ARTICLE IN PRESS

Available at www.sciencedirect.com

Developmental and Comparative Immunology (]]]]) ], ]]]–]]]

0145-305X/$ - see frodoi:10.1016/j.dci.20

�Corresponding aSciences, Instituteversity of Leeds, LS2fax: +44 113 343 2835

E-mail address: a

Please cite this artDev Comp Immuno

journal homepage: www.elsevier.com/locate/devcompimm

SHORT COMMUNICATION

Variation of success of Leptopilina boulardi inDrosophila yakuba: The mechanisms explored

Aurore Dubuffeta,�, Geraldine Douryb, Carole Labroussea,Jean-Michel Drezena, Yves Cartonc, Marylene Poiried

aUniversite Franc-ois-Rabelais, IRBI, Parc Grandmont, 37200 Tours, FrancebLaboratoire de Biologie des Entomophages, Faculte des Sciences, Universite de Picardie-Jules Verne, 33 rue Saint Leu,80039 Amiens, FrancecLaboratoire Evolution, Genomes et Speciation, CNRS, Avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, FrancedUMR UNSA-INRA ‘‘ROSE’’, Universite de Nice-Sophia Antipolis-INRA, 400 Route des Chappes, 06903 Sophia Antipolis, France

Received 10 July 2007; received in revised form 26 October 2007; accepted 27 October 2007

KEYWORDSParasitoid;Leptopilina boulardi;Drosophila yakuba;Encapsulation;Haemocytes;Venom gland;Virulence strategy;Intra-specificvariation ofimmunosuppression

nt matter & 200707.10.009

uthor. Present adof Integrative an9JT Leeds, UK. T.

urore_dubuffet@y

icle as: Dubuffet Al (2007), doi:10.10

AbstractDespite an increasing knowledge of insect immune defences and virulence strategies usedby parasitoids to escape them, the mechanisms underlying variation of success betweenparasitoid strains are still poorly understood. We have investigated this point using twolines of the parasitoid wasp Leptopilina boulardi that differ in virulence towardsDrosophila yakuba.By injecting oil drops in D. yakuba larvae parasitized by virulent ISy females and thendissecting the larvae at different times following injection, we demonstrate that the ISyline alters host encapsulation ability but only during the early parasitism period. Thiseffect is mimicked by injecting venom gland extracts, indicating that venom proteins arelikely involved in immunosuppression. By contrast, the ISm line, unsuccessful on D. yakuba,has no immunosuppressive effect. This variation in virulence may be explained by thestriking difference we report in haemocytic profiles between ISm- and ISy-parasitizedlarvae. We discuss our results in the light of our knowledge of the strategies evolved byLeptopilina species to counteract the D. melanogaster immune system as well as the roleof parasitoid venoms in intra-specific variation of parasitoid virulence.& 2007 Elsevier Ltd. All rights reserved.

Elsevier Ltd. All rights reserved.

dress: Faculty of Biologicald Comparative Biology, Uni-el.: +44 113 343 3076;

ahoo.fr (A. Dubuffet).

, et al. Variation of success of Lep16/j.dci.2007.10.009

1. Introduction

Diverse factors are known to influence the outcome ofinteractions between hosts and parasites, including hostimmunocompetence, parasite ability to circumvent the host

topilina boulardi in Drosophila yakuba: The mechanisms explored.

ARTICLE IN PRESS

A. Dubuffet et al.2

immune response (virulence ability), or specific interactionsbetween host and parasite genotypes [1,2]. Endoparasitoidwasps, which develop within and eventually kill theirarthropod host, are also under the influence of such sourcesof success variation [3].

