-specic gene
cloned and characterized. It encodes a protein of 129
amino acids (including a signal sequence of 22 amino
acids) and carries a cysteine-rich motif similar to that
described from Campoletis sonorensis polydnavirus
(Dahlman et al., 2003). Like native protein, the
recombinant TSP14 inhibited host protein synthesis
selectively in a dose dependent manner (Rana et al.,
2002). However, there is no signicant sequence
similarity between TSP14 and TSC of the present study
suggesting the existence of host-specic differences in
the primary structure of major teratocyte-specic genes.
Likewise, the biological role of teratocytes in host-
parasitoid relationships may differ from one system to
another (Dahlman, 1990).
It is interesting to note the distinct sequence difference
of TSC and other carboxylesterases. Although TSC
shares homologies with the members of the carboxyles-
terases and exhibits domain structures that are typical
for this group, it signicantly differs from them by a
unique insertion of 294 amino acids. A NCBI conserved
domain search (rpsblast) reveals a clear demarcation in
the TSC protein domain structure. While the amino acid
positions 27110 and positions 405829 display domain
similarity (with E value ranging from 0.008 to 4e-77) to
the members of the carboxylesterase family, the inter-
vening amino acids (111404) do not show any putative
domain. In addition, an alignment of the complete
amino acids of TSC with other insect carboxylesterases
shows that the inserted 294 amino acids leaves a long
gap in the alignment, indicating that this non-aligned
sequence is unique to TSC and that they are anked by
carboxylesterase domains. The physiological signi-
cance of this insertion is uncertain. Interestingly this
region is enriched with glutamine and glycine. BLAST
search (protein-protein) of these unique amino acids
alone shows no known protein domains but reveals
similarity to immunoglobulin binding proteins. Since
insects have no antigenic immunity, this non-aligned
part of the sequence may function in host immune
suppression through other means or be involved in other
function(s) which warrants further investigation. While
the size range of most esterases is 6070 kDa (Cygler et
al., 1993), TSC is 94 kDa and this relatively large
molecular size invites speculation that in addition to a
possible ester bond hydrolysis, TSC may also perform
other function(s). Although the precise physiological
functions of TSC remain unclear, our data on the
cloning and sequence analysis clearly indicate the
presence of a unique carboxylesterase in the teratocytes
of a parasitic wasp.
Acknowledgements
G.R. is thankful to Japan Science and Technology
Agency for a postdoctoral fellowship. We thank Dr.
Kazuo Masaki for his help in the peptide sequencing,
Drs. Takahiro Kikawada, Takahiro Shiotsuki, Yuichi
Nakahara, and Yusuke Kato for their kind help in
various ways. We also thank Dr. Barbara Barratt for
her critical reading of the manuscript.
References
Barratt, B.I.P., Sutherland, M., 2001. Development of teratocytes
associated with Microctonus aethiopoides Loan (Hymenoptera:
Braconidae) in natural and novel host species. J. Insect Physiol. 47,
257262.
Beckage, N.E., Gelman, D.B., 2004. Wasp parasitoid disruption of
host development: implications for new biologically based strate-
gies for insect control. Ann. Rev. Entomol. 49, 299330.
Buron, I.de., Beckage, N.E., 1997. Developmental changes in
teratocytes of the braconid wasp Cotesia congregata in larvae of
the tobacco hornworm, Manduca sexta. J. Insect Physiol. 43,
915930.
Cygler, M., Schrag, J.D., Sussman, J.L., Harel, M., Silman, I., Gentry,
M.K., Doctor, B.P., 1993. Relationship between sequence con-
servation and three-dimensional structure in a large family of
esterases, lipases, and related proteins. Protein Sci 2, 366382.
Dahlman, D.L., 1990. Evaluation of teratocyte functions: an overview.
Arch. Insect. Biochem. Physiol. 13, 159166.
Dahlman, D.L., Vinson, S.B., 1993. Teratocytes: developmental and
biochemical characteristics. In: Beckage, N.E., Thompson, S.N.,
Federici, B.A. (Eds.), Parasites and Pathogens of Insects, vol. 1.
Academic Press, New York, pp. 145165.
Dahlman, D.L., Rana, R.L., Schepers, E.J., Schepers, T., Diluna,
F.A., Webb, B.A., 2003. A teratocyte gene from a parasitic wasp
that is associated with inhibition of insect growth and development
inhibits host protein synthesis. Insect Biochem. Mol. Biol. 12,
527534.
ARTICLE IN PRESS
Falabella, P., Tremblay, E., Pennacchio, F., 2000. Host regulation by
the aphid parasitoid Aphidius ervi: the role of teratocytes. Entomol.
Exp. Appl. 97, 19.
Field, L.M., Devonshire, A.L., Forde, B.G., 1988. Molecular evidence
that insecticide resistance in peach-potato aphid (Myzus persicae)
results from amplication of an esterase gene. Biochem. J. 251,
309312.
Heyman, E., 1980. In: Jakoby, W.B. (Ed.), Enzymatic Basis of
Detoxication, Vol. II. Academic Press, New York, pp. 291323.
Kadono-Okuda, K., Sakurai, H., Takeda, S., Okuda, T., 1995.
Synchronous growth of a parasitoid, Perilitus coccinellae and the
teratocyte with the development of the host, Coccinella septem-
punctata. Entomol. Exp. Appl. 75, 145149.
Kadono-Okuda, K., Weyda, F., Okuda, T., 1998. Dinocampus (
Perilitus) coccinellae teratocyte polypeptide: accumulative prop-
erty, localization and characterization. J. Insect Physiol. 44,
10731080.
Karunaratne, S.H.P.P., Jayawardena, K.G.I., Hemingway, J., Ketter-
man, A.J., 1993. Characterization of a B-type esterase involved in
insecticide resistance from the mosquito Culex quinquefasciatus.
Biochem. J. 294, 575579.
Kawooya, J.K., van der Host, D.J., van Heusden, M.C., Brigot,
B.L.J., van Antwerpen, R., Law, J.H., 1991. Lipophorin structure
analyzed by invitro treatment with lipases. J. Lipid. Res. 32,
17811788.
Kitano, H., Wago, H., Arakawa, T., 1990. Possible role of teratocytes
of the gregarious parasitoid, Cotesia ( Apanteles) glomerata in
the suppression of phenoloxidase activity in the larval host, Pieris
rapae crucivora. Arch. Insect Biochem. Physiol. 13, 177185.
Kozak, M., 1984. Compilation and analysis of sequences upstream
from the translational start site in eukaryotic mRNAs. Nucleic
Acids Res 12, 857872.
Kroetz, D.L., McBridge, O.W., Gonzalez, F.J., 1993. Glycosylation-
dependent activity of baculovirus-expressed human liver carbox-
ylesterase: cDNA cloning and characterization of two highly
similar enzyme forms. Biochemistry 32, 1160611617.
Lawrence, P.O., 1990. Serosal cells of Biosteres longicaudatus
(Hymenoptera: Braconidae): ultrastructure and release of polypep-
tides. Arch. Insect Biochem. Physiol. 13, 199216.
Medda, S., Proia, R.L., 1992. The carboxylesterase family exhibits C-
terminal sequence diversity reecting the presence or absence of
endoplasmic-reticulum-retention sequences. Eur. J. Biochem. 206,
801806.
Nakamatsu, Y., Fujii, S., Tanaka, T., 2002. Larva of an endopar-
asitoid, Cotesia kariyai (Hymenoptera: Braconidae), feed on the
host fat body directly in the second stadium with the help of
teratocytes. J. Insect Physiol. 48, 10411052.
Newcomb, R.D., Campbell, P.M., Ollis, D.L., Cheah, E., Russel, R.J.,
Oakeshot, J.G., 1997. A single amino acid substitution converts a
carboxylesterase to an organophosphorous hydrolase and confers
insecticide resistance on a blowy. Proc. Natl. Acad. Sci. USA 94,
74647468.
Oakeshott, J.G., Claudianos, C., Russel, R.J., Robin, G.C., 1999.
Carboxyl/cholinesterases: A case study of the evolution of a
successful multigene family. Bioessays 21, 10311042.
Okuda, T., Hodek, I., 1983. Response to constant photoperiods in
Coccinella septempunctata bruckii population from central Japan
(Coleoptera: Coccinellidae). Acta Ent. Bohemoslov. 80, 7475.
Okuda, T., Kadono-Okuda, K., 1995. Perilitus coccinellae teratocyte
polypeptide: Evidence for production of a teratocyte-specic
540 kDa protein. J. Insect Physiol. 41, 819825.
Pennacchio, F., Vinson, S.B., Tremblay, E., 1992. Host regulation
effects of Heliothis virescens (F) larvae induced by teratocytes of
Cardiochiles nigriceps Viereck (Lepidoptera: Noctuida-Hymenop-
tera: Braconidae). Arch. Insect Biochem. Physiol. 19, 177192.
Quin, Q., Gong, H., Ding, T., 2000. Two collagenases are secreted by
teratocytes from Microplitis mediator (Hymenoptera: Braconidae)
cultured in vitro. J. Invert. Pathol. 76, 7980.
Rana, R.L., Dahlman, D.L., Webb, B.A., 2002. Expression and
characterization of a novel teratocyte protein of the braconid,
Microplitis croceipes (Cresson). Insect Biochem. Mol. Biol. 32,
15071516.
Sambrook, J., Fritisch, E.F., Maniatis, T., 1989. Molecular Cloning: A
Laboratory Manual, second ed. Cold Spring Harbor Lab. Press,
New York.
Satoh, T., 1987. Reviews in Biochemical Toxicology. In: Hodgson, E.,
Bend, J.R., Philpot, R.M. (Eds.), Reviews in Biochemical
Toxicology, Vol. 8. Elsevier, New York, pp. 155181.
Shapiro, J.P., Keim, P.S., Law, J.H., 1984. Structural studies on
lipophorin, an insect lipophorin. J. Biol. Chem. 259, 36803685.
Strand, M.R., Wong, E.A., 1991. The growth and role of Microplitis
demolitor teratocytes in parasitism of Pseudoplusia includens.
J. Insect Physiol. 37, 503515.
Takagi, Y., Morohashi, K., Kawabata, S., Go, M., Omura, T., 1988.
Molecular cloning and nucleotide sequence of cDNA of micro-
somal carboxylesterase E1 of rat liver. J. Biochem. 104, 801806.
D.G., 1997. The ClustalX windows interface: exible strategies for
multiple sequence alignment aided by quality analysis tools.
Nucleic Acids Res 24, 48764882.
Tursch, B., Doloze, D., Dupont, M., Hootele, C., Kaisin, M., Pasteels,
J.M., Zimmermann, D., 1971. Coccinellin, the defensive alkaloid of
the beetle Coccinella septempunctata. Chimia 25, 307.
Zhang, D., Dahlman, D.L., 1989. Microplitis croceipes teratocytes
cause developmental arrest of Heliothis virescens larvae. Arch.
Insect. Biochem. Physiol. 12, 5161.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
The extensible alloscutal cuticle of the tick, Ixodes ricinus
Svend Olav Andersen
a,
, Peter Roepstorff
b
a
August Krogh Institute, University of Copenhagen, Universitetsparken 13, DK-2100 Copenhagen O, Denmark
b
Department of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark
Received 28 October 2004; received in revised form 19 May 2005; accepted 25 May 2005
Abstract
The proteins in the distensible alloscutal cuticle of the blood-feeding tick, Ixodes ricinus, have been characterized by
electrophoresis and chromatography, two of the proteins were puried and their total amino acid sequence determined. They show
sequence similarity to cuticular proteins from the spider, Araneus diadematus, and the horseshoe crab, Limulus polyphemus, and to a
lesser extent to insect cuticular proteins. They contain a conserved sequence region, which is closely related to the chitin-binding
RebersRiddiford consensus sequence present in many insect cuticular proteins.
Only a fraction of the alloscutal proteins can be readily dissolved, and the dissolved proteins are difcult to separate by
electrophoresis and column chromatography. The insoluble fraction can only be dissolved after degradation to smaller peptides. The
mixture of extractable proteins as well as hydrolysates of the insoluble fraction are uorescent when exposed to ultraviolet light, and
the uorescence corresponds in excitation and emission maxima to the uorescence of the rubber-like arthropodan protein, resilin,
and to the amino acid dityrosine. Small amounts of dityrosine were obtained from ticks in the early phase of a blood meal when the
cuticle weighs less than 4 mg; increasing amounts were obtained from animals in the initial period of feeding, during which the
cuticular weight increases from 4 to 11 mg, whereas little increase in dityrosine content was observed during the nal period of
engorgement. Cuticle from fully distended ticks contains about 6080 nmole dityrosine per tick, corresponding to 23 mg/mg cuticle.
It is suggested that the major part of the cuticular proteins is made inextractable by cross-linking by dityrosine residues, and that
dityrosine plays a role in stabilizing the cuticular structure during the extensive distension occurring during a blood meal.
Small amounts of 3-monochlorotyrosine and 3,5-dichlorotyrosine were obtained from the distended tick cuticle, corresponding to
chlorination of between 0.5% and 1.5% of the tyrosine residues. It is suggested that the chlorotyrosines are a side-product of
oxidative processes in the cuticle.
Keywords: Cuticle; Ticks; Amino acid sequence; Dityrosine; Chlorotyrosine
1. Introduction
The adult female sheep tick, Ixodes ricinus, sucks
blood from its host for a period of 810 days. During
the feeding period the tick ingests a large volume of
blood; its body weight increases from about 2 mg to
more than 250 mg, and the alloscutal cuticle is distended
to allow for the necessary volume increase (Lees, 1952).
The epicuticular layer of the alloscutal cuticle is deeply
folded in unfed ticks, and during feeding the folds are
attened and the apparent surface area of the alloscu-
tum is increased to about 15 times the initial value,
although the area of the epicuticle is not increased (Lees,
1952; Hackman and Filshie, 1982). The underlying
procuticle is not sclerotized, it has been classied as
endocuticle, and it is extensively stretched during
feeding. The endocuticle of unfed female ticks is thin;
it increases in thickness during the feeding period due to
addition of more material and at the same time a layer
of inner endocuticle is deposited, which has a more open
structure than the initial layer of endocuticle. After
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.05.009
Corresponding author. Tel.: +45 35321703; fax: +45 35321567.

