Heredity (2009) 103, 248–256 & 2009 Macmillan Publishers Limited All rights reserved 0018-067X/09 $32.00 www.nature.com/hdy ORIGINAL ARTICLE A new case of dependence in the genus Asobara: evidence for parthenogenesis induction in Asobara japonica

N Kremer1, D Charif1, H Henri1, M Bataille1, G Pre´vost2, K Kraaijeveld3 and F Vavre1 1Universite´ de Lyon, F-69000, Lyon; Universite´ Lyon 1; CNRS, UMR5558, Laboratoire de Biome´trie et Biologie Evolutive, F-69622, Villeurbanne, France; 2Laboratoire de Biologie des Entomophages (EA 3900), Universite´ de Picardie Jules Verne, F-80039, Amiens, France and 3Department of Ecology, Institute of Biology Leiden, Leiden University, Leiden, The Netherlands

Wolbachia is a maternally inherited bacterium that is widely oogenesis. However, they produced only sons, showing that distributed among , in which it manipulates the wAjap induces thelytokous parthenogenesis. Analyses of reproduction of its hosts. Although generally facultative for its mating behavior and offspring production of individuals from hosts, Wolbachia has recently become obligatory in Asobara Wolbachia-infected populations showed that while males tabida (: ) in which it is required for were still sexually functional, females no longer attract males, the completion of oogenesis. Here, we describe a new making Wolbachia an obligate partner for daughter produc- Wolbachia strain (wAjap) that is associated with the genus tion in thelytokous populations. The fact that Wolbachia has Asobara and infects Asobara japonica. wAjap was detected become independently obligatory in two species of the same in all female-biased populations of A. japonica found in the genus tends to show that dependence evolution can be main islands of Japan, but not in the arrhenotokous common and swift, although no clear benefit for the populations from the southern islands. Using phylogenetic parasitoid can be attributed to this dependence. Although analyses based on multi-locus sequence typing (MLST), we dependence should lead to co-divergence between Wolba- show that this strain is closely related to wAtab3 (the strain chia and its hosts, the very few cases of co-speciation required for oogenesis in A. tabida), even though they differ observed in host–Wolbachia associations question the on Wolbachia surface protein (WSP) and WO phage stability of these obligatory associations. sequences. Using antibiotic treatments, we show that cured Heredity (2009) 103, 248–256; doi:10.1038/hdy.2009.63; thelytokous females are not dependent on Wolbachia for published online 10 June 2009