Until the present, a lot of progress has been madeconcerning the mechanisms of insect immune defence, aswell as strategies used by parasitoid species to escape it(i.e. virulence strategies) [4,5]. Towards parasitoids,the insect immune response generally consists of theformation of a multicellular, melanized capsule around theintruder. In Drosophila larvae, the number of circulatinghaemocytes first increases following non-self recognition.Then, plasmatocytes and lamellocytes collaborate to form acellular sheath around the intruder, while the blackening ofthe cellular capsule requires the activation of the phenolox-idase cascade. Free oxygen radicals and quinones arebelieved to promote the final killing of the entrappedintruder [4,6]. Parasitoids use a huge variety of mechanismsto escape the encapsulation process, ranging from posses-sing surface features that prevent their recognition(immunoevasion) to altering specific components of thehost immune system (immunosuppression) [7]. In manycases, avoidance of the immune response is achieved thanksto maternal secretions injected into the host duringoviposition, such as venoms or polydnaviruses-containingovarian fluids [8,9].

If some virulence strategies used by successful parasitoidsare well described, little is known about the mechanismsunderlying intra-specific variations in virulence ability ofparasitoids [10–12]. Such variations, reported in severalspecies, are of great importance since they reflect adapta-tion of parasitoid populations to their local host species[13–15]. A thrilling objective, both for immunologistsand evolutionary biologists, is to bridge the gap betweenfield variations of parasite success and the underlyingphysiological and molecular mechanisms. In this study, weinvestigate the virulence strategy used by the parasitoidwasp Leptopilina boulardi towards the host Drosophilayakuba and highlight variations in immunosuppressiveeffects of two lines of this parasitoid that differ in theirvirulence ability.

2. Materials and methods

2.1. Insects

The two lines of L. boulardi, ISy (Gif stock, no. 486)and ISm (Gif stock, no. 431), were described previously [3].The ISm line derives from a single female collected inthe Nasrallah oasis (Tunisia) and the ISy line derives from asingle female originating from a mass culture of a Brazza-ville (Congo) population. They have opposite abilities toevade encapsulation by the 1880-D line of D. yakuba: all ISmeggs are encapsulated (n ¼ 87), whereas only 2275% of ISyeggs are encapsulated (n ¼ 81). They were reared on asusceptible D. melanogaster strain (Gif stock no. 1333) at25 1C. The D. yakuba reference strain used duringthe experiments is the isofemale line 1880-D, which wasraised at 25 1C.

Please cite this article as: Dubuffet A, et al. Variation of success of LepDev Comp Immunol (2007), doi:10.1016/j.dci.2007.10.009

2.2. Effects of parasitism on host encapsulationability

We investigated the effects of the two parasitoid lines onthe general encapsulation ability of their D. yakuba hosts.For this, we compared the dynamics of encapsulation of oildrops injected into second instar larvae parasitized byeither ISy or ISm females with that of oil drops injected inunparasitized larvae. An Eppendorf transjector (FemtoJets,needles Femtotips IIs, 0.7 mm outer diameter) was used,and a single paraffin oil drop (2–5 nl) was injected per larva.Patches of larvae were then placed at 25 1C for variousperiods (24, 30, 40 and 48 h) until dissection. The degrees ofencapsulation of the oil drops were then categorized asfollows: ‘absence of haemocytes’ (total absence of haemo-cytes at the surface of the drop), ‘presence of somehaemocytes’ (some haemocytes observed at the surface ofthe drop), ‘partial cellular encapsulation and spots ofmelanization’ (presence of a multilayered capsule coveringa large part of the foreign body, also partially melanized),‘completed capsule’ (cellular capsule totally covering thesurface of the drop and strongly melanized) (see Supple-mentary Material for illustrations).

2.3. Effect of the venom gland content onencapsulation

To determine whether the venom has immunosuppressiveeffects, we compared the encapsulation degree of oil dropsco-injected with venom from the virulent line ISy or theavirulent line ISm with that of oil drops co-injected with asaline solution. Injections of 10 nl paraffin oil together with10 nl of either 1 mg/ml venom gland extracts (0.015 waspgland-equivalent) or Ringer’s solution (182mM KCl; 46mMNaCl; 3mM CaCl2; 10mM Tris–HCl) were performed in 3-day-old D. yakuba larvae using a glass micro-needle of 25 mmdiameter. Injected larvae were then placed at 25 1C for 24 or44 h until dissection. Degrees of encapsulation of the oildrops were categorized as described above.