E-mail address: soandersen@aki.ku.dk (S.O. Andersen).
about 6 days of feeding, the tick has increased both the
weight of its cuticle and its body weight by a factor of
about ten, and the cuticular surface has increased its
area by a factor of three to four. During the following
day the tick ends its meal by a brief but heavy
engorgement of blood, resulting in a body surface area
which is increased about 15 times, and a body weight
which is about 100200 times larger than in the unfed
tick. The cuticle as such does not increase in weight
during the last part of the meal; it is stretched and
becomes thinner. At the end of the blood meal the tick
releases itself from the host and drops to the ground in
preparation of egg laying (Lees, 1952).
The mechanism of the distension of the cuticle is still
an unsolved problem. Lees (1952) argued that at least
during the rst part of the feeding process the stretching
of the cuticle cannot be due to passive distension caused
by pressure of the ingested blood, since the body
remains attened until the last day of engorgement.
Dillinger and Kesel (2002) describe the stretchable tick
cuticle as being both elastic and plastic, allowing it to
expand due to increased internal pressure. Based on the
autouorescence of the material, Dillinger and Kesel
(2002) suggested that the rubber-like protein, resilin, is
present both in the alloscutal cuticle and in various
cuticular arthrodial membranes of the tick, its elasticity
allowing for stretching of the cuticle and attening of
the epicuticular folds during cuticular distension. Resilin
is an elastic protein present in specialized cuticular
regions in various arthropods; it is characterized by
nearly perfect rubber-like elasticity and a pronounced
blue uorescence due to the cross-linking amino acids,
dityrosine and trityrosine (reviewed by Andersen and
Weis-Fogh, 1964; Andersen, 1971, 2003).
To obtain a better understanding of what happens in
the alloscutal cuticle during the distension, it is
necessary to know more about the individual cuticular
components and how they interact with each other. For
that purpose we have initiated a study of the proteins in
the alloscutal cuticle of I. ricinus and the presence of the
cross-linking amino acid, dityrosine, which is respon-
sible for the uorescence of resilin.
During the summer months, I. ricinus was collected
from dogs which were allowed to play in local forests.
Ticks at various stages of the feeding process were
obtained by removing them from the dog after various
feeding periods and by collecting them from the oor
after they had dropped from the host when fully
engorged. Before being used for analysis, the ticks were
in most cases stored in 96% ethanol, which caused the
ingested blood to coagulate, but some ticks were frozen
and stored at 20 1C without being exposed to organic
solvents. The surface of the ticks was dried in tissue
paper and the ticks were weighed before the cuticle was
isolated. The cuticles were obtained by bisecting the
ticks longitudinally, removing the coagulated blood
from both halves, washing the remains in 1% potassium
tetraborate, and cleaning the cuticles by removal of the
remnants of epidermal cells and muscles by means of
ne forceps, then their purity was checked by micro-
scopy. The cleaned cuticles were washed briey in
distilled water and ethanol, air dried over night, and
weighed on a Sartorius microbalance, model MC 5.
Soluble cuticular proteins were obtained by extracting
the cuticles by either 1 M acetic acid or 6 M urea (one ml
per tick cuticle) over night at ambient temperature,
whereby about 25% of the cuticular dry weight was
dissolved, and the dissolved material was fractionated
by gel ltration on a column of Sephacryl S200HR
(1.0 30 cm, Amersham Biosciences, Uppsala, Sweden).
Elution was performed with 0.1% triuoroacetic acid
(TFA) at a rate of 0.5 ml /min, the eluate was collected in
fractions of 0.5 ml, and the absorbancy was recorded at
280 nm. Afterwards, the fractions were diluted to 2.5 ml
with 0.1 M sodium bicarbonate, and the uorescence
was measured on a Kontron Spectrouorometer SFM23
using an excitation wavelength of 320 nm and emission
wavelength of 405 nm.
The proteins were puried by means of reverse phase
high-performance liquid chromatography (RP-HPLC)
of samples from the protein peak obtained in the gel-
ltration separation. The column (Source 5RPC ST 4.6/
150, Pharmacia Biotech, Uppsala, Sweden) was equili-
brated with 10% acetonitrile, 0.1% TFA in ultrahigh-
quality water (A-buffer). The B-buffer was 90%
acetonitrile, 0.1% TFA in ultrahigh-quality water.
During the rst 2 min of elution, the concentration of
B-buffer was increased linearly from 0% to 30%, and
during the following 40 min, it was increased linearly to
60%. The major protein-containing fractions were
collected manually and lyophilized.
To obtain peptides of a reasonable size for sequencing
studies, the puried proteins were digested with one or
more of the following proteolytic enzymes: trypsin and
endoproteinase Glu-C from Promega (Madison, WI), and
chymotrypsin and endoproteinase Asp-N from Boehringer
(Mannheim, Germany). The digestions were performed as
described previously (Andersen, 1998). The resulting
peptides were separated by RP-HPLC on a Vydac C
4
-
column as described previously (Andersen, 2000).
Amino acid analyses and MALDI-MS of puried
proteins were performed at the Institute of Molecular
Biology and Biochemistry, University of Southern
Denmark, as described previously (Jensen et al., 1998).
Plasma desorption mass spectrometry (PDMS) of
peptides was performed on a BioIon 20 time-of-ight
mass spectrometer (BioIon, Uppsala, Sweden) as
described previously (Andersen, 1998).
ARTICLE IN PRESS
S.O. Andersen, P. Roepstorff / Insect Biochemistry and Molecular Biology 35 (2005) 11811188 1182
The intact proteins and selected peptides were
sequenced by Edman degradation using an Applied
Biosystems 476A Protein Sequencer. Degradation,
conversion and identication of the liberated phe-
nylthiohydantoin amino acids were performed as
described by the supplier.
Acid hydrolysis of cuticle and quantitative determina-
tion of dityrosine were performed as described in
Andersen (2004b). Contents of chlorotyrosines were
measured by Dr. R. Senthilmohan, Department of
Anatomy, Christchurch School of Medicine, New
Zealand, using stable isotope dilution mass spectro-
metry according to Buss et al. (2003).
3. Results
When alloscutal cuticle from fully engorged ticks were
extracted in either 6 M urea or 1 M acetic acid, most of
the cuticular protein remained in the residue and about
25% of the cuticular material was dissolved. No
signicant differences were observed between the amino
acid compositions of intact alloscutal cuticle and the
protein mixture extracted from the cuticle (Table 1).
Two-dimensional gel electrophoresis of the extracts
showed a number of distinct, relatively low-molecular
weight protein spots in the acidic range and a
pronounced smear in the extreme basic range of the
gel (Fig. 1). According to staining intensity, the smear
contains signicantly more protein than is present in the
spots. No signicant differences were observed between
patterns of proteins from extracts in urea or acetic acid.
Gel ltration of the extracts gave only a poor
separation of the dissolved proteins, which were eluted
closely together in a single peak with a tailing shoulder
(Fig. 2). All fractions which according to ultraviolet
absorbtion spectra contained protein gave a blue
uorescence when exposed to ultraviolet light. The
uorescence was more intense at pH 8 than at pH 4, and
the excitation and emission maxima corresponded to
those of dityrosine. A small blue uorescent peak was
eluted late from the gel-ltration column; it has not been
fully characterized, but its elution position and uor-
escent properties suggest it represents small dityrosine-
containing peptides.
Samples from the main part of the protein-containing
peak could not be resolved into distinct proteins by
ARTICLE IN PRESS
Table 1
Amino acid compositions of intact alloscutal cuticle from fully engorged female ticks, proteins extracted with 6 M urea, and the two sequenced
proteins
Amino acid Intact cuticle Cuticular extract Ir-ACP10.9 Ir-ACP16.8
n 4 n 4
Asx 2.5670.06 3.1270.12 11.76 9.94
Thr 8.0370.12 7.9070.11 9.80 9.94
Ser 5.7970.10 6.1170.23 11.76 10.56
Glx 3.1570.15 3.6170.33 10.78 7.45
Gly 15.5370.86 16.3271.20 5.88 4.97
Ala 21.2970.76 20.6470.46 13.73 14.91
Cys 0 0 0 0
Val 10.4470.21 10.3470.29 4.90 8.07
Met 0 0 0 0
Ile 0.7970.12 0.9770.11 3.92 4.35
Leu 4.8870.09 5.0870.22 0.98 2.48
Tyr 11.3370.12 9.2470.37 7.84 5.59
Phe 1.7370.10 2.0370.02 2.94 1.24
His 4.4470.19 4.5470.26 1.96 4.35
Lys 0.8170.02 0.9070.10 4.90 3.11
Arg 2.3170.01 2.4470.13 1.96 2.48
Pro 6.9470.22 6.7670.31 6.86 10.56
Fig. 1. Two-dimensional electrophoretogram of proteins extracted
from the alloscutal cuticle of an engorged female I. ricinus. About
0.3 mg protein was applied. The positions of molecular weight markers
are shown. Arrows indicate the positions of the two sequenced
proteins.
RP-HPLC (Fig. 3A), only a very broad peak was
obtained, which apparently contained several poorly
separated proteins. RP-HPLC of samples from the
shoulder of the main peak gave a few narrow peaks
followed by a broad peak (Fig. 3B). None of the narrow
peaks showed visible uorescence, while the broad peaks
showed weak blue uorescence, resembling that of
dityrosine. Several attempts were made to purify
components from the broad peaks by means of gel
electrophoresis, ion-exchange resins or by electrofocus-
ing, but in vain.
Two of the narrow peaks in Fig. 3B were selected for
more detailed characterization. Their relative masses
were determined to 10903.9 and 16812.2, respectively,
by MALDI mass spectrometry, and Edman degradation
of the intact proteins revealed that they are both N-
terminally blocked. The proteins were degraded by
trypsin digestion, and the sequences of all peptides
except two, which appeared to be N-terminally blocked,
were obtained by combined Edman degradation and
PDMS mass spectrometry. The blocked peptides were
deblocked by digestion with pyroglutamic aminopepti-
dase, indicating that the blocking group is a pyrogluta-
mine residue, and they were then completely sequenced
by Edman degradation. The order of the tryptic peptides
in the intact proteins was determined from the masses
and sequences of peptides obtained by digestion of the
proteins with chymotrypsin, endopeptidases Glu-C and
endoproteinase Asp-N, as described previously for
proteins puried from insect cuticles (Andersen, 1998).
The complete sequences of the two proteins are shown
in Fig. 4. The calculated relative masses of the proteins
are 10903.6 and 16812.3, respectively, in good agree-
ment with the masses obtained by mass spectrometry.
The two proteins were called Ir-ACP10.9 and Ir-
ACP16.8. The isoelectric points of the two proteins
were calculated to 4.2 and 5.3, respectively. The protein
sequence data has been deposited in the Swiss-Prot and
TrEMBL knowledgebase under the accession numbers
P84251 for Ir-ACP10.9 and P84252 for Ir-ACP16.8.
Based upon the autouorescence of the alloscutal
cuticle, Dillinger and Kesel (2002) suggested that resilin
is present in the cuticle, and resilin is characterized by
the presence of the cross-linking amino acids, di and
trityrosine. For quantitative determination of dityrosine
in cuticle from ticks at various stages of feeding, their
cuticles were hydrolyzed in HCl in two steps (Andersen,
2004a): rst hydrolysis in 1 M hydrochloric acid for 3 h
to degrade the proteins to small soluble peptides without
degrading the cuticular chitin, followed by further
hydrolysis of the supernatant in 6 M hydrochloric acid
for 20 h to degrade the peptides to amino acids. The
insoluble residue left after hydrolysis in 1 M hydro-
chloric acid accounted for 8.3% of the cuticular dry
weight; it was not analyzed further, presumably it
consisted of chitin plus epicuticular material. RP-HPLC
analysis of the complete hydrolysate showed that up to
10 nmol dityrosine per mg cuticle could be obtained
ARTICLE IN PRESS
Fig. 2. Gel ltration on Sephacryl S200HR of urea extract of
alloscutal cuticle from a single fully engorged female I. ricinus.
Absorbance at 280 nm was recorded (), and uorescence at
405 nm of the individual fractions were measured (K).
Fig. 3. (A) RP-HPLC on Source 5RPC of aliquot from fraction 16 at
8.0 ml (Fig. 2). (B) RP-HPLC on Source 5RPC of aliquot from
fraction 21 at 10.5 ml (Fig. 2). Ir-ACP10.9 eluted at 19 ml and Ir-
ACP16.8 eluted at 20 ml, as indicated by crosses.
Fig. 4. Complete amino acid sequences for I. ricinus proteins Ir-
ACP10.9 and Ir-ACP16.8. N-terminal pyroglutamine is indicated as
pQ.
from the cuticle of fully fed female ticks, whereas only
trace amounts of trityrosine were detected. Comparison
between unextracted and urea-extracted cuticles showed
that they have similar dityrosine contents, indicating
that it is only a minor fraction of the cuticular dityrosine
that is readily extractable.
Fig. 5 relates the weight of cuticle to the total weight
of the ticks. During the early phase of the blood meal,
the weight of the animals increase from a few mg to
about 100 mg, and the weight of the cuticle increases
from about 0.3 mg to about 7 mg. Little, if any, increase
in cuticular weight is observed during the period of
engorgement, during which the ticks reach body weights
of about 300 mg, indicating that deposition of cuticle
does not occur during engorgement in agreement with
the observations of Lees (1952).
Fig. 6 shows the increase in cuticular dityrosine
content during the period of deposition of endocuticle.
Tick cuticles weighing less than 4 mg contained about
25 nmol dityrosine, and the amounts of dityrosine
increased gradually from about 5 to 70 nmol per cuticle
while the cuticular weight increased from 4 to 11 mg.
Cuticles of nearly the same weight could vary signi-
cantly in dityrosine content.
Comparing the content of dityrosine to the weight of
the more or less distended ticks (Fig. 7) shows that the
cuticular dityrosine content increased gradually from a
level of a few nmol in unfed ticks weighing a few
milligram to about 70 nmol per animal in ticks weighing
about 300 mg; the scattering of the measured values was
pronounced.
RP-HPLC of acid hydrolysates of tick cuticles
revealed the presence of a compound eluting in the
same position as authentic 3-monochlorotyrosine. The
compound may be formed in vivo in the cuticle or it may
be formed as an artefact by chlorination of tyrosine
residues during hydrolysis (Hunt, 1984). Dr. R. Senthil-
mohan, Christchurch School of Medicine, New Zeal-
and, kindly analyzed three samples of tick cuticle by
stable isotope dilution mass spectrometry according to
Buss et al. (2003), and the results showed that
chlorotyrosines are present in the intact cuticle. The
measured values ranged from 5.4 to 11.5 mmol mono-
chlorotyrosine per mole tyrosine, from 0.3 to 3.9 mmol
dichlorotyrosine per mole tyrosine, and from 1.4 to
1.6 mmol monobromotyrosine per mole tyrosine. Arti-
factually formed derivatives were in all cases less than
10% of the real values.
Cuticles from ticks, which had not been exposed to
alcohol or to other agents which could inuence the
cuticular properties, were cleaned and immersed in
saline at pH 7, and an indication of their deformability
and elasticity was obtained by using ne forceps for
stretching and bending them while they were being
observed under microscope. The cuticles were easily
bent, and they returned spontaneously to their original
shape when the force was released, but they resisted any
signicant stretching. After being transferred to saline at
pH values between 5 and 6, the samples swelled, became
much softer, and tended to break when exposed to
relatively weak stretching forces. Thus the mechanical
properties of the cuticles appear to be pH-dependent,
but the material did not show long-range elasticity like
that observed in resilin.
ARTICLE IN PRESS
Fig. 5. Relationship between cuticular weight and body weight of
individual ticks.
Fig. 6. Relationship between cuticular content of dityrosine and
cuticular weight of individual ticks.
Fig. 7. Relationship between cuticular content of dityrosine and body
weight of individual ticks.
4. Discussion
4.1. Cuticular proteins
About 25% of the total protein content can be
extracted from tick cuticle without degradation, indicat-
ing that the main part of the proteins have been
somehow cross-linked or stabilized. The amino acid
composition of the solubilized fraction is similar to the
composition of the intact cuticle; the contents of alanine
and glycine are high, also valine and tyrosine are
abundant, whereas none of the sulphur-containing
amino acids, cysteine and methionine, were observed.
The two cuticular proteins which have been sequenced
are minor components in the mixture of extractable
proteins, and their amino acid compositions show little
similarity to the composition of the intact cuticle,
indicating that they are not representative for the bulk
