Báo cáo khoa học: Pherokine-2 and -3 Two Drosophila molecules related to pheromone/odor-binding proteins induced by viral and bacterial infections ppt

Pherokine-2 and -3
Two
Drosophila
molecules related to pheromone/odor-binding proteins induced
by viral and bacterial infections
Laurence Sabatier
1,
*, Emmanuelle Jouanguy
1,
*†, Catherine Dostert
1
, Daniel Zachary
1
, Jean-Luc Dimarcq
2
,
Philippe Bulet
1,
‡ and Jean-Luc Imler
1
1
CNRS UPR9022, Institut de Biologie Mole
´
culaire et Cellulaire, Strasbourg, France;
2
Entomed SA; Illkirch, France
Drosophila is a powerful model system to study the regula-
tory and effector mechanisms of innate immunity. To iden-
tify molecules induced in the course of viral infection in this
insect, we have developed a model based on intrathoracic
injection of the picorna-like Drosophila Cvirus(DCV).We

have used MALDI-TOF mass spectrometry to compare
the hemolymph of DCV infected flies and control flies.
By contrast with the strong humoral response triggered by
injection of bacteria or fungal spores, we have identified only
one molecule induced in the hemolymph of virus infected
flies. This molecule, pherokine-2 (Phk-2), is related to OS-D/
A10 (Phk-1), which was previously characterized as a
putative odor/pheromone binding protein specifically
expressed in antennae. The virus-induced molecule is also
similar to the product of the gene CG9358 (Phk-3), which is
induced by septic injury. Both Phk-2 and Phk-3 are strongly
expressed during metamorphosis, suggesting that they may
participate in tissue-remodeling.
Keywords: host-defense; antiviral; Drosophila C virus; odor-
binding protein; tissue remodelling.
Innate immunity enables multicellular organisms to detect
and fight infectious microbes. In vertebrates, the innate
immune system also participates in the induction and
shaping of the subsequent adaptive immune response
carried by lymphocytes. The innate immune system involves
pattern recognition receptors (PRRs) which recognize
conserved molecular patterns from broad classes of
microorganisms, such as lipopolysaccharide (LPS) from
Gram-negative bacteria, peptidoglycan (PGN) from Gram-
positive bacteria, or double-stranded (ds)RNA from viruses.
Activation of PRRs triggers a host response to control the
infection either by acting directly on the microorganisms by
phagocytosis or the production of toxic compounds such
as nitric oxide and antimicrobial peptides, or by inducing
the production of cytokines or costimulatory molecules

(reviewed in [1–3]).
The fruitfly Drosophila melanogaster is a good model to
decipher the molecular mechanisms governing innate
immunity in animals because of its well-characterized
genetics, and its lack of an adaptive immune system. The
significant progress in our understanding of the response of
Drosophila to infection by bacteria and fungi made in the
past years have revealed a number of molecular similarities
between the pathways regulating innate host-defense in
flies and mammals [4,5]. The best characterized aspect of
the Drosophila response to infection is the inducible
synthesis and secretion in the hemolymph of a cocktail of
potent antimicrobial peptides active against bacteria and/or
fungi. Transcriptional induction of the genes encoding
these peptides involves two pathways: infections by fungi
and Gram-positive bacteria trigger the Toll pathway,
named after the transmembrane receptor Toll, whereas
infections by Gram-negative bacteria activate the Imd
pathway, named after the immune deficiency (imd) gene.
The Toll and Imd pathways exhibit similarities with the
interleukin-1 and the TNFa pathways, respectively [4,5].
Following the demonstration of the critical role played by
the Toll receptor in Drosophila, a family of related
molecules was identified in mammals. These Toll-like
receptors (TLRs) are involved in cell activation by
microbial molecules such as LPS (TLR4), PGN (TLR2)
or bacterial DNA (TLR9) [6,7].
By contrast, nothing is known about the response to
virus infection in Drosophila. In mammals, dsRNA from
viruses has long been known to activate enzymes such as

protein kinase R (PKR) or oligo A 2.5 synthase, and
cytokines such as interferon-b. However, our understanding
of the mechanisms operating during the innate antiviral
response remain sketchy, as illustrated by the recent
identification of TLR3 as a transmembrane receptor for
dsRNA [8]. In order to analyze the Drosophila host-defense
against viral infection, we developed a model based on
Correspondence to J L. Imler, CNRS UPR9022, Institut de Biologie
Mole
´
culaire et Cellulaire, 15 rue Rene
´
Descartes, 67000 Strasbourg,
France. Fax: + 33 388 60 69 22, Tel.: + 33 388 41 70 36
E-mail:
Abbreviations:DCV,Drosophila Cvirus;PRR,patternrecognition
receptors; LPS, lipopolysaccharide; PGN, peptidoglycan;
TLR, Toll-like receptors.
*These two authors contributed equally to the work.
Present address: INSERM U550; 156 rue de Vaugirard; 75015 Paris,
France.
àPresent address: Atheris laboratories; Case Postale 314; CH-1233
Bernex, Switzerland.
(Received 28 April 2003, revised 12 June 2003,
accepted 18 June 2003)
Eur. J. Biochem. 270, 3398–3407 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03725.x
Drosophila C virus (DCV). DCV is a nonenveloped small
single stranded (+) RNA virus, which differs from
picornaviridae by its specific genome organization, and
the presence of two open reading frames [9]. Drosophila is