2.4. Haemograms

D. yakuba larvae parasitized by either ISy or ISm females andunparasitized controls were individually bled at 27 h post-parasitization (PP) (76-h-old larvae) or 48 h PP (96-h-oldlarvae) to collect their haemolymph on a Thoma haemocyt-ometer, as previously described [10]. The total number ofhaemocytes per mm3 of haemolymph (THC) and the numberof each type of haemocyte per mm3 (i.e. plasmatocytes,crystal cells and lamellocytes) were counted in 21–26 larvaeper treatment under a phase-contrast microscope. The ISmline is known to induce a modification of the shape oflamellocytes, which enables a successful immunosuppres-sion of the host species D. melanogaster [10,16]. For thisreason, we also identified and counted the lamellocyteswith an apparent abnormal shape (i.e. blunt or flared endsinstead of a normal disc shape) in D. yakuba larvae. Thepercentage of ‘bipolar lamellocytes’ was calculated in thelamellocyte population and this percentage was arcsinetransformed for statistical analyses [17]. ANOVA analyseswere performed, followed by pairwise comparisons using

topilina boulardi in Drosophila yakuba: The mechanisms explored.

ARTICLE IN PRESS

Variation of success of L. boulardi in D. yakuba 3

Fisher’s least significant difference (LSD) test, to determinewhether there were treatment-dependent differences inhaemocyte counts and percentages of bipolar lamellocytes(Software: Systast 10, SPSS Inc., Chicago, IL).

3. Results

3.1. Effects of parasitism and venom gland extractson encapsulation response

ISy parasitization has a significant effect on the encapsula-tion degree of oil drops, but the strength of this effectdeclined in the time-course of encapsulation process(Figure 1). Indeed, at 24 h post-injection (PI), the percen-tage of completely encapsulated drops in ISy-parasitizedlarvae was 2.3 times lower in comparison with unparasitizedlarvae while at 30 and 40 h PI, it was 1.7 and 1.4 times lower,respectively. Effects of parasitization were no longerstatistically significant at 48 h PI. By contrast, no effect ofISm parasitization on D. yakuba encapsulation towards oildrops has been detected, neither at 24 h PI nor at 48 h PI.

Venom extracts from the virulent ISy line were able toprotect oil drops from encapsulation (Figure 2), and, again,the strength of this effect declined in the time-course ofencapsulation process. At 24 h PI, the percentage ofcompletely encapsulated drops was a seventh of that inthe control but it was only a half at 44 h PI. Injection ofvenom gland extracts thus reproduced parasitism effects ofthe ISy line. As expected, venom extracts from the avirulentISm line were unable to protect oil drops from encapsula-tion, neither at 24 h PI nor at 44 h PI.

Figure 1 Encapsulation degrees of paraffin oil drops injected in D.line (P(ISy)) or the ISm line (P(ISm)) of L. boulardi. Larvae were disResults are expressed in percentages of oil drops. Values within bracThe different samples were compared using w2 contingency compar

Please cite this article as: Dubuffet A, et al. Variation of success of LepDev Comp Immunol (2007), doi:10.1016/j.dci.2007.10.009

3.2. Effects of ISy and ISm parasitization on cellularimmunity

Haemocyte profiles of unparasitized, ISy-parasitized andISm-parasitized larvae differed at both 27 and 48h PP (Figure 3).

At 27 h PP, the total haemocyte count (THC) wassignificantly higher in parasitized larvae compared withunparasitized controls. This difference corresponded to anincrease in both the number of plasmatocytes and lamello-cytes but not of crystal cells. However, the number oflamellocytes (P ¼ 0.001), plasmatocytes (Po0.001) and theresulting THC (Po0.001) were significantly lower inD. yakuba larvae parasitized by ISy females than in larvaeparasitized by ISm females. The most important effect wasobserved for plasmatocytes, whose number was twice lowerin ISy-parasitized larvae compared with ISm-parasitizedlarvae.