of cuticular proteins. The two proteins are similar in
several respects: like the intact cuticle and most insect
cuticular proteins they are devoid of sulphur-containing
amino acids and tryptophan, they are both N-terminally
blocked by a pyroglutamine residue, and they contain a
sequence closely related to the conserved RebersRiddi-
ford consensus sequence (Rebers and Riddiford, 1988),
which is present in many insect cuticular proteins
(Andersen et al., 1995) and which has been demon-
strated to bind to chitin (Rebers and Willis, 2001). Fig. 8
shows an alignment of the small tick protein (Ir-
ACP10.9) and the N-terminal half of the larger protein
(Ir-ACP16.8) to regions of endocuticular proteins from
the spider, Araneus diadematus, the horseshoe crab,
Limulus polyphemus, the migratory locust, Locusta
migratoria, and the silk moth, Bombyx mori. The
RebersRiddiford consensus sequence is indicated in
the alignment. Many other cuticular proteins show
sequence similarities to the tick proteins according to
BLAST searches of Entrez and Uniprot protein
databases, but the proteins included in the alignment
are those being most similar to the tick proteins. Ticks
as well as spiders and horseshoe crabs belong to the
subphylum Chelicerata, and it is not surprising that the
two tick proteins are more closely related to cuticular
proteins from spiders and horseshoe crabs than to
cuticular proteins from insects.
The small tick protein (Ir-ACP10.9) consists almost
exclusively of an extended RebersRiddiford sequence,
and Ir-ACP16.8 consists of a similar sequence plus a
longer C-terminal extension, which shows little similar-
ity to other cuticular sequences. As the extended
RebersRiddiford sequence has been shown to have
afnity for chitin (Rebers and Willis, 2001), it appears
likely that the two tick proteins bind to cuticular chitin,
although tick cuticle is poor in chitin. The alloscutal
cuticle of Ixodes contains about 8% chitin, and many of
the cuticular proteins are probably not bound directly to
chitin and may not contain a RebersRiddiford
sequence. They are probably intertacting with the
proteins which are directly bound to chitin, and the
linkages may be stable covalent linkages, such as
dityrosine, or weak secondary linkages, which can easily
be broken and reformed. The mechanical properties of
cuticle will depend upon both the deformability of the
individual protein chains and the strength and number
of interactions between protein chain segments. Accord-
ingly, formation of covalent cross-links between protein
chains will inuence the mechanical properties of the
cuticular material, but the properties of the individual
proteins will also be of importance.
4.2. Dityrosine analyses
According to the measurements of Lees (1952), the
feeding period for female ticks can be divided into two
phases: an initial period of about 6 days, dominated by
deposition of endocuticle and a moderate increase in
body weight, followed by the second phase, lasting for
about a day, and characterized by a drastic increase in
body weight due to heavy engorgement with blood.
During this second phase, the epicuticular foldings are
ARTICLE IN PRESS
Fig. 8. Alignments of sequence regions from Ir-ACP10.9 and Ir-ACP16.8 to Ad-ACP15.7 from the spider A. diadematus (Norup et al., 1996), Lp-
CP14a from the horseshoe crab L. polyphemus (Ditzel et al., 2003), Lm-NCP19.8 from the migratory locust L. migratoria (Andersen, 2000), and Bm-
WCP6 from wing discs of the silk moth B. mori (Takeda et al., 2001). The RebersRiddiford consensus sequence (Rebers and Riddiford, 1988) is
shown below the cuticular sequences. Amino acid residues identical to the corresponding positions in both Ir-ACP10.9 and Ir-ACP16.8 are indicated
in bold and by double underlining.
straightened out and the endocuticle is stretched, but
does not increase in mass.
In contrast to the results of Lees (1952), the present
determinations of cuticular dityrosine content cannot be
directly related to the duration of the feeding period, but
only to the weight of animals xed in alcohol and to the
cuticular weight. The weight difference between live
animals and alcohol-xed animals appears to be
negligible, as the maximal values for tick body weight
recorded by Lees (1952) were slightly above 300 mg, and
the maximal body weight observed for alcohol-xed
ticks were 325 mg, indicating that the live ticks have
removed most of the engorged water, in agreement with
the reported ability of ticks to excrete excess water and
to inject water back into the host during feeding
(Sonenshine, 1991). Good agreement between the
maximum cuticular weights (about 12 mg) reported by
Lees and those obtained in the present work (about
11 mg) indicates that a comparison between the two
investigations is reliable. In the present work, the growth
of cuticle was found to occur only during the early part
of the feeding period, when the ticks increased their
body weight from a few mg to about 5060 mg, and no
signicant cuticular growth was observed during the
later period where body weight increased from 60 to
about 300 mg. These observations are in agreement with
the results of Lees (1952) that the increase in cuticular
weight stops shortly before the start of the period of
heavy engorgement.
Little dityrosine was obtained from cuticle of ticks in
the early phase of deposition of endocuticle; only after
the ticks had reached a cuticular weight of 4 mg was a
signicant formation of dityrosine observed, and for-
mation of dityrosine continued at a nearly constant rate
as long as deposition of endocuticle occurred. Large
variations in cuticular content of dityrosine were
observed for ticks in the engorgement period, when
the deposition of endocuticle had stopped and the
weight of the ticks increased from 60 to about 300 mg,
making it difcult to decide whether a slight increase in
dityrosine content occurred during this period. The
cuticular dityrosine content for fully distended ticks
varied from 10 to about 100 nmole per tick, indicating
that it is not the content of dityrosine which determines
the extent to which the cuticle can be stretched.
Comparing these results to the time course for tick
feeding determined by Lees (1952) shows that little
dityrosine formation occurs during the early period of
endocuticle deposition (before day 5 after attachment to
the host), and that the most intense formation of
cuticular dityrosine occurs during the late period of
endocuticular deposition (days 57), just before the
heavy engorgement begins.
The highest values for dityrosine content obtained
(about 10 nmole/mg tick cuticle) is about a factor ten
lower than the dityrosine content reported for locust
resilin (Andersen, 1966, 2004a), and in contrast to resilin
tick cuticle contains only trace amounts of trityrosine.
Compared to resilin, the tick alloscutal cuticle is a lightly
cross-linked material.
Dityrosine is formed by coupling of two free radicals
of tyrosine, and such radicals can be formed by
oxidation of tyrosine residues by means of hydrogen
peroxide and the enzyme peroxidase. A few peroxidases,
such as the mammalian myeloperoxidase, can oxidize
chloride ions to hypochlorite ions, which readily react
with tyrosine residues to form chlorotyrosines (Kettle,
1996; Winterbourn and Kettle, 2000). The presence of
both dityrosine and mono and dichlorotyrosine in tick
cuticle suggests that an enzyme related to myeloperox-
idase could be present, a suggestion which deserves to be
investigated in detail. Signicant amounts of mono and
dichlorotyrosine have been reported from the cuticle of
various insects (Andersen, 2004b), and it is uncertain
whether these amino acids have any function in the
cuticle, or whether they are side-products from some
oxidative processes.
The elasticity of resilin is due to the tendency of the
peptide chains to occur in a nearly random conforma-
tion, which can easily and reversibly be deformed by
external forces, and inter-chain cross-links prevent the
peptide chains from sliding past each other when the
material is stretched. Without cross-links resilin would
be a visco-elastic material, which would not have
complete recoil, but show lasting deformations. The
mechanical behaviour of tick alloscutal cuticle during
the feeding phase has to my best knowledge never been
properly determined, so we do not know with certainty
whether it is rubber like or whether it is more like a
plastic material which ows under tension. Manual
manipulation of pieces of female alloscutal cuticle from
animals at various stages of feeding indicated that the
cuticle does not have long-range elasticity, and that its
stiffness can be changed by small changes in surround-
ing pH, similar to what has been described for the cuticle
of the blood-sucking reduviid bug, Rhodnius prolixus
(Reynolds, 1974, 1975). The role of dityrosine in tick
alloscutal cuticle may therefore be to link the cuticular
proteins chains together into an insoluble network, the
mechanical properties of which can be modulated by
small changes in pH and ionic composition. But precise
measurements of the mechanical properties of tick
cuticle are needed before any reliable conclusions can
be drawn.
Acknowledgements
Our thanks are due to Karen Dissing for running two-
dimensional electrophoresis of cuticular extracts, to
Lene Skou and Kate Rafn for mass spectrometric
measurements, and to Peter Hjrup for amino acid
ARTICLE IN PRESS
analyses. Dr. R. Senthilmohan, Department of Anat-
omy, Christchurch School of Medicine, New Zealand, is
thanked for using stable isotope dilution mass spectro-
metry to analyze for chlorotyrosines. Ticks were kindly
provided by Peter Roepstorffs dogs, Honey and Lulu.
Economic support from the Carlsberg Foundation
and the Novo Nordisk Foundation is gratefully
acknowledged.
References
Andersen, S.O., 1966. Covalent crosslinks in a structural protein,
resilin. Acta Physiol. Scand. 66 (Suppl. 263), 181.
Andersen, S.O., 1971. Resilin. In: Florkin, C.M., Stotz, E.H. (Eds.),
Comprehensive Biochemistry, vol. 26. Elsevier, Amsterdam,
pp. 633657.
Andersen, S.O., 1998. Amino acid sequence studies on endocuticular
proteins from the desert locust, Schistocerca gregaria. Insect
Andersen, S.O., 2000. Studies on proteins in post-ecdysial nymphal
cuticle of locust, Locusta migratoria, and cockroach, Blaberus
craniifer. Insect Biochem. Mol. Biol. 30, 569577.
Andersen, S.O., 2003. Structure and function of resilin. In: Shewry,
P.R., Tatham, A.S., Bailey, A.J. (Eds.), Elastomeric Proteins.
Cambridge University Press, UK, pp. 259278.
Andersen, S.O., 2004a. Regional differences in degree of resilin cross-
linking in the desert locust, Schistocerca gregaria. Insect Biochem.
Mol. Biol. 34, 459466.
Andersen, S.O., 2004b. Chlorinated tyrosine derivatives in insect
cuticle. Insect Biochem. Mol. Biol. 34, 10791084.
Andersen, S.O., Weis-Fogh, T., 1964. Resilin. A rubberlike protein in
arthropod cuticle. Adv. Insect Physiol. 2, 165.
Andersen, S.O., Hjrup, P., Roepstorff, P., 1995. Insect cuticular
proteins. Insect Biochem. Mol. Biol. 25, 153176.
Buss, I.H., Senthilmohan, R., Darlow, B.A., Mogridge, N., Kettle,
A.J., Winterbourn, C.C., 2003. 3-Chlorotyrosine as a marker of
protein damage by myeloperoxidase in tracheal aspirates from
preterm infants: association with adverse respiratory outcome.
Pediatr. Res. 53, 455462.
Dillinger, S.C.G., Kesel, A.B., 2002. Changes in the structure of the
cuticle of Ixodes ricinus L. 1758 (Acari, Ixodidae) during feeding.
Arthropod Struct. Dev. 31, 95101.
Ditzel, N., Andersen, S.O., Hjrup, P., 2003. Cuticular proteins from
the horseshoe crab, Limulus polyphemus. Comp. Biochem. Physiol.
B 134, 489497.
Hackman, R.H., Filshie, B.K., 1982. The tick cuticle. In: Obenchain,
F.D., Galun, R. (Eds.), Physiology of Ticks. Pergamon Press, New
York, pp. 142.
Hunt, S., 1984. Halogenated tyrosine derivatives in invertebrate
scleroproteins: isolation and identication. Methods Enzymol.
107, 413438.
Jensen, C., Andersen, S.O., Roepstorff, P., 1998. Primary structure of
two major cuticular proteins from the migratory locust, Locusta
migratoria, and their identication in polyacrylamide gels by mass
spectrometry. Biochim. Biophys. Acta 1429, 151162.
Kettle, A.J., 1996. Neutrophils convert tyrosine residues in albumin to
chlorotyrosine. FEBS Lett. 379, 103106.
Lees, A.D., 1952. The role of cuticle growth in the feeding process of
ticks. Proc. Zool. Soc. Lond. 121, 759772.
Norup, T., Berg, T., Stenholm, H., Andersen, S.O., Hjrup, P., 1996.
Purication and characterization of ve cuticular proteins from the
spider Araneus diadematus. Insect Biochem. Mol. Biol. 26, 907915.
Rebers, J.E., Riddiford, L.M., 1988. Structure and expression of a
Manduca sexta larval cuticle gene homologous to Drosophila
cuticle genes. J. Mol. Biol. 203, 411423.
Rebers, J.E., Willis, J.H., 2001. A conserved domain in arthropod
cuticular proteins binds chitin. Insect Biochem. Mol. Biol. 31,
10831093.
Reynolds, S.E., 1974. A post-ecdysial plasticization of the abdominal
cuticle in Rhodnius. J. Insect Physiol. 20, 19571962.
Reynolds, S.E., 1975. The mechanism of plasticization of the
abdominal cuticle of Rhodnius larvae. J. Exp. Biol. 62, 8198.
Sonenshine, D.E., 1991. Biology of Ticks. Oxford University Press,
Oxford.
Takeda, M., Mita, K., Quan, G.-X., Shimada, T., Okano, K., Kanke,
E., Kawasaki, H., 2001. Mass isolation of cuticle protein cDNAs
from wing discs of Bombyx mori and their characterizations. Insect
Winterbourn, C.C., Kettle, A.J., 2000. Biomarkers of myeloperox-
idase-derived hypochlorous acid. Free Radical Biol. Med. 29,
403409.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Uptake and turn-over of glucosinolates sequestered in the
sawy Athalia rosae
Caroline Mu ller
a,
, Ute Wittstock
b
a
Julius-von-Sachs-Institut fur Biowissenschaften, Universitat Wurzburg, Julius-von-Sachs-Platz 3, D-97082 Wurzburg, Germany
b
Max Planck Institute for Chemical Ecology, Hans-Knoll-Str. 8, D-07745 Jena, Germany
Received 5 April 2005; received in revised form 1 June 2005; accepted 3 June 2005
Abstract
Larvae of the sawy Athalia rosae sequester glucosinolates from their various host plants of the Brassicaceae into their
hemolymph for defensive purposes. We found that the glucosinolate concentration in the insect varies in a uctuating manner
during larval development. Analyses of larvae which had been offered diets with different glucosinolate proles showed that there is
an equilibrium between a rapid uptake of glucosinolates into the hemolymph and a continuous turn-over. Injection of
glucotropaeolin into the hemolymph and ingestion of the same amount resulted in similar levels of intact glucosinolates recovered
from larvae after different periods of time. This indicates that hemolymph glucosinolates are the principal source for glucosinolate
degradation. Feeding experiments with [
14
C]-labeled glucotropaeolin revealed that the majority of the ingested glucosinolate is
excreted as one or more unidentied metabolite(s) within 14 h. We found no indication for the presence of an insect myrosinase, or
sulfatase in A. rosae, which have been shown to be involved in glucosinolate metabolism in other specialists feeding on Brassicaceae.
Furthermore, the metabolism of sinalbin in A. rosae seems to result in different products than its metabolism in the caterpillar Pieris
rapae. Obviously, A. rosae has yet another way of coping with the glucosinolates.
Keywords: Tenthredinidae; Glucosinolate; Sequestration; Hemolymph; Metabolism; Myrosinase; Sulfatase
1. Introduction
Several insects have been shown to sequester bio-
active compounds from the plant in their body-tissue
(e.g. Duffey, 1980; Nishida and Fukami, 1990; Pasteels
et al., 2003). The underlying physiological mechanisms
have been well studied especially for insects sequestering
pyrrolizidin-alkaloids from host species of the Aster-
aceae or phenol glucosides such as salicin from
Salicaceae (Ehmke et al., 1999; Bru ckmann et al.,
2000; Hartmann et al., 2001; Pasteels et al., 2003;
Narberhaus et al., 2003). However, little is known about
sequestration of glucosinolates in specialists on Brassi-
caceae. Members of the Brassicaceae are well character-
ized by the presence of a binary defense mechanism, the
glucosinolate-myrosinase system or the so-called mus-
tard oil bomb (Matile, 1980). Glucosinolates (Fig. 1),
amino acid derived b-thioglucosides, occur throughout
all plant tissues and are nontoxic (Wittstock and
Halkier, 2002), but can have deterrent or feeding
stimulatory properties on generalists or specialists,
respectively (Blau et al., 1978; Renwick, 2002). Myr-
osinases, thioglucosidases capable of metabolizing
glucosinolates, are localized in myrosin cells scattered
in plant tissue (Koroleva et al., 2000; Andre asson et al.,
2001). Upon tissue disruption, myrosinases come into
contact with their substrate and degrade the glucosino-
lates to different mainly volatile hydrolysis products,
such as isothiocyanates (Fig. 2a), nitriles, and others
(Fahey et al., 2001). Several types of hydrolysis-products
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.06.001
Corresponding author. Tel.: +49 931 888 62 21;