a natural host for DCV, which is transmitted horizontally
through contact or ingestion [10–12]. In a first step to
characterize the response of Drosophila to virus infection,
we attempted to identify molecules induced by DCV
infection that could serve as markers of the antiviral
response. We used MALDI-TOF mass spectrometry (MS)
differential analysis between the hemolymph of DCV-
infected flies vs. unchallenged Drosophila. This differential
MS approach was developed by Uttenweiler-Joseph and
colleagues to study the antibacterial response, and allowed
the identification of more than 24 small peptides, named
Drosophila immune-induced molecules [13,14]. Using the
same approach, we have identified only one peptide which
is induced upon virus infection. This peptide presents strong
sequence similarity to OS-D/A10, a molecule previously
characterized as a putative odor-binding protein [15,16]. A
third Drosophila molecule belonging to this small family is
induced by septic injury.
Experimental procedures
Plasmids
The attacin A promoter in the pCasper transformation
vector pJL166 [17] was replaced by a XbaI–NheIPCR
fragment containing 2.6 kb of phk-2 5¢ untranslated
sequences (GenBank AE003462 nt250436–253041) to
obtain pJL265. This fragment includes exon 0 of phk-2,
the first intron, and the first exon to the ATG, which is used
to initiate translation of GFP S65T. The transforming
vector pJL264 contains a shorter fragment of phk-2 5¢
untranslated sequences (GenBank AE003462 nt2516916–
253041), and yielded identical results (data not shown). The

phk-2 cDNA (EST clone GH24283) was obtained from the
Berkeley Drosophila Genome Project (Berkeley, CA; http://
www.fruitfly.org). The phk-2 cDNA was subcloned as an
EcoRI–XhoI fragment between the corresponding sites in
the pP{UAST} vector [18].
Fly strains and bacterial infections
Oregon-R and ywDD1; cnbw wild-type flies were used
throughout this study [19]. Flies were maintained on a
standard cornmeal medium at 25 °C. Transgenic lines were
generated by P element transformation of a w
–
strain.
Standard crosses with flies carrying appropriate balancers
were performed to establish stable heterozygous or homo-
zygous lines, as well as to determine the chromosome
carrying the insertion. At least three independent lines were
analyzed for each construct. To overexpress Phk-2, females
carrying the UAS-phk-2 transgene were crossed with males
carrying the P{GAL4-YP1.JMR} (yolk protein 1 gene
promoter-Gal4) driver [20]. Bacterial infections were per-
formed by pricking adult flies with a thin needle, previously
dipped in a concentrated culture of Escherichia coli and
Micrococcus luteus. RNA extraction, Northern blot analysis
and RT-PCR on total RNA were performed as described
previously [21].
Preparation of the DCV stock
An isolate of DCV was kindly provided by X. Jousset and
M. Bergoin (INRA-CNRS URA2209, St Christol-Lez-
Ale
`

s, France). A concentrated viral suspension was prepared
by successive rounds of amplification in infected adult flies.
This was purified on a caesium chloride gradient and
analyzed by electron microscopy as described [22]. Briefly,
4000 flies were injected with DCV and collected and frozen
after death. Flies were crushed in 10 m
M
Tris/HCl (pH 7.5),
followed by sonication (20 kHz; three times for 5 s). The
extract was deposited on a 20% (w/v) sucrose solution, and
centrifuged (25 000 g,1h30min,15°C).Theviralpellet
was resuspended in 10 m
M
Tris/HCl (pH 7.5). After
sonication as above, the viral suspension was added to a
tube containing two layers of caesium chloride (5% and
40%), and ultracentrifuged for 16 h at 36 000 g at 15 °C.
The virus band was collected and DCV was recovered by
centrifugation (25 000 g,1h30min,15°C). The purified
viral pellet was resuspended in 1 mL 10 m
M
Tris/HCl
(pH 7.5), sonicated as above, aliquoted and stored at
)80 °C. The viral titer was estimated to be 10
11.5
LD
50
ÆmL
)1
, using the Reed–Muench endpoint method

and a 7 day incubation period in adult flies [23]. For the
infection experiments described, 5 nL containing 10
4.5
LD
50
were injected in the thorax of 4–6-day-old adult flies. For
survival experiments, groups of 25 flies were kept on
standard medium at 22 °C, and counted daily.
Microscopic observations
For observation of GFP expression patterns, live flies and
larvae were anaesthetized with ether or on ice, and viewed
under epifluorescent illumination (excitation filter, 480 nm;
dichroic filter, 505 nm; and emission filter, 510 nm) with a
Leica MZFLIII dissecting microscope and images were
recorded using a digital charge-coupled device Spot RT
color camera (Diagnostic Instruments). For histology
analysis, dissected female flies were fixed in 4% (v/v)
glutaraldehyde in 0.1
M
sodium phosphate buffer (pH 7.3)
for 1 h at 4 °C, postfixed with osmium tetroxide, embedded
in araldite/epon, and sectioned for optic or electron
microscopy. A toluidin blue coloration was performed on
semithin sections. Briefly, after a 2 min treatment with
sodium methoxide, the sections were incubated in a 50%
methanol/benzene mixture (v/v) during 90 s, followed by
acetone for 1 min, before rinsing in distilled water. The
slides were then stained for 5 min in a toluidin blue solution
(0.1% toluidin blue, 1% borax; pH 11), rinsed with distilled
water and dehydrated. For transmission electron micro-

scopy, preparations of dissected tissues were fixed in 4% (v/
v) glutaraldehyde in 0.1
M
sodium phosphate buffer (pH 7.3)
for 1 h at 4 °C, and postfixed with osmium tetroxide.
Samples embedded in araldite/epon were sectioned and
counterstained with lead citrate and uranyl acetate.
MALDI-TOF MS analysis
For mass spectrometry analysis, hemolymph of DCV-,
or buffer-injected Drosophila was collected and directly
deposited on the target. Samples were prepared
according to the sandwich method using the matrix
Ó FEBS 2003 Drosophila humoral response to virus infection (Eur. J. Biochem. 270) 3399
a-cyano-4-hydroxycinnamic acid [24]. MALDI-TOF mass
spectrometry was performed with a Bruker BIFLEX III
TM
(Bremen, Germany) mass spectrometer operating in a
positive linear mode using an external calibration and
synthetic peptides (MH
+
2199.6; 3046.4; 4890.5).
Purification of Phk-2
Hemolymph from 430 flies was collected in 0.1% (v/v)
trifluoroacetic acid 72 h after DCV infection. After centri-
fugation (10 000 g, 20 min), the supernatant was subjected
to gel permeation HPLC using two serially linked columns
(Ultraspherogel SEC 3000 and SEC 2000 columns,
7.5 · 300 mm, Beckman). Elution was performed under
isocratic conditions with 30% acetonitrile in 0.05% (v/v)
trifluoroacetic acid at a flow rate of 0.4 mLÆmin