At 48 h PP, the THC of ISm-parasitized larvae waslower than at 27 h PP (Po0.001), and similar to theTHC of unparasitized larvae (P ¼ 0.340). In contrast, theTHC of ISy-parasitized larvae was higher (Po0.001)than at 27 h PP, an effect due to an increase in theplasmatocyte number only, and higher than that ofunparasitized larvae (Po0.001). ISy-parasitized larvae alsohad a higher THC, lamellocyte number and plasmatocytenumber than ISm-parasitized larvae (Po0.001; P ¼ 0.004;P ¼ 0.003, respectively) but a similar number of crystal cells(P ¼ 0.582).

Only a small percentage of bipolar lamellocytes wasobserved in each category (o5%), and this percentage wasnot statistically different between the different conditions(ANOVA: F ¼ 0.692; df ¼ 5; P ¼ 0.631).

yakuba larvae unparasitized (U) or parasitized by either the ISysected at different times after injection (24, 30, 40 and 48 h).kets indicate numbers of examined oil drops for each category.isons tests.

topilina boulardi in Drosophila yakuba: The mechanisms explored.

ARTICLE IN PRESS

Figure 2 Encapsulation degrees of paraffin oil drops co-injected in D. yakuba larvae together with a saline solution or with venomgland extracts from the ISy or ISm line of L. boulardi. Injected larvae were dissected at 24 or 44 h after injection. Results areexpressed in percentages of oil drops. Values within brackets indicate numbers of examined oil drops for each category. The sampleswere compared using w2 contingency comparisons tests.

A. Dubuffet et al.4

4. Discussion

4.1. Virulence strategy of the ISy line

Among the 50 known species of Drosophila parasitoids, thosebelonging to the genus Leptopilina, Asobara and Ganaspishave been studied most intensively for their virulencestrategies. Except for the species A. tabida, which embedsits eggs within the host tissues, all these parasitoids wereshown to alter the encapsulation ability of their hosts[18–21]. In Leptopilina species, this immunosuppressioninvolves products from the female venom gland [20,22,23]and is usually correlated with a change of lamellocytemorphology from a discoidal to a bipolar shape, whichdiminishes their ability to adhere and form a capsule[10,20,23]. These changes are still observed 48 h followingparasitism and, accordingly, parasitism by L. heterotomafemales protects from encapsulation avirulent parasitoideggs laid 30 h later in the same host [24].

Likewise, we report here that the ISy line of L. boulardiuses immunosuppression as a virulence strategy on the hostD. yakuba, and that venom gland products alter hostencapsulation ability. However, this immunosuppression isnot constant during all the time-course of parasitism. It isstrong at 24 h PP, but not observed anymore at 48 h PP. Thissuggests that the encapsulation process is either slowed ortransiently inhibited rather than totally suppressed. Asparasitoid eggs hatch within the recovery period (3271 hPP, data not shown), early immunosuppression might protect

Please cite this article as: Dubuffet A, et al. Variation of success of LepDev Comp Immunol (2007), doi:10.1016/j.dci.2007.10.009

the parasitoid eggs until completion of their embryonicdevelopment, where the parasitoid larvae might then use alocal immunoevasion strategy. Such a succession of virulencestrategies, evidenced for parasitoids of lepidopteran hosts[25,26], is suggested here for the first time for a parasitoidof Drosophila.

The strategy of the ISy line of L. boulardi also differs fromthe ‘usual’ virulence strategy of this genus since thepercentage of abnormal lamellocytes does not differbetween parasitized and unparasitized larvae neither inD. yakuba nor in D. melanogaster [10]. One interestingfeature is the fact that the ISm line of L. boulardi inducessuch lamellocyte modification in D. melanogaster, itssuitable host [10], but not in D. yakuba, in which it isencapsulated. Modifying the lamellocyte morphology seemsthus to be a strategy that has host-species specificity, themechanisms of which should be explored, and it is not astrategy shared by all Leptopilina parasitoids.