fax: +49 931 888 62 35.
E-mail address: cmueller@botanik.uni-wuerzburg.de (C. Mu ller).
have been shown to be repellent or toxic, respectively, to
insects, nematodes, fungi and bacteria (Tierens et al.,
2001; Agrawal and Kurashige, 2003; Zasada and Ferris,
2003; Wittstock et al., 2003). In contrast, they can
attract specialist herbivores (Chew, 1988; Renwick,
2002) as well as their parasitoids (Bradburne and
Mithen, 2000) to the host plants.
Recently, mechanisms have been elucidated by which
insects specialized on Brassicaceae can detoxify or at least
tolerate the glucosinolate-myrosinase system (Ratzka et al.,
2002; Wittstock et al., 2004). Larvae of the diamondback
moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae),
possess a glucosinolate sulfatase. This enzyme converts
glucosinolates into desulfoglucosinolates (Fig. 2c) which are
not degradable by myrosinases and are excreted with the
feces (Ratzka et al., 2002). In larvae of the small cabbage
white, Pieris rapae (L.) (Lepidoptera: Pieridae), glucosino-
lates are converted to the corresponding nitriles due to the
presence of a gut nitrile-specier protein, which promotes
the formation of nitriles instead of isothiocyanates upon
myrosinase-catalyzed glucosinolate hydrolysis (Fig. 2b).
The nitriles seem to be less toxic and are excreted with
the feces (Wittstock et al., 2004). After uptake of sinalbin
(4-hydroxybenzylglucosinolate), an additional sulfatation
leads to the formation of 4-hydroxybenzylcyanide sulfate in
P. rapae (Wittstock et al., 2004). Insects that sequester
intact glucosinolates exemplify a special case, such as the
aphids Brevicoryne brassicae (L.) and Myzus persicae
(Sulzer) (Homoptera: Aphididae) (Weber et al., 1986), the
sawy Athalia rosae L. (Hymenoptera: Tenthredinidae)
(Mu ller et al., 2001), and the harlequin bug, Murcantia
histrionica Hahn (Heteroptera: Pentatomidae) (Aliabadi
et al., 2002). The aphid B. brassicae contains its own
myrosinase that is, like in the plant, compartmentalized
from the glucosinolate storage in the insect body (Pontop-
pidan et al., 2001; Bridges et al., 2002; Jones et al., 2002).
This study focuses on the characterization of gluco-
sinolate sequestration by the sawy A. rosae. The larvae
can feed on various species of Brassicaceae and thereby
sequester different glucosinolates, for example, sinalbin
(Fig. 1b) from Sinapis alba, sinigrin (2-propenylgluco-
sinolate) from Brassica nigra, and glucobarbarin [(S)-2-
hydroxy-2-phenylethylglucosinolate] (Fig. 1c) from Bar-
barea stricta into their hemolymph (Mu ller et al., 2001).
In the present work, glucosinolate contents were
quantied in sawies after shift to food plants with
unique glucosinolate proles and after injection of a
glucosinolate into the hemolymph, and the storage
capacity for glucosinolates and their metabolites was
investigated using a radiolabeled glucosinolate. Newly
ingested glucosinolates rapidly replaced the previous,
indicating constant sequestration but also catabolism of
sequestered glucosinolates. Glucosinolate metabolite(s)
were not stored in the body but excreted. With further
biochemical assays it was investigated whether A. rosae
might use any of the so far known metabolic pathways
for dealing with the glucosinolate-myrosinase system.
2.1. Insects and plants
A stock of A. rosae was established from a eld
collection in Germany and kept in culture at 20 1C, a
relative humidity of 70% and 168 h l-d. Insects were
reared on 45 weeks old, pre-owering plants of either
S. alba L. (cultivar: SALVO, seeds from Advanta Seeds
B.V., The Netherlands), B. stricta Andrz. (seeds
obtained from wild plants), or Brassica pekinensis
(Lour.) Rupr. (var. Hong-Kong F1, seeds obtained
commercially).
2.2. Sampling for glucosinolate and metabolite detection
in A. rosae
To investigate where glucosinolates and possible
metabolites (desulfoglucosinolates, 4-hydroxybenzylcya-
nide sulfate) are localized in A. rosae, selected samples
ARTICLE IN PRESS
X
X
X
N
H
X
HO
HO H
Glucose
S R
N
O-SO
3
-
X =
(a) (c) (d) (b)
Fig. 1. Structures of glucosinolate backbone (x) with various side
chains (R): (a) glucotropaeolin, (b) sinalbin, (c) glucobarbarin and (d)
glucobrassicin.
(a)
(b)
(c)
Fig. 2. Routes of degradation for glucosinolates. (a) Activity of plant
myrosinases commonly results in the formation of isothiocyanates; (b)
in caterpillars of Pieris rapae, the corresponding nitriles are formed
instead of isothiocyanate due to a larval nitrile-specier protein; (c)
sulfatase activity causes the formation of desulfoglucosinolates in the
caterpillars of Plutella xylostella.
C. Muller, U. Wittstock / Insect Biochemistry and Molecular Biology 35 (2005) 11891198 1190
were collected from fth instar larvae fed on S. alba.
Hemolymph was taken from larvae as described else-
where (Mu ller et al., 2001) and the bled body kept as
separate sample. As the hemolymph could not be
collected exhaustively from a larva, the bled body still
contained some hemolymph as well. From additional
larvae, lled guts were dissected in 50 mM Tris-HCl
buffer, pH 7.5 and washed 3 times in new buffer to
remove any hemolymph contamination. Furthermore,
feces were collected and all samples frozen at 80 1C for
further analysis.
2.3. Tracking of glucosinolate levels in developing larvae
To investigate glucosinolate levels in larvae during
their development, adult females (parent generation)
were allowed to oviposit for 24 h on S. alba. To ensure
identical sex of offspring, unmated females were used,
producing haploid eggs, which result in male larvae.
Once hatched, 24 larvae were taken every day, frozen
at 80 1C and analyzed for glucosinolates (see below).
The glucosinolate recovered from a larva was used as a
measure for the amount of glucosinolate in the
hemolymph as the majority is located here (see below
and Mu ller et al., 2001). Additionally, shed exuviae of
larvae (2 20 exuviae pooled) as well as adults of the
F1-generation were taken for analysis. The experiment
was carried out three times independently.
2.4. Short-term metabolism of glucosinolates in larvae
To investigate how fast glucosinolates are sequestered
in the hemolymph, fth instar larvae reared on S. alba
were transferred to leaf discs (diam. 14 mm) of known
weight of B. stricta, both plants containing different
glucosinolate proles. Leaf discs were placed on
moistened lter paper in Petri dishes (diam. 9 cm). After
30, 60, 90 min, and 3, 6, 16 or 20 h, larvae were removed
and frozen at 80 1C. From half of the larvae, at least
15 ml hemolymph were collected separately before
freezing. For calculation of total glucosinolate amounts
per larva, amounts of hemolymph and bled body were
added again. Remaining leaves were weighted to
calculate the amount of tissue consumed by a larva.
The remaining leaf discs of the 3 h feeding experiment
were frozen to measure their average glucosinolate
concentration. The experiment was carried out at four
different times, using B. stricta plants of different age.
2.5. Long-term metabolism of glucosinolates in larvae
To investigate, how long glucosinolates are main-
tained in the hemolymph, larvae were transferred for
different numbers of days to a new host and analyzed
for their old and new glucosinolate contents.
Groups of larvae were reared on either S. alba or
B. stricta. When larvae reached the fourth instar, they
were transferred to a plant of the other species. In this
developmental stage, they also start moving to other
plants in the eld (personal observation). After 1, 2, 3, 4,
and 5 days, individual larvae were removed from the
plant, frozen and taken for glucosinolate analysis (4
replicates for each switch and time point). From half of
the larvae, 15 ml hemolymph were collected separately
before freezing.
2.6. Feeding versus injection of glucotropaeolin
To follow the turn-over of dened amounts of a
glucosinolate taken up orally or injected into larvae, the
following assays were performed. Larvae were reared on
B. pekinensis, which contains only low amounts of
glucosinolates and lacks glucotropaeolin (benzylgluco-
sinolate, Fig. 1a). Fifth instar larvae were used for the
experiments. One group of larvae was offered a leaf disc
(diameter 1 cm) of B. pekinensis treated with 50 or
250 nmol glucotropaeolin (from www.glucosinolates.-
com, Frederiksberg, Denmark; dissolved in 50 mM Tris-
HCl buffer, pH 7.5). These larvae were injected with 2 or
5 ml of 50 mM Tris-HCl buffer into the hemolymph. The
pH was chosen according to the pH of the hemolymph
(measured with indicator paper). Another group of
larvae received B. pekinensis leaves treated with buffer
only, but were injected with 50 or 250 nmol glucotro-
paeolin dissolved in buffer (in a 2 or 5 ml volume).
Larvae do survive a puncture into their integument
without harm, as easy bleeding is part of their defense
strategy (Mu ller and Brakeeld, 2003). Larvae were
taken at different time points after injection of
glucosinolate or after nishing the glucosinolate-treated
food (5 h after the start of the experiment) and frozen
for analysis of their glucosinolate content. In addition,
leaves treated with glucotropaeolin were taken after 5 h
and analyzed to evaluate their glucosinolate concentra-
tions in order to detect any losses due to myrosinase or
other metabolic activities. Recovery rates in larvae were
correlated to this amount.
2.7. Feeding experiment with [
14
C]glucotropaeolin
To investigate the fate of a glucosinolate after
ingestion, a further approach was taken. [
14
C]Gluco-
tropaeolin (200 Bq/nmol, 33 Bq/ml), [U-
14
C]-labeled in
the aglycon moiety, was prepared as described (Chen
and Halkier, 2000) by using transgenic Arabidopsis
thaliana CaMV35S::CYP79A2 (Wittstock and Halkier,
2000). Individual fth instar larvae of A. rosae were
offered leaf discs (diameter 1 cm) of B. pekinensis treated
with 10 ml [
14
C]glucotropaeolin. After complete con-
sumption of the leaf discs and some feeding afterwards
on untreated B. pekinensis (in total after 14 h) feces were
collected and larvae were dissected. The head was cut
ARTICLE IN PRESS
and the gut separated from the remaining body. Feces
and body parts from 48 larvae were pooled for one
sample (two independent experiments with two replicate
groups each). All samples were extracted in a mixture of
Tris-HCl buffer (50 mM, pH 7.5) and acetone (9:1 v:v)
and analyzed by liquid scintillation counting on a Tri-
carb 2300TR Liquid Scintillation Analyzer (Packard)
using Lumasafe Liquid Scintillation Cocktail (Packard).
2.8. Preparation of protein extracts and test for enzyme
activities in insect tissue
Protein extracts of larvae were prepared to test for
potential enzyme activities for glucosinolate degradation.
The following samples were collected and pooled from
four fth-instar larvae fed on B. pekinensis: hemolymph,
gut tissue and the remaining body (dissected in 50mM
Tris-HCl buffer, pH 7.5). Gut content was removed and
gut and body tissues were washed three times with buffer
to clean them from any contamination with ingested
plant tissue. All samples were homogenized in Tris-HCl
buffer, pH 7.5. The 20,000g supernatants were used as
crude extracts. The protein content was determined with
Bio-Rad Protein Reagent using BSA as standard, and
concentration of all samples adjusted to 2 mg/ml. Protein
extracts were incubated with substrate for 1.5 h at room
temperature, and formation of metabolites was measured
as described below (in three independent experiments
with 3 replicates each).
To test for sulfatase activity, 25 ml protein extracts
(50mg protein) were incubated with 50ml 5mM glucotro-
paeolin or 5mM sinalbin (from www.glucosinolates.com,
Frederiksberg, Denmark). As control, additional pro-
tein extracts were incubated with the glucosinolates and
externally added sulfatase (50 mg). The reactions were
stopped by adding 80% MeOH, and analysis for
desulfoglucosinolates was carried out as described
below.
To test for myrosinase-activity in insect tissue, forma-
tion of hydrolysis-products was measured. Assays were
done in 50mM Tris-HCl buffer, pH 7.5, containing
1 mM glucotropaeolin in a total volume of 500 ml. The
reaction was started by addition of either 50ml protein
and 25ml water extract (test samples) or 50ml protein
extract and 25ml myrosinase (control samples, 1 mg/ml,
278 units/g); Sigma, thioglucosidase from S. alba, EC
3.2.3.1). After incubation, 50ml phenylcyanide-solution
(Merck, 100 ng/ml in methanol) were added as internal
standard, and the reaction was stopped by adding 500 ml
dichloromethane. Samples were shaken vigorously and
solvents allowed to separate. The water phase was
extracted twice with an equal volume of dichloro-
methane, and the pooled organic phases dried over a
Na
2
SO
4
-column. After concentration to 150 ml in a
nitrogen stream, samples were analyzed by GC-MS and
GC-FID using an Agilent 6890 series gas chromatograph
with an HP5MS column (30m0.25mm0.25mm
lm), splitless injection at 200 1C and a temperature
program of 351C for 3 min, a 121C/min ramp to 961C
and a 181C/min ramp to 240 1C (with a 6 min nal hold).
For product identication by MS, the column was
coupled to an Agilent 5973N quadrupole mass detector
with helium as the carrier gas, and EI at 280 1C, 30V
repeller, 34.6 mA emission, 70eV electron energy, 230 1C
source temperature. Retention times and mass spectra
were compared to those of synthetic references (Fluka).
2.9. Analysis of glucosinolates, desulfoglucosinolates, and
4-hydroxybenzylcyanide sulfate
Analysis of glucosinolates and 4-hydroxybenzylcya-
nide sulfate was elaborated by HPLC of the desulfated
derivatives. Frozen samples were lyophilized and
extracted three times in 70% boiling methanol. At the
rst extraction, 20 ml 5 mM sinigrin (Merck) were added
as internal standard. For samples, in which desulfoglu-
cosinolate content of larval extracts were to be analyzed,
20 ml 5 mM desulfosinigrin (from www.glucosinolates.
com, Frederiksberg, Denmark) were added as second
internal standard. Combined extracts were applied on a
DEAE Sephadex A25-column (0.1 g powder equlibrated
in 0.5 M acetic acid NaOH, pH 5). The column was
washed with 1 ml methanol and 4 1 ml water. These
wash-fractions were collected, concentrated and ana-
lyzed by HPLC as described below to test for the
presence of desulfoglucosinolates in larval extracts. To
analyze the content of glucosinolates and 4-hydroxy-
benzylcyanide sulfate in larval extracts, the fraction of
the extracts that bound to the column was treated with
100 mg sulfatase (E.C. 3.1.6.1) designated type H-1,
from Helix pomatia, 16,400 units/gram solid (Sigma),
dissolved in 400 ml 0.02 M acetic acid NaOH, pH 5.
After over-night incubation, columns were eluted with
1 ml methanol and 5 1 ml water. The eluates were
pooled, concentrated and analyzed by HPLC.
Desulfoglucosinolates and 4-hydroxybenzylcyanide
were analyzed on a 1100 Series chromatograph (Hew-
lett-Peckard, Waldbronn, Germany) equipped with a
quaternary pump, a Supelcosil LC-18 column
(250 4.6 mm, 5 mm, Supelco), and a 1040M diode
array detector. Elution was accomplished with a
gradient (solvent A: water, solvent B: methanol) of
05% B (10 min), 538% B (24 min), followed by a
cleaning cycle (38100% B in 0.5 min, 3 min hold,
1000% B in 0.5 min, 7 min hold). Peaks were quantied
by the peak area at 229 nm (bandwidth 4 nm) relative to
the area of the internal standard peak applying the
response factors as in (Mu ller et al., 2001, 2003a). The
identities of the major glucosinolates sinalbin, gluco-
barbarin, and glucobrassicin (indol-3-ylmethylglucosi-
nolate) were conrmed based on the UV-spectra of their
desulfated derivatives (Mu ller et al., 2001, Fig. 1).
ARTICLE IN PRESS
Glucotropaeolin was identied by comparison with an
authentic standard (Phytoplan).
3. Results
3.1. Localization of glucosinolates in A. rosae
In all samples of A. rosae where hemolymph had been
collected separately, the concentration of glucosinolates
was ve to 25 times higher in the hemolymph samples
compared to the remaining bodies (still containing some
hemolymph as well). In lled guts and feces, only traces
of sinalbin of o0.01 and o0.02 mmol/g fresh weight
(fw), respectively, were detected. This demonstrates that
the glucosinolates are mainly present in the hemolymph.
3.2. Fluctuation of glucosinolate levels in developing
larvae
For larvae reared on S. alba, the amount of the main
glucosinolates sinalbin (Fig. 1b) and glucotropaeolin
(Fig. 1a) increased with increasing size and age of the
larvae until day 17 of development (Fig. 3A). The
amount of glucotropaeolin was at maximum one tenth
of the content of sinalbin, in several larvae it was not
detectable (this reects the pattern in leaves of S. alba
where glucotropaeolin is sometimes absent). At the last
day of larval development, the amount of glucosinolates
dropped back to the amount reached at day 16, and also
larval body mass dropped. At day 21, larvae molted to
eonymphs (last larval instar that is not feeding). In
adults, the amount of sinalbin was 5.172.0 nmol (n 6)
which is one-tenth of the amount found in the last
feeding larval instar. Glucotropaeolin was not detect-
able in adults. In exuviae about 0.2 nmol sinalbin and
0.05 nmol glucotropaeolin were found.
The concentration of glucosinolates varied in a
uctuating manner (Fig. 3B). There was a high variation
in glucosinolate contents and concentrations between
individual larvae from independent experiments. How-
ever, the overall patterns matched well in independent
experiments.
3.3. Short-term uptake of glucosinolates
By transferring larvae from S. alba (main glucosino-
late sinalbin, Fig. 1b) to B. stricta (main glucosinolate
glucobarbarin, Fig. 1c), the time course of glucosinolate
uptake and turn-over was followed. After 30 min of
feeding on B. stricta (corresponding to about 2 mg
consumed leaf) glucobarbarin was already detected in
the hemolymph of some A. rosae larvae and accounted
for 1036% of total glucosinolate content (Fig. 4A).
After 20 h, it accounted for 7096%. Within this time,
larvae had consumed between 29 and 125 mg leaf
material. Larvae accepted B. stricta with different time
delays, resulting in a high variation of the amount of
ingested leaf material. Glucobrassicin (Fig. 1d) could be
detected as well after 30 min feeding and increased to a
maximum of 7% of the total glucosinolate content (Fig.
4B) of a larva but was not located in the hemolymph.
The concentration of glucobarbarin and glucobrassi-
cin varied in the B. stricta material (taken at 3 h after
feeding) of the four replicates from 0.07 to 2.53 mmol/g
fw for glucobarbarin and 0.07 to 0.38 mmol/g fw for
glucobrassicin due to different leaf age. This resulted in
a high variation of the glucosinolate concentrations in
the larvae. The slope of the increase of the glucobarbar-
in proportion in the larvae did not seem to reect the
glucobarbarin concentration in the corresponding host
leaves. The relative abundance of sinalbin decreased
from the rst to the last sampling time to about 10% of
the total glucosinolate content.
ARTICLE IN PRESS
Age of larva (days)
G
l
u
c
o
s
i
n
o
l
a
t
e

