Wolbachia in Parasitoids Attacking Native European and Introduced Eastern Cherry Fruit Flies in Europe

Abstract The eastern cherry fruit fly, Rhagoletis cingulata Loew (Diptera: Tephritidae), is an economically important pest of cherries in North America. In 1983 it was first reported in Europe where it shares its ecological niche with the native European cherry fruit fly, Rhagoletis cerasi L. (Diptera: Tephritidae). Their coexistence in Europe led to the recent horizontal transmission of the Wolbachia strain wCer1 from R. cerasi to R. cingulata. Horizontal Wolbachia transmission is mediated by either sharing of ecological niches or by interacting species such as parasitoids. Here we describe for the first time that two braconid wasps, Psyttalia rhagoleticola Sachtleben (Hymenoptera: Braconidae) and Utetes magnus Fischer (Hymenoptera: Braconidae), naturally parasitizing R. cerasi, use the invasive R. cingulata in Europe as a new host. In contrast, no parasitoids that parasitize R. cingulata in its native American range were detected in the introduced European range. Diagnostic Wolbachia PCR screening and sequence analyses demonstrated that all P. rhagoleticola individuals were infected with the newly described Wolbachia strain wRha while all U. magnus individuals were uninfected. wRha is different from wCer1 but had an Wolbachia surface protein (wsp) gene sequence that was identical to wCer2 of R. cerasi and wCin2 of R. cingulata. However, multi locus sequence typing revealed differences in all loci between wRha and the tephritid's strains. The horizontal transmission of wCer1 between the two tephritid species did not result in fixed heritable infections in the parasitoids. However, the parasitoids may have acted as a transient wCer1 vector.

The introduction of exotic insects into new environments poses a major threat to global biodiversity (Traveset and Richardson 2006). Although just a small proportion of introduced species may be established in any new environment and become invasive (Williamson 1996), human activity and globalization lead to an ever-growing number of translocated species (Mack et al. 2000, Levine andD'Antonio 2003). Once in new environments, alien species can influence native species and their ecosystems mainly by competition over resources (Dorcas et al. 2012), hybridization with native species (Fitzpatrick et al. 2010), and introduction of parasites and pathogens that can impact native species (Stout andMorales 2009, Meeus et al. 2011). Natural enemies are hypothesized to play a major role in biological invasions (Dunn et al. 2012). The release from natural enemies during the invasion process can allow fast adaptation and rapid spread of alien species within new environments (Keane and Crawley 2002). The lack of enemies in new habitats can facilitate alien species establishment (Torchin et al. 2003) and may enhance its ability to impact native species via competition (Tilman 1999). Conversely, invasive species can provide new food sources for natural enemies (Strauss et al. 2012). For example, invasive insect species may be a new resource for native parasitoids; however, only few studies have investigated this (Klug et al. 2008).
The European cherry fruit fly Rhagoletis cerasi L. (Diptera, Tephritidae) is the major pest of sweet and sour cherries (Prunus spp.) in Europe (Boller et al. 1970, Daniel andGrunder 2012). In addition to cherries, R. cerasi also infests berries of honeysuckle, Lonicera spp. Bush 1974, Schwarz et al. 2003). It occurs in all cherry-growing countries in Europe and is also present in Russia and western Asia (Fimiani 1989, Namin andRasoulian 2009). It has a univoltine life cycle with an obligatory diapause (Boller and Prokopy 1976). Females oviposit usually one egg per cherry, and emerging larvae feed in the pulp of the growing fruit for 4-10 wk and overwinter as pupae in the soil (Boller and Prokopy 1976). After oviposition, R. cerasi larvae and pupae are exposed to up to 12 parasitoid species (Hoffmeister 1990(Hoffmeister , 1992. Parasitization rates fluctuate significantly between different life stages of R. cerasi. Larvae can be attacked only by well-synchronized parasitoids, and pupae by more generalist parasitoids. Considering the short time of larval occurrence in the field, generalist pupal parasitoids do usually not require to be synchronized since pupae are available for most of the year (Hoffmeister 1990). Parasitization rates also differ between the two host plant genera and geographically (Hoffmeister 1990).