)1
. Fractions
were hand-collected according to the absorbance at 225 nm
and analyzed by MALDI-TOF mass spectrometry. The
fraction containing the induced molecule Phk-2 was further
purified by reverse phase HPLC on a microbore Aquapore
RP300 C
8
column (1 · 100 mm, Brownlee
TM
Perkin Elmer)
using a linear biphasic gradient of acetonitrile in 0.05% (v/v)
trifluoroacetic acid from 2 to 25% over 10 min and from 25
to 35% over 50 min, at a flow rate of 80 lLÆmin
)1
.
Structure identification
Purified Phk-2 was treated with trypsin (modified sequen-
cing grade, Roche Diagnostics, Mannheim, Germany)
using the conditions recommended by the supplier.
Digestion was carried out at 37 °Cfor16hin0.1
M
Tris/HCl pH 8.9 supplemented with 10% (v/v) aceto-
nitrile. The reaction was stopped by acidification and the
peptide fragments were separated on a capillary FUS-
15–03-C18-PepMap column (0.3 · 150 mm, LC Packings,
Amsterdam, the Netherlands) using a linear gradient of
acetonitrile in 0.05% (v/v) trifluoroacetic acid from 5 to
40% over 40 min at a flow rate of 4 lLÆmin
)1

at the
temperature of 30 °C. The column effluent was monitored
by absorbance at 214 nm and the fractions were hand-
collected and analyzed by MALDI-TOF MS. Three
purified fragments were submitted to automated Edman
degradation on a pulse liquid automatic sequenator
(Applied Biosystems Model Procise cLC).
Cell culture experiments
S2 cells were purchased from Invitrogen, and maintained in
Schneider’s medium supplemented with 10% (v/v) fetal
bovine serum; 60 mgÆL
)1
penicillin and 100 mgÆL
)1
strepto-
mycin. 20-Hydroxyecdysone was added to the cells (10
)6
M
)
48 h prior to stimulation with 10 lgÆmL
)1
LPS (E. coli
serotype 055:B5, Sigma).
Results
Systemic infection of
Drosophila
by DCV
We prepared a stock of DCV by serial passages in flies (see
Experimental procedures). When aliquots of this suspension
were injected into flies, we observed a rapid lethality that

was dose-dependent (Fig. 1A). Histological analysis
revealed morphological defects associated with the fat body
appearing two to three days postinfection as cells started to
shrink and stained more strongly with toluidin blue
(Fig. 1B,D). Similar changes were detected in cells of the
perioveolar epithelial sheath. Consistent with these obser-
vations, transmission electron microscopy revealed high
levels of DCV particles in cells from the fat body and the
perioveolar sheath (Fig. 1E,F). Virus particles were also
detected in cells from tracheae, muscles and the digestive
tract. The quantity of virus present in these tissues increased
over time, indicating that productive infection had occurred
(data not shown).
DCV infection triggers a discrete humoral response
We have analyzed the hemolymph of single flies injected
with a suspension of DCV or Tris buffer through a
differential display analysis by MALDI-TOF MS. We did
not observe induction of any molecules in the 0–10 kDa size
range in the mass fingerprint, at any of the time points
analyzed (24 h, 48 h, 72 h; Fig. 2A and data not shown). In
particular, the antimicrobial peptides which rapidly appear
in the hemolymph upon bacterial or fungal infections [14]
are not present in the hemolymph of DCV-infected flies.
However, one molecule at a measured average molecular
mass of 12 820 Da is clearly induced in the hemolymph of
flies infected with the virus 48 h after the beginning of the
infection (Fig. 2A). This molecule is not present in the
hemolymph of flies injected with buffer, or challenged by
septic injury with a mixture of Gram-negative and Gram-
positive bacteria, or upon natural infection with the

entomopathogenic fungus Beauveria bassiana (data not
shown and [14]). Hemolymph from DCV-injected flies was
analyzed by HPLC and the molecule induced by the viral
infection was detected by MALDI-TOF MS (Fig. 2B). The
molecule was purified to homogeneity by gel permeation
and reversed-phase chromatography and submitted to
proteolysis for structural characterization. The digest was
purified, and the recovered fragments sequenced by Edman
degradation. Database analysis using one of these sequen-
ces, namely YIIENKPEEWK, revealed that the virus-
induced molecule corresponds to the product of the gene
PebIII [25] (also called CG11390; itfly.org/)
(Fig. 2C). The Drosophila genome contains two additional
genes related to PebIII/CG11390 (Fig. 2D): OS-D/A10 has
been described previously on the basis of its tissue-specific
expression in the olfactory region of antennae [15,16];
CG9358 has not been described previously. Interestingly,
expression of the latter gene appears to be upregulated in
response to bacterial or fungal infections [26–28] (see
below). We therefore propose to call these molecules
induced by infection which may function as odor/phero-
mone binding proteins pherokines (Phk-1, OS-D/A10;
Phk-2, PebIII/CG11390; Phk-3, CG9358).
Sequencing of the fragment with a measured mass of
MH
+
1170.9 showed a difference (Asp/Glu) between the
sequence from the Drosophila genome (CG11390) and the
sequence of Phk-2 (Fig. 2C). The start of the mature protein
after cleavage of the signal peptide was ascertained by the