(
m
o
l
/
g

f
r
e
s
h

w
e
i
g
h
t
)
0
2
4
6
8
Age of larva (days)
8 10 12 14 16 18 20
8 10 12 14 16 18 20
G
l
u
c
o
s
i
n
o
l
a
t
e

(
m
o
l
/
l
a
r
v
a
)

0.00
0.05
0.10
0.15
(A)
(B)
Fig. 3. Glucosinolate levels in Athalia rosae over development. Larvae
were reared on Sinapis alba. The glucosinolate amount of a larva
reects almost exclusively the glucosinolate sequestered in the
hemolymph (see text). Glucosinolate amount in the larvae (A), and
glucosinolate concentration (B) are shown in dependence of age. Black
bars: sinalbin; grey bars: glucotropaeolin. Results of one out of four
independent experiments are shown. Each bar represents the mean7se
of four individual larvae. Arrows indicate molting event to next larval
instar.
3.4. Long-term changes of glucosinolate levels
When transferring larvae from S. alba to B. stricta,
the glucosinolate from the original host, sinalbin,
dropped to less than 6% after 24 h (Fig. 5A), and was
not detectable after 24 days (Fig. 5A, B). Also over a
longer time period, sinalbin and glucobarbarin were
mainly located in the hemolymph while the indole
glucosinolate glucobrassicin was detectable in bled body
tissue only. When transferring larvae from B. stricta to
S. alba, the original major glucosinolate, glucobarbarin,
dropped less drastically than sinalbin, after 24 h to a
relative abundance of 1527% (Fig. 5C). It could still be
detected after 4 days in all larvae. Glucobrassicin was
detectable for up to 23 days (Fig. 5C, D). The amounts
of glucosinolates that were taken up from the new
host did not steadily increase over the period of 5 days
but uctuated around the level they reached at the
second day after transfer (Fig. 5B, D).
3.5. Glucosinolate turn-over: feeding versus injection of
glucotropaeolin
In larvae that were injected into their hemolymph
with 250 nmol of glucotropaeolin, a rapid decrease of
this glucosinolate down to about 10% of the injected
amount was found after 24 h (Fig. 6A). The turn-over
rate of glucotropaeolin was on average only slightly
lower in larvae that were injected with glucotropaeolin
than in larvae that had taken up the same amount orally
(Fig. 6B), but did not differ signicantly (P40:8 for
both experiments, MannWhitney U-test). The same
pattern was found in four independent experiments,
however, levels of glucosinolate recovered varied (in-
dependently of the volumes used to dissolve glucotro-
paeolin).
3.6. Fate of [
14
C]glucotropaeolin
To study, whether glucosinolates and their metabo-
lites are stored in the body or excreted, larvae were fed
leaf discs with [
14
C]glucotropaeolin. After 14 h they were
dissected and analyzed for radioactivity. The radio-
activity in guts and feces accounted for 84% of the total
radioactivity recovered from larvae and feces.
3.7. Search for glucosinolate-metabolizing enzymes in
larvae
To test, whether A. rosae possesses a sulfatase to
convert glucosinolates to desulfoglucosinolates, two
approaches were taken. On the one hand, larval parts
of A. rosae reared on S. alba were analyzed for the
presence of desulfoglucosinolates. Desulfosinalbin was
detectable in hemolymph and bled larvae (0.370.05 and
0.0570.01 mmol/g fw, respectively), however, the
ARTICLE IN PRESS
(A)
(B)
(C)
Fig. 4. Short-term metabolism of glucosinolates in larvae of Athalia
rosae. Relative abundance of glucobarbarin (A), glucobrassicin (B),
and sinalbin (C) in percent of the total glucosinolate content in
complete larvae (hemolymph and bled body) of A. rosae. While the
aromatic glucosinolates were mainly found in the hemolymph, the
indol glucosinolate was present in the bled body only. Larvae fed on
Barbarea stricta after a transfer from Sinapis alba for periods between
0 and 20 h. Concentration of glucobarbarin and glucobrassicin (in
parenthesis) in leaf discs of B. stricta after 3 h: replicate 1: 0.43
(0.07), repl. 2: 2.53 (0.33), repl. 3: 0.07 (0.18) and repl. 4: 1.40
(0.38) mmol/g fw.
concentration was about twenty times lower than that of
intact sinalbin (6.670.8 and 1.970.2 mmol/g fw, respec-
tively, n 6). In feces, desulfosinalbin was present in
low concentrations between 0.05 and 0.15 mmol/g fw,
while intact sinalbin was present only in traces
(0.0270.01 mmol/g fw).
On the other hand, sulfatase activity was evaluated by
incubation of protein extracts of larval tissue using
glucotropaeolin or sinalbin as substrates. No desulfo-
glucosinolates were formed within 1.5 h. In control
assays, which contained authentic H. pomatia sulfatase,
in addition to the A. rosae protein extracts, desulfoglu-
cosinolates were formed.
To test for myrosinase activity, protein extracts were
incubated with benzylglucosinolate and formation of
hydrolysis-products was followed. Myrosinase activity
was not detectable in extracts of hemolymph, gut tissue
and the remaining body. When external myrosinase was
present in addition to the protein extracts, benzyl
isothiocyanate and traces of benzylcyanide were formed
as hydrolysis products.
No 4-hydroxybenzylcyanide sulfate was detectable in
samples of larvae fed on S. alba or their feces, indicating
that the metabolism of sinalbin in A. rosae occurs via a
different route than in P. rapae.
4. Discussion
Larvae of A. rosae sequester intact glucosinolates,
except for indole glucosinolates, probably more or less
exclusively in the hemolymph. The small amounts of
glucosinolates found in bled bodies are likely due to
the fact, that the hemolymph cannot be collected
ARTICLE IN PRESS
(A) (B)
(C) (D)
Fig. 5. Long-term metabolism of glucosinolates in larvae of A. rosae. Content of specic glucosinolates expressed as relative abundance in percent of
the total glucosinolate content (A, C) and as amount in larvae (B, D) of A. rosae. Larvae were transferred from Sinapis alba to Barbarea stricta (A,
B), and B. stricta to S. alba (C, D) and were analyzed after different numbers of days. While the aromatic glucosinolates were mainly found in the
hemolymph, the indol glucosinolate was present in the bled body only.
exhaustively from the larva. In lled guts and feces only
traces of intact glucosinolates were detected. Therefore,
the glucosinolate content of complete larvae was used in
this study as an approximate measure for hemolymph
glucosinolate content.
The larvae of A. rosae sequester glucosinolates from
their host plant as soon as they hatch from the egg,
which is laid into leaf tissue. Along the development, the
glucosinolate concentration of larvae varied in a
uctuating manner around 5 mmol/g fw. Fluctuation in
uptake of allelochemicals might be a common feature
for sequestering insect species. For example, it has been
reported previously for caterpillars of Utethesia ornatrix
(L.) (Lepidoptera: Arctiidae) feeding on diet containing
pyrrolizidine alkaloids (Kelley et al., 2002).
The decrease in glucosinolate concentration and
amount at the last feeding instar of A. rosae is not
related to an incorporation of glucosinolates in the
molted exuvia. In these, only traces of glucosinolate
were found. Adults still contained some of the glucosi-
nolates, although in a ten times lower concentration
than the larvae. This is in contrast to a previous study
where we showed similar concentrations of sinalbin in
larvae and adults (Mu ller et al., 2001). In the current
development experiments carried out in winter, the
pupal stage took about three times as long as in the
previously reported experiment. Within this long pupal
period some of the glucosinolates might have been
metabolized. The high variation of glucosinolate con-
centration in A. rosae samples of different experiments is
due to a high variation in plant glucosinolate levels
(which can vary 10100 times) but also possible
variation in sequestration ability between the larvae
(Mu ller et al., 2003b).
The sequestration ability of A. rosae larvae is very
efcient. After feeding of only about 2 mg leaf within
30 min, a newly offered glucosinolate is already detect-
able in the hemolymph (Fig. 4). While the amount of a
new glucosinolate increased within the rst 20 h in A.
rosae, it remained (or uctuated) at about a certain level
over the next four days of continuous feeding, for
glucobarbarin below 0.12 mmol/larva and for sinalbin
below 0.17 mmol/larva in this experiment (Fig. 5).
Within the rst 20 h the amount of old glucosinolates
decreased rapidly and then was slowly further diluted.
This indicates an equilibrium between a continuous and
efcient transport of glucosinolates into the hemo-
lymph, as well as a constant degradation. Glucobrassi-
cin was not found in the hemolymph, however, it is also
not excreted as intact glucosinolate (Mu ller et al., 2001).
Thus, this glucosinolate must be metabolized elsewhere.
The hemolymph is not only the main storage place for
intact glucosinolates (except the indole glucosinolate).
Hemolymph glucosinolates must also be the principle
source for glucosinolate degradation, as the degradation
rate was similar in larvae that fed on a leaf disc treated
with glucotropaeolin and larvae that were injected with
glucotropaeolin (Fig. 6B). It is not known yet, whether
glucosinolates are degraded in the hemolymph or
whether they are excreted in the hind gut and degraded
there. The slightly higher mean amount of glucosino-
lates recovered in larvae that took up glucosinolates
orally in replicates 2 and 3 is probably only due to the
fact that injection happened at time-point zero, while
feeding was continuous over 23 h.
The feeding experiment with labeled [
14
C]glucotro-
paeolin revealed that more than 80% of the radio-
activity were excreted within 14 h. As only traces of
intact glucosinolates were detectable in lled guts and
feces in experiments with unlabeled glucosinolates (see
above), the radioactivity in guts and feces must be due to
the product(s) of glucosinolate metabolism. The [
14
C]-
feeding experiment indicates furthermore that metabo-
lite(s) of the glucosinolates are not stored in the body.
The storage of glucosinolates (and their metabolites) is
thus not very efcient. In contrast, among others
several highly adapted Longitarsus spp. (Coleoptera:
ARTICLE IN PRESS
(A)
(B)
Time after injection (h)
Fig. 6. Turn-over of injected versus fed glucosinolate. Percentage of
glucotropaeolin recovered in larvae at 0, 1, 5 and 24 h after injection of
250 nmol into hemolymph (mean7se of n 8 per time point) (A) and
of injected (black bars) versus fed (gray bars) glucotropaeolin (50 or
250 nmol) after 5 h (n.s.not signicant, MannWhitney U-test,
number of replicates is given above each column (B). (1), (2): 50,
250 nmol dissolved in a volume of 5 ml, (3) 250 nmol in a volume of 2 ml.
Chrysomelidae) show an efcient storage of pyrrolizi-
dine alkaloids for at least 2 weeks without major losses
(Narberhaus et al., 2004).
To investigate possible turn-over mechanisms of
glucosinolates, it was investigated whether A. rosae
might possess glucosinolate-metabolizing enzyme acitiv-
ities known from other insects specialized on Brassicaceae
(Fig. 2). Although small amounts of desulfoglucosino-
lates were detected in larvae and feces, a sulfatase activity
could not be veried. No conversion of sinalbin and
glucotropaeolin into the corresponding desulfo-deriva-
tives by larval protein extracts occurred within an
incubation time of up to 1.5 h, while a clear formation
of desulfoglucosinolate was shown for protein extracts of
P. xylostella already within 3 min (Ratzka et al., 2002).
However, we cannot exclude that the sulfatase assay has
to be optimized to enable detection of sulfatase activity in
A. rosae. In contrast to P. xylostella larvae that need a
high glucosinolate sulfatase activity to deplete the plant
myrosinase of substrate (Ratzka et al., 2002), a low
sulfatase activity might be sufcient for A. rosae. It can be
assumed that the intact glucosinolates are harmless
compounds for A. rosae as long as they circulate in the
hemolymph separately from myrosinase. Thus, the small
amounts of desulfoglucosinolates in A. rosae larvae and
feces might be an intermediate in the glucosinolate turn-
over pathway or a by-product of a minor degradation
pathway.
Caterpillars of P. rapae possess a nitrile-specier
protein that converts glucosinolates into their corre-
sponding nitriles which are excreted (Wittstock et al.,
2004). When feeding on sinalbin-containing plant
material, the caterpillars excrete 4-hydroxybenzylcya-
nide sulfate (Mu ller et al., 2003a). In contrast, in feces of
the sawy larvae fed on S. alba this metabolite was not
found. Thus, at least the metabolism of sinalbin in A.
rosae does not result in the formation of the same
metabolite as in P. rapae.
Also, there was no myrosinase activity detectable in
protein extracts of hemolymph, gut tissue or the
remaining body of A. rosae. Volatile repellent products
such as isothiocyanates are not involved in the bleeding
defense mechanism of this sawy (Mu ller and Brake-
eld, 2003). In contrast to glucosinolates, isothiocya-
nates have frequently been reported to be toxic to insects
and other animals (Newman et al., 1992; Li et al., 2000).
Some specialist aphids produce their own myrosinases
which are stored in crystalline microbodies, apart from
the sequestered glucosinolates (Pontoppidan et al., 2001;
Bridges et al., 2002). In one predator of these aphids, the
hovery Episyrphus balteatus De Geer (Diptera: Syrphi-
dae), glutathione S-transferases are induced as detox-
ication enzymes when feeding on B. brassicae
(Vanhaelen et al., 2001).
To summarize, the following scenario for A. rosae can
be proposed. When feeding, most glucosinolates are
transported rapidly into the hemolymph probably at the
beginning of the digestive tract. A transporter must be
present, as glucosinolates are rather polar and can thus
not diffuse passively via the gut membrane. Further-
more, there has to be a continuous turn-over of
glucosinolates in the larval hemolymph to keep a
steady-state level. Plant myrosinases are either inhibited
or out-competed by larval enzymes that produce a
transport form. Alternatively, transport of intact
glucosinolates into the hemolymph is so fast that the
activity of plant myrosinases is severely impaired due to
a lack of substrate. The specialist A. rosae offers a
convenient system to characterize the sequestration and
metabolism as it is possible to switch larvae between
plants with different glucosinolate proles. Further
feeding experiments with labeled compounds might help
to identify the metabolite(s) of glucosinolates in A.
rosae. Nevertheless, we were able to exclude degradation
strategies that are known from other specialists on
Capparales for glucosinolate metabolism in A. rosae.
Thus, during evolution specialists have obviously found
various ways to handle the mustard oil bomb and A.
rosae must proceed yet another route.
Acknowledgements
The authors thank N. Martens and M. Hoffstadt for
technical assistance; M. Riederer for access to the HPLC
equipment in Wu rzburg and discussions; J. Gershenzon
for discussions and supporting the stay of C. Mu ller at
the MPI; B. Halkier (Royal Veterinary and Agricultural
University, Copenhagen) for transgenic lines; and N.
Agerbirk for useful comments on an earlier draft of this
manuscript. This research was in part funded by the
Sonderforschungsbereich 567 Interspezische Interak-
tionen of the Deutsche Forschungsgemeinschaft. Fi-
nancial support by the Max Planck Society is gratefully
acknowledged.
References
Agrawal, A.A., Kurashige, N.S., 2003. A role for isothiocyanates in
plant resistance against the specialist herbivore Pieris rapae. J.
Chem. Ecol. 29, 14031415.
Aliabadi, A., Renwick, J.A.A., Whitman, D.W., 2002. Sequestration
of glucosinolates by Harlequin bug Murcantia histrionica. J. Chem.
Ecol. 28, 17491762.
Andre asson, E., Jrgensen, L.B., Hoglund, A.S., Rask, L., Meijer, J.,
2001. Different myrosinase and idioblast distribution in Arabidop-
sis and Brassica napus. Plant Physiol. 127, 17501763.
Blau, P.A., Feeny, P., Contardo, L., 1978. Allylglucosinolate and
herbivorous caterpillars, a contrast in toxicity and tolerance.
Science 200, 12961298.
Bradburne, R.P., Mithen, R., 2000. Glucosinolate genetics and the
attraction of the aphid parasitoid Diaeretiella rapae to Brassica.
Proc. R. Soc. Lond. B Biol. Sci. 267, 8995.
ARTICLE IN PRESS
Bridges, M., Jones, A.M.E., Bones, A.M., Hodgson, C., Cole, R.,
Bartlet, E., Wallsgrove, R., Karapapa, V.K., Watts, N., Rossiter,
J.T., 2002. Spatial organization of the glucosinolate-myrosinase
system in brassica specialist aphids is similar to that of the host
plant. Proc. R. Soc. Lond. B Biol. Sci. 269, 187191.
Bru ckmann, M., Trigo, J.R., Foglio, M.A., Hartmann, T., 2000.
Storage and metabolism of radioactively labeled pyrrolizidine
alkaloids by butteries and larvae of Mechanitis polymnia
(Lepidoptera: Nymphalidae, Ithomiinae). Chemoecology 10, 2532.
Chen, S., Halkier, B.A., 2000. In vivo synthesis and purication of
radioactive p-hydroxybenzylglucosinolate in Sinapis alba L. Phy-
tochem. Anal. 11, 174178.
Chew, F.S., 1988. Biological effects of glucosinolates. In: Cutler, H.G.
(Ed.), Biologically Active Natural ProductsPotential Use in
Agriculture. American Chemical Society Symposium, Washington,
DC, pp. 155181.
Duffey, S.S., 1980. Sequestration of plant natural products by insects.
Annu. Rev. Entomol. 25, 447477.
Ehmke, A., Rahier, M., Pasteels, J.M., Theuring, C., Hartmann, T.,
1999. Sequestration, maintainance, and tissue distribution of
pyrrolizidine alkaloid N-oxide in larvae of the two Oreina species.
J. Chem. Ecol. 25, 23852395.
Fahey, J.W., Zalcmann, A.T., Talalay, P., 2001. The chemical diversity
and distribution of glucosinolates and isothiocyanates among
plants. Phytochemistry 56, 551.
Hartmann, T., Theuring, C., Witte, L., Pasteels, J.M., 2001.
Sequestration, metabolism and partial synthesis of tertiary
pyrrolizidine alkaloids by the neotropical leaf-beetle Platyphora
boucardi. Insect Biochem. Molec. Biol. 31, 10411056.
Jones, A.M.E., Winge, P., Bones, A.M., Cole, R., Rossiter, J.T., 2002.
Characterization and evolution of a myrosinase from the cabbage
aphid Brevicoryne brassicae. Insect Biochem. Molec. Biol. 32,
275284.
Kelley, K.C., Johnson, K.S., Murray, M., 2002. Temporal modulation
of pyrrolizidine alkaloid intake and genetic variation in perfor-
mance of Utetheisa ornatrix caterpillars. J. Chem. Ecol. 28,
669685.
Koroleva, O.A., Davies, A., Deeken, R., Thorpe, M.R., Tomos, A.D.,
Hedrich, R., 2000. Identication of a new glucosinolate-rich cell
type in Arabidopsis ower stalk. Plant Physiol. 124, 599608.
Li, Q., Eigenbrode, S.D., Stringham, G.R., Thiagarajah, M.R., 2000.
Feeding and growth of Plutella xylostella and Spodoptera eridania
on Brassica juncea with varying glucosinolate concentrations and
myrosinase activities. J. Chem. Ecol. 26, 24012419.
Matile, P., 1980. The mustard oil bombcompartmentation of the
myrosinase system. Biochem. Physiol. Panzen 175, 722731.
Mu ller, C., Brakeeld, P.M., 2003. Analysis of a chemical defense in
sawy larvae: easy bleeding targets predatory wasps in late
summer. J. Chem. Ecol. 29, 26832694.
Mu ller, C., Agerbirk, N., Olsen, C.E., Boeve , J.-L., Schaffner, U.,
Brakeeld, P.M., 2001. Sequestration of host plant glucosinolates
in the defensive hemolymph of the sawy Athalia rosae. J. Chem.
Ecol. 27, 25052516.
Mu ller, C., Agerbirk, N., Olsen, C.E., 2003a. Lack of sequestration of
host plant glucosinolates in Pieris rapae and P. brassicae.
Chemoecology 13, 4754.
Mu ller, C., Zwaan, B.J., de Vos, H., Brakeeld, P.M., 2003b. Chemical
defence in a sawy: genetic components of variation in relevant life-
history traits. Heredity 90, 468475.
Narberhaus, I., Theuring, C., Hartmann, T., Dobler, S., 2003. Uptake
and metabolism of pyrrolizidine alkaloids in Longitarsus ea
beetles (Coleoptera: Chrysomelidae) adapted and non adapted to
alkaloid containing host plants. J. Comp. Physiol. B 173,
483491.
Narberhaus, I., Theuring, C., Hartmann, T., Dobler, S., 2004. Time
course of pyrrolizidine alkaloids sequestration in Longitarsus
ea beetles (Coleoptera: Chrysomelidae). Chemoecology 14,
1723.
Newman, R.M., Hanscom, Z., Kerfoot, W.C., 1992. The watercress
glucosinolate-myrosinase system: a feeding deterrent to caddisies,
snails and amphipods. Oecologia 92, 17.
Nishida, R., Fukami, H., 1990. Sequestration of distasteful com-
pounds by some pharmacophagous insects. J. Chem. Ecol. 16,
151164.
Pasteels, J.M., Theuring, C., Witte, L., Hartmann, T., 2003.
Sequestration and metabolism of protoxic pyrrolizidine alkaloids
by larvae of the leaf beetle Platyphora boucardi and their transfer
via pupae into defensive secretions of adults. J. Chem. Ecol. 29,
337355.
Pontoppidan, P., Ekbom, B., Eriksson, S., Meijer, J., 2001. Purica-
tion and characterization of myrosinase from the cabbage aphid
(Brevicoryne brassicae), a brassica herbivore. Eur. J. Biochem. 268,
10411048.
Ratzka, A., Vogel, H., Kliebenstein, D.J., Mitchell-Olds, T., Kroy-
mann, J., 2002. Disarming the mustard oil bomb. Proc. Natl. Acad.
Sci. USA 99, 1122311228.
Renwick, J.A.A., 2002. The chemical world of crucivores: lures, treats
and traps. Entomol. Exp. Appl. 104, 3542.
Tierens, K.F.M.J., Thomma, B.P.H., Brouwer, M., Schmidt, J.,
Kistner, K., Porzel, A., Mauch-Mani, B., Cammue, B.P.A.,
Broekaert, W.F., 2001. Study of the role of antimicrobial
glucosinolate-derived isothiocyanates in resistance of Arabidopsis
to microbial pathogens. Plant Physiol. 125, 16881699.
Vanhaelen, N., Haubruge, E., Lognay, G., Francis, F., 2001. Hovery
glutathione S-transferases and effect of Brassicaceae secondary
metabolites. Pestic. Biochem. Physiol. 71, 170177.
Weber, G., Oswald, S., Zo llner, U., 1986. Die Wirtseignung von
Rapssorten unterschiedlichen Glucosinolatgehaltes fu r Brevicoryne
brassicae (L.) und Myzus persicae (Sulzer) (Hemiptera, Aphididae).
Z. Panzenkr. Panzenschutz 93, 113124.
Wittstock, U., Halkier, B.A., 2000. Cytochrome P450CYP79A2 from
Arabidopsis thaliana L. catalyzes the conversion of L-phenylalanine
to phenylacetaldoxime in the biosynthesis of benzylglucosinolate. J.
Biol. Chem. 275, 1465914666.
Wittstock, U., Halkier, B.A., 2002. Glucosinolate research in the
Arabidopsis era. Trends Plant Sci. 7, 263270.
Wittstock, U., Kliebenstein, D.J., Lambrix, V., Reichelt, M.,
Gershenzon, J., 2003. Glucosinolate hydrolysis and its impact on
generalist and specialist insect herbivores. In: Romeo, J.T. (Ed.),
Recent Advances in Phytochemistry. Pergamon, Amsterdam, pp.
101125.
Wittstock, U., Agerbirk, N., Stauber, E.J., Olsen, C.E., Hippler, M.,
Mitchell-Olds, T., Gershenzon, J., Vogel, H., 2004. Successful
herbivore attack due to metabolic diversion of a plant chemical
defense. Proc. Natl. Acad. Sci. USA 101, 48594864.
Zasada, I.A., Ferris, H., 2003. Sensitivity of Meloidogyne javanica and
Tylenchulus semipenetrans to isothiocyanates in laboratory assays.
Phytopathology 93, 747750.
ARTICLE IN PRESS
Insect
Biochemistry
and
Molecular
Biology
Mutant Mos1 mariner transposons are hyperactive in Aedes aegypti
David W. Pledger
a,b
, Craig J. Coates
b,
a
Department of Biology (MSC-158), Texas A&M University, Kingsville, TX 78363, USA
b
Department of Entomology (MS-2475), Texas A&M University, College Station, TX 77843, USA
Received 2 April 2005; received in revised form 23 May 2005; accepted 10 June 2005
Abstract
The development of genetic strategies to control the spread of mosquito-borne diseases through the use of class II transposons has
been hampered by suboptimal rates of transformation and the absence of post-integration mobility for all transposons evaluated to
date. Two Mos1 mariner transposase mutants were produced by the site-directed mutagenesis of amino acids, E137 and E264, to K
and R, respectively. The effects of these mutations on the transpositional activities of Mos1-derived transposon constructs were
evaluated by interplasmid transposition assays in Escherichia coli and Aedes aegypti. The transpositional activities of two Mos1
transposons, one with imperfect wild type inverted terminal repeats (ITRs) and another that contained two perfectly matched 3
0
ITRs, were increased when the mutant transposases were supplied in trans in E. coli. The use of the perfect repeat transposon with
wild type transposase did not result in an increase in transposition frequency in Ae. aegypti. However, an improvement in the
integrity of the transposition process did occur, as evidenced by a lower rate of recombination events in which the transgene was
transferred. An increase in the transpositional activity of the perfect repeat transposon was observed in the mosquito in the presence
of either mutant transposase, and in the case of the E264R transposase, the observed increase in transposition frequency was also
accompanied by a further improvement in the integrity of transposition. We discuss the possible contributions of these mutant
residues to the transposition of the perfect repeat Mos1 transposon, the implications of these results with respect to the molecular
evolution of Mos1, and the potential uses of the perfect repeat transposon and mutant transposases for the improvement of Mos1
mediated germ line transformation of Ae. aegypti.
Keywords: Mos1; Mariner; Aedes aegypti; Transposon; Transposable element; Transposition assay
1. Introduction
In an attempt to stem the global rise in the incidence of
mosquito-borne diseases, a variety of avenues are being
explored in an effort to develop an efcient method of
altering the ability of a mosquito species to harbor and/
or transmit a specic disease causing pathogen (Ito et al.,
2002; Moreira et al., 2002; James, 2003; Dean and
Dobson, 2004; Kim et al., 2004; Travanty et al., 2004).
The production of transgenic mosquitoes expressing
refractory genes represents one approach, and the use of
plasmid-borne transposons to produce transgenic insects
has been well documented (for a review see OBrochta
and Atkinson, 2004). The mariner family of transposons
are class II transposable elements that have been
identied in a variety of insects, with a partial list
including Drosophila mauritiana (Jacobson and Hartl,
1985), Haematobia irritans, Oncopeltus fasciatus, Hyalo-
fora cecropia, Apis melifera, Anopheles gambiae, Ephestia
cautella (Robertson, 1993), Bactrocera tryoni (Green and
Frommer, 2001), Bombyx mori (Kumaresan and Matha-
van, 2004), Musca domestica and Blattella germanica
(Liu et al., 2004). Other mariner elements have identied
in non-arthropod invertebrates (Garcia-Fernandez et al.,
1995), as well as vertebrates (Morgan, 1995) and plants
(Jarvik and Lark, 1998).
ARTICLE IN PRESS
doi:10.1016/j.ibmb.2005.06.002

