Environmental Microbiology Reports (2016) 00(00), 00–00 doi:10.1111/1758-2229.12425

Wolbachia endosymbionts in haplodiploid and diploid scolytine (Coleoptera: : Scolytinae)

Yuuki Kawasaki,1*† Hannes Schuler,2,3 revealed different Wolbachia strains in scolytines. Christian Stauffer,2 Ferenc Lakatos4 and Comparisons between the and Wolbachia phy- Hisashi Kajimura1 logenies revealed that closely related beetles were 1Graduate School of Bioagricultural Sciences, Nagoya infected with genetically different Wolbachia strains. University, Furo, Chikusa, Nagoya, 464-8601, Japan. These results suggest the horizontal transmission of 2Department of Forest and Soil Sciences, Boku, multiple Wolbachia strains between scolytines. We University of Natural Resources and Life Sciences, discuss these results in terms of the evolution of dif- Vienna, Austria. ferent sex determination systems in scolytine beetles. 3Department of Biological Sciences, University of Notre Dame, Galvin Life Science Building, Notre Dame, IN 46556, USA. Introduction 4Institute of Silviculture and Forest Protection, University of West Hungary, Sopron, H-9400, Hungary. Sex is mostly determined by sex chromosomes, but an alternative system in ants, wasps and bees is haplodiploidy (Cook, 1993). True haplodiploidy is defined as females Summary developing as diploids from fertilized eggs and males as Haplodiploidy is a sex determination system in which haploids from unfertilized ones. A second type of haplodi- fertilized diploid eggs develop into females and ploidy is paternal genome elimination (PGE) (Haig, 1993) unfertilized haploid eggs develop into males. The in which males develop from fertilized eggs but the paternal evolutionary explanations for this phenomenon chromosomes are eliminated during the early ontogenetic include the possibility that haplodiploidy can be rein- stages. PGE is frequently treated as functional haplodi- forced by infection with endosymbiotic bacteria, ploidy (Brun et al., 1995a). In both types of haplodiploidy, such as Wolbachia. The subfamily Scolytinae con- only maternal genome sets are transmitted to offspring. tains species with haplodiploid and diploid sex deter- The evolutionary mechanisms that explain haplodi- mination systems. Thus, we studied the association ploidy are still unclear. Some ecological and adaptive with Wolbachia in 12 diploid and 11 haplodiploid sco- hypotheses have been proposed (Engelstadter,€ 2008), lytine beetles by analyzing wsp and multilocus as follows: (i) females that can produce haploid males sequence typing (MLST) of five loci in this endosym- have twice the genetic fitness advantage compared with biont. Wolbachia genotypes were compared with females that cannot (Brown, 1963; Smith, 2000); (ii) mitochondrial (COI) and nuclear (EF) genotypes in recessive deleterious mutations will be purged effectively the scolytines. Eight of the 23 scolytine species when selection occurs in haploid males (Goldstein, were infected with Wolbachia, with haplodiploids at 1994) and (iii) sex ratio control by fertilization is more significantly higher rates than diploid species. Clon- advantageous under local mate competition (Hamilton, ing and sequencing detected multiple infections with 1967). Recent theoretical studies have shown that up to six Wolbachia strains in individual species. maternally transmitted endosymbiotic bacteria may facili- Phylogenetic analyses of wsp and five MLST genes tate haplodiploid evolution (Normark, 2004; Engelstadter€ and Hurst, 2006; Ubeda and Normark, 2006). Received 4 May, 2016; accepted 9 May, 2016. *For correspon- Endosymbiotic bacteria, such as Wolbachia, are found dence. E-mail [email protected]; Tel. 181-52-789- in a wide range of (Werren et al., 2008; Zug and 4113. †Present address: Office of Research Development and Sponsored Projects, Keio University, 1S4 in Institute of Integrated Hammerstein, 2012). They are asymmetrically (mater- Medical Research, 35, Shinano-machi, Shinjuku, Tokyo, 160-8582, nally not paternally) transmitted, so they selfishly manip- Japan. ulate host reproduction in order to spread effectively and [Corrections added on 03 August 2016, after first online publication. Some values in Figure 3 and Table 1 were misprinted and are now maintain their infection in host populations (Werren updated]. et al., 2008) via cytoplasmic incompatibility (CI;

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd 2 Y. Kawasaki et al.