Keywords: Wolbachia; Asobara; thelytokous parthenogenesis; dependence; symbiosis

Introduction isopods (Werren, 1997). Feminization of genetic males into functional neo-females is mainly observed in isopod Wolbachia (Anaplasmataceae) is a cytoplasmically inher- crustaceans (Bouchon et al., 1998), but also occurs in ited bacterium that infects a wide variety of hosts Lepidoptera and Hemiptera (Kageyama et al., 2002; ranging from arthropods (Werren, 1997) to nematodes Negri et al., 2006). Male killing independently appeared (Bandi et al., 2001). The invasion of this bacterium into in phylogenetically unrelated groups, such as populations is because of its ability to Lepidoptera and Coleoptera, in which high levels of manipulate host reproduction. Four major types of host competition between siblings occurs (Hurst and Jiggins, manipulation by Wolbachia are known: cytoplasmic 2000). PI is often associated with Wolbachia infection in incompatibility (CI), feminization, male killing and haplodiploid species, especially mites and Hymenoptera parthenogenesis induction (PI). These effects either (Huigens and Stouthamer, 2003). In uninfected haplodi- decrease the offspring production of uninfected females ploids, males develop from unfertilized haploid eggs, or bias the sex ratio of infected females toward females. whereas females develop from fertilized diploid eggs Both result in the enhanced transmission of Wolbachia,as (arrhenotoky). In contrast, only diploid females are Wolbachia is maternally transmitted through its host produced in thelytokous reproduction. When thelytoky oocytes. The more widespread effect is CI, which is is Wolbachia-induced, it usually results from the diploi- found in all major orders of , and also in mites and dization of non-fertilized eggs during the first mitotic division (Stouthamer et al., 1990; Stouthamer and Kazmer, 1994, but see Weeks and Breeuwer, 2001). Correspondence: N Kremer, Universite´ de Lyon, F-69000, Lyon; Universite´ All these reproductive manipulations are generally Lyon 1, CNRS, UMR5558, Laboratoire de Biome´trie et Biologie Evolutive, facultative to the host. However, in the case of Wolbachia- F-69622, Villeurbanne, France. E-mail: [email protected] induced thelytoky, any genes involved in sexual repro- Received 14 November 2008; revised 16 March 2009; accepted duction can accumulate mutations without being 3 May 2009; published online 10 June 2009 selected against. Thus, the sexual functionality of males Wolbachia dependence in the genus Asobara N Kremer et al 249 and females from thelytokous populations is expected (Japan) were maintained at 20 1C, under 70% relative to deteriorate over time, culminating in Wolbachia humidity and 12/12 light/dark (LD) cycle on Wolbachia- dependence for daughter production, which has been free D. melanogaster reared on axenic nutritive medium reported in various species (Zchori-fein et al., 1992, (David, 1962). In the following, T(w) and T(0) will refer 1995; Pijls et al., 1996; Arakaki et al., 2000; Pannebakker to infected and uninfected individuals from the et al., 2005). thelytokous population, respectively, and A(0) to unin- Another case of Wolbachia dependence occurs in the fected individuals from the arrhenotokous population. wasp Asobara tabida (Hymenoptera, Braconidae) in which Wolbachia is necessary for host oogenesis, rendering Molecular characterization of symbionts and host aposymbiotic females sterile (Dedeine et al., 2001). This Wasps were individually crushed for 30 s (25 Hz) in obligatory dependence is believed to be because of a loss 150 ml of 5% Chelex solution (Bio-Rad, Hercules, CA, of host control over oogenesis (Pannebakker et al., 2007). USA) using a TissueLyser (Qiagen, Hilden, Germany). Alternatively, it may be because of an antidote/poison Samples were kept at 56 1C overnight and then for 15 min mechanism imposing the ‘sterilization of aposymbiotic at 95 1C before centrifugation. sisters’ (Charlat and Mercot, 2001), similar to the PCRs were carried out in 25 ml volumes containing modification/rescue model of CI (Breeuwer and Werren, 1 Â Taq Buffer, 200 mM dNTP, 1.5 mM MgCl , 200 nM 1993). In this context, it is interesting to examine the 2 primers, 0.5 IU Taq DNA polymerase (Eurobio, Les Ulis, different species of the genus Asobara to (i) characterize France) and 2 ml of DNA solution. PCR was carried out the potential infections by Wolbachia and their effect on using the thermocycler PTC-100 (MJ Research Inc., host reproduction and (ii) determine whether obligatory Waltham, MA, USA). Thermal cycling conditions were dependence to Wolbachia occurs more widely in this 1 min at 95 1C, a total of 35 cycles of 30 s at 95 1C, 1 min at genus. the specific Tm (melting temperature for each primer set In the genus Asobara, A. citri and A. persimilis are shown in Supplementary file S1), 1 min 30 s at 72 1C, and uninfected (Dedeine et al., 2005). A. tabida is infected by a final extension step of 10 min at 72 1C. two CI-inducing bacterial strains (wAtab1 and wAtab2) The Asobara species were characterized by sequencing and a third one (wAtab3) that is essential for oogenesis the ITS2 gene (Internal Transcribed Spacer 2) and completion (Dedeine et al., 2001, 2004). A. rufescens is Wolbachia by sequencing a set of genes for MLST (Baldo partially infected by a strain closely related to the et al., 2006). In the case of A. tabida, which naturally CI-inducing strain wAtab1 (Dedeine et al., 2005). In a harbors three different Wolbachia strains, only the strain, recent survey of larval parasitoids of frugivorous wAtab3 that has been isolated was characterized Drosophilidae in Japan, Mitsui et al. (2007) established (Dedeine et al. 2004). In addition to MLST typing, the geographical distribution of A. japonica. They showed Wolbachia were characterized by sequencing the WSP that field sex ratios are typical of arrhenotokous gene (Wolbachia Surface Protein). The WO_B phage was reproduction in three populations on subtropical islands, characterized by sequencing the ORF7 gene (Gavotte whereas those of all five populations sampled on the et al., 2004). DNA sequences have been deposited in the main (northern) islands are highly female biased. This EMBL database under accession numbers FM872332– suggests that the latter could be infected by a reproduc- FM872352. tive manipulator. We tested for the presence of the other endosymbionts Here, we show that thelytokous populations of Arsenophonus, Cardinium, Rickettsia, Hamiltonella and A. japonica are infected by Wolbachia. We characterize Spiroplasma by diagnostic PCR (Supplementary file S1). this Wolbachia strain, called wAjap, using phylogenetic analyses based on multi-locus sequence typing (MLST), Wolbachia surface protein (WSP) and WO phage and Phylogenetic analysis determine the phenotype it induces on its host. We Additional MLST and WSP sequences used in this study show that parthenogenesis in northern populations of were derived from the Wolbachia MLST database 2-07-1 A. japonica is caused by Wolbachia. We then examine the (http://pubmlst.org/wolbachia/). MLST and WSP extent to which sexual function has decayed as a result of sequences are available for 37 Wolbachia strains in this prolonged parthenogenesis. Finally, we show that database, including 35 strains belonging to supergroup A. japonica is dependent on Wolbachia for female A or B (Baldo et al., 2006). Sequences are accessible from offspring production, but in contrast to A. tabida it can the GenBank database under accession numbers complete oogenesis in the absence of the symbiont. DQ842268–DQ842486. Characteristics of sequences used for phylogenetic analyses are summarized in Supplementary file S2. Materials and Methods Sequences were aligned using Muscle software (Edgar, 2004). Nucleotides from the four hyper variable regions Biological system (HVRs) (positions: 66–156, 192–288, 342–444, 498–596) Asobara japonica (Hymenoptera:Braconidae:) is were used for WSP reconstruction. With regard to the a solitary endoparasitoid laying its eggs into the first or nucleotide positions used for the other genes, see Baldo second instar larvae of Diptera and especially et al. (2006) for MLST genes and Gavotte et al. (2004) for species (Mitsui et al., 2007; Ideo et al., 2008). Wasps were ORF7 gene. The PhyML software (Guindon and Gascuel, trapped in the field using banana and natural fruits 2003) was used for tree reconstruction with 500 boot- found in some localities (see Mitsui et al. 2007). Wasps straps. Phylogenetic trees were built using maximum from mass-reared lines were kindly provided by MT likelihood (ML) inference, with the appropriate evolu- Kimura. A thelytokous population from Sapporo (Japan) tion model estimated with MrAIC.pl 1.4.3 (Nylander, and an arrhenotokous population from Amami-Oshima 2004). The best likelihood score was evaluated using the