fax: +1 979 845 6305.
E-mail address: c-coates@tamu.edu (C.J. Coates).
Mariner transposons transpose by a conservative cut-
and-paste mechanism whereby the transposase, the
coding sequence for which is contained within the
transposon, catalyzes the excision of the transposon
from one DNA molecule and its subsequent insertion
into another DNA molecule at the target dinucleotide,
TA. Of the hundreds of mariners identied, only
Famar1, Himar1 and Mos1 are known to be active,
and only Mos1 has been observed to be active in
eukaryotic cells (Jacobson and Hartl, 1985; Lampe
et al., 1996; Barry et al., 2004). Mos1, the best
characterized of the three, was originally identied in
the genome of D. mauritiana (Jacobson and Hartl, 1985).
Mos1 requires no host specic factors for transposition
(Tosi and Beverley, 2000), and has been used for
transforming a range of organisms including protozoa
(Gueiros-Filho and Beverley, 1997); vertebrates (Fadool
et al., 1998; Leal et al., 2004); non-arthropod invertebrates
(Bessereau et al., 2001); and arthropods (Garza et al.,
1991; Lidholm et al., 1993; Coates et al., 1998). Chief
among these in the context of efforts to control mosquito-
borne diseases is the transformation of Aedes aegypti
(Coates et al., 1998; Coates et al., 2000), a major mosquito
vector of the yellow fever and dengue viruses (Clark and
Quiroz Martinez, 2001; Gubler, 2004; Tomori, 2004).
In addition to Mos1, the unrelated transposons,
piggyBac and Hermes, have also been used for the
successful germ line transformation of Ae. aegypti, with
rates of transformation reported to be 510% (Coates
et al., 1998; Jasinskiene et al., 1998; Moreira et al., 2000;
Kokoza et al., 2001; Lobo et al., 2002; Adelman et al.,
2004). These rates are sufcient for producing a small
number of transgenic lines, but due to the costly and
time-consuming nature of mosquito transgenesis work,
they are suboptimal for the development of gene
discovery and analysis technologies for Ae. aegypti,
such as insertional mutagenesis, which require the
production of numerous transgenic lines in order to be
effective on a broad scale. Furthermore, none of these
transposons have been reported to remobilize to any
signicant degree in transgenic individuals of this species
(Wilson et al., 2003). Remobilization would improve the
efciency of these methodologies by circumventing the
need for subsequent microinjection experiments and
may also provide a genetic drive mechanism for
propagating refractory genes throughout wild popula-
tions (Tabachnick, 2003). Unlike piggyBac and Hermes,
mariner transposons have been observed to be active in
Escherichia coli, a characteristic that makes them
particularly amenable to mutagenesis studies for the
purpose of increasing their transpositional activity.
One study of Himar1 showed that hyperactive
transposase mutants could be produced by random
mutagenesis (Lampe et al., 1999). In a previous study of
cis-acting elements of Mos1, we mutated the DNA
sequence of the 5
0
inverted terminal repeat (ITR) to
perfectly match that of the 3
0
ITR (Pledger et al., 2004).
While this mutant transposon did result in an increase in
transpositional activity in E. coli using transposase
produced in trans, an increase was not observed in Ae.
aegypti. Evidence suggests that the divergence of
mariner elements during the course of their evolution
is due to changes in both ITR and transposase coding
DNA sequences (Lampe et al., 2001). As a result of such
mutations altering the recognition properties of the
transposasetransposon interaction, it is likely that a
concomitant down regulation of transposition and/or
remobilization might also occur since repeated upregu-
lation over time would likely prove to be deleterious to
the host. These observations indicate that efforts to
achieve increases in Mos1 transpositional activity in Ae.
aegypti might benet from the simultaneous utilization
of mutations in both the cis- and trans-acting elements
of Mos1 transposition.
Himar1 and Mos1 appear to be closely related, each
sharing a consensus ITR sequence and a characteristic
DD(34)D catalytic triad in their transposase amino acid
sequences (Tu and Coates, 2004). Given their related-
ness, we aligned the Himar1 and Mos1 transposase
amino acid sequences in order to select those Mos1
residues which corresponded to the mutated residues in
the hyperactive Himar1 transposases, and the mutations
were duplicated in the Mos1 transposase. These mutant
transposases were supplied in trans, and their effects
were evaluated using a Mos1 mutant transposon
containing the cis-acting perfectly matched ITR pro-
duced in our previous study (Pledger et al., 2004). The
results presented herein indicate that, when used in
concert, these mutations can bring about increases in
Mos1 transposition frequency in Ae. aegypti. In addition
to determining the transpositional activity under these
conditions, the character of the integration events were
also examined. This analysis demonstrated that muta-
tions in the ITR and transposase coding DNA sequence
improved the integrity of the Mos1 transposition
process.
2.1. Mutagenesis
The amino acid sequences of the mariner transpo-
sases, Himar1 (GenPept AAM56714) and Mos1 (Gen-
Bank X78906), were aligned with NCBI BLASTP 2.2.10
using the BLOSUM62 matrix at default settings. Mos1
residues, E137 and E264, corresponded to Himar1
residues, E137 and H267, and were thus selected for
mutation to K and R, respectively, based on the results
of the previous Lampe et al. (1999) study. The two
mutations were created separately in the Mos1 transpo-
sase open reading frame (ORF) using mutagenic
ARTICLE IN PRESS
D.W. Pledger, C.J. Coates / Insect Biochemistry and Molecular Biology 35 (2005) 11991207 1200
oligonucleotide primers and the Gene Editor in vitro
site-directed Mutagenesis System (Promega, Madison,
WI), according to the suppliers instruction. The E137K
mutation was produced by incorporating a single base
substitution, G to A (underlined), at nucleotide position
409 within the Mos1 transposase ORF using the
mutagenic primer E137K, 5
0
CGCAAAAACACATG
CAAAATTTTGCTTTCA 3
0
(K137 codon boldfaced).
The E264R mutation was produced by incorporating
two base substitutions, GA to CG (underlined), at
nucleotide positions 790 and 791 within the transposase
ORF using the mutagenic primer E137R, 5
0
CGCGA
CACGTTGCGAACACTCAATTGG 3
0
(R264 codon
boldfaced).
Briey, the 1035 bp Mos1 transposase coding se-
quence was excised from pBADMos (Pledger et al.,
2004) with EcoRI and HindIII (Promega) and ligated
into EcoRI+HindIII digested pGEM-11fz(+) using T4
DNA ligase (Promega). After alkaline denaturation, the
ligation product, pGEM-Mos, was used as the template
for single strand DNA synthesis using each mutagenic
primer and T4 DNA polymerase to produce hetero-
duplex DNA. Double stranded mutant DNA was
produced by propagating the heteroduplex construct in
BMH 71-18 mutS E. coli cells, which suppress mismatch
nucleotide base pairing repair. The mixture of mutant
and wild type constructs were puried from BMH 71-18
mutS cells and further propagated in JM109 E. coli cells
for clonal selection and DNA sequencing to verify the
presence of the desired base substitutions.
DNA sequencing was performed using the ABI Prism
Big Dye Terminator Cycle Sequencing Ready Reaction
Kit (Applied Biosystems, Foster City, CA). The DNA
sequences of the wild type and mutant transposase
ORFs were translated and analyzed using the Vector
NTI Suite 7 computer software (Invitrogen, Carlsbad,
CA) to determine theoretical values of isoelectric point
and charge at pH 7. The E137K and E264R mutant
coding sequences were excised from pGEM-11fz(+)
with EcoRI and HindIII and cloned back into the EcoRI
and HindIII sites of the pBAD24 bacterial expression
vector (Invitrogen) to produce the mutant helper
plasmids, pBADMosE137K and pBADMosE264R, for
use in mating out assays in E. coli. The double mutant
was constructed by ligating the 500 bp SacI+HindIII
fragment from the E264R construct with the 5.25 kb
SacI+HindIII fragment from pBADMosE137K to
produce pBADMosKR.
2.2. Plasmid construction
The construction of the two donor plasmids used in
the mating out assays in this study is published
elsewhere (Pledger et al., 2004). Both plasmids contained
the Mos1 transposon with an inserted kanamycin
resistance gene (Kan
R
) interrupting the transposase
ORF. The pAC5
0
3
0
MosKan donor possessed the wild
type ITR conguration in which the ITRs differ at four
nucleotide positions. The pAC3
0
3
0
MosKan donor pos-
sessed perfectly matched ITRs which both contained the
sequence of the 3
0
ITR. A 641 bp fragment from each of
the mutant Mos1 transposase coding sequences was
excised with AatII (BioLabs, Beverly, MA) and SphI
(Biolabs) and ligated into AatII+SphI digested
pKhsp82MOS (Coates et al., 1997) to produce the
mutant helper plasmids, pKhsp82MosE137K and
pKhsp82MosE264R, for use interplasmid transpositions
assays in Ae. aegypti embryos. The DNA sequences of
all mutant and wild type transposase ORFs were
determined by automated uorescent DNA sequencing.
The donor plasmids, pDMSKO and pSL3
0
3
0
MKO, used
for the assays in Ae. aegypti are also described elsewhere
(Pledger et al., 2004).
2.3. Transposition assays
Mating out assays were performed as previously
described (Lampe et al., 1999; Pledger et al., 2004) using
the pAC5
0
3
0
MosKan and pAC3
0
3
0
MosKan donor plas-
mids and the pBADMos, pBADMosE137K, pBAD-
MosE264R, and pBADMosKR helper plasmids
providing the transposase in trans. Interplasmid trans-
position assays in Ae. aegypti were performed as
described (Sarkar et al., 1997a; Pledger et al., 2004)
using the donor plasmids, pDMSKO and pSL3
0
3
0
MKO,
and the helper plasmids, pKhsp82MOS, pKhsp82Mo-
sE137 K, and pKhsp82MosE264R.
2.4. Analysis of transposition events
Miniprep DNA from rescued target plasmid clones
that were Kan+Cam resistant and sensitive to Amp was
used as template DNA in a PCR screening assay
designed to determine the presence of recombination
and aberrant transposition products in order to exclude
these from insertion site junction DNA sequencing.
PCR was performed using a single primer, 5
0
TAT
CAGGTGTACAAGTATGAAATGTCGTTT 3
0
, that
anneals to both the 5
0
and 3
0
ITRs of Mos1. PCR
amplication used 40 pmol of primer, an extension time
of 4 min, and a nal extension of 10 min. Lack of an
approximate 3.5 kb product from this reaction indicated
a recombination or aberrant transposition event that
resulted in transfer of the Kan
R
selective marker to the
target plasmid and the loss of at least one of the Mos1
ITRs. The presence of full length donor construct PCR
product indicated the need for insertion site junction
DNA sequencing.
The insertion site junction DNA sequences were
determined by automated uorescent dye terminator
DNA sequencing as described above. The sequencing
primer used for the left end of Mos1 was 5
0
ARTICLE IN PRESS
TGGTGGTTCGACAGTCAAGG 3
0
, which anneals at
nucleotide positions 120 through 101. The right inser-
tion site junction was sequenced using the primer,
5
0
ACAAATTGCCCGAGAGATGG 3
0
, which anneals
at Mos1 nucleotide positions 1151 through 1170. The
potential sequencing read for a canonical transposition
event included the internal Mos1 anking sequence, the
respective ITR, and the anking sequence of the
recipient DNA molecule. Statistical analyses of all
transposition data from the E. coli and Ae. aegypti
studies were performed using the Origin 7.5 computer
software (OriginLab Corp., Northhampton, MA). All P
values were calculated by two sample student t-test. All
relative changes in transposition and recombination in
both E. coli and Ae. aegypti were calculated relative to
those values observed for assays utilizing wild type
transposase and the respective donor construct contain-
ing the wild type ITR conguration.
3. Results
BLASTP alignment of mariner transposons Mos1
and Himar1 indicated 30% identity and 49% similarity
(Fig. 1). Mos1 residue R131 aligned with Himar1 residue
Q131. Because Q131 was mutated to R in the Himar1
study (Lampe et al., 1999), Mos1 residue R131 was not
selected for mutation. Himar1 residues E137 and H267
aligned with Mos1 residues E137 and E264, respectively.
Mos1 E137 fell within an area of only moderate
similarity, and Mos1 E264 anked a region of high
similarity and was positioned between the second and
third aspartate residues of the DD(34)D catalytic triad
characteristic of mariner transposases. Based on the
Himar1 mutations E137 K and H267R that were
produced in the Lampe et al. (1999) study, Mos1
residues E137 and E264 were selected for mutation to
K and R, respectively.
Transposition frequencies observed in mating out
assays in E. coli (Table 1) are expressed as the average
ratios of the total number of transposition events (minus
total false positive events) to the total number of mating
events for each of the ve matings performed in each
mating out assay. The relative increases in transposition
frequency observed in the mating out assays were
calculated by dividing the average transposition fre-
quency of the respective assay by the average frequency
of the assay performed using the wild type transposase
and the 5
0
3
0
MosKan donor construct. Percent false
positive values were less than 2% for all mating out
assays performed based on colony counting of negative
controls in which the helper plasmid did not contain the
transposase coding sequence. Due to loss of transposi-
tional activity observed for the double mutant, that
construct was not selected for further evaluation in Ae.
aegypti.
Results of interplasmid transposition assays in Ae.
aegypti embryos are shown in Table 2. Canonical and
non-canonical transposition frequencies observed in Ae.
aegypti embryos are shown in Table 3. For comparison
ARTICLE IN PRESS
Fig. 1. Amino acid sequence alignment of Himar1 and Mos1 mariner transposases. Identity and similarity (+) are indicated on the consensus line
between the Himar1 and Mos1 sequences. Amino acid residues of the DD(34)D catalytic triad are boxed. The mutated residues in the Himar1 and
Mos1 transposases are shown above and below the alignment, respectively.
purposes herein, the donor constructs contained in
pDMSKO and pSL3
0
3
0
MKO will be referred to as
5
0
3
0
MosKO and 3
0
3
0
MosKO in order to reect their
respective wild type (imperfect) and perfect ITR
congurations. Based on relative increases in transposi-
tion observed in E. coli (Pledger et al., 2004), only the
3
0
3
0
MosKO donor construct was selected for testing of
the mutant transposases in Ae. aegypti. Canonical
transposition events are herein dened as integrative
events in which the DNA sequences anking the intact
Mos1 ITR were determined by insertion site junction
sequencing to be of pGDV1 target plasmid origin. The
average canonical transposition frequencies observed in
Ae. aegypti are expressed as the ratio of the number of
canonical transposition events to the total number of
donor plasmids recovered. This denition also applies to
the results of assays using wild type transposase with
5
0
3
0
MosKO and 3
0
3
0
MosKO donor constructs which
were obtained previously (Pledger et al., 2004).
Non-canonical transposition frequencies observed in
Ae. aegypti are expressed as the ratio of the number of
non-canonical transposition events to the total number
of donor plasmids recovered. Clones for which non-
canonical insertion site characteristics were indicated by
PCR screening and those with anking insertion
junction sequences determined to be of donor plasmid
origin were considered non-canonical transposition
events. Greater than 90% of all non-canonical transpo-
sition events were determined by left and right insertion
site junction DNA sequencing (data not shown). The
results of insertion site sequencing showed that the
character of the right insertion site junction differed
from that of the left in less than 5% of all the clones that
were sequenced, and this relationship was observed to be
ARTICLE IN PRESS
Table 1
Mos1 transposition in Escherichia coli
Mos1 transposase Mos1 donor construct Transposition frequency
a
(10
7
)
P value Relative increase
Wild type 5
0
3
0
MosKan 4.5470.870
Wild type 3
0
3
0
MosKan 287714.9 0.00327 63.2
E137K 3
0
3
0
MosKan 504773.9 0.00975 111
E264R 3
0
3
0
MosKan 342744.2 0.00881 75.3
KR 3
0
3
0
MosKan 189727.3 0.00546 0.35
a
Average value from ve matings.
Table 2
Interplasmid transposition assays in Aedes aegypti embryos
Mos1 transposase Mos1 donor
construct
Embryos injected
a
Donor plasmids
recovered
a
(10
5
)
Canonical events
a
Non-canonical
events
a
Wild type 5
0
3
0
MosKan 388
b
11.4
b
54
b
66
Wild type 3
0
3
0
MosKan 234
b
4.97
b
29
b
16
E137K 3
0
3
0
MosKan 203 12.1 196 40
E264R 3
0
3
0
MosKan 201 7.97 215 19
a
Total values from two independents experiments.
b
Data from Pledger et al. (2004).
Table 3
Mos1 transposition in Aedes aegypti embryos
Mos1
transposase
Mos1 donor
construct
Canonical
frequency
a
(10
5
)
P value Relative
increase
Non-canonical
frequency
a
(10
5
)
value
P value Relative
decrease
Wild type 5
0
3
0
MosKO 4.670.44
b
5.670.45
Wild type 3
0
3
0
MosKO 5.971.3
b
0.43 3.370.27 0.047 1.7
E137K 3
0
3
0
MosKO 1772.3 0.041 3.7 3.370.095 0.0042 1.7
E264R 3
0
3
0
MosKO 2674.5 0.033 5.6 2.370.28 0.0076 2.4
a
Average value from two independent experiments.
b
Data from Pledger et al. (2004).
similar regardless of the donor construct or transposase
used (data not shown).
4. Discussion
We attempted to increase the transposition frequency
of Mos1 in Ae. aegypti using a two pronged approach
which investigated both DNA (cis-acting) and protein
(trans-acting) elements of the Mos1 transformation
system (transformation by microinjection of helper
and donor plasmid). Site-directed mutagenesis of the 5
0
ITR of Mos1 was performed in order to produce a
perfect repeat donor construct containing two 3
0
ITRs.
Site-directed mutagenesis was also performed in order to
mutate those residues of the Mos1 transposase aligning
with the mutated residues of the hyperactive Himar1
mariner mutants produced in a previous study and
evaluated in E. coli (Lampe et al., 1999). The transposi-
tion characteristics of the Mos1 transposase mutants
were evaluated in E. coli and Ae. aegypti using the
perfect repeat donor in order to determine if there was
an increase in transposition frequency. In addition,
insertion site junctions were analyzed in order to
determine to what extent, if any, these mutations
affected the frequency of non-canonical transposition
events.
While the use of the perfect repeat Mos1 donor
construct resulted in an approximate 60-fold relative
increase in transposition in E. coli, no signicant
increase in canonical transposition was observed in Ae.
aegypti relative to that of a similar donor construct with