Hoffmann and Turelli, 1997) and sex ratio distortion beetle, Xyleborus ferrugineus, which reproduces partheno- (male killing, feminization of genetic males and thelyto- genetically in association with Gram-positive endosymbiotic kous parthenogenesis; e.g. Engelstadter€ and Hurst, bacteria. No further studies of the endosymbiont-scolytine 2009). For example, male-killing bacteria disable the beetle relationship were reported until Stauffer and paternal genome sets in male eggs. Under sib mating, colleagues (1997), and Vega and colleagues (2002) the female offspring produced by infected females can described Wolbachia infections in two scolytine bark bee- use more resources than the offspring of uninfected tles, i.e., Ips typographus and Hypothenemus hampei mothers because the death of sibling males increases respectively. This endosymbiont has also been described the availability of resources (Hurst and Jiggins, 2000). in Coccotrypes dactyliperda (Zchori-Fein et al., 2006) and Females that can produce haploid males will be strongly Pityogenes chalcographus (Arthofer et al., 2009). These favoured in a population with few males, and this ability four bark beetles differ in terms of their biology, wherein will spread rapidly in the population. If endosymbiotic H. hampei exhibits PGE, C. dactyliperda has a haplodi- bacteria drive the evolution of a haplodiploid sex deter- ploid sex determination system, and I. typographus and mination system, haplodiploidy should originate after the P. chalcographus are diploid. Haplodiploidy is closely asso- bacterial infection. Hence, the phylogenetic relationships ciated with ambrosia beetles, but Xylosandrus germanus between insects and endosymbionts are expected to is the only known species to be infected with Wolbachia co-diverge even if some bacterial losses occur during (Kawasaki et al., 2010). co-diversification (Frost et al., 2010). The objective of this study was to investigate whether The most efficient mode for the transmission of endo- haplodiploid scolytines are more frequently associated symbionts, such as Wolbachia, is vertical transmission with Wolbachia than diploid species. Therefore, we from the mother to offspring via eggs. Therefore, these screened for Wolbachia in 424 individuals that belonged parasites manipulate their host’s reproduction to enhance to 12 diploid and 11 haplodiploid species of scolytine their transmission to the next generation, e.g., through beetles. Moreover, we characterized wsp and performed the induction of CI or by killing males (Engelstadter€ and multilocus sequence typing (MLST) in Wolbachia using Hurst, 2009). The rate of vertical Wolbachia transmission a subset of species, which were also characterized by is an important factor in the maintenance of endosym- sequencing the mitochondrial COI and nuclear EF1a bionts in the host population (Kawasaki et al., 2014). In addition, incongruence between the phylogeny of genes. Our results provide new insights into the occur- Wolbachia and its host shows that these bacteria can rence of Wolbachia in haplodiploid and diploid scolytine switch species boundaries and move horizontally beetles by highlighting the divergent patterns between between species (Baldo et al., 2008; Watanabe et al., Wolbachia and host beetles. 2012). The close ecological relationship between differ- ent species, such as shared habitats or shared parasi- toids, can facilitate the horizontal transmission of Results Wolbachia (Heath et al., 1999; Stahlhut et al., 2010; PCR screening for Wolbachia infection in different Schuler et al., 2013; 2016; Ahmed et al., 2015). scolytine beetles The subfamily Scolytinae (Coleoptera: Curculionidae), In total, 424 individuals from 23 scolytine beetle spe- also known as bark and ambrosia (fungi-growing) bee- tles, include diploid species and haplodiploid species cies (Table 1) were screened for Wolbachia with wsp (Kirkendall, 1983; 1993) as well as PGE species (Brun primers. Wolbachia was found in eight species: Taph- et al., 1995b). Scolytinae is a large taxonomic group rorychus bicolor, Euwallacea interjectus, Euwallacea with approximately 6000 species worldwide (Alonso- validus, Xyleborus schaufussi, Xyleborus seiryorensis, Zarazaga and Lyal, 2009), and one-fifth are likely to Xyleborus dispar, Xylosandrus crassiusculus and X. exhibit haplodiploidy (Normark et al., 1999). The evolu- germanus (Fig. 1). Seven of 11 haplodiploid species tion of arrhenotokous haplodiploidy occurred at least had Wolbachia infections, whereas 1 of 12 diploid bee- once from diploidy in scolytine beetles (Normark et al., tles was infected with this endosymbiont. The infection 1999). The causal mechanism has not been elucidated, rate at the species level differed significantly between but endosymbiotic bacteria might have played an haplodiploid and diploid beetles (GLM, P 5 0.011; Fig. important role in the evolution of haplodiploidy (Normark 1). The Wolbachia infection rates ranged from 80% in et al., 1999; Normark, 2004). However, little is known T. bicolor to 100% in E. interjectus, X. seiryorensis, X. about the interactions between scolytine beetles and dispar and X. germanus (Fig. 1). The infection rates their endosymbionts. did not differ significantly among eight Wolbachia- Endosymbiotic bacteria were first detected in scolytine infected species (proportion test for multiple compari- beetles by Peleg and Norris (1972a,b) in the ambrosia sons, P 5 0.069).