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 250 Akaike Information Criterion. The models selected were on standard diet with or without antibiotics for T(w) GTR þ I þ G for WSP, and the concatenated MLST gene females to produce T(0) and T(w) females, respectively. data set, according to Baldo et al. (2006). With regard to In the third generation, we carried out three different ITS2 and ORF7, the evolution models selected for the treatments (30 replicates each) with four females allowed reconstruction were GTR þ G. All trees were drawn to oviposit on 150 Drosophila larvae on standard diet: (i) using the Dendroscope software (Huson et al., 2007). virgin T(w) females as a control treatment, (ii) virgin T(0) females and (iii) T(0) females with T(0) males (mating allowed for 48 h). For each treatment, the male and Antibiotic treatments female offspring were counted and the sex ratio scored Cured A. japonica were obtained by antibiotic treatments as the percentage of females. as in Dedeine et al. (2001). Briefly, 150 mlofa2% rifampicin (Hoechst) solution were added to 1.5 g of Influence of Wolbachia on oocyte load Drosophila standard diet (2 mg g–1) on which 70 Drosophila eggs were placed. As parasitoid larvae feed on the Asobara females are mainly pro-ovogenic and thus emerge hemolymph of their Drosophila hosts, the antibiotic was with most of their oocytes. The oocyte load of thelytokous transmitted to the developing wasps, thereby rendering and arrhenotokous females was estimated in standard them aposymbiotic. As individuals were exposed to the conditions or after antibiotic treatment (that is, T(w), A(0), antibiotic during their larval development, embryos T(0) and A(0)* females, respectively). Newly emerged were still infected when their sex was determined. females were kept for 5 days with water and honey to Hence, adult parasitoids emerging after this treatment allow the completion of oocyte maturation. One ovary was were uninfected females, that is, T(0) females for dissected in PBS and gently squashed between the slide thelytokous populations that produced T(0) males when and the cover glass to disperse its content. Oocytes were allowed to reproduce as virgins. then counted under the microscope (AxioCam Imager Z.1, Zeiss, Oberkochen, Germany). Oocyte load was estimated for 20 ovaries for each treatment. Influence of Wolbachia on parasitoid offspring production and sex ratio Influence of Wolbachia on host reproduction and sexual To investigate whether A. japonica is dependent on behavior Wolbachia for oogenesis and the production of daughters T(0) males obtained after antibiotic treatments were in thelytokous populations, we employed the cross- placed together with T(0) females (in mass). Males and generational treatment outlined in Figure 1a. In the first females were observed for more than 1 h during which generation, T(w) females were reared with or without we scored male courtship (wing vibrations), female antibiotics to obtain T(0) and T(w) females, respectively. receptivity (females stand motionless when males try to Both types of females were allowed to oviposit as virgins: copulate) and matings. on standard diet for T(0) females to produce T(0) males, In addition, we looked for the presence of spermato- zoids in the testes of T(0) males. Testes were dissected in A-buffer ([KCl] ¼ 25 mM, [MgCl2] ¼ 10 mM, [Sucro- ¼ M ¼ M ¼ T(0) se] 250 m , [Tris] 35 m ,pH 7.5), treated with x glacial acetic acid 60% (3 min), fixed with Carnoy and 1 T(w) T(w) A(0) mounted in 20 ml of Vectashield HardSet containing –1 0 +ab 1.5 mgml DAPI (4 -6-Diamidino-2-phenylindole, Vector Laboratories, Burlingame, CA, USA). Preparations were T(0) observed using a fluorescence microscope (AxioCam Imager Z.1, Zeiss) at the excitation wavelength of 350 nm. T(0) x 2 T(0) T(w) A(0) To evaluate the degree of sexual degradation, T(w), +ab T(0) and A(0) females were crossed either with A(0) or T(0) males (Figure 1b). To test for possible effects of the T(0) antibiotic treatment, we also tested A(0) females that x were treated with antibiotics (A(0)*). Ten couples (one 3 T(0) T(0) A(0) A(0) male with one female) of each combination were observed for 5 min to (i) detect possible courtship or mating behaviors and (ii) determine the latency to T(0) courtship (if any) evaluated as the time from the start ? ? ? 4 A(0)* x of the exposure to the first wing vibration by the male. A(0) Presence of a female in the progeny was checked to Figure 1 Experimental procedures to determine the phenotypic determine the sexual functionality of the different effect of Wolbachia on A. japonica.(a) Characterization of phenotypic crosses. effect of Wolbachia. Offspring production and sex-ratio of each cross were estimated at the fourth generation. (b) Evaluation of sexual degradation. Sexual behavior, latency to courtship, and mating Results efficiency were estimated for each cross. Full symbols indicate Wolbachia-infected individuals (w). Population of origin is indicated Highly female-biased sex ratio is associated with infection by A (arrhenotokous population from Amami–Oshima) or T by Wolbachia in A. japonica populations (thelytokous population from Sapporo). Full arrows indicate standard treatment; dotted arrows indicate antibiotic treatment Individuals from female-biased populations were all (ab). Star corresponds to an antibiotic treatment carried out on an infected by Wolbachia (Sapporo (n ¼ 6 females tested), uninfected female A(0). Hirosaki (n ¼ 5), Sendai (n ¼ 3), Tokyo (n ¼ 5) and