the wild type ITR conguration. These results, together
with those of in vitro studies of Mos1 ITR binding,
could indicate the existence of an Ae. aegypti factor that
interacts with Mos1, thus impeding transposition. The
observation that the rate of non-canonical events
decreased 1.7-fold when the perfect repeat donor was
used may provide evidence that improved left ITR
binding by the transposase had occurred, resulting in an
improvement in the integrity of the transposition
process. Taken collectively, the transposition data for
the perfect repeat donor seem to point to the mechanism
of the suspected Ae. aegypti factor functioning in a
manner independent of the interaction between the
transposase and the ITRs, and that the occurrence of
non-canonical transposition events is at least to some
extent mediated by Mos1 cis-acting elements.
There are a multitude of mechanisms by which the
E137K and E264R mutations may have brought about
their respective increases in canonical transposition in
Ae. aegypti embryos. Thus, conclusions that can be
drawn from the data collected in this study alone are
necessarily limited. Improved nuclear localization of the
mutant transposases could, however, likely be ruled out
since similar relative increases in transposition frequency
were observed for each mutation in both the prokaryotic
and eukaryotic hosts. Lampe et al. (1999) proposed that
the corresponding mutations in Himar1 resulted in
greater non-specic DNA binding by the transposase.
This same scenario could also apply for the Mos1
mutants since their theoretical net charge (10.8 for both
mutants) at physiological pH was considerably more
basic than that of the wild type transposase (8.82).
Interestingly, while the E137K mutation resulted in no
decrease in the frequency of non-canonical events
relative to that of the wild type transposase, E264R
did bring about a further decrease of 0.7-fold. There-
fore, though both mutations may have increased non-
specic DNA binding, the overall affect brought
about by each individual change was unique in its
properties.
The Mos1 rate of transformation of Ae. aegypti was
observed to increased approximately two-fold when
puried transposase was co-injected in place of the
helper plasmid (Coates et al., 2000). However, DNA
sequencing of the genomic insertion site junctions
revealed that the observed increase in transformation
was not due to an increase in the frequency of precise
transposition, but rather to an approximate four-fold
increase in the frequency of recombination events in
which the donor construct was integrated into the host
genome. Experiments using puried transposase were
performed with recombinant protein produced in
E. coli. Therefore, it is possible that the observed
increase in this type of recombination event was a result
of poorly folded bacterially expressed transposase
protein. However, insertion site junction analysis of
germ line transformants in another study in which co-
injected helper plasmid was used indicated that 25% of
transposition events had resulted in the transfer of
donor plasmid anking sequences along with the
transposon (Coates et al., 2000).
Interestingly, this type of transfer has been observed
for Hermes, and to some extent piggyBac mediated germ
line transformantion of Ae. aegypti as well (Jasinskiene
et al., 2000; Adelman et al., 2004). These ndings are
somewhat surprising when considering that signicant
non-canonical transposition of Mos1, Hermes, and
piggyBac had not been not been reported for any of
the three transposons in other species for which the
character of integration was examined (Sarkar et al.,
1997b; Handler and Harrell, 2001; Lobo et al., 1999;
Thibault et al., 1999; Hediger et al., 2001; Michel et al.,
2001; Granger et al., 2004; Li et al., 2005). These
observations strongly suggest that there exists some Ae.
aegypti host factor(s) that may affect the mobility of
these transposons. In addition, the observations that
various Drosophila proteins interact with P element may
further indicate that similar proteins in Ae. aegypti may
be responsible for altering the transpositional mechan-
isms of Mos1, Hermes, and piggyBac in this mosquito
ARTICLE IN PRESS
species (Rio and Rubin, 1988; Beall et al., 1994; Staveley
et al., 1995).
These observations notwithstanding, caution must be
exercised with respect to extrapolating the observed
improvement in the integrity of interplasmid transposi-
tion for Mos1 mutants to chromosomal integration. For
example, though all Hermes chromosomal integrations
in Ae. aegypti were determined to be non-canonical
transpositions (Jasinskiene et al., 2000), previous studies
of Hermes interplasmid transposition in Ae. aegypti
embryos indicated that only canonical transposition
events had occurred (Sarkar et al., 1997a). Similarly,
while some non-canonical chromosomal integrations of
piggyBac-based donor constructs have been reported for
Ae. aegypti, only canonical interplasmid transposition
have been observed in Ae. aegypti embryos (Lobo et al.,
1999). However, non-canonical transposition by piggy-
Bac at a rate of approximately 44% was observed for
interplasmid transposition in embryos of the mosquito,
A. gambiae (Grossman et al., 2000). Likewise, there
exists no strong quantitative data that demonstrates a
direct relationship between interplasmid transposition
and chromosomal integration. The increases in canoni-
cal transposition observed for the Mos1 E137K and
E264R mutants are nonetheless encouraging. It is
possible that any improvement in the mechanisms of
Mos1 interplasmid transposition as a result of these
mutations could have similar effects on chromosomal
integration events in Ae. aegypti, as may be borne out by
further evaluation of these mutant transposases in
future germ line transformation experiments.
While imprecise, or non-canonical, transposition does
achieve the desired endpoint of transgene integration, it
could also introduce potential downstream problems
since the effects of the presence of donor anking
sequences in the host genome may not be foreseeable.
This type of recombination event could pose particularly
serious problems with respect to the post-integration
mobility of transposons in Ae. aegypti if host genome
anking sequences are likewise transferred during
remobilization. It is also possible that non-canonical
transposition events could result in the loss of donor
construct elements, including portions the transgene.
These types of events would likely be difcult to detect
by phenotypic screening and would therefore be passed
over for further analysis. Improvements in transposon
mediated transformation rates of Ae. aegypti would
therefore ideally be the result of increasing transposition
frequency while simultaneously minimizing non-canoni-
cal events.
In summary, the perfect repeat donor, when used in
concert with these mutant transposases, shows much
potential for increasing the current rates of Mos1
mediated germ line transformation of Ae. aegypti, and
possibly other species. Such advancements would in turn
benet efforts to develop efcient gene discovery and
analysis technologies based on the Mos1 transposon. If
remobilization of Mos1 transposon constructs can be
accomplished in Ae. aegypti, the use of the perfect repeat
donor in conjunction with the E264R transposase may
make possible the development of a Mos1 based genetic
drive system for this mosquito species. Moreover, these
ndings indicate that future efforts to improve Mos1
mediated transformation rates of Ae. aegypti using
plasmid-borne transposons should focus not only on the
cis-acting DNA elements of the transposon, but also on
the trans-acting transposase.
Acknowledgements
We thank Yuqing Fu for her technical expertise and
the anonymous referees for their helpful comments. This
work was supported by National Institutes of Health
grant AI47303.
References
Adelman, Z.N., Jasinskiene, N., Vally, K.J.M., Peek, C., Travanty,
E.A., Olson, K.E., Brown, S.E., Stephens, J.L., Knudson, D.L.,
Coates, C.J., James, A.A., 2004. Formation and loss of large,
unstable tandem arrays of the piggyBac transposable element in the
yellow fever mosquito, Aedes aegypti. Insect Biochem. Mol. Biol.
13, 411425.
Barry, E.G., Witherspoon, D.J., Lampe, D.J., 2004. A bacterial genetic
screen identies functional coding sequences of the insect mariner
transposable element Famar1 amplied from the genome of the
earwig, Forcula auricularia. Genetics 166, 823833.
Beall, E.L., Admon, A., Rio, D.C., 1994. A Drosophila protein
homologous to the human p70 Ku autoimmune antigen interacts
with the P transposable element inverted repeats. Proc. Natl. Acad.
Sci. USA 91, 1268112685.
Bessereau, J.-L., Wright, A., Williams, D.C., Schuske, K., Davis,
M.W., Jorgensen, E.M., 2001. Mobilization of a Drosophila
transposon in the Caenorhabditis elegans germ line. Nature 413,
7074.
Clark, G.G., Quiroz Martinez, H., 2001. Mosquito vector control and
biology in Latin Americaan eleventh symposium. J. Am. Mosq.
Control Assoc. 17, 166180.
Coates, C.J., Jasinskiene, N., Miyashiro, L., James, A.A., 1998.
Mariner transposition and transformation of the yellow fever
mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. USA 95,
37483751.
Coates, C.J., Jansinskiene, N., Morgan, D., Tosi, L.R., Beverly, S.M.,
James, A.A., 2000. Puried mariner (Mos1) transposase catalyzes
the integration of marked elements into the germ-line of the yellow
fever mosquito, Aedes aegypti. Insect Biochem. Mol. Biol. 30,
10031008.
Coates, C.J., Turney, C.L., Frommer, M., OBrochta, D.A., Atkinson,
P.W., 1997. Interplasmid transposition of the mariner transposable
element in non-drosophilid insects. Mol. Gen. Genet. 253, 728733.
Dean, J.L., Dobson, S.L., 2004. Characterization of Wolbachia
infections and interspecic crosses of Aedes (Stegomyia) polyne-
siensis and Ae. (Stegomyia) riversi (Diptera: Culicidae). J. Med.
Entomol. 41, 894900.
Fadool, J.M., Hartl, D.L., Dowling, J.E., 1998. Transposition of the
mariner element from Drosophila mauritiana in zebrash. Proc.
Natl. Acad. Sci. USA 95, 51825186.
ARTICLE IN PRESS
Garcia-Fernandez, J., Bayascas-Ramirez, J.R., Marfany, G., Munoz-
Mormol, A.M., Casali, A., Baguna, J., Salo, E., 1995. High copy
number of highly similar mariner-like transposons in planarian
(Platyhelminthe): evidence for a trans-phyla horizontal transfer.
Mol. Biol. Evol. 12, 421431.
Garza, D., Medhora, M., Koga, A., Hartl, D.L., 1991. Introduction of
the transposable element mariner in the germline of Drosophila
melanogaster. Genetics 128, 303310.
Granger, L., Martin, E., Segalat, L., 2004. Mos as a tool for genome-
wide insertional mutagenesis in Caenorhabditis elegans: results of a
pilot study. Nucleic Acids Res. 32, e117.
Green, C.L., Frommer, M., 2001. The genome of the Queensland fruit
y Bactrocera tyroni contains multiple representatives of the
mariner family of transposable elements. Insect Mol. Biol. 10,
371386.
Grossman, G.L., Rafferty, C.S., Fraser, M.J., Benedict, M.Q., 2000.
The piggyBac element is capable of precise excision and transposi-
tion in cells and embryos of the mosquito, Anopheles gambiae.
Insect Biochem. Mol. Biol. 30, 909914 (Erratum in: Insect.
Biochem. Mol. Biol. 2002. 32, 487).
Gubler, D.J., 2004. The changing epidemiology of yellow fever and
dengue, 1900 to 2003: full circle? Comp. Immunol. Microbiol.
Infect. Dis. 27, 319330.
Gueiros-Filho, F.J., Beverley, S.M., 1997. Trans-kingdom transposi-
tion of the Drosophila element mariner within the protozoan
Leishmania. Science 276, 17161719 (Erratum in: Science, 1997.
277, 753).
Handler, A.M., Harrell II, R.A., 2001. Transformation of the
Caribbean fruit y, Anastrepha suspensa, with a piggyBac vector
marked with polyubiquitin-regulated GFP. Insect Biochem. Mol.
Biol. 31, 199205.
Hediger, M., Niessen, M., Wimmer, E.A., Dubendorfer, A., Bopp, D.,
2001. Genetic transformation of the housey Musca domestica with
the lepidopteran derived transposon piggyBac. Insect Mol. Biol. 10,
113119.
Ito, J., Ghosh, A., Moreira, L.A., Wimmer, E.A., Jacobs-Lorena, M.,
2002. Transgenic anopheline mosquitoes impaired in transmission
of a malaria parasite. Nature 417, 452455.
Jacobson, J.W., Hartl, D.L., 1985. Coupled instability of two X-linked
genes in Drosophila mauritiana: germinal and somatic mutability.
Genetics 111, 5765.
James, A.A., 2003. Blocking malaria parasite invasion of mosquito
salivary glands. J. Exp. Biol. 206, 38173821.
Jarvik, T., Lark, K.G., 1998. Characterization of Soymar1, a mariner
element in soybean. Genetics 149, 15691574.
Jasinskiene, N., Coates, C.J., Benedict, M.Q., Cornel, A.J., Rafferty,
C.S., James, A.A., Collins, F.H., 1998. Stable transformation
of the yellow fever mosquito, Aedes aegypti, with the Hermes
element from the housey. Proc. Natl. Acad. Sci. USA 95,
37433747.
Jasinskiene, N., Coates, C.J., James, A.A., 2000. Structure of Hermes
integrations in the germline of the yellow fever mosquito, Aedes
Kim, W., Koo, H., Richman, A.M., Seeley, D., Vizioli, J., Klocko,
A.D., OBrochta, D.A., 2004. Ectopic expression of a cecropin
transgene in the human malaria vector mosquito Anopheles
gambiae (Diptera: Culicidae): effects on susceptibility to Plasmo-
dium. J. Med. Entomol. 41, 447455.
Kokoza, V., Ahmed, A., Wimmer, E.A., Raikhel, A.S., 2001. Efcient
transformation of the yellow fever mosquito Aedes aegypti using
the piggyBac transposable element vector pBac[3xP3-EGFP afm].
Kumaresan, G., Mathavan, S., 2004. Molecular diversity and
phylogenetic analysis of mariner-like transposons in the
genome of the silkworm, Bombyx mori. Insect Mol. Biol. 13,
259271.
Lampe, D.J., Churchill, M.E.A., Robertson, H.M., 1996. A puried
mariner transposase is sufcient to mediate transposition in vitro.
EMBO J. 15, 54705479.
Lampe, D.J., Akerley, B.J., Rubin, E.J., Mekalanos, J.J., Robertson,
H.M., 1999. Hyperactive transposase mutants of the
Himar1 mariner transposon. Proc. Natl. Acad. Sci. USA 96,
1142811433.
Lampe, D.J., Walden, K.K., Robertson, H.M., 2001. Loss of
transposase-DNA interaction may underlie the divergence of
mariner family transposable elements and the ability of more than
one mariner to occupy the same genome. Mol. Biol. Evol. 18,
954961.
Leal, S., Acosta-Serrano, A., Morris, J., Cross, G.A., 2004.
Transposon mutagenesis of Trypanosoma brucei identies glycosy-
lation mutants resistant to concanavalin A. J. Biol. Chem. 279,
2897928988.
Lidholm, D.-A., Lohe, A.R., Hartl, D.L., 1993. The transposable
element mariner mediates germline transformation in Drosophila
melanogaster. Genetics 134, 859868.
Li, X., Harrell, R.A., Handler, A.M., Beam, T., Hennessy, K., Fraser
Jr., M.J., 2005. piggyBac internal sequences are necessary for
efcient transformation of target genomes. Insect Mol. Biol. 14,
1730.
Liu, N., Pridgeon, J.W., Wang, H., Liu, Z., Zhang, L., 2004.
Identication of mariner elements from house ies (Musca
domestica) and German cockroaches (Blattella germanica). Insect
Mol. Biol. 13, 443447.
Lobo, N., Li, X., Fraser Jr., M.J., 1999. Transposition of the piggyBac
element in embryos of Drosophila melanogaster, Aedes aegypti and
Trichoplusia ni. Mol. Gen. Genet. 261, 803810.
Lobo, N.F., Hua-Van, A., Li, X., Nolen, B.M., Fraser Jr., M.J., 2002.
Germ line transformation of the yellow fever mosquito, Aedes
aegypti, mediated by transpositional insertion of a piggyBac vector.
Michel, K., Stamenova, A., Pinkerton, A.C., Franz, G., Robinson,
A.S., Gariou-Papalexiou, A., Zacharopoulou, A., OBrochta,
D.A., Atkinson, P.W., 2001. Hermes-mediated germ-line transfor-
mation of the Mediterranean fruit y Ceratitis capitata. Insect Mol.
Biol. 10, 155162.
Moreira, L.A., Edwards, M.J., Adhami, F., Jasinskiene, N., James,
A.A., Jacobs-Lorena, M., 2000. Robust gut-specic gene expres-
sion in transgenic Aedes aegypti mosquitoes. Proc. Natl. Acad. Sci.
USA 97, 1089510898.
Moreira, L.A., Ito, J., Ghosh, A., Devenport, M., Zieler, H.,
Abraham, E.G., Crisanti, A., Nolan, T., Catteruccia, F., Jacobs-
Lorena, M., 2002. Bee venom phospholipase inhibits malaria
parasite development in transgenic mosquitoes. J. Biol. Chem. 277,
4083940843.
Morgan, G.T., 1995. Identication in the human genome of mobile
elements spread by DNA-mediated transposition. J. Mol. Biol. 254,
15.
OBrochta, D.A., Atkinson, P.W., 2004. Transformation systems in
insects. In: Miller, W.J., Capy, P. (Eds.), Mobile Genetic Elements.
Methods in Molecular Biology, vol. 260. Humana Press, Totowa,
NJ, pp. 227254.
Pledger, D.W., Fu, Y.Q., Coates, C.J., 2004. Analyses of cis-acting
elements that affect the transposition of Mos1 mariner transposons
in vivo. Mol. Genet. Genomics 272, 6775.
Rio, D.C., Rubin, G.M., 1988. Identication and purication of a
Drosophila protein that binds to the terminal 31-base-pair inverted
repeats of the P transposable element. Proc. Natl. Acad. Sci. USA.
85 (23), 89298933.
Robertson, H.M., 1993. The mariner transposable element is wide-
spread in insects. Nature 362, 241245.
Sarkar, A., Yardley, K., Atkinson, P.W., James, A.A., OBrochta,
D.A., 1997a. Transposition of the Hermes element in embryos of
ARTICLE IN PRESS
the vector mosquito, Aedes aegypti. Insect Biochem. Mol. Biol. 27,
359363.
Sarkar, A., Coates, C.J., Whyard, S., Willhoeft, U., Atkinson, P.W.,
OBrochta, D.A., 1997b. The Hermes element from Musca
domestica can transpose in four families of cyclorrhaphan ies.
Genetica 99, 1529.
Staveley, B.E., Heslip, T.R., Hodgetts, R.B., Bell, J.B., 1995. Protected
P-element termini suggest a role for inverted-repeat-binding
protein in transposase-induced gap repair in Drosophila melanoga-
ster. Genetics 139, 13211329.
Tabachnick, W.J., 2003. Reections on the Anopheles gambiae genome
sequence, transgenic mosquitoes and the prospect for controlling
malaria and other vector borne diseases. J. Med. Entomol. 40,
597606.
Thibault, S.T., Luu, H.T., Vann, N., Miller, T.A., 1999. Precise
excision and transposition of piggyBac in pink bollworm embryos.
Tomori, O., 2004. Yellow fever: the recurring plague. Crit. Rev. Clin.
Lab. Sci. 41, 391427.
Tosi, L.R.O., Beverley, S.M., 2000. cis and trans factors affecting
Mos1 mariner evolution and transposition in vitro, and its
potential for functional genomics. Nucleic Acids Res. 28,
784790.
Travanty, E.A., Adelman, Z.N., Franz, A.W.E., Keene, K.M., Beaty,
B.J., Blair, C.D., James, A.A., Olson, K.E., 2004. Using
RNA interference to develop dengue virus resistance in
genetically modied Aedes aegypti. Insect Biochem. Mol. Biol.
34, 607613.
Tu, Z., Coates, C., 2004. Mosquito transposable elements. Insect
Wilson, R., Orsetti, J., Klocko, A.D., Aluvihare, C., Peckham, E.,
Atkinson, P.W., Lehane, M.J., OBrochta, D.A., 2003. Post-
integration behavior of a Mos1 mariner gene vector in Aedes
aegypti. Insect Biochem. Mol. Biol. 33, 853863.
ARTICLE IN PRESS