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 V C 06SceyfrApidMcoilg n onWly&Sn Ltd, Sons & Wiley John and Microbiology Applied for Society 2016

Table 1. Profiles of scolytine beetles examined in the present study

Subtribe Scientific name Feed Sex determination§ Mating system† Sex ratio¶ N Collection site

Tomicina Tomicus destuens Bark D R 1:1 4 Austria Tomicus piniperda Bark D R 1:1 4 Austria Drycoetina Taphrorychus bicolor Bark D R 1:1 5 Austria Polygraphina Polygraphus sp. Bark D R 1:1 4 Austria Ipina Pityogenes bidentatus Bark D R 1:1 2 Austria Ips cembrae Bark D R 1:1 4 Austria Ips sexdentatus Bark D R 1:1 4 Austria Ips typographus Bark D R 1:1 15 Japan (Furano), Austria Xyleterina Indocrypharus pubipennis Ambrosia D R 1:1 20 Japan (Aichi) Trypodendron signatum Ambrosia D R 1:1 12 Japan (Aichi) Scolytoplatypodina Scolytoplatypus mikado Ambrosia D R 1:1 29 Japan (Aichi, Wakayama) niomna irbooyReports Microbiology Environmental Scolytoplatypus tycon Ambrosia D R 1:1 16 Japan (Aichi) Xyleborina Euwallacea interjectus Ambrosia HD S FB 20 Japan (Aichi, Wakayama) Euwallacea validus Ambrosia HD S FB 21 Japan (Aichi) Xyleborinus saxeseni Ambrosia HD S FB 22 Japan (Austria, Toyama) Xyleborinus schaufussi Ambrosia HD S FB 11 Japan (Aichi) Xyleborus seiryorensis Ambrosia HD S FB 7 Japan (Aichi) Xyleborus seriatus Ambrosia HD S FB 12 Japan (Aichi) Xyleborus dispar Ambrosia HD S FB 5 Austria

Xylosandrus crassiusculus Ambrosia HD S FB 47 Japan (Sapporo, Aichi) Wolbachia Xylosandrus brevis Ambrosia HD S FB 25 Japan (Aichi) Xylosandrus germanus Ambrosia HD S FB 130 Germany, Hungary, with Kawasaki et al. (2010) Cnestus mutilatus Ambrosia HD S FB 5 Japan (Wakayama) neto nsoyiebeetles scolytine in infection

, 7 subtribes 13 genera; 8 bark 15 12 diploidy 12 random mating 12 1:1 sex-ratio 405 7 sites 00, 23 species ambrosia 11 haplodiploidy 11 sib mating 11 female-biased 00–00 §D: Diploidy; HD: Haplodiploidy; †R: Random mating; S: Sib mating; ¶FB: Female-biased sex ratio 3 4 Y. Kawasaki et al.

The phylogenetic analyses of wsp showed that the Wolbachia strains infecting haplodiploid scolytine beetles belonged to the A supergroup (Fig. 2), whereas the wTbi that infected the diploid T. bicolor belonged to the B supergroup (Fig. 2). Three major clades were recon- structed based on wsp. The first clade included wEi, wXse, wXge1 and wXge6, and the second comprised wXc and wXge4. The third clade contained wEv and wXsc. The other strains (wXge2, wXge3, wXge5, wXdi1- 4 and wTbi) were located at distinct nodes. Five MLST genes from six different species were con- catenated and aligned with strains from the Wolbachia MLST database. The MLST phylogenetic relationships (Fig. S1) detected two major clades, one of which com- prised wEi, wXse and wXge1. Similar to the wsp gene, these strains were identical based on the ftsZ, coxA and gatB loci, but they exhibited polymorphisms in the hcpA and fbpA genes (Fig. S2). By contrast, wEv and wXsc were identical based on wsp but also for all five MLST loci. The wXc, wXge2, wXge3 and wXge5 strains exhib- ited polymorphisms at almost all of the MLST loci.