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 251 Kagoshima (n ¼ 5)), whereas individuals from non- lighting the incongruence between MLST and WSP- biased populations were never infected (Amami-Oshima based phylogenies, as already shown in Baldo et al. 2005. (n ¼ 8) and Iriomote-Jima (n ¼ 10)). Sequencing WSP Analysis of the entire proteic WSP sequence of the four showed infection by a single Wolbachia strain, which species clustered in MLST allowed detection of muta- we called wAjap. Diagnostic PCR for other classical tions at sites under positive selection (Jiggins et al., 2002) endosymbionts showed no amplification for Arsenopho- and different potential recombination events between nus, Cardinium, Rickettsia, Hamiltonella and Spiroplasma. hypervariable regions (Supplementary file S3). In addition, we checked for the presence of WO_B phages and detected two different phages (Supplemen- Phylogenetic position of wAjap tary file S4). The Ajap 2 phage was also closely related to The phylogenetic relationship between wAjap and other the phage Atab_C, known to infect wAtab3, again Wolbachia strains was first assessed by using concate- suggesting that wAjap and wAtab3 might have a recent nated MLST loci (Figure 2). The Wolbachia strain present common ancestor. However, the other phage of wAjap in A. japonica (wAjap) clusters with Wolbachia strains and the two others of wAtab3 are not closely related, infecting other parasitoids of Diptera (A. tabida, Musci- suggesting some recombination or phage transfer events difurax uniraptor and Nasonia vitripennis). such as in WSP. Furthermore, we analyzed Wolbachia strains using the Finally, the ITS2 gene was sequenced to determine the WSP gene (Figure 3). The phylogenetic tree indicates that phylogenetic relationships among the different species of wAjap was neither directly associated with the Wolbachia Asobara. The phylogenetic tree reconstructed indicates strains of A. tabida nor with the strains of M. uniraptor that A. japonica is more related to A. persimilis than to the and N. vitripennis (which are grouped together), high- other Asobara species (Figure 4). The topology of this tree

0.01

wAjap wDneo wAtab3 wKueA wCpen wDori wSinvA wNvitA wRi 66 wDbif wIsny wUni 61 65 97 100 63 wNgir wAspa wNlon 55 wmel 79 90 wAlbA 84 wAu wDinn wDrec

wClec 97 100 wBm

100 wGfir wAepo wNvitB

wAenc 88 93 wMa 100 wNo 63 wTtai wKueB wOsca wVul wCpipq wPip 91 wTcon wEfor wCalt wTdei wPsia Figure 2 Maximum Likelihood inference phylogeny based on the concatenated sequences of five genes (MLST) of 39 Wolbachia strains. Numbers at tree nodes represent bootstrap support values (500 replicates) higher than 50. The codes of Wolbachia strains correspond to the name of their host (see Supplementary file S2 for more details).

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 252 0.1 wAalbA wDori wIsny wNvitA wAtab1 wDneo 99 wUni 100 72 wDrec wAjap 86 wLhet3 wDbif wMel wDinn 94 wCpen wAtab3 wAu wNgir wAtab2 wLhet2 wNlon

wLhet1 wRi 100 wKueA 54 wAspa 55 wBm

97

99 wCqui wNo wCalt wPip wMa 91 wVul wNvitB 96 75 wKueB 92 wLcla wSinvA 70 wTcon wTtai wGfir wEfor wPsia wAepo wOsca wAenc wClec Figure 3 Maximum Likelihood inference phylogeny based on WSP sequences of 44 Wolbachia strains. Numbers at tree nodes represent bootstrap support values (500 replicates) higher than 50. The codes of Wolbachia strains correspond to the name of their host (see Supplementary file S2 for more details).

À7 was congruent with those obtained with 18S gene (data the treatment (ANOVA; treatment, F1,76 ¼ 40.23, Po10 ; not shown). strain, F1,76 ¼ 0.03, P ¼ 0.85; Interaction, F1,76 ¼ 2.55, P ¼ 0.11). Consequently, Wolbachia is not necessary for oogenesis in A. japonica as opposed to A. tabida. Wolbachia induces thelytokous parthenogenesis All together, these results show that Wolbachia only in A. japonica induces thelytokous parthenogenesis in A. japonica. To determine the effect of Wolbachia on A. japonica more Finally, T(0) females produced only sons even when precisely, we treated females from a female-biased put together with T(0) males. This can be explained by population with antibiotics. As shown in Figure 5, T(0) the complete absence of courtship and mating as females produced only males, whereas T(w) females observed during the first hour of contact between them. produced mostly females (SR ¼ 0.99±0.01). There was no In these populations, Wolbachia is thus obligatory for influence of the infection status on the total offspring daughter production. production (Kruskal–Wallis rank-sum test; w2 ¼ 4.31, d.f. ¼ 2, P-value ¼ 0.12), suggesting that the variation in sex ratio was not because of preimaginal mortality. Females have lost the ability to sexually reproduce, Furthermore, quantification of oocyte load showed a not males slight reduction in T(0) females (T(w) ¼ 61.40±1.14 vs To estimate the magnitude of sexual degradation in T(0) ¼ 54.35±1.26, t ¼ 9.7, d.f. ¼ 18, Po0.001), but this males and females originating from thelytokous popula- variation was also observed after antibiotic treatment of tions, we reciprocally crossed arrhenotokous and A(0) females (A(0) ¼ 63.50±1.71 vs A(0)* ¼ 51.70±1.75, thelytokous males and females (treated or not with t ¼ 11.80, d.f. ¼ 18, Po 0.001), suggesting a direct effect of antibiotics) (Figure 1b, n ¼ 10 for each cross). Females