Recently, the North American eastern cherry fruit fly R. cingulata Loew (Diptera: Tephritidae) was introduced from North America into Europe (Lampe et al. 2005. It now co-occurs with R. cerasi in cherries, while it is not known to infest honeysuckle. Rhagoletis cingulata has a similar univoltine life cycle, with adult emergence from the overwintering pupal stage in the soil 3 wk after its European relative (Vogt et al. 2010). In Europe, R. cingulata was reported for the first time in Switzerland in 1983 and thereupon in Austria, Belgium, Croatia, France, Germany, Hungary, Italy, the Netherlands, and Slovenia (EPPO 2007(EPPO , 2010(EPPO , 2013(EPPO , 2014. However, while R. cingulata was mostly detected in Germany, Hungary, and the Netherlands, detection in other countries was infrequent (Smit andDijkstra 2008, Schuler et al. 2013).
Wolbachia is an alphaproteobacterium widespread in different insect taxa, several other arthropods and nematodes ). This endosymbiont is predominantly inherited maternally through the egg cytoplasm. Wolbachia strains are usually able to enhance their spread by mechanisms promoting the reproductive success of infected females, like parthenogenesis, male-killing, feminization, or the induction of cytoplasmic incompatibility (Engelst€ adter and Hurst 2009). However, phylogenetic incongruence between Wolbachia and host species demonstrated that this endosymbiont can occasionally move horizontally between species (O'Neill et al. 1992, Vavre et al. 1999, Kawasaki et al. 2016, and between individuals of a species (Kraaijeveld et al. 2011. Although the mechanisms of horizontal transmission in nature are not fully understood, sharing the same ecological niche and the interaction with natural enemies such as parasitoids and predators are expected to be key factors for horizontal transmission (Raychoudhury et al. 2009, Stahlhut et al. 2010, Gehrer and Vorburger 2012. Therefore, the introduction and establishment of a new species could result in an opportunity for horizontal transmission of endosymbionts between native and introduced species . Although Wolbachia is one of the best-studied endosymbionts, knowledge on its influence on invasive species is scarce (Nguyen et al. 2016). For example, prevalence of Wolbachia in the European paper wasp, Polistes dominulus, is similar in its native and introduced ranges, suggesting that this endosymbiont has been cointroduced (Stahlhut et al. 2006). In contrast, invasive populations of the Argentine ant Linepithema humile and the garden ant Lasius neglectus showed significantly lower Wolbachia prevalences than in their native ranges (Tsutsui et al. 2003, Cremer et al. 2008 or even endosymbiont loss (Reuter et al. 2004). A similar scenario was recently described for the loss of Wolbachia but not another insect endosymbiont, Cardinium, in the invasive range of a thrips species while native populations have both endosymbionts (Nguyen et al. 2016).
Currently, at least five Wolbachia strains (wCer1-wCer5) are described from R. cerasi (Riegler and Stauffer 2002, Augustinos et al. 2014. American populations of R. cingulata are only infected by one Wolbachia strain, wCin2, a strain that, based on available sequence information, appears identical to the wCer2 strain of R. cerasi. However, European populations of R. cingulata also have a second strain, wCin1 (Schuler et al. 2009. Characterization of the five multi locus sequence typing (MLST) genes and the Wolbachia surface protein (wsp) gene showed that this strain is identical to wCer1, a strain that is omnipresent in R. cerasi (Riegler and Stauffer 2002, Arthofer et al. 2011. While these data highlight that the shared ecological niche of the native and introduced Rhagoletis species resulted in a horizontal transmission of wCer1 from R. cerasi to R. cingulata within a short time period , the mechanism how this endosymbiont had crossed host species boundaries remained unclear.
Here we studied the Wolbachia infection of parasitoids attacking both Rhagoletis species. We examined whether the introduced fruit fly species R. cingulata in Europe is attacked by any parasitoids, either from North America or from its European congeneric R. cerasi. Furthermore, we tested whether these parasitoids are also infected by wCer1, and therefore constitute intermediate Wolbachia hosts with a potential role in horizontal wCer1 transmission between the native R. cerasi and the introduced R. cingulata in Europe.

Materials and Methods
Insect Sampling, DNA Extraction, DNA Barcoding, and Sequence Analysis Infested cherries were collected in four localities in Austria (AT: Prinzersdorf), the Czech Republic (CZ: Brno), and Germany (Ger1: Ingelheim; Ger2: Dresden). Infested fruits were transported to the laboratory where Rhagoletis larvae were allowed to emerge and pupate. In this way we obtained pupae of R. cerasi from Austria and the Czech Republic where R. cingulata is very rare (Egartner et al. 2010, EPPO 2014. From the two localities in Germany we obtained pupae of R. cingulata. Pupae were kept at diapause conditions of 4 C and 65-70% relative humidity for 5 mo (Vallo et al. 1976) followed by incubation at 24 C. Emerging parasitoids were collected directly after eclosion and stored in ethanol. They were then morphologically grouped into two morphospecies.