Fig. 1. Infection rates of Wolbachia in 12 diploid and 11 haplodi- Discordance between Wolbachia and genetic diversity of ploid scolytine beetle species. Gray bars show the infection rates the hosts per species. Numbers at the right of the graph indicate the number of beetles examined. In this comparison, we only used ambrosia beetles cap- tured in Japan to exclude the effects of geographic and ecological traits. The molecular phylogenetic relation- Phylogenetic characterization of Wolbachia strains ships of the beetles were reconstructed using the puta- based on wsp and MLST tive amino acid sequences of the combined EF1a and We characterized 11 wsp sequences from 8 scolytine COI nucleotide sequences (Fig. 3A). The different beetle species. Six species (T. bicolor, E. interjectus, E. clades indicated monophyly for each subtribe (Table 1). validus, X. schaufussi, X. seiryorensis and X. crassius- Mapping the distribution of Wolbachia against its host culus) had clear single-peak chromatograms, which sug- (Fig. 3) indicated that there was a discordant relation- gested the presence of only a single Wolbachia strain. ship, which suggested frequent horizontal transfer of The strains were designated as wTbi from T. bicolor Wolbachia within the subfamily Scolytinae. For example, (GenBank Accession No. LC041010), wEi from E. inter- E. interjectus and E. validus are closely related sister jectus (AB588925), wEv from E. validus (AB588926), species, but they were infected with two genetically dif- wXsc from X. schaufussi (AB588928), wXse from ferent Wolbachia strains. By contrast, X. seiryorensis X. seiryorensis (AB588929) and wXc from X. crassius- and E. interjectus are genetically separated haplodiploid culus (AB588930). Wolbachia strains that infect beetles, but they were infected with genetically identical X. germanus have already been named as wXge1 to Wolbachia strains. Related species, such as X. crassius- wXge5 (Kawasaki et al., 2010), but an additional novel culus and X. germanus, differed in terms of their infec- Wolbachia strain from X. germanus was designated as tion status, wherein X. crassiusculus was infected with wXge6 (LC041015). Unlike the previously described the wXc strain whereas X. germanus had six different sequences, the chromatograms of wsp from X. dipar strains. X. crassiusculus and X. germanus also shared had double peaks, which indicated the presence of mul- two genetically related strains. Five additional Wolbachia tiple Wolbachia strains. Thus, we segregated the alleles strains were found only in X. germanus. Xyleborus of different strains by cloning and sequencing the PCR schaufussi is infected by Wolbachia, but we could not products. Seven plasmids from two X. dispar individuals detect any Wolbachia strains in the closely related spe- had four different strains designated as wXdi1 to wXdi4 cies X. seriatus. Therefore, stable trends in the Wolba- (LC041011–LC041014). Details of the host beetles and chia infections of haplodiploid beetles were not the accession numbers are summarized in Table S1. observed, thereby suggesting that the horizontal

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 Wolbachia infection in scolytine beetles 5

Fig. 2. Molecular phylogenetic tree obtained for Wolbachia strains infecting haplodiploid scolytine beetles based on the wsp gene using neighbour-joining (NJ) and maximum-likelihood (ML) methods. Each node shows the strain name with the host beetle. Bootstrap values > 50% estimated using 1000 replicates are shown above (NJ) and below (ML) each node. An arrow shows the bootstrap values obtained by NJ and ML between wXge1 and wEi. Wolbachia strains infecting Aedes albopictus and Brugia malayi are markers of the B and D supergroups respectively.

Fig. 3. Phylogenetic trees for (A) scolytine species and (B) Wolbachia strains. A. The phylogeny of the scolytines was estimated using the combined amino acid sequences of EF1a and COI (total length 5 470 amino acids). The analysis included 13 scolytine species. Bootstrap val- ues > 50% inferred from 1000 replicates are shown above (neighbour-joining) and below (maximum-likelihood) each node. B. The simplified phylogenetic relationships between Wolbachia strains infecting scolytine species were drawn based on Figs 2 and S1. Relationships between host beetles and infecting Wolbachia strains are indicated by the corresponding coloured lines (Fig S1). The outgroup was wBm in Brugi malayi, which belongs to the D supergroup.