02 Insect Biochemistry Molecular

Încărcat de

Informații document

Descriere originală:

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

02 Insect Biochemistry Molecular

Încărcat de

Drepturi de autor:

Formate disponibile

Insect Biochemistry and Molecular Biology

Volume 35, Issue 10, Pages 1073-1208 (October 2005)

Corresponding author. Present address: Department of Genetics,

Corresponding author. Tel.: +49 5313 915681;

Department of Ecology, Lund University, Ecology building, SE-22362, Sweden

Corresponding author. Tel.: +46 46 222 9344;

Corresponding author. Tel.: +81 3 5841 8011;

, L.F. Lopes, V.E. do Rosa rio, H. Silveira

Corresponding author. Tel.: +351 213622458;

Corresponding author. Tel.: +82 55 751 5951;

resolution. Biochemistry 30, 603608.

resolution. J. Mol. Biol. 198, 499513.

Corresponding author. Tel.: +1 301 402 2748;

resolution. J. Mol. Biol. 89,

Corresponding author. Tel.: +1 409 747 2426;

Corresponding author. Tel.: +81 29 838 6157;

Corresponding author. Tel.: +45 35321703; fax: +45 35321567.

Corresponding author. Tel.: +49 931 888 62 21;

Corresponding author. Tel.: +1 979 458 1219;

S-ar putea să vă placă și