The Journal of Experimental Biology 203, 2503–2509 (2000) 2503 Printed in Great Britain © The Company of Biologists Limited 2000 JEB2738

WOLBACHIA INFECTION AND CYTOPLASMIC INCOMPATIBILITY IN THE TELEOGRYLLUS TAIWANEMMA

SATORU KAMODA, SHINJI MASUI, HAJIME ISHIKAWA AND TETSUHIKO SASAKI* Department of Biological Sciences, Graduate School of Science, University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan *Author for correspondence (e-mail: [email protected])

Accepted 25 May; published on WWW 20 July 2000

Summary Wolbachia are cytoplasmically inherited bacteria found chain reaction assays; they were detected in the antennae, in many . They induce various reproductive heads, forewings, hindwings, testes, ovaries, Malpighian alterations in their hosts, including cytoplasmic tubules, foot muscles and fat bodies. Quantitative incompatibility, thelytokous parthenogenesis, feminization polymerase chain reaction assays showed that the bacterial and male-killing. In this study, we examined Wolbachia density was highest in the fat bodies, followed by the infection and its effects on the host cricket Teleogryllus ovaries and testes. Wolbachia were not detected in the taiwanemma. In a phylogenetic study based on the wsp gene haemolymph or in mature spermatozoa. The spermatozoa coding for a Wolbachia surface protein, the Wolbachia of the infected male may be modified by the presence of strain harboured by T. taiwanemma was clustered together Wolbachia during its development. To examine this with those harboured by Laodelphax striatellus, Tribolium possibility, we compared the profiles of sperm proteins confusum, Acraea encedon, Trichogramma deion and Adalia between the infected and uninfected males using two- bipunctata. Crossing experiments using the Wolbachia- dimensional gel electrophoresis. However, no differences in infected and uninfected strains of cricket showed that the the protein profiles were observed. infection is associated with the expression of unidirectional cytoplasmic incompatibility: the egg hatch rate in the incompatible cross between the infected males and Key words: Wolbachia, cricket, Teleogryllus taiwanemma, uninfected females was 20.3 %. We also examined the cytoplasmic incompatibility, bacterial density, two-dimensional gel distribution of Wolbachia within the host using polymerase electrophoresis.

Introduction Wolbachia are intracellular bacteria present in a wide range cross is compatible. In addition, bidirectional incompatibilities of arthropods, particularly in . In a previous field survey are observed when males and females carry different by Werren et al. (1995b), Wolbachia were found in more than Wolbachia strains. It has been reported that cytoplasmic 16 % of the insects sampled. Phylogenetic analyses of the incompatibility arises from defects in paternal chromatin Wolbachia harboured by various host species have shown that condensation during mitosis in Drosophila simulans (O’Neill the Wolbachia clade can be divided into two major groups, and Karr, 1990; Lassy and Karr, 1996; Callaini et al., 1997), designated A and B (Werren et al., 1995a). Nasonia vitripennis (Reed and Werren, 1995) and Culex Wolbachia are known to induce various reproductive pipiens (Jost, 1971). Cytological examinations in several alterations in their hosts, including cytoplasmic species have revealed that the bacteria in spermatids are incompatibility, thelytokous parthenogenesis, feminization released during spermatogenesis and are absent from the and male-killing (for reviews, see Werren, 1997; Stouthamer mature spermatozoa (Wright and Barr, 1980; Binnington and et al., 1999). Cytoplasmic incompatibility is a crossing Hoffmann, 1989; O’Neill, 1989). It is assumed that Wolbachia incompatibility that results in embryonic mortality in diploid modify the spermatozoa during spermatogenesis and that this species. In haplodiploid species such as wasps, in which modification is responsible for the expression of cytoplasmic haploid embryos develop into males, cytoplasmic incompatibility. However, few investigations have attempted incompatibility causes male-biased sex ratios. The effect of to detect the modification in the spermatozoa of Wolbachia- cytoplasmic incompatibility on crossability is typically infected insects, mainly because it is difficult to obtain sperm unidirectional: the cross between infected males and samples from insects in which cytoplasmic incompatibility has uninfected females is incompatible, whereas the reciprocal been expressed. 2504 S. KAMODA AND OTHERS