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 253 0.01 A. japonica (Japan)

A. persimilis (Sydney, Australia) A. persimilis (Adelaïde, Australia)

100

100 A. citri (Ivory Coast)

100

100 A. rufescens (Wervicq, France) 100 A. tabida (Pierrefeu, France) A. tabida (Kos, Greece) A. tabida (Seattle, USA) Figure 4 Maximum Likelihood inference phylogeny based on ITS2 sequences of five species of the genus Asobara. Numbers at tree nodes represent bootstrap support values (500 replicates) higher than 50.

Virgin infected Virgin uninfected Uninfected females T(0) females T(w) females T(0) with males T(0)

100 SR: 0.99 ±0.01 100 SR: 0 ±0.0 100 SR: 0 ±0.0 80 80 80

60 60 60

40 40 40

20 20 20 Offspring production 0 0 0 Males Females Males Females Males Females Figure 5 Influence of Wolbachia on female offspring production. Offspring production of (a) infected T(w) females (n ¼ 28) (b) virgin uninfected females T(0) (n ¼ 30) (c) uninfected females T(0) in presence of males T(0) (n ¼ 30). Box-and-whisker plot shows the extreme of the lower whisker, the lower hinge, the median, the upper hinge, and the extreme of the upper whisker for each group.

treated and not treated with antibiotics showed the same vibration and sometimes complete mating sequence, as 2 pattern regarding courtship behavior (ANOVA, efficiently as A(0) males (Figure 6a, w(2;0.05) ¼ 3, P ¼ 0.22). F1,27 ¼ 0.03, P-value ¼ 0.88), showing that antibiotic treat- Latency to courtship was also not significantly different ment and infection status do not affect the measured between A(0) and T(0) males (ANOVA, F1,27 ¼ 0.22, traits. These data were then pooled for a graphical P-value ¼ 0.64, Figure 6b). Furthermore, numerous representation (Figure 6a). T(w) or T(0) females put spermatozoids in differentiation were observed by DAPI together with T(0) and A(0) males never showed court- staining in the lumen of the testis (Supplementary file S5) ship behavior, showing that thelytokous females were in T(0) males. Finally, A(0) females crossed with T(0) not able to reproduce sexually. In contrast to females, males produced daughters (average proportion males T(0) males courted A(0) females by showing wing 0.27, n ¼ 3), although relatively fewer than A(0) males

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 254 No courtship Wing vibration Mating 100% 90% 250 80% 200 70% 60% 150 50% 40% 100 30% 20% 50 10% 0% Latency to courtship (sec) 0 T(w) T(w) A(0) A(0) A(0) A(0)

T(0) A(0) T(0) A(0) T(0) A(0)

Figure 6 Influence of Wolbachia on sexual behavior degradation. (a) Courtship behavior between arrhenotokous (A) and thelytokous (T) individuals: no courtship behavior (white), wing vibrations (gray), mating (black). (n ¼ 2 Â 10 for each crossing; antibiotic-treated and untreated females pooled). (b) Latency for thelytokous and arrhenotokous males to court an arrhenotokous female (in seconds). (n ¼ 6 þ 9, 8 þ 6, respectively; antibiotic-treated and untreated females pooled). Population of origin is indicated by A (arrhenotokous population from Amami–Oshima) or T (thelytokous population from Sapporo). Box-and-whisker plot shows the extreme of the lower whisker, the lower hinge, the median, the upper hinge and the extreme of the upper whisker for each group.