All emerged parasitoid individuals were DNA extracted using the GenElute Mammalian DNA Mini-Prep Kit (Sigma) according the manufacturer's protocol. The DNA was dissolved in 50 ml elution buffer and stored at 4 C. To confirm the morphological identification, we amplified and sequenced a 611 bp fragment of the mitochondrial cytochrome oxidase I (COI) gene, a region used for DNA barcoding using the primers LCO1490 and HCO2198 (Folmer et al. 1994). PCR was performed in 10 ml reactions containing 1Â NH 4 buffer, 2 mM MgCl 2 , 100 lM dNTPs, 0.2 lM of each primer, 0.2 U Taq polymerase (Thermo Scientific), and 1 ll of template DNA. PCR amplifications were performed in a 2720 thermal cycler (Applied Biosystems) with the following cycling conditions: 2 min at 94 C followed by 5 cycles at 94 C for 30 s, 45 C for 90 s, and 72 C for 1 min and 35 cycles at 94 C for 30 s, 51 C for 1 min, and 72 C for 1 min with a final extension at 72 C for 5 min. PCR products were then Sanger sequenced by Eurofins MWG Operon (Ebersberg, Germany). COI sequences were analyzed and a phylogenetic tree constructed with the inclusion of other parasitoid sequences from Rugman-Jones et al.

Wolbachia Characterization Using wsp and MLST
All parasitoids were screened for the presence of Wolbachia using the wsp primers 81F and 691R (Braig et al. 1998). Positive samples were further characterized for the five MLST genes gatB, coxA, hcpA, ftsZ, and fbpA (Baldo et al. 2006). PCR was set up using the same conditions as described above. Cycling conditions were 2 min denaturation at 95 C, followed by 35 cycles at 94 C for 30 s, 55 C for 45 s, and 72 C for 1 min, and a final extension at 72 C for 15 min. PCR conditions of all primer sets were identical except for ftsZ with an annealing temperature of 50 C. PCR products were then Sanger sequenced by Eurofins MWG Operon (Ebersberg, Germany).

Phylogenetic Analyses of Wolbachia
Sequence chromatograms were carefully screened for ambiguous sites to exclude the presence of multiple infections, edited manually, and assembled by CodonCode Aligner vers. 3.7 (Codon Code Corporation). To exclude PCR artifacts, all genotypes were confirmed by independent PCRs. DNA sequences were determined by BLAST analysis and Wolbachia sequences were compared with the MLST database (http://pubmlst.org/wolbachia). Obtained sequences were deposited in GenBank and the Wolbachia MLST database. The pairwise genetic distances between the different Wolbachia strains were calculated using the Kimura 2 parameter model in Mega 5.2 (Tamura et al. 2011).
The sequences of the five MLST genes were concatenated and aligned with the MLST sequences of wCer1, wCer2, wCer4, and wCer5 in R. cerasi (Arthofer et al. 2011), and wCin1 and wCin2 in R. cingulata . wCer3, a strain that arose from recombination between wCer2 and wCer5, is present in low titers and at low prevalence in R. cerasi ). Therefore, it was not included in this study. The optimal substitution model was determined by jModeltest (Posada and Crandall 1998), and maximum likelihood analyses of wsp and MLST alignments were performed using the Hasegawa-Kishino-Yano (HKYþGþI) substitution model in MEGA 5.2 (Tamura et al. 2011).

Identification of Parasitoids in R. cerasi and R. cingulata
In total, we investigated 28 parasitoid wasps that emerged from R. cerasi and R. cingulata from four different localities of three countries ( Table 1). The wasps were grouped into two morphospecies that were then identified as Psyttalia rhagoleticola Sachtleben (1934) and Utetes magnus Fischer (1958), both of the braconid subfamily of Opiinae. A total of 17 individuals were identified as P. rhagoleticola-seven in Austria, four in Czech Republic, and six in Ingelheim (Ger1). Eleven individuals were identified as U. magnus-five in Ingelheim (Ger1), five in Dresden (Ger2), and one in Austria (Table 1). Psyttalia rhagoleticola was collected exclusively from R. cerasi pupae in Austria and Czech Republic, and exclusively from R. cingulata pupae in both German populations. Utetes magnus was less common and found in one R. cerasi pupa in Austria and in ten R. cingulata pupae from both German populations.