sequencing of the N-terminal peptide (measured mass
MH
+
2204.2). The concentration of Phk-2 in the hemo-
lymph of DCV-infected Drosophila was estimated at 40 n
M
.
3400 L. Sabatier et al.(Eur. J. Biochem. 270) Ó FEBS 2003
Fig. 1. Systemic infection following intrathoracic injection of DCV in Drosophila. (A) Adult flies (Oregon-R) were injected with 5 nL of different
dilutions of a DCV stock (10
11.5
LD
50
ÆmL
)1
), or with buffer (Tris). Surviving flies were counted daily. (B–D) Histology of the fat body of flies 2 (C)
or 4 (B,D) days after injection of Tris (B) or DCV (C,D). The scale bar represents 20 lm (E,F). Accumulation of viral particles in the cytoplasm of
cells from the fat body (E) or the perioveolar sheath (F) 4 days after injection of DCV. Crystal-like arrangements of virus particles are indicated by
stars and enlarged in insets. The scale bars represent 0.3 lminEand0.5lm in F; m, mitochondria; n, nucleus; ov, ovary; ch, chorion.
Ó FEBS 2003 Drosophila humoral response to virus infection (Eur. J. Biochem. 270) 3401
Phk-2 is regulated by DCV infection at the
post-transcriptional level
Northern blot analysis revealed that the quantity of
transcripts encoding Phk-2 does not increase following
viral infection (data not shown). This result suggests
either that phk-2 is not regulated at the transcriptional
level, or that the induction of phk-2 in some tissues is
masked by constitutive expression in others. To identify
the expression domains of this gene, we constructed
transgenic strains of Drosophila expressing GFP under

the control of 2.2 kb of 5¢ untranscribed sequences from
phk-2. We observed green fluorescence in noninfected
larvae in both anterior and posterior spiracles, in the
digestive tract, the ring gland, the antenna buds and
testis (Fig. 3A,B and data not shown). In adult flies,
GFP was expressed in several tissues, including the legs,
the wing veins, the male and female reproductive tracts,
the digestive tract and the labellum (Fig. 3C–F). Analysis
by mass spectrometry, RT-PCR (Fig. 3G–I), and in situ
hybridization [25] confirmed that the endogenous gene is
expressed in fluorescent tissues. These results are sum-
marized in Fig. 3J,K. Importantly, we did not observe
any modification of the fluorescence pattern in DCV
infected flies, confirming that Phk-2 induction by DCV
infection is not mediated at the transcriptional level (data
not shown).
Constitutive production of Phk-2 does not protect flies
from DCV infection
In the absence of mutant strains for the phk-2 gene, we
constructed strains of Drosophila constitutively expressing
Phk-2 in the hemolymph, using the UAS-Gal4 system [18].
Fig. 2. Identification of Phk-2 in the hemo-
lymph of DCV infected flies. (A) MALDI-TOF
mass spectrometry analysis of the hemolymph
of single flies 3 days after injection of buffer
(Tris) or a DCV suspension. The position of
the 12 820 Da DCV-induced molecule (Phk-
2) is indicated. The peaks present in the Tris-
injected fly but not in the DCV-injected fly are
not reproducibly observed and correspond to

previously described Drosophila induced mole-
cules transiently upregulated following injury
[14]. (B) HPLC chromatography of the
hemolymph of 40 flies 3 days after injection of
a DCV suspension. The column is a microbore
Aquapore RP300 C
8
column (1 · 100 mm,
Brownlee Labs) eluted with a linear gradient
of acetonitrile (2–80%, v/v) in acidified water
at a flow rate of 80 lLÆmin
)1
at 35 °C. The
position of Phk-2, detected by MALDI-TOF
mass spectrometry, is indicated. (C) Sequence
of the cDNA clone GH24283 which encodes
Phk-2. The amino acid sequence is shown
below the nucleotide sequence. The sequences
of the tryptic peptides sequenced after purifi-
cation of Phk-2 are underlined. The only
difference between these sequences is the
replacementofanasparticacidatposition108
by a glutamic acid. The signal peptide
sequence is boxed, and the two disulfide
bridges are indicated as determined experi-
mentally. (D) Alignment of the mature
sequences of Drosophila pherokines with
related molecules from other insects.
3402 L. Sabatier et al.(Eur. J. Biochem. 270) Ó FEBS 2003
Overexpression of phk-2 using the ubiquitous drivers