4.2. Intra-specific variation of virulence ability

Intra-specific variation in parasitism success has beenreported in some parasitoid species, but its underlyingmechanisms are rarely investigated. Here, we report thefirst insights of the mechanisms that might explain variationof success of L. boulardi on the host D. yakuba. We showthat the ISy line, which is not encapsulated by D. yakuba,delays or transiently alters the host encapsulation ability

topilina boulardi in Drosophila yakuba: The mechanisms explored.

ARTICLE IN PRESS

Figure 3 Total haemocyte count (THC), differential count of plasmatocytes, lamellocytes and crystal cells in unparasitizedDrosophila yakuba larvae and in larvae parasitized by ISy or ISm females of Leptopilina boulardi, at 27 and 48 h after parasitization.Mean values (7SE) are given for each category and numbers within brackets indicate numbers of replicates. ANOVA tests wereperformed for each cellular category and, when significant, followed by Fisher’s LSD test. Significant differences between means areindicated by different letters above the histogram bars.

Variation of success of L. boulardi in D. yakuba 5

whereas the unsuccessful ISm line does not. As thisdifference is observed at least from 24 h PP, a large part ofthe success of this parasitoid species is determined veryearly during its development.

The difference in immunosuppressive effects ofL. boulardi lines may be explained at least partly by adifferent cellular response to ISy or ISm infestation inD. yakuba larvae. Indeed, in ISm-parasitized larvae, theTHC first increases drastically in comparison to unparasi-tized larvae and then decreases to reach the same THC as inunparasitized larvae. This ‘up and down’ change in THC,similar to variations observed in D. melanogaster larvaeinfested by ISy or ISm females [10], may correspond to thecellular reaction ‘normally’ induced in Drosophila larvae.Interestingly, in D. yakuba larvae parasitized by ISy, theincrease of THC is less important than in ISm-parasitizedlarvae at 27 h PP and no decrease is observed at 48 h PP. Onthe contrary, the THC is more important at 48 h PP than at27 h PP, which suggests that ISy infestation delays thecellular response of D. yakuba larvae. This lag might thenexplain why oil drops are encapsulated at 48 h PP but notat 27 h PP.

The effect of ISy infestation is particularly strong onplasmatocytes, their number being twice lower inISy-parasitized larvae than in ISm-parasitized larvae at 27 h

Please cite this article as: Dubuffet A, et al. Variation of success of LepDev Comp Immunol (2007), doi:10.1016/j.dci.2007.10.009

PP but significantly increased between 27 and 48 h PP. Thesehaemocytes, known to be immune surveillance cells, are thefirst that adhere to foreign bodies [27,28]. Limiting thenumber of plasmatocytes may thus delay the first step ofcapsule formation. Obviously, ISy effects on other hostimmune components, such as phenoloxidase activity, willhave to be investigated in the future to determine if intra-specific differences in L. boulardi effects are restricted ornot to the host cellular immune response.

Crosses between the two parasitoid lines ISy and ISm havepreviously shown that the success of a parasitoid within ahost species is not determined by its own genotype but bythe genotype of its mother, suggesting that maternalsecretions greatly influence parasitoid success [29]. Ourpresent study clearly demonstrates the role of the venominjected by ISy females in early immunosuppression. As thevenom proteinic profiles of the two parasitoid lines are quitedifferent [30], we hypothesize that quantitative and/orqualitative differences in venoms may explain the variationsof success of L. boulardi on its Drosophila hosts. Suchassumption is at least verified for the variation of virulencetowards D. melanogaster. Indeed, a major venom virulencefactor used by the ISm line to alter the lamellocytemorphology and suppress the encapsulation response, aRhoGAP protein, is undetectable with antibodies in the

topilina boulardi in Drosophila yakuba: The mechanisms explored.

ARTICLE IN PRESS

A. Dubuffet et al.6

venom glands of the ISy line, avirulent on resistant D.melanogaster flies [30]. Moreover, differences in proteinicprofiles have also been observed in the calyx of braconidwasp biotypes that differ in their virulence ability towards alepidopteran host [31]. Further investigations will deter-mine if this feature can be generalized, and will state themolecular basis of these variations.