(average proportion males 0.17, n ¼ 6, glm with binomial the male–killer, wBol1, was shown to induce CI when error: deviance ¼ 6.06, P ¼ 0.014). Hence, males from infecting male-killing resistant hosts (Hornett et al., 2008). infected populations are still sexually functional, In contrast to the pattern observed using MLST whereas females are not. reconstruction, genetic variations were observed in other parts of the Wolbachia genome (WSP and bacteriophage WO), showing that this genome is a mosaic of regions Discussion evolving at very different rates. If genes involved in phenotypic effects are rapidly evolving, this could also Phylogenetic relationships between Wolbachia strains, explain why strains, which are closely related based on host spectrum and induced phenotype MLST, induce different effects. The close relationship among these four Wolbachia In addition to A. japonica, 66 Hymenoptera species have strains is also particularly interesting as all of them are been reported to be infected by a PI-microorganism. Of parasitoids of Diptera. It has been suggested that these, 46 were infected by Wolbachia (Huigens and horizontal transfer of Wolbachia is more efficient among Stouthamer, 2003). However, PI-Wolbachia do not form closely related species (Bouchon et al., 1998), which could a monophyletic group within the Wolbachia phylogeny partly explain why these parasitoids are infected with (review in Huigens and Stouthamer, 2003), and the same related Wolbachia strains. However, these parasitoids pattern was found using MLST (Baldo et al., 2006). belong to very different families of Hymenoptera and On the basis of MLST genes, wAjap was phylogeneti- thus other factors, such as sharing of similar hosts, may cally close to the Wolbachia strain of A. tabida necessary explain this pattern. This would illustrate how the for oogenesis completion (wAtab3). Similarly, one of the combination of ecological connections and phylogenetic two phages of wAjap was closely related to a phage of relatedness of hosts can shape the efficiency of horizontal wAtab3. These phylogenetic associations might reflect transmission of Wolbachia. a shared ancestry between A. japonica and A. tabida, although they are not the closest species of the Asobara genus. Alternatively, a recent horizontal transfer event Induction of parthenogenesis by Wolbachia and its could have occurred as A. japonica and A. tabida are consequences for sexual reproduction ecologically related and share some of the same Infected females produced female offspring without Drosophila hosts. Our data do not allow us to differentiate having mated and aposymbiotic females produced only between these hypotheses. males. Both findings indicate that Wolbachia induces In any case, these closely related strains induce very parthenogenesis in A. japonica. Populations of A. japonica different effects (PI, oogenesis completion), indicating in the subtropical islands are sexual, whereas popula- rapid evolution of an induced phenotype. This result tions in the main Japanese islands are asexual is reinforced by the phylogenetic association between (Mitsui et al., 2007). We have shown that this situation the wAjap and the Wolbachia strains of M. uniraptor corresponds to the presence of a PI-Wolbachia in (wUni) and N. vitripennis (wNvitA), inducing PI and CI, the parthenogenetic populations. In addition, reciprocal respectively. Hence, transitions among phenotypes can crosses between individuals from these conspecific occur between closely related strains. This could reflect a allopatric populations allowed us to show that host effect, as previously described in the Wolbachia (i) thelytokous males induced by antibiotics from strain of Ostrinia scapularis inducing feminization in its parthenogenetic populations mate with females from natural host, but causing male killing when transferred sexual populations and produce daughters; (ii) females in Ephestia kuehniella (which is naturally infected by a from parthenogenetic populations are never courted by CI-inducing Wolbachia strain) (Fujii et al., 2001). Similarly, males. The latter finding suggests a reduction/absence of