daughterless-Gal4 and actin5C-Gal4 or the fat body specific
driver yolk-Gal4, was not lethal and did not induce any
obvious phenotype (data not shown). Phk-2 was constitu-
tively present in the hemolymph of female flies expressing
both the yolk-Gal4 driver and the UAS-phk-2 transgene
(Fig. 4A). However, susceptibility of these flies to DCV
infection did not significantly differ from that of control flies
(Fig. 4B).
Regulation of pherokine expression during infection
and development
We confirmed by Northern blot analysis that the quantity of
phk-3 transcripts increases in response to septic injury, with
a peak at 3–6 h postinfection (Fig. 5A). Under the same
conditions, phk-1 and phk-2 are not upregulated. phk-3
remains inducible in Dif, dorsal, spaetzle and kenny mutant
flies [4], suggesting that it is not regulated by the Toll or Imd
Fig. 3. Expression pattern of phk-2. (A,B) phk-2-GFP transgenic larvae exhibit green fluorescence in the ganglia of the antenno-maxillary organ
(A, arrow), in the anterior (A) and posterior (B) spiracles (arrowheads), in the ring gland (A, asterisk) and in the hindgut (B, dot). (C–F) GFP
expression in the legs (C), wing veins (D) and reproductive tract (E,F) of phk-2-GFP transgenic flies. In males (E), GFP is expressed in the
ejaculatory bulb (arrow) and restricted areas of the seminal vesicles (arrowheads). In females (F), green fluorescence can be detected in the uterus
(arrow) and seminal receptacle (arrowhead). The hindgut and rectum are shown in bracket (E,F). (G,H) MALDI-TOF mass spectrometry analysis
of dissected ejaculatory bulb (G) and legs (H) showing expression of a 12.8 kDa molecule (arrow). (I) Fluorescent (hindgut) and nonfluorescent
(midgut) parts of the gut of phk-2-GFP transgenic flies were dissected, and used to extract mRNA. Expression of rp49 and phk-2 was monitored by
RT-PCR. For rp49, 25 cycles of amplification were performed. For phk-2, 35 cycles of PCR were performed, followed by 25 cycles on an aliquot
(2 lL) of this reaction with nested primers. (J,K) Summary of the expression pattern of phk-2 in noninfected larvae and adult flies.
Ó FEBS 2003 Drosophila humoral response to virus infection (Eur. J. Biochem. 270) 3403
pathways (Fig. 5B and data not shown). phk-3 is the only
member of the family expressed in the macrophage-like S2
cells. Furthermore, its expression is upregulated upon
treatment with LPS (Fig. 5C). Pherokines also display

interesting developmental expression patterns. Expression
of phk-3 is first detectable at the end of embryogenesis and
in larvae, and the highest expression level is observed in
white pupae (0–72 h). Expression decreases in black pupae
(72–96 h) and adults. phk-2 is also expressed in larvae and
pupae, with a strong peak of expression in black pupae.
Both phk-2 and phk-3 are more strongly expressed in male
than female adult flies. Finally, phk-1 expression starts in
black pupae, when the development of olfactory sensilla is
essentially complete, and remains constant in adult flies
(Fig. 5D). Consistent with this regulated expression pattern
during metamorphosis, we observed that treatment of S2
cells with the molting steroid hormone 20-hydroxyecdysone
completely suppresses phk-3 expression (Fig. 5C). By con-
trast, 20-hydroxyecdysone treatment strongly potentiates
the immune-inducibility of the gene encoding the antibac-
terial peptide Diptericin, as described previously [21].
Discussion
Antiviral response in
Drosophila
Our results reveal striking differences in the response of
Drosophila to infection with the virus DCV compared to
bacteria or fungi. Indeed, one hallmark of the response to
bacterial or fungal infections is the inducible secretion into
the hemolymph of a cocktail of antimicrobial peptides [4,5].
In addition, a large number of Drosophila immune-induced
molecules are also induced in the hemolymph following
septic injury [13,14,29]. By contrast, none of these molecules
are induced upon DCV infection at the time points
analyzed, and we only identified a single induced molecule,

Phk-2, in the hemolymph of DCV-infected flies. Constitu-
tive overexpression of Phk-2 does not protect flies against a
DCV challenge, suggesting that it is not directly involved in
the antiviral response. Rather, this molecule may be
involved in tissue-repair, or in the behavior of infected flies
(see below). In agreement with these biochemical data, we
observed that Dif (Toll pathway) and key (Imd pathway)
mutant flies exhibit the same sensitivity to DCV infection as
wild-type flies (data not shown). Importantly, bacterial
challenge 48 h after the injection of DCV led to normal
induction of antimicrobial peptide genes, indicating that the
Toll and Imd pathways in fat body cells are not affected by
DCV infection, at least in the first three to four days of
infection. Altogether, these experiments suggest that the
host-defense mechanisms against virus infection are differ-
ent from the mechanisms operating during bacterial or
fungal infections in flies. In future work, it will be interesting
to compare the response of Drosophila to other types of
viruses such as Sigma virus [30], to confirm that the
pathways regulating antibacterial and antifungal responses
Fig. 4. Constitutive expression of Phk-2 in the hemolymph of transgenic
flies is not sufficient to protect them against DCV infection. (A) MALDI-
TOF mass spectrometry analysis of the hemolymph from a single
female fly containing the UAS-phk-2 transgene and the fat-body spe-
cific yolk-Gal4 driver (left panel). The analysis of the hemolymph from
a control female fly containing the UAS-phk-2 transgene but not the
driver is shown in the right panel. (B) Flies of the indicated genotypes
and gender (the yolk promoter is only active in the fat body of female
flies) were infected with DCV (10
4.5

LD
50
), and survival was moni-
tored daily. Two independent experiments are shown. Flies constitu-
tively expressing Phk-2 in the hemolymph are indicated with squares,
andcontrolflieswithcircles.Transgeniclineswereestablishedinaw
–
(w) background (see Experimental procedures). Note that the genetic
background of these flies differs from the Oregon-R flies used in Fig. 1,
which explains the different susceptibility to DCV infection (our
unpublished data).
Fig. 5. Expression of phk genes in response to infection and during
development. (A) phk-3 Transcripts are transiently upregulated fol-
lowing infection with a mixture of Gram-positive and Gram-negative
bacteria. Drosomycin was used as a positive control and rp49 as a
loading control. (B) ywDD1; cnbw wild-type flies (WT), Dif (Toll
pathway) and key (Imd pathway) mutant flies were infected by septic
injury, and expression of phk genes was analyzed by Northern blot. (C)
phk-3 Expression is upregulated by LPS (+) and repressed by the
molting hormone ecdysone in S2 tissue-culture cells. (D) Develop-
mental expression profile of pherokines. Poly(A)+ RNA was extrac-
ted from embryos, third instar larvae, L(3), 0–72 h white pupae (w),
72–96 h black pupae (b), and male or female adults, and analyzed by
Northern blot using the indicated probes.
3404 L. Sabatier et al.(Eur. J. Biochem. 270) Ó FEBS 2003
differ from those activated by viral infection. It will also be
interesting to study flies infected through the respiratory or
digestive tracts [12].
Pherokines and chemosensation
We describe in this report two new molecules which are