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 6 Y. Kawasaki et al. transmission of Wolbachia has occurred frequently, Wolbachia, this endosymbiont should have become fixed rather than co-speciation between the beetles and in these species. However, our data do not support this Wolbachia. assumption because we found that all E. interjectus, X. seiryorensis, X. dispar and X. germanus individuals were Discussion infected with Wolbachia, whereas the endosymbionts were not ubiquitous in E. validus, X. schaufussi and X. Our comparative study shows that the reproductive bac- crassiusculus, and four species did not have infections terial endosymbiont Wolbachia is widespread in haplodi- with Wolbachia. It is not clear whether these species ploid and diploid scolytine beetles. However, the were never infected by Wolbachia or if a historic infection infections are significantly biased toward haplodiploid was lost (Morrow et al., 2015). Moreover, we cannot species rather than diploid mating species (Fig. 1). Three exclude the possibility that certain Wolbachia strains possibilities may explain the correlation of the biased were present only at low titers, so they were not detecta- infection rate with ecological traits, as follows. First, it is ble using our conventional sequencing approach. In gen- possible that the evolution of haplodiploidy is attributable eral, we cannot exclude the possibility that Wolbachia to Wolbachia infection. According to theoretical studies played a role during haplodiploid evolution in scolytine (Normark, 2004; Engelstadter€ and Hurst, 2006), repro- species, but our data do not support this hypothesis. It is ductive manipulators, such as Wolbachia, can accelerate also possible that other endosymbionts, such as Cardi- the evolution of haplodiploidy (including PGE) from dip- nium, Rickettsia or Spiroplasma (e.g. Duron et al., 2008), loidy (Engelstadter€ and Hurst, 2009). Normark and col- could have played roles in the haplodiploid evolution of leagues (1999) stated that endosymbiotic bacteria are scolytine beetles. factors that can induce haplodiploidy in scolytine beetles. The second possibility is that haplodiploidy is favour- Similar to the present study, Vega and colleagues (2002) able, or that diploidy is unfavourable, for Wolbachia also detected Wolbachia in the PGE beetle H. hampei, infection and maintenance. In particular, differences in and they suggested that the endosymbiont may cause the sex ratio with these two reproduction systems the elimination of the paternal genome. These results (Kirkendall, 1993) should have an important effect on are supported by a theoretical model, which showed that Wolbachia infection (Wenseleers et al., 1998). Most of haplodiploidizing endosymbionts can become beneficial the forms of reproductive manipulation (male killing, fem- for their female hosts (Normark, 2004). Hence, our inization and parthenogenesis) caused by Wolbachia are results support these previous studies. However, if Wol- known to bias the sex ratio toward females (Werren bachia is the evolutionary key to haplodiploidy in scoly- et al., 2008), and haplodiploid scolytine species are tine beetles, Wolbachia infection should have occurred known to have a female-biased sex ratio (Mizuno and prior to the evolution of haplodiploidy before the subse- Kajimura, 2002; Peer and Taborsky, 2004; Biedermann, quent co-divergence between Wolbachia and host bee- 2010). If the distortion of the sex ratio is not caused by tles. However, we found no indication of the coevolution Wolbachia, then this endosymbiont does not necessarily of Wolbachia and its hosts (Fig. 3). manipulate host reproduction, and their relationship All seven Wolbachia-infected haplodiploid species had might no longer be considered parasitic. Therefore, a Wolbachia strains that belonged to the A supergroup, but female-biased sex ratio may make Wolbachia advanta- the only diploid species infected with this endosymbiont geous. As a result, Wolbachia can infect haplodiploid had a Wolbachia strain from the B supergroup (Fig. 2). scolytine species at higher rates than diploid species. We can explain this observation by the independent hori- However, it is also possible that we underestimated the zontal transmission of different Wolbachia strains. We occurrence of Wolbachia in diploid scolytine species. found two different clades in the A supergroup in which First, Wolbachia may be present in populations at fre- wXsc and wEv were genetically identical based on wsp quencies ranging from 100% (e.g. Schuler et al., 2013) and the MLST genes, whereas wXge1, wEi and wXse to 1.3% (Sun et al., 2007). The low number of individual were separated by just a few SNPs but highly diverged diploid scolytines may have influenced our results. For from wXsc and wEv (Figs S1 and S2). This suggests example, Stauffer and colleagues (1997) found that Wol- that haplodiploid beetles are infected by two different bachia infected all of the I. typographus individuals genetic lineages of Wolbachia. However, our comparison tested in their study, whereas we could not detect this with the genetic background of their hosts showed that endosymbiont in the present study. Our results agree closely related strains were present in distantly related with a recent study, which did not detect Wolbachia in species of haplodiploid beetles (Fig. 3). This suggests populations of I. typographus from Western Carpathia, that Wolbachia probably has horizontal routes, rather Slovakia (Michalkova et al., 2012). Arthofer and col- than coevolving with its host. In addition, if the evolution leagues (2009) and Vega and colleagues (2002) also of haplodiploidy in scolytine beetles was driven by detected Wolbachia in scolytine beetles using more