The COI sequences of the 17 P. rhagoleticola individuals contained between one to six SNPs (over a length of about 600 bp) and were assigned to six haplotypes (GenBank KX503389-KX503394). These six haplotypes were monophyletic when compared with COI sequences of other Psyttalia species (Rugman-Jones et al. 2009), thereby confirming the species status based on morphological identification (Fig. 1). According to the phylogenetic analysis the closest relative was Psyttalia ponerophaga from Pakistan. The abundance of the P. rhagoleticola haplotypes varied, as they occurred as single individual (Psy4 and Psy8) or up to eight individuals (Psy1) ( Table 1). The most common haplotype, Psy1, was found in all three localities where P. rhagoleticola occurred. German P. rhagoleticola was most polymorphic and had four haplotypes found across six individuals (Psy1, Psy3, Psy4, Psy6). Three haplotypes (Psy3, Psy4, Psy6) were found exclusively in Germany. Austrian P. rhagoleticola had three different haplotypes (Psy1, Psy2, Psy5) across seven individuals. All three haplotypes were also found in the Czech Republic, while only Psy1 was shared with German populations. In contrast to the diverse P. rhagoleticola, all 11 U. magnus specimens shared a single haplotype (GenBank KX503395).

Characterization of Wolbachia in Parasitoids of Rhagoletis
Irrespective of host fruit fly species, all individuals of P. rhagoleticola were positive for Wolbachia, while Wolbachia was not detected in any specimens of U. magnus. Characterization of the wsp gene and the MLST loci revealed that all P. rhagoleticola individuals were infected by the same Wolbachia strain. This strain was named wRha according to . BLAST search of the wsp fragment revealed identical sequences for wRha and wCer2 in R. cerasi (Riegler and Stauffer 2002) and wCin2 in R. cingulata (Schuler et al. 2009; Fig. 2a). However, all wRha MLST alleles were different to the MLST alleles of wCer2 from R. cerasi (Arthofer et al. 2011) and wCin2 from R. cingulata , resulting in different phylogenetic placements of the strains (Fig. 2b). Pairwise distances between MLST alleles of wRha, wCer2, and wCin2 ranged from 0.012 on fbpA to 0.023 on hcpA (Table 2).
Furthermore, none of the wsp and MLST alleles of wRha were identical with wCin1, the strain horizontally acquired by R. cingulata from R. cerasi in Europe . The comparison of the wsp sequence of wRha with the other strains in R. cerasi, wCer4 and wCer5, showed a sequence divergence of 0.736 and 0.928. However, in three of the five MLST loci wRha and wCer4 were closely related, with sequence divergences ranging from 0.002 in ftsZ to 0.05 in fbpA. wRha and wCer5 were more divergent, with sequence divergences ranging from 0.103 in ftsZ to 0.281 in fbpA (Table 2). A comparison with the MLST database showed that the newly described wRha strain is a novel strain with unique ftsZ, coxA, and gatB alleles according to comparisons with alleles deposited in the MLST database. All sequences of wsp and the five MLST alleles were assigned to wRha and entered in GenBank (KX503396-KX503401) and the MLST database (ID 1826).

Discussion
The introduction and establishment of alien species in new environments is a significant threat to biodiversity worldwide (Mack et al. 2000). Many studies described the successful invasions of various species by focusing on the genetic (Mooney and Cleland 2001) and ecological dimensions (Davis 2009). Within the latter, many focused on direct effects of alien species on ecosystems, for example via predation or competition over resources, while fewer studies looked at native diversity impacting invasive species in their newly established invasive range. Here we studied parasitoids of the native European cherry fruit fly R. cerasi and the introduced North American eastern cherry fruit fly R. cingulata in Europe. We examined the parasitoids' Wolbachia infection status and potential role as vectors in the horizontal wCer1 transmission recently detected between their native and introduced host fruit fly species . While our analysis of four population samples did not detect any American parasitoids of R. cingulata in their introduced range in Europe, we found that two European parasitoids of R. cerasi, P. rhagoleticola and U. magnus, were able to attack and develop in the introduced congener. We also surveyed and characterized Wolbachia infections across the two trophic levels, and found Wolbachia in P. rhagoleticola but not in the other parasitoid species. This newly described strain, wRha, has a wsp gene sequence that is identical with wsp of wCer2 found in both fruit fly species but is different from wCer2 and other strains in all MLST genes. Therefore, the horizontal transmission of the Wolbachia strain wCer1 across both fruit fly species has not resulted in the establishment of identical infections in their parasitoids.