induced by septic injury. Phk-2 is induced by DCV infection,
whereas Phk-3 is induced by bacterial challenge. The third
member of this family in Drosophila, Phk-1, is not induced
and is specifically expressed in the olfactory segment of
antennae [15,16]. Pherokines belong to a family of small
hydrophilic secreted peptides isolated from several insect
species on the basis of their tissue-specific expression in the
olfactory organs (e.g. [31–34]). They are characterized by
four cysteines involved in two disulfide bridges and forming a
CX
6
CX
18
CX
2
C signature motif, and differ from the mem-
bers of the major family of odorant-binding proteins in
Drosophila, which are characterized by six conserved cysteine
residues [35]. Based on their tissue-specific expression, these
molecules have been suggested to participate in sensing odors
and/or pheromones. Our data thus raise the provocative
prospect that the sensorial system may play a role in host-
defense in Drosophila, as previously reported in social insects
[36,37]. In another invertebrate, Caenorhabditis elegans,the
Toll receptor CeTol-1 was recently shown to participate in
chemosensory behavior, enabling worms to avoid ingestion
of pathogenic bacteria [38]. Pherokines may be involved in a
similar type of chemosensory behavior in Drosophila.
Another interesting possibility is that pherokines may
participate in the control of reproduction in Drosophila.

Indeed, the cabbage armyworm Mamestra brassicae mole-
cule MbraAOBP2, which shares 50% identity with Phk-2,
has been shown to bind the pheromone vaccenyl acetate [32].
Interestingly, phk-2 is expressed in the ejaculatory bulb of
Drosophila males, which contains cis-vaccenyl-acetate. This
pheromone is transferred by males to females during
copulation, and has an antiaphrodisiac effect on male
courtship [39]. Phk-2 may act as a carrier in this process.
Thus, the induction of Phk-2 by DCV infection may be
connected to modification of the fly’s reproduction dynam-
ics. This could represent an efficient host-defense strategy, as
DCV is not transmitted vertically. Importantly, DCV-
infected flies have been shown to have higher fecundity and
fertility than DCV-free animals [10,12]. However, we have so
far failed to detect changes in the reproductive dynamics of
flies overexpressing phk-2 (data not shown).
Pherokines and host-defense
The fact that one member of the family, namely Phk-1, may
function as an odor/pheromone-binding factor does not
necessarily imply that the other members exhibit similar
functions. In agreement with this possibility, we found that
phk-2 is expressed in many tissues not linked to chemo-
sensory functions. There is at least one report in which a
pherokine-related molecule has been isolated in a context
different from olfaction. In the larval stage, the cockroach
Periplaneta americana can regenerate lost tissues or organs
such as the eyes, the antennae or the legs. The P. americana
protein p10, which shares 50% identity with Phk-2, is
strongly and transiently upregulated during the regeneration
of the legs in larvae [40]. Thus pherokines may have a general

role in tissue remodeling in response to injury or in a
developmental context. In keeping with this hypothesis, we
have shown that the phk-2 and phk-3 genes are highly
expressed during metamorphosis in Drosophila. In addition,
we have shown that the phk-2 promoter is active in the ring
gland in larvae, a neuroendocrine center which produces the
hormones controlling molting, metamorphosis, reproduc-
tion and organ growth. Finally, our finding that phk-3 is
downregulated by ecdysone treatment in S2 cells was recently
confirmed in a genome-wide analysis of steroid-induced cell
death which showed that expression of both phk-2 and phk-3
is strongly reduced by ecdysone in vivo [41]. These data
support the hypothesis that Phk-2 and Phk-3 may interact
with ligands different from Phk-1, and carry other functions
than chemosensation. Similar observations were made in
mammals, where some odor-binding proteins, which are
specifically expressed in olfactory epithelia, are structurally
related to molecules involved in the binding and transport of
other molecules. This is the case, for example, for OBP, which
belongs to the same structural family as the retinol-binding
protein and the cholesterol-binding protein apoD [42], or of
RYA3, which exhibits significant sequence homology to the
LPS-binding protein [43]. This latter example finally raises
the possibility that all pherokines, including Phk-1, serve a
primary defense function by recognizing and/or neutralizing
invading microorganisms. The openings of the chemosen-
sory sensillae clearly represent an easy entry for microbes,
and mechanisms to maintain sterility of the sensillar fluid are
likely to exist, possibly including expression of phk-1.The
fact that the antenno–maxillary complex, which mediates

olfaction in larvae, expresses two antimicrobial peptides
upon exposure of larvae to bacteria confirms the existence of
host-defense mechanisms associated with olfactory tissues in
Drosophila [17].
In summary, we have identified a family of molecules that
are expressed in a regulated manner during infection and
development. Some members of this family are expressed in
a tissue-specific manner in olfactory organs, where they may
function as odor- or pheromone-binding molecules. Other
members may function as ligand-binding molecules for
other factors regulating tissue repair or remodeling. Future
studies using the powerful genetics of Drosophila will help to
clarify the exact physiological roles of pherokines. Our data
further suggest that the response to virus infection involves
mechanisms different from those operating to control
bacterial or fungal infections.
Acknowledgements
WewouldliketothankRene
´
Lanot for help with the microscopy
analysis; Estelle Santiago for expert technical assistance; Sebahat
Ozkan for help with transgenesis; Xavie
`
re Jousset and Max Bergoin for
providing virus stocks and much useful advice in the early stages of this
project and Liliane Gloeckler and Anne-Marie Aubertin for assistance
in producing virus stocks; Dominique Ferrandon and Jules Hoffmann
for critical reading of the manuscript and stimulating discussions. This
project was funded by CNRS, Entomed, as well as a grant from the
Ministe