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 Wolbachia infection in scolytine beetles 7 sensitive PCR techniques, such as nested PCR. There- but we cannot exclude a potential role of this endosym- fore, it is also possible that Wolbachia is present in sco- biont. For example, the historic gain and loss of Wolba- lytine beetles at lower densities, and, thus, we may have chia could have obscured the original pattern. In overlooked these bacteria in our survey. addition, future studies should focus on other endosym- The incongruence of the Wolbachia phylogeny with bionts, such as Cardinium, Rickettsia or Spiroplasma,to that of its hosts suggests that haplodiploid scolytine bee- understand the evolutionary roles of endosymbionts in tles have acquired different Wolbachia strains horizon- the sex determination systems of scolytine beetles. tally from independent sources. Previous studies have shown that the ecological overlap of different species by Acknowledgements sharing the same host provides the opportunity for Wol- We would like to thank F.E. Vega for critical comments bachia exchange between different species (Stahlhut on an earlier version of the manuscript and S.L. O’Neill for et al., 2010; Schuler et al., 2013; 2016). Different fac- valuable comments to the study. We are very grateful to W. tors, such as shared parasitoids (Heath et al., 1999; Arthofer, R.A. Beaver, P.H.W. Biedermann, A.I. Cognato, Vavre et al., 1999; Ahmed et al., 2015), cannibalism (Le J. Hulcr and M. Schebeck for sharing their precious knowl- Clec’h et al., 2013) and the host plant itself (Sintupa- edge. We are very grateful to A. Ueda, M. Nishimura, chee et al., 2006), can facilitate the switch between dif- T. Hogen, Y. Imaizumi, N. Yamaguchi, N. Takabe and M. Ito ferent species boundaries. All of the haplodiploid for collecting and providing specimens. This study was species tested in this study are also known as ambrosia financially supported by Grants-in-Aid for Scientific beetles. Ambrosia beetles live in galleries within logs for Research (18405012, 20405025 and 26292083 to H.K.), almost their entire life cycle. During larval growth within the IFO (Institute for Fermentation, Osaka) Foundation the galleries, the adults block the entrances of the gal- (2007) to H.K., the Showa-Houkoukai (Ito Chube’e) Foun- leries in order to prevent attack by natural enemies, dation (2008) to H.K., the Austrian Science Fund FWF pro- such as parasitoids and predators. Therefore, low para- ject (J-3527-B22) to H.S., and the FWF project ID10219 to sitism rates with the wasps that attack ambrosia beetles C.S. The stay of Y.K. in Vienna was supported by the Nagoya University International Academic Exchange have been reported (Kenis et al., 2004). Scholarship for Oversea Study Program. Common habitats and/or resource availability appear to be a more plausible explanation for the horizontal References transmission of Wolbachia in haplodiploid scolytine bee- tles. The ranges of the host trees used by ambrosia Ahmed, M.Z., Li, S.J., Xue, X., Yin, X.J., Ren, S.X., Jiggins, beetles depend partially on the species, but ambrosia F.M., et al. (2015) The intracellular bacterium Wolbachia beetles can usually exploit a wide range of different host uses parasitoid wasps as phoretic vectors for efficient trees provided that their symbiotic fungi can grow in the horizontal transmission. PLoS Pathog 10: e1004672. tree. For example, X. germanus and E. validus can co- Alonso-Zarazaga, M.A., and Lyal, C.H.C. (2009) A cata- logue of family and genus group names in Scolytinae and occur on Japanese cypress (Chamaecyparis obtusa)(Y. Platypodinae with nomenclatural remarks (Coleoptera: Kawasaki, pers. obs.). However, these beetles are Curculionidae). Zootaxa 2258: 1–134. infected with different Wolbachia strains. Maples (Acer Arthofer, W., Riegler, M., Avtzis, D., and Stauffer, C. (2009) japonicum) can be attacked by five different ambrosia Evidence for low-titre infections in symbiosis: Wol- beetles: X. crassiusculus, X. germanus, X. seiryorensis, bachia in the bark beetle Pityogenes chalcographus Scolytoplatypus mikado and S. tycoon. While the latter (Coleoptera, Scolytinae). Environ Microbiol 11: 1923– two species were not infected by Wolbachia, but the first 1933. three harboured different Wolbachia strains. Therefore, Baldo, L., Hotopp, J., Jolley, K., Bordenstein, S.R., Biber, S.A., Choudhury, R.R., et al. (2006) Multilocus sequence a subsequent study should focus on sampling different typing system for the endosymbiont Wolbachia pipientis. beetles from the same trees to determine whether the Appl Environ Microbiol 72: 7098–7110. host might play a role in Wolbachia transfer. Baldo, L., Ayoub, N.A., Hayashi, C.Y., Russell, J.A., In conclusion, for the first time, we showed that Wol- Stahlhut, J.K., and Werren, J.H. (2008) Insight into the bachia is present in seven haplodiploid and one diploid routes of Wolbachia invasion: high levels of horizontal scolytine species. We found no indication of co- transfer in the spider genus Agelenopsis revealed by Wol- phylogeny between Wolbachia and its hosts, but the bachia strain and mitochondrial DNA diversity. Mol Ecol occurrence of genetically different Wolbachia strains in 17: 557–569. Biedermann, P. (2010) Observations on sex ratio and genetically similar species suggests that the horizontal behavior of males in Xyleborinus saxesenii Ratzeburg acquisition of Wolbachia has occurred frequently in dis- (Scolytinae, Coleoptera). Zookeys 56: 253–267. tant phylogenetic clades. Our results do not indicate that Brown, S.W. (1963) The comstockiella system of chromo- Wolbachia drove the evolution of the haplodiploid and some behavior in armored scale insects (Coccoidea – diploid sex determination systems in scolytine beetles, Diaspididae). Chromosoma 14: 360–406.

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 8 Y. Kawasaki et al.