Acknowledgements

We are grateful to P. Eslin and G. Prevost for fruitfuldiscussions and to F. Frey for technical assistance. This workwas supported by the French GDR 2153. Aurore Dubuffetreceived financial support from the Franc-ois RabelaisUniversity (Tours).

Appendix A. Supporting Information

Supplementary data associated with this article can be foundin the online version at doi:10.1016/j.dci.2007.10.009.

References

[1] Schmid-Hempel P, Ebert D. On the evolutionary ecology ofspecific immune defence. Trends Ecol Evol 2003;18(1):27–32.

[2] Lambrechts L, Fellous S, Koella JC. Coevolutionary interactionsbetween host and parasite genotypes. Trends Parasitol 2006;22(1):12–6.

[3] Dubuffet A, Dupas S, Frey F, Drezen J-M, Poirie M, Carton Y.Genetic interactions between the parasitoid wasp Leptopilinaboulardi and its Drosophila hosts. Heredity 2007;98(1):21–7.

[4] Lemaitre B, Hoffmann J. The host defense of Drosophilamelanogaster. Annu Rev Immunol 2007;25(1):697–743.

[5] Pennacchio F, Strand MR. Evolution of developmental strategiesin parasitic hymenoptera. Annu Rev Entomol 2006;51:233–58.

[6] Carton Y, Nappi AJ. Drosophila cellular immunity againstparasitoids. Parasitol Today 1997;13(6):218–27.

[7] Schmidt O, Theopold U, Strand M. Innate immunity and itsevasion and suppression by hymenopteran endoparasitoids.BioEssays 2001;23(4):344–51.

[8] Moreau SJM, Guillot S. Advances and prospects on biosynthesis,structures and functions of venom proteins from parasiticwasps. Insect Biochem Mol Biol 2005;35:1209–23.

[9] Espagne E, Dupuy C, Huguet E, Cattolico L, Provost B, MartinsN, et al. Genome sequence of a polydnavirus: insights intosymbiotic virus evolution. Science 2004;306(5694):286–9.

[10] Russo J, Brehelin M, Carton Y. Haemocyte changes in resistantand susceptible strains of D. melanogaster caused by virulentand avirulent strains of the parasitic wasp Leptopilinaboulardi. J Insect Physiol 2001;47(2):167–72.

[11] Moreau SJ, Doury G, Giordanengo P. Intraspecific variation inthe effects of parasitism by Asobara tabida on phenoloxidaseactivity of Drosophila melanogaster larvae. J Invertebr Pathol2000;76(2):151–3.

[12] Gitau CW, Gundersen-Rindal D, Pedroni M, Mbugi JP, Dupas S.Differential expression of the CrV1 haemocyte inactivation-associated polydnavirus gene in the African maize stem borerBusseola fusca (Fuller) parasitized by two biotypes of theendoparasitoid Cotesia sesamiae (Cameron). J Insect Physiol2007;53:676–84.

[13] Dupas S, Carton Y, Poirie M. The genetic dimension of thecoevolution of virulence resistance in Drosophila–parasitoidwasp relationships. Heredity 2003;90(1):84–9.

Please cite this article as: Dubuffet A, et al. Variation of success of LepDev Comp Immunol (2007), doi:10.1016/j.dci.2007.10.009

[14] Kraaijeveld AR, Van Alphen JJ, Godfray HC. The coevolution ofhost resistance and parasitoid virulence. Parasitology1998;116(Suppl.):S29–45.

[15] Ngi-Song AJ, Overholt WA, Stouthamer R. Suitability ofBusseola fusca and Sesamia calamistis (Lepidoptera: Noctui-dae) for the development of two populations of Cotesiasesamiae (Hymenoptera: Braconidae) in Kenya. Biol Control1998;12(3):208–14.