`
re de la Recherche et de la Technologie (ACI Physiologie
Inte
´
grative). EJ was supported by a postdoctoral fellowship from the
Ligue contre le Cancer. CD is supported by a fellowship from the
Ministe
`
re de la Recherche du Grand-Duche
´
de Luxembourg.
Ó FEBS 2003 Drosophila humoral response to virus infection (Eur. J. Biochem. 270) 3405
References
1. Girardin, S.E., Sansonetti, P.J. & Philpott, D.J. (2002)
Intracellular vs extracellular recognition of pathogens – common
concepts in mammals and flies. Trends Microbiol. 10, 193–199.
2. Hoffmann, J.A., Kafatos, F.C., Janeway, C.A. & Ezekowitz, R.A.
(1999) Phylogenetic perspectives in innate immunity. Science 284,
1313–1318.
3. Kimbrell, D.A. & Beutler, B. (2001) The evolution and genetics of
innate immunity. Nat. Rev. Genet. 2, 256–267.
4. Hoffmann, J.A. & Reichhart, J.M. (2002) Drosophila innate
immunity: an evolutionary perspective. Nat. Immunol. 3, 121–126.
5. Tzou, P., De Gregorio, E. & Lemaitre, B. (2002) How Drosophila
combats microbial infection: a model to study innate immunity
and host–pathogen interactions. Curr. Opin. Microbiol. 5,
102–110.
6. Akira, S., Takeda, K. & Kaisho, T. (2001) Toll-like receptors:
critical proteins linking innate and acquired immunity. Nat.
Immunol. 2, 675–680.

7. Imler, J. & Hoffmann, J.A. (2001) Toll receptors in innate
immunity. Trends Cell Biol. 11, 304–311.
8. Alexopoulou, L., Holt, A.C., Medzhitov, R. & Flavell, R.A.
(2001) Recognition of double-stranded RNA and activation of
NF-kappaB by Toll-like receptor 3. Nature 413, 732–738.
9. Johnson, K.N. & Christian, P.D. (1998) The novel genome
organization of the insect picorna-like virus Drosophila Cvirus
suggests this virus belongs to a previously undescribed virus
family. J. Gen. Virol. 79, 191–203.
10. Gravot, E., Thomas-Orillard, M. & Jeune, B. (2000) Virulence
variability of the Drosophila C virus and effects of the micro-
parasite on demographic parameters of the host (Drosophila
melanogaster). J. Invertebr. Pathol. 75, 144–151.
11. Jousset, F.X. & Plus, N. (1975) [Study of the vertical transmission
and horizontal transmission of ÔDrosophila melanogasterÕ and
ÔDrosophila immigransÕ picornavirus (author’s transl)]. Ann.
Microbiol (Paris). 126, 231–249.
12. Thomas-Orillard, M. (1988) Interaction between a picornavirus
and a wild population of Drosophila melanogaster. Oecologia 75,
516–520.
13. Bulet, P. & Uttenweiler-Joseph, S. (2000) A MALDI-TOF mass
spectrometry approach to investigate the defense reactions in
Drosophila melanogaster, an insect model for the study of innate
immunity. Proteome and Protein Analysis (Kamp, R.M.,
Kyriakidis, D. & Choli-Papadopoulos, T., eds), pp. 157–174.
Springer, Berlin.
14. Uttenweiler-Joseph, S., Moniatte, M., Lagueux, M., Van
Dorsselaer, A., Hoffmann, J.A. & Bulet, P. (1998) Differential
display of peptides induced during the immune response of Dro-
sophila: a matrix-assisted laser desorption ionization time-of-flight

mass spectrometry study. Proc. Natl Acad. Sci. USA 95, 11342–
11347.
15. McKenna, M.P., Hekmat-Scafe, D.S., Gaines, P. & Carlson, J.R.
(1994) Putative Drosophila pheromone-binding proteins expressed
in a subregion of the olfactory system. J. Biol. Chem. 269, 16340–
16347.
16. Pikielny, C.W., Hasan, G., Rouyer, F. & Rosbash, M. (1994)
Members of a family of Drosophila putative odorant-binding
proteins are expressed in different subsets of olfactory hairs.
Neuron 12, 35–49.
17. Tzou, P., Ohresser, S., Ferrandon, D., Capovilla, M., Reichhart,
J.M.,Lemaitre,B.,Hoffmann,J.A.&Imler,J.L.(2000)Tissue-
specific inducible expression of antimicrobial peptide genes in
Drosophila surface epithelia. Immunity 13, 737–748.
18. Brand, A.H. & Perrimon, N. (1993) Targeted gene expression as a
means of altering cell fates and generating dominant phenotypes.
Development 118, 401–415.
19. Jung, A.C., Criqui, M.C., Rutschmann, S., Hoffmann, J.A. &
Ferrandon, D. (2001) Microfluorometer assay to measure the
expression of beta-galactosidase and green fluorescent protein
reporter genes in single Drosophila flies. Biotechniques 30, 594–8,
600–601.
20. Georgel, P., Naitza, S., Kappler, C., Ferrandon, D., Zachary, D.,
Swimmer,C.,Kopczynski,C.,Duyk,G.,Reichhart,J.M.&
Hoffmann, J.A. (2001) Drosophila immune deficiency (IMD) is a
death domain protein that activates the antibacterial response and
can promote apoptosis. Dev. Cell. 1, 503–514.
21. Dimarcq, J., Imler, J., Lanot, R., Ezekowitz, R., Hoffmann, J.,
Janeway, C. & Lagueux, M. (1997) Treatment of l (2) mbn Dro-
sophila tumorous blood cells with the steroid hormone ecdysone