The success of establishment of introduced pest species is often correlated with the release from parasites and pathogens present in the native range of invasive hosts Power 2003, Torchin andMitchell 2004). In its native range in North America, R. cingulata is host to five different parasitoid species: Coptera cingulatae, Coptera occidentalis, Diachasma ferrugineum, Diachasmimorpha mellea, and Utetes frequens (Wharton and Marsh 1978, Muesebeck 1980, Rull et al. 2011, Hood et al. 2015, Hamerlinck et al. 2016. While Utetes attack egg or early larval stages, Diachasma and Diachasmimorpha oviposit into late instar larvae in fruits. Coptera species, however, attack their host after pupation in the soil. The introduction of R. cingulata to Europe could have been linked with a potential cointroduction of its natural enemies attacking either the egg or larval stages. However, we did not detect any American native parasitoids in R. cingulata in Europe, and thus the establishment of R. cingulata in Europe may have profited from a release from its natural American enemies prior or during its establishment in Europe. In contrast to previous studies that described 12 species of parasitoids in R. cerasi (Hoffmeister 1990(Hoffmeister , 1992, in this study we detected just two, P. rhagoleticola and U. magnus. The low diversity captured in our study may be due to our sampling strategy. In previous studies, Rhagoletis pupae were exposed to a wider parasitoid community in the field, including species that specifically target the fly puparium, while in our study we collected fruits with infested larvae that then pupated in a laboratory environment; therefore, we effectively excluded puparium parasitoids. Both P. rhagoleticola and U. magnus are parasitoids of the last larval instar of R. cerasi and complete their life cycle in the pupal stage of their host. Nine of the 12 described parasitoid species attack R. cerasi during the pupal stage in the soil (Hoffmeister 1990) and could therefore not be detected in our study. Furthermore, Hoffmeister (1990Hoffmeister ( , 1992 described that some parasitoids are associated exclusively to one of the two different host forms on cherry and honeysuckle: while eleven different parasitoid species were reared from Lonicera infesting flies, just six parasitoid species were collected from Prunus infesting flies. Similarly, Monaco (1984) found exclusively U. magnus in wild cherries (Prunus mahaleb), and no parasitoids in cultivated cherries while we found two parasitoid species in cultivated cherries.
The different biology of R. cerasi and R. cingulata could potentially influence the different occurrence of the two parasitoid species. While P. rhagoleticola was reared from both species, all except a single individual of U. magnus emerged from R. cingulata pupae. It has previously been found that adult U. magnus appears in the field $2 wk later than adult P. rhagoleticola (Hoffmeister 1990), and this could result in more parasitization of R. cingulata, which   , 2016, Vol. 0, No. 0 has its peak flight activity 3 wk after R. cerasi (Vogt et al. 2010). Furthermore, both parasitoids are generalists that attack larvae of different Rhagoletis species. The main host of P. rhagoleticola are R. cerasi and Myoleia lucida, a tephritid fly that also infests Lonicera fruits. In contrast, U. magnus mainly attacks Rhagoletis meigenii and Rhagoletis alternata developing in Berberis and Rosa species that usually fruit at least one month after cherries (Hoffmeister 1990(Hoffmeister , 1992. The late emergence of U. magnus adults may therefore constitute a preadaption to introduced R. cingulata that also occurs later in the season. Adults of P. rhagoleticola, however, are active for long enough time periods to attack larvae of both R. cerasi and R. cingulata. The detection of two shared parasitoid species in native R. cerasi and introduced R. cingulata may suggest that they are responsible for the horizontal transmission of the Wolbachia strain wCer1 from R. cerasi to R. cingulata . Laboratory studies demonstrated that the close interaction of parasitoids with their hosts could result in horizontal Wolbachia transmission. Parasitoids can have three different roles in horizontal transmission: 1) They can acquire Wolbachia from an infected host and then transmit it vertically to their offspring. For example, parasitic Leptopilina boulardi wasps acquired Wolbachia by attacking infested Drosophila pupae. Subsequently, the bacterium was vertically transmitted within the new host species (Heath et al. 1999). 2) Wolbachia can be transmitted from one parasitoid to another by attacking the same host. For example, Trichogramma larvae, developing in Wolbachiauninfected moths acquired the bacterium by super-and multiparasitism with infected larvae of the same (Huigens et al. 2000) or closely related parasitoid species (Huigens et al. 2004). Finally, 3) parasitoids can act as vectors by transmitting Wolbachia without being infected. For example, parasitic wasps attacking Wolbachia-infested Bemisia tabaci were able to transmit the bacterium with contaminated mouthparts and ovipositors to uninfected hosts but did not acquire it, as it was not detected in the parasitoid tissues and gonads (Ahmed et al. 2015).