[16] Labrosse C, Eslin P, Doury G, Drezen J-M, Poirie M. Haemocytechanges in D. melanogaster in response to long glandcomponents of the parasitoid wasp Leptopilina boulardi:a Rho-GAP protein as an important factor. J Insect Physiol2005;51(2):161–70.

[17] Anscombe FJ. The transformation of Poisson, binomial, andnegative binomial data. Biometrika 1948;35:246–54.

[18] Moreau SJM, Eslin P, Giordanengo P, Doury G. Comparativestudy of the strategies evolved by two parasitoids of the genusAsobara to avoid the immune response of the host, Drosophilamelanogaster. Dev Comp Immunol 2003;27(4):273–82.

[19] Chiu H, Sorrentino RP, Govind S. Suppression of the Drosophilacellular immune response by Ganaspis xanthopoda. Adv ExpMed Biol 2000;484:161–7.

[20] Morales J, Chiu H, Oo T, Plaza R, Hoskins S, Govind S.Biogenesis, structure, and immune-suppressive effects ofvirus-like particles of a Drosophila parasitoid, Leptopilinavictoriae. J Insect Physiol 2005;51(2):181–95.

[21] Rizki TM, Rizki RM, Carton Y. Leptopilina heterotoma and L.boulardi: strategies to avoid cellular defense responses ofDrosophila melanogaster. Exp Parasitol 1990;70(4):466–75.

[22] Labrosse C, Carton Y, Dubuffet A, Drezen JM, Poirie M. Activesuppression of D. melanogaster immune response by long glandproducts of the parasitic wasp Leptopilina boulardi. J InsectPhysiol 2003;49(5):513–22.

[23] Rizki RM, Rizki TM. Parasitoid virus-like particles destroyDrosophila cellular immunity. Proc Natl Acad Sci USA 1990;87:8388–92.

[24] Streams FA, Greenberg L. Inhibition of the defense reaction ofDrosophila melanogaster parasitized simultaneously by thewasps Pseudocoila bochei and Pseudocoila mellipes. J Inver-tebr Pathol 1969;13:371–7.

[25] Lavine MD, Beckage NE. Temporal pattern of parasitism-induced immunosuppression in Manduca sexta larvae para-sitized by Cotesia congregata. J Insect Physiol 1996;42(1):41–51.

[26] Yin L, Zhang C, Qin J, Wang C. Polydnavirus of Campoletischlorideae: characterization and temporal effect on hostHelicoverpa armigera cellular immune response. Arch InsectBiochem Physiol 2003;52:104–13.

[27] Russo J, Dupas S, Frey F, Carton Y, Brehelin M. Insect immunity:early events in the encapsulation process of parasitoid(Leptopilina boulardi) eggs in resistant and susceptible strainsof Drosophila. Parasitology 1996;112(Part 1):135–42.

[28] Williams MJ, Ando I, Hultmark D. Drosophila melanogasterRac2 is necessary for a proper cellular immune response. GenesCells 2005;10:813–23.

[29] Dupas S, Carton Y. Two non-linked genes for specific virulenceof Leptopilina boulardi against Drosophila melanogaster andD. yakuba. Evol Ecol 1999;13:211–20.

[30] Labrosse C, Stasiak K, Lesobre J, Grangeia A, Huguet E, DrezenJ-M, et al. A RhoGAP protein as a main immune suppressivefactor in the Leptopilina boulardi (Hymenoptera, Figitidae)–Drosophila melanogaster interaction. Insect Biochem Mol Biol2005;35:93–103.

[31] Gitau CW, Dupas S, Ngi-Song AJ, Mbugi JP, Schulthess F. Calyxfluid proteins of two Cotesia sesamiae (Cameron) (Hymenop-tera: Braconidae) biotypes in Kenya: implications to biologicalcontrol of the stem borer Busseola fusca (Fuller) (Lepidoptera:Noctuidae). Ann Soc Entomol Fr 2006;42(3–4):433–41.

topilina boulardi in Drosophila yakuba: The mechanisms explored.