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 255 sexual pheromone production in thelytokous females failure of the thelytokous females to elicit male courtship. similar to the situation described in Trichogramma In such populations, Wolbachia is necessary for female cordubensis (Silva and Stouthamer, 1997). Given that offspring production. Consequently, A. japonica is females from thelytokous populations from which dependent on Wolbachia in these populations, but not Wolbachia had been removed still failed to attract in populations free of infection. The existence within the matings, the unattractiveness must be because of the same species of Wolbachia-free individuals with indivi- changes in the wasp genome, rather than a direct effect of duals that are dependent on Wolbachia for reproduction Wolbachia infection. It follows that only thelytokous illustrates how rapidly this situation can evolve. In females are no longer capable of sexual reproduction. addition, A. japonica is the second species in the genus Wolbachia is thus obligatory for daughter production in Asobara showing dependence on Wolbachia, the other infected populations, therefore rendering the wasp being A. tabida, which depends on Wolbachia for oogen- completely dependent on its symbiont. esis (Dedeine et al., 2001). It is interesting to note that Wolbachia-induced thelytokous parthenogenesis is these two strains are phylogenetically close, but the generally fixed in natural populations, except in some mechanisms underlying this dependence are completely rare cases in which the effect of Wolbachia is counter- different and have clearly evolved independently. Again, balanced, notably by the presence of other selfish this illustrates that dependence on Wolbachia can quickly elements (Stouthamer et al., 2001; Huigens and Stoutha- evolve and may occur frequently. mer, 2003). So far, it has not been possible to restore In these two cases, Wolbachia symbiosis is necessary for sexual reproduction in lines derived from populations in the host but does not provide a new additional function which thelytoky is fixed (review in Pannebakker, 2004). that would potentially be beneficial to the host, as An explanation for this pattern could be that when reproductive functions pre-existed Wolbachia infection. thelytokous parthenogenesis reaches fixation, genes Therefore, the relationship between the wasp and its involved in sexual reproduction are no longer under Wolbachia strain seems to be an obligatory dependence selection and can accumulate mutations (Carson et al., derived from a parasitic association rather than a real 1982; Zchori-Fein et al., 1992). However, sexual decay mutualistic association. Growing evidence that depen- seems stronger in females than in males in A. japonica, dence can evolve rapidly without having a clear as in the three other species in which asexual infected mutualistic effect sheds new light on the initial stage of and sexual uninfected populations occur (Apoanagyrus numerous associations between insects and bacteria. diversicornis, Pijls et al., 1996; Telenomus nawai, Arakaki Whether mutualism has evolved secondarily to depen- et al., 2000; Leptopilina clavipes, Pannebakker et al., 2005). dence is difficult to show in ancient associations, but it is This cannot be explained by neutral accumulation of suggested by these examples in which recent evolution mutations, because these should affect both sexes of dependence is not beneficial to the host. equally. Additional hypotheses have been proposed. When complete interdependence has evolved, co- First, some traits involved in female sexual reproduction speciation between host and symbiont is expected. are likely to be expressed in thelytokous populations. However, almost all cases of host dependence on Such traits could be counter-selected if there is a cost to Wolbachia are recent at the evolutionary time, and only maintain them, like pheromone production (Pijls et al., very few cases of co-speciation occur in host–Wolbachia 1996). Second, fixation of Wolbachia and loss of sexual associations such as in filarial nematodes (Fenn and reproduction could be explained by the ‘virginity Blaxter, 2004). This could suggest that although evolution mutation’ hypothesis: in a population partly infected of host dependence can occur frequently and rapidly, by Wolbachia, mutations allowing females to produce these associations are not stable at the evolutionary time more males (by preventing mating or egg fertilization) scale, either because interdependence is reversible, or will spread rapidly in the population until the mutation because these interactions are more prone to extinction. and the infection are fixed (Huigens and Stouthamer, 2003; Jeong and Stouthamer, 2005). These mutations can be selected for two reasons. First, infected thelytokous Acknowledgements females are still supposed to be able to be fertilized by We thank H Mitsui and MT Kimura for supplying males and produce females through sexual reproduction different populations of A japonica, G Gueguen for in the beginning of the association, as already shown in having confirmed by PCR the absence of other endo- different Trichogramma species in which both thelytokous symbionts and BA Pannebakker for helpful comments and arrhenotokous females coexist (Stouthamer and on the paper. This work was supported with funds Kazmer, 1994). Second, infected females produce males from UMR CNRS 5558, IFR 41, ANR EndoSymbArt and when transmission of Wolbachia is not perfect. These two GDR 2153. factors allow gene flow between infected and uninfected individuals in the population. In both scenarios, sexual decay in females is selected for, which could lead to a References rapid loss of the ability to reproduce sexually, and a swift evolution toward dependence on Wolbachia in these Arakaki N, Noda H, Yamagishi K (2000). Wolbachia-induced populations. parthenogenesis in the egg parasitoid Telenomus nawai. Entomol Exp Appl 96: 177–184. Baldo L, Lo N, Werren JH (2005). Mosaic nature of the Wolbachia Independent evolution of Wolbachia dependence in the surface protein. J bacteriol 187: 5406–5418. Baldo L, Dunning Hotopp JC, Jolley KA, Bordenstein SR, Biber genus Asobara SA, Choudhury RR et al. (2006). Multilocus sequence typing In populations of A. japonica in which wAjap is fixed, system for the endosymbiont Wolbachia pipientis. Appl Environ sexual reproduction is no longer possible because of the Microbiol 72: 7098–7110.