Sequence analysis of the wsp gene of wRha of P. rhagoleticola indicated that it is identical to wsp of wCer2, currently spreading in R. cerasi in Europe Stauffer 2002, Schuler et al. 2016), and wsp of wCin2, omnipresent in R. cingulata across North America and Europe . The characterization of the MLST genes, however, demonstrated that wRha is genetically distinct from Wolbachia strains found in R. cerasi and R. cingulata. From this perspective, our data do not unequivocally support the hypothesis that parasitoids are responsible for the horizontal transmission of wCer1 from native R. cerasi to introduced R. cingulata. A previous study that described the horizontal transmission of Wolbachia via parasitoids in whiteflies showed that nonendogenous Wolbachia can be transmitted by parasitoids within 48 h following the parasitoids contact with host Wolbachia via ovipositor and mouthparts (Ahmed et al. 2015). In our study, we collected parasitoids directly after eclosion from their hosts and they did not have access to any new hosts prior to DNA extraction. Therefore, we cannot rule out that under field conditions the parasitoid species can transfer Wolbachia via their ovipositor or mouthparts. A follow-up study should therefore test if parasitoids can transmit Wolbachia after attacking infected hosts.
The strain wRha is genetically distinct from the Wolbachia strains of the fruit fly hosts from which P. rhagoleticola emerged. Therefore, it is an inherited infection of P. rhagoleticola and we can exclude it as a contamination or not inherited somatic infection present in P. rhagoleticola. However, the close genetic relationship of the Wolbachia strain found in P. rhagoleticola with Wolbachia strains in fruit fly hosts, in particular the identical wsp sequences, suggest a shared evolutionary history, potentially also involving recombination across MLST loci.
The acquisition of Wolbachia can influence the mitochondrial diversity of its host (Hurst and Jiggins 2005). Reproductive advantages of Wolbachia-infected individuals result in the spread of the associated mitochondrial genomes replacing the haplotypes of uninfected individuals . Therefore, Wolbachia-infected species are assumed to display fewer mitochondrial lineages than uninfected ones (Hurst and Jiggins 2005). However, we found that all 11 individuals of U. magnus that were not infected by Wolbachia shared a single mitochondrial haplotype, while 17 Wolbachia-infected P. rhagoleticola individuals showed high mitochondrial diversity with six different haplotypes. This could indicate either an ancient Wolbachia infection of P. rhagoleticola or that the wRha infection is a result of multiple horizontal transmission events into or within this host species. The incidence and prevalence of Wolbachia in one out of two tephritid parasitoid species in our study is similar to a previous study that detected Wolbachia in one out of two parasitoid species of Australian tephritids (Morrow et al. 2014). In contrast, a study about parasitoids of Malaysian tephritids reported all five tested braconid parasitoid species infected with Wolbachia (Muhamad et al. 2015).
In summary, our results showed that the two parasitoids P. rhagoleticola and U. magnus naturally attacking R. cerasi were adapted to introduced R. cingulata in Europe. While we could not detect Wolbachia in U. magnus, all individuals of P. rhagoleticola were infected by this endosymbiont. Characterization of the wsp and MLST loci showed that the parasitoid harbored a different Wolbachia strain than its Rhagoletis hosts. Although we cannot exclude a potential role of the parasitoids to have transiently vectored Wolbachia between the two Rhagoletis species, our data do not support the hypothesis that parasitoids are hosts for a Wolbachia strain that has been transferred between the two Rhagoletis species.