Brun, L.O., Borsa, P., Gaudichon, V., Stuart, J.J., Aronstein, Kirkendall, L. (1983) The evolution of mating systems in K., Coustau, C., and French-Constant, R.H. (1995a) bark and ambrosia beetles (Coleoptera, Scolytidae and Functional haplodiploidy. Nature 374: 506. Platypodidae). Zool J Linn Soc 77: 293–352. Brun, L.O., Stuart, J., Gaudichon, V., Aronstein, K., and Kirkendall, L. (1993) Ecology and evolution of biased sex ratios French-Constant, R.H. (1995b) Functional haplodiploidy – in bark and ambrosia beetles. In Evolution and Diversity of a mechanism for the spread of insecticide resistance in Sex Ratio in Insects and Mites. Wrensch, D., and Ebbert, M. an important international insect pest. Proc Natl Acad Sci (eds). New York: Chapman and Hall, pp. 235–345. USA 92: 9861–9865. Le Clec’h, W., Chevalier, F.D., Genty, L., Bertaux, J., Bou- Cook, J.M. (1993) Sex determination in the Hymenoptera – chon, D., and Sicard, Ml. (2013) Cannibalism and preda- a review of models and evidence. Heredity 71: 421–435. tion as paths for horizontal passage of Wolbachia Duron, O., Bouchon, D., Boutin, S., Bellamy, L., Zhou, L., between terrestrial isopods. PLoS ONE 8: e60232. Engelstadter,€ J., and Hurst, G.D. (2008) The diversity of Michalkova, V., Krascsenitsova, E., and Kozanek, M. (2012) reproductive parasites and : Wolbachia do not On the pathogens of the spruce bark beetle Ips typogra- walk alone. BMC Biol 6: 27. phus (Coleoptera: Scolytinae) in the western Carpathi- Engelstadter,€ J. (2008) Constraints on the evolution of asex- ans. Biologia 67: 217–221. ual reproduction. BioEssays 30: 1138–1150. Mizuno, T., and Kajimura, H. (2002) Reproduction of the ambro- Engelstadter,€ J., and Hurst, G.D.D. (2006) Can maternally sia beetle, Xyleborus pfeili (Ratzeburg) (Col., Scolytidae), on transmitted endosymbionts facilitate the evolution of hap- semi-artificial diet. J Appl Entomol 126: 455–462. lodiploidy? J Evol Biol 19: 194–202. Morrow, J.L., Frommer, M., Royer, J.E., Shearman, D.C.A., Engelstadter,€ J., and Hurst, G.D.D. (2009) The ecology and and Riegler, M. (2015) Wolbachia pseudogenes and low evolution of microbes that manipulate host reproduction. prevalence infections in tropical but not temperate Austra- Annu Rev Ecol Evol Syst 40: 127–149. lian tephritid fruit flies: manifestations of lateral gene trans- Frost, C., Fernandez-Marin, H., Smith, J., and Hughes, W. fer and endosymbiont spillover? BMC Evol Biol 15: 202. (2010) Multiple gains and losses of Wolbachia symbionts Normark, B.B. (2004) Haplodiploidy as an outcome of across a tribe of fungus-growing ants. Mol Ecol 19: coevolution between male-killing cytoplasmic elements 4077–4085. and their hosts. Evolution 58: 790–798. Goldstein, D.B. (1994) Deleterious mutations and the evolu- Normark, B.B., Jordal, B.H., and Farrell, B.D. (1999) Origin tion of male haploidy. Am Nat 144: 176–183. of a haplodiploid beetle lineage. Proc Roy Soc 266: Haig, D. (1993) The evolution of unusual chromosomal sys- 2253–2259. tem in coccoids – extraordinary sex-ratios revisited. Peer, K., and Taborsky, M. (2004) Female ambrosia beetles J Evol Biol 6: 69–77. adjust their offspring sex ratio according to outbreeding Hamilton, W. (1967) Extraordinary sex ratios. A sex-ratio opportunities for their sons. J Evol Biol 17: 257–264. theory for sex linkage and inbreeding has new implica- Peleg, B., and Norris, D.M. (1972a) Bacterial symbiote acti- tions in cytogenetics and entomology. Science 156: vation of insect parthenogenetic reproduction. Nat New 477–488. Biol 236: 111–112. Heath, B., Butcher, R., Whitfield, W., and Hubbard, S. Peleg, B., and Norris, D.M. (1972b) Symbiotic interrelation- (1999) Horizontal transfer of Wolbachia between phyloge- ships between microbes and ambrosia beetle VII. Bacte- netically distant insect species by a naturally occurring rial symbionts associated with Xyleborus ferrugineus. mechanism. Curr Biol 9: 313–316. J Invertebrate Pathol 20: 59–65. Hoffmann, A.A., and Turelli, M. (1997) Cytoplasmatic Schuler, H., Bertheau, C., Egan, S.P., Feder, J.L., Riegler, incompatibility in insects. In Influential Passengers: Inher- M., Schlick-Steiner, B.C., et al. (2013) Evidence for a ited Microorganisms and Reproduction. O’Neill, recent horizontal transmission and spatial spread of Wol- S.L., Hoffmann, A.A., and Werren, J.H. (eds). Oxford: bachia from endemic Rhagoletis cerasi (Diptera: Tephriti- Oxford University Press, pp. 42–80. dae) to invasive Rhagoletis cingulata in Europe. Mol Ecol Hurst, G.D., and Jiggins, F.M. (2000) Male-killing bacteria in 22: 4101–4111. insects: mechanisms, incidence, and implications. Emerg Schuler, H., Koppler,€ K., Daxbock-Horvath,€ S., Rasool, B., Infect Dis 6: 329–336. Krumbock,€ S., Schwarz, D., et al. (2016) The hitchhiker’s Kawasaki, Y., Ito, M., Miura, K., and Kajimura, H. (2010) guide to Europe: the infection dynamics of an ongoing Superinfection of five Wolbachia in the alnus ambrosia Wolbachia invasion and mitochondrial selective sweep in beetle, Xylosandrus germanus (Blandford) (Coleoptera: Rhagoletis cerasi. Mol Ecol 25: 1595–1609. Curuculionidae). Bull Entomol Res 100: 231–239. Sintupachee, S., Milne, J., Poonchaisri, S., Baimai, V., and Kawasaki, Y., Ito, H., and Kajimura, H. (2014) Equilibrium Kittayapong, P. (2006) Closely related Wolbachia strains frequency of endosymbionts in multiple infections based within the pumpkin arthropod community and the poten- on the balance between vertical transmission and cyto- tial for horizontal transmission via the plant. Microb Ecol plasmic incompatibility. PLoS One 9: e94900. 51: 294–301. Kenis, M., Wermelinger, B., and Gregoire, J.C. (2004) Smith, N.G.C. (2000) The evolution of haplodiploidy under Research on parasitoids and predators of Scolytidae, a inbreeding. Heredity 84: 186–192. review. In Bark and Wood Boring Insects in Living Trees Stahlhut, J.K., Desjardins, C.A., Clark, M.E., Baldo, L., in Europe: A Synthesis. Lieutier, F., Day, K., Battisti, A., Russell, J.A., Werren, J.H., and Jaenike, J. (2010) The Gregoire, J.-C., and Evans, H. (eds). Dordrecht, the Neth- mushroom habitat as an ecological arena for global erlands: Kluwer Academic Publishers, pp. 237–314. exchange of Wolbachia. Mol Ecol 19: 1940–1952.