amplifies the inducibility of antimicrobial peptide gene expression.
Insect Biochem. Molec. Biol. 27, 877–886.
22. Jousset, F.X., Bergoin, M. & Revet, B. (1977) Characterization of
the Drosophila Cvirus.J. Gen. Virol. 34, 269–283.
23. Reed, L. & Muench, H. (1938) A simple method of estimating fifty
percent endpoints. Am. J. Hygiene 27, 493–497.
24. Kussmann, M., Nordhoff, E., Rahbeck-Nielsen, H., Haebel, S.,
Rossel-Larsen, M., Jakobsen, L., Mirgorodskaya, E., Kroll-
Kristensen, A., Palm, L. & Roepstorff, P. (1997) Matrix-assisted
laser desorption/ionization mass spectrometry sample preparation
designed for various peptides and protein analytes. J. Mass
Spectrometry 32, 593–601.
25. Dyanov, H.M. & Dzitoeva, S.G. (1995) Method for attachment of
microscopic preparations on glass for in situ hybridization, PRINS
and in situ PCR studies. Biotechniques 18, 822–4, 826.
26. Boutros, M., Agaisse, H. & Perrimon, N. (2002) Sequential acti-
vation of signaling pathways during innate immune responses in
Drosophila. Dev. Cell. 3, 711–722.
27. De Gregorio, E., Spellman, P.T., Rubin, G.M. & Lemaitre, B.
(2001) Genome-wide analysis of the Drosophila immune response
by using oligonucleotide microarrays. Proc.NatlAcad.Sci.USA
98, 12590–12595.
28. Irving, P., Troxler, L., Heuer, T.S., Belvin, M., Kopczynski, C.,
Reichhart, J.M., Hoffmann, J.A. & Hetru, C. (2001) A genome-
wide analysis of immune responses in Drosophila. Proc. Natl Acad.
Sci. USA 98, 15119–15124.
29. Carte, L., Nukhet-Cavusoglu, N., Leize, E., Charlet, M., Bulet, P.
& Van Dorsselaer, A. (2001) De novo sequencing by nanoES-MSn
of an immune induced peptide of Drosophila melanogaster. Eur.
J. Mass Spectrometry 7, 399–408.

30. Contamine, D., Petitjean, A.M. & Ashburner, M. (1989) Genetic
resistance to viral infection: the molecular cloning of a Drosophila
gene that restricts infection by the rhabdovirus sigma. Genetics
123, 525–533.
31. Angeli, S., Ceron, F., Scaloni, A., Monti, M., Monteforti, G.,
Minnocci, A., Petacchi, R. & Pelosi, P. (1999) Purification,
structural characterization, cloning and immunocytochemical
localization of chemoreception proteins from Schistocerca
gregaria. Eur. J. Biochem. 262, 745–754.
32. Bohbot, J., Sobrio, F., Lucas, P. & Nagnan-Le Meillour, P. (1998)
Functional characterization of a new class of odorant-binding
proteins in the moth Mamestra brassicae. Biochem. Biophys. Res.
Commun. 253, 489–494.
33. Maleszka, R. & Stange, G. (1997) Molecular cloning, by a novel
approach, of a cDNA encoding a putative olfactory protein in the
labial palps of the moth Cactoblastis cactorum. Gene 202, 39–43.
34. Picimbon, J.F., Dietrich, K., Angeli, S., Scaloni, A., Krieger, J.,
Breer, H. & Pelosi, P. (2000) Purification and molecular cloning of
chemosensory proteins from Bombyx mori. Arch. Insect Biochem.
Physiol. 44, 120–129.
35. Graham, L.A. & Davies, P.L. (2002) The odorant-binding pro-
teins of Drosophila melanogaster: annotation and characterization
of a divergent gene family. Gene 292, 43–55.
3406 L. Sabatier et al.(Eur. J. Biochem. 270) Ó FEBS 2003
36. Moret, Y. & Schmid-Hempel, P. (2001) Immune defence in
bumble-bee offspring. Nature 414, 506.
37. Traniello, J.F., Rosengaus, R.B. & Savoie, K. (2002) The devel-
opment of immunity in a social insect: evidence for the group
facilitation of disease resistance. Proc. Natl Acad. Sci. USA 99,
6838–6842.

38. Pujol, N., Link, E.M., Liu, L.X., Kurz, C.L., Alloing, G., Tan,
M., Ray, K.P., Solari, R., Johnson, C.D. & Ewbank, J.J.
(2001) A reverse genetic analysis of components of the Toll
signaling pathway in Caenorhabditis elegans. Curr. Biol. 11,
809–821.
39. Brieger, G. & Butterworth, F.M. (1970) Drosophila melanogaster:
identity of male lipid in reproductive system. Science 167, 1262.
40. Kitabayashi, A.N., Arai, T., Kubo, T. & Natori, S. (1998)
Molecular cloning of cDNA for p10, a novel protein that increases
in the regenerating legs of Periplaneta americana (American
cockroach). Insect Biochem. Mol. Biol. 28, 785–790.
41. Lee, C.Y., Clough, E.A., Yellon, P., Teslovich, T.M., Stephan,
D.A. & Baehrecke, E.H. (2003) Genome-wide analyses of steroid-
and radiation-triggered programmed cell death in Drosophila.
Curr. Biol. 13, 350–357.
42. Snyder, S.H., Sklar, P.B. & Pevsner, J. (1988) Molecular
mechanisms of olfaction. J. Biol. Chem. 263, 13971–13974.
43. Dear,T.N.,Boehm,T.,Keverne,E.B.&Rabbitts,T.H.(1991)
Novel genes for potential ligand-binding proteins in subregions of
the olfactory mucosa. EMBO J. 10, 2813–2819.
Ó FEBS 2003 Drosophila humoral response to virus infection (Eur. J. Biochem. 270) 3407