Heredity Wolbachia dependence in the genus Asobara N Kremer et al 256 Bandi C, Dunn AM, Hurst DD, Rigaud T (2001). Inherited Jeong G, Stouthamer R (2005). Genetics of female functional microorganisms, sex-specific virulence and reproductive virginity in the parthenogenesis-Wolbachia infected parasi- parasitism. Trends parasitol 17: 88–94. toid wasp Telenomus nawai (Hymenoptera: Scelionidae). Bouchon D, Rigaud T, Juchault P (1998). Evidence for Heredity 94: 402–407. widespread Wolbachia infection in isopod crustaceans: Jiggins FM, Hurst GD, Yang Z (2002). Host-symbiont conflicts: molecular identification and host feminization. Proc Biol Sci positive selection on an outer membrane protein of parasitic 265: 1081–1090. but not mutualistic Rickettsiaceae. Mol Biol Evol 19: Breeuwer JAJ, Werren JH (1993). Effect of genotype on 1341–1349. cytoplasmic incompatibility between 2 species of Nasonia. Kageyama D, Nishimura G, Hoshizaki S, Ishikawa Y (2002). Heredity 70: 428–436. Feminizing Wolbachia in an insect, Ostrinia furnacalis Carson HL, Chang LS, Lyttle TW (1982). Decay of female sexual (Lepidoptera: Crambidae). Heredity 88: 444–449. behavior under parthenogenesis. Science 218: 68–70. Mitsui H, Van Achterberg K, Nordlander G, Kimura MT (2007). Charlat S, Mercot H (2001). Did Wolbachia cross the border? Geographical distributions and host associations of larval Trends Ecol Evol 16: 540–541. parasitoids of frugivorous Drosophilidae in Japan. J Nat Hist David J (1962). A new medium for rearing Drosophila in axenic 41: 1731–1738. conditions. Drosophila Info Serv 36: 128. Negri I, Pellecchia M, Mazzoglio PJ, Patetta A, Alma A (2006). Dedeine F, Bouletreau M, Vavre F (2005). Wolbachia requirement Feminizing Wolbachia in Zyginidia pullula (Insecta, Hemi- for oogenesis: occurrence within the genus Asobara (Hyme- ptera), a leafhopper with an XX/XO sex-determination noptera, Braconidae) and evidence for intraspecific variation system. Proc Biol Sci 273: 2409–2416. in A. tabida. Heredity 95: 394–400. Nylander JAA (2004). MrAIC.pl. Program distributed by the Dedeine F, Vavre F, Fleury F, Loppin B, Hochberg ME, author, Evolutionary Biology Centre, Uppsala University, Bouletreau M (2001). Removing symbiotic Wolbachia bacteria Sweden. specifically inhibits oogenesis in a parasitic wasp. Proc Natl Pannebakker BA (2004). Evolutionary Consequences of Wolbachia- Acad Sci USA 98: 6247–6252. Induced Parthenogenesis in the Parasitoid Leptopilina Clavipes, Dedeine F, Vavre F, Shoemaker DD, Bouletreau M (2004). Intra- Leiden University, The Netherlands. individual coexistence of a Wolbachia strain required for host Pannebakker BA, Schidlo NS, Boskamp GJ, Dekker L, van oogenesis with two strains inducing cytoplasmic incompat- Dooren TJ, Beukeboom LW et al. (2005). Sexual functionality ibility in the wasp Asobara tabida. Evolution Int J Org Evolution of Leptopilina clavipes (Hymenoptera: Figitidae) after 58: 2167–2174. reversing Wolbachia-induced parthenogenesis. J Evol Biol 18: Edgar RC (2004). MUSCLE: multiple sequence alignment with 1019–1028. high accuracy and high throughput. Nucleic Acids Res 32: Pannebakker BA, Loppin B, Elemans CP, Humblot L, Vavre F 1792–1797. (2007). Parasitic inhibition of cell death facilitates symbiosis. Fenn K, Blaxter M (2004). Are filarial nematode Wolbachia Proc Natl Acad Sci USA 104: 213–215. obligate mutualist symbionts? Trends Ecol Evol 19: 163–166. Pijls JWAM, van Steenbergen HJ, van Alphen JJM (1996). Fujii Y, Kageyama D, Hoshizaki S, Ishikawa H, Sasaki T (2001). Asexuality cured: the relations and differences between Transfection of Wolbachia in Lepidoptera: the feminizer of the sexual and asexual Apoanagyrus diversicornis. Heredity 76: adzuki bean borer Ostrinia scapulalis causes male-killing in 506–513. the Mediterranean flour moth Ephestia kuehniella. Proc Biol Sci Silva IMMS, Stouthamer R (1997). To mate or not to mate. Can 268: 855–859. sex pheromones be used as a taxonomic tool in Trichogramma Gavotte L, Vavre F, Henri H, Ravallec M, Stouthamer R, spp? Proc Exp Appl Entomol NEVAmsterdam 8: 41–46. Boule´treau M (2004). Diversity, distribution and specificity of Stouthamer R, Kazmer DJ (1994). Cytogenetics of microbe- WO phage infection in Wolbachia of four insect species. Insect associated parthenogenesis and its consequences for gene Mol Biol 13: 147–153. flow in Trichogramma wasps. Heredity 73: 317–327. Guindon S, Gascuel O (2003). A simple, fast, and accurate Stouthamer R, Luck RF, Hamilton WD (1990). Antibiotics cause algorithm to estimate large phylogenies by maximum like- parthenogenetic Trichogramma (Hymenoptera/Trichogram- lihood. Syst Biol 52: 696–704. matidae) to revert to sex. Proc Natl Acad Sci USA 87: Hornett EA, Duplouy AM, Davies N, Roderick GK, Wedell N, 2424–2427. Hurst GD et al. (2008). You can’t keep a good parasite down: Stouthamer R, van Tilborg M, de Jong JH, Nunney L, Luck RF evolution of a male-killer suppressor uncovers cytoplasmic (2001). Selfish element maintains sex in natural populations incompatibility. Evolution 62: 1258–1263. of a . Proc Biol Sci 268: 617–622. Huigens ME, Stouthamer R (2003). Parthenogenesis associated Weeks AR, Breeuwer JA (2001). Wolbachia-induced partheno- with Wolbachia. In: Bourtzis K, Miller TA (eds). Insect genesis in a genus of phytophagous mites. Proc Biol Sci 268: Symbiosis, vol. 1. CRC press: Boca Raton, FL, USA. pp 247– 2245–2251. 266. Werren JH (1997). Biology of Wolbachia. Annu Rev Entomol 42: Hurst GDD, Jiggins FM (2000). Male-killing bacteria in insects: 587–609. mechanisms, incidence, and implications. Emerg Infect Dis 6: Zchori-Fein E, Faktor O, Zeidan M, Gottlieb Y, Czosnek H, 329–336. Rosen D (1995). Parthenogenesis-inducing microorganisms Huson DH, Richter DC, Rausch C, Dezulian T, Franz M, Rupp R in Aphytis (Hymenoptera: Aphelinidae). Insect Mol Biol 4: (2007). Dendroscope: an interactive viewer for large phylo- 173–178. genetic trees. BMC Bioinformatics 8: 460. Zchori-Fein E, Roush RT, Hunter MS (1992). Male production Ideo S, Watada M, Mitsui H, Kimura MT (2008). Host range of induced by antibiotic treatment in Encarsia formosa (Hyme- Asobara japonica (Hymenoptera: Braconidae), a larval para- noptera, Aphelinidae), an asexual species. Experientia 48: sitoid of drosophilid flies. Entomol Sci 11: 1–6. 102–105.

Supplementary Information accompanies the paper on Heredity website (http://www.nature.com/hdy)

Heredity