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00 Wolbachia infection in scolytine beetles 9

Stauffer, C., van Meer, M., and Riegler, M. (1997) The pres- depends on symbiotic bacteria. Physiolog Entomol 31: ence of the protobacteria Wolbachia in European Ips 164–169. typographus (Col., Scolytidae) populations and the con- Zug, R., and Hammerstein, P. (2012) Still a host of hosts sequences for genetic data. Proc Ger Soc Gen Appl for Wolbachia: analysis of recent data suggests that 40% Entomol 11: 709–711. of terrestrial arthropod species are infected. PLoS One 7: Sun, X., Cui, L., and Li, Z. (2007) Diversity and phylogeny e38544. of Wolbachia infecting Bactrocera dorsalis (Diptera: Teph- ritidae) populations from China. Environ Entomol 36: Supporting Information 1283–1289. Ubeda, F., and Normark, B.B. (2006) Male killers and the Additional Supporting Information may be found in the origins of paternal genome elimination. Theor Pop Biol online version of this article at the publisher’s website: 70: 511–526. Vavre, F., Fleury, F., Lepetit, D., Fouillet, P., and Bouletreau, Fig. S1. Molecular phylogenetic tree of Wolbachia strains M. (1999) Phylogenetic evidence for horizontal transmis- infecting scolytine species based on the concatenated sion of Wolbachia in host-parasitoid associations. Mol sequences of five MLST genes (2079 base pairs) using the Biol Evol 16: 1711–1723. neighbor-joining (NJ) method. The strains in scolytine bee- Vega, F.E., Benavides, P., Stuart, J.A., and O’Neill, S.L. tles are shown in bold letters. Bootstrap values > 50% (2002) Wolbachia infection in the coffee berry borer (Cole- obtained by the NJ method are shown above or below the optera: Scolytidae). Ann Entomol Soc Am 95: 374–378. branches. The analysis included 20 additional Wolbachia Watanabe, M., Tagami, Y., Miura, K., Kageyama, D., and strains, most of which were detected in coleopteran species Stouthamer, R. (2012) Distribution patterns of Wolbachia (Table S2). endosymbionts in the closely related flower bugs of the Fig. S2. Neighbor-joining trees for individual Wolbachia genus Orius: implications for coevolution and horizontal MLST genes with a simplified Wolbachia genome map transfer. Microb Ecol 64: 537–545. modified from Baldo and colleagues (2006). Wolbachia Wenseleers, T., Ito, F., van Borm, S., Huyubrechts, R., strains infecting scolytine species are described by the Volckaert, F., and Billen, J. (1998) Widespread occur- strain names given in bold letters and other Wolbachia rence of the micro-organism Wolbachia in ants. Proc Roy strains by ST (Table S2). Bootstrap values > 50% are Soc 265: 1447–1452. shown above the branches. Werren, J.H., Baldo, L., and Clark, E. (2008) Wolbachia: Table S1. Summary of strain names, hosts, accession num- master manipulators of invertebrate biology. Nat Rev bers and sequence types for Wolbachia strains used in the Microbiol 6: 741–751. present study Zchori-Fein, E., Borad, C., and Harari, A. (2006) Oogenesis Table S2. Profiles of the sequence types and allelic num- in the date stone beetle, Coccotrypes dactyliperda, bers used in the present study

VC 2016 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 00, 00–00