Males of orthopteran insects release spermatophores GAAAC-3′) and wsp691R (5′-AAAAATTAAACGCTACT- containing pure spermatozoa when they mate. Because of this CCA-3′). These primers amplify a DNA fragment of habit, samples of pure spermatozoa can be collected from the approximately 600 base pairs (bp). The cycling conditions males of this group. However, despite this advantage, no consisted of 5 min at 94 °C, followed by 35 cycles of 30 s at laboratory strain of an orthopteran insect has been established 94 °C, 30 s at 52 °C, and 1 min at 72 °C, and a final cycle of as an insect model for the study of cytoplasmic incompatibility. 5 min at 72 °C. The PCR products were directly ligated into Masui et al. (1997) previously reported that a laboratory pGEM-T vectors (Promega), and five independent clones were strain of the cricket Teleogryllus taiwanemma was infected with sequenced using an ABI PRISM 310 genetic analyzer (PE B-group Wolbachia, although the effect of the infection was Applied Biosystems). The wsp sequence has been submitted to unclear. In this study, we further investigated Wolbachia the GenBank database under accession number AB035514. infection in T. taiwanemma. The phylogenetic position of the This sequence was aligned to the wsp sequences reported by Wolbachia was estimated on the basis of the wsp gene that Zhou et al. (1998) and Hurst et al. (1999). A 41 bp region codes for a Wolbachia surface protein. To clarify the phenotype (positions 519–559) corresponding to the third hypervariable induced by the Wolbachia, we performed crossing experiments region of the gene (Braig et al., 1998) was omitted from the and demonstrated that cytoplasmic incompatibility is expressed, analysis because it could not be aligned with confidence (Zhou giving relatively high embryonic mortality. We also examined et al., 1998). Sites with gaps were also excluded from the the presence of Wolbachia in various tissues of the host using aligned data set. The resulting alignment of approximately 500 the polymerase chain reaction (PCR). In addition, the bacterial bases was used to construct a neighbour-joining tree (Saitou density in each tissue was determined by quantitative PCR and Nei, 1987) with Kimura’s two-parameter distance assays. After we had confirmed that the spermatozoa did not (Kimura, 1980) using the program package Clustal W contain Wolbachia, we compared the profiles of sperm proteins (Thompson et al., 1994). A bootstrap test (Felsenstein, 1981) between the infected and uninfected males. was performed with 1000 replications.

Detection of Wolbachia Materials and methods The presence of Wolbachia in various tissues of T. Insects taiwanemma was tested by PCR using the primer sets ftsZBf A strain of the cricket Teleogryllus taiwanemma infected with (5′-CCGATGCTCAAGCGTTAGAG-3′) and ftsZBr (5′- a B-group Wolbachia strain (wTai) was maintained under 16 h:8 h CCACTTAACTCTTTCGTTTG-3′), which specifically light:dark cycle at 25 °C. The crickets were fed an insect food amplify the ftsZ cell cycle gene of B-group Wolbachia (Werren purchased from Oriental Yeast, Tokyo, Japan, and provided with et al., 1995a). The PCR cycling conditions consisted of 5 min tap water. To establish a Wolbachia-free strain, the infected at 94 °C, followed by 35 cycles of 30 s at 94 °C, 30 s at 55 °C, insects were given an aqueous solution of tetracycline and 1 min at 72 °C, and a final cycle of 5 min at 72 °C. To hydrochloride (0.5 % w/v) for two generations. This strain was confirm successful DNA extraction, the samples were also reared without tetracycline for at least two generations subjected to PCR using primers specific for the insect (approximately 5 months) before being subjected to experiments. pgi gene that codes phosphoglucose isomerase (Katz and Harrison, 1997). The primers used were pgiF DNA extraction (5′-AACAAGGAGATATGGAATCTAATGG-3′) and pgiR The antennae, heads, forewings, hindwings, testes, ovaries, (5′-TTCCAGGTTCACCCCACA-3′). The cycling conditions Malpighian tubules, foot muscles and fat bodies were obtained were the same as those for the amplification of the ftsZ gene. by dissection from adult insects. Haemolymph, which oozed out through the incision when a hindleg was cut, was collected in Quantitative PCR a small pipette. To obtain pure sperm samples, we cut the end To determine the Wolbachia density in each tissue, of a spermatophore and immersed it in phosphate-buffered quantitative PCR was performed in a LightCycler (Roche −1 −1 saline (PBS; 137 mmol l NaCl, 8.1 mmol l Na2HPO4, Diagnostics). A 294 bp fragment of the internal region −1 −1 2.68 mmol l KCl, 1.47 mmol l KH2PO4, pH 7.5). The of the VirD4 gene (GenBank accession number spermatozoa released into PBS was collected by centrifugation. AB041342) was amplified using the primers VirD4F (5′- These samples were homogenized in saline/Tris/EDTA (STE; ATCAGAGAAAGACATACGAAAAGCAGG) and VirD4R 100 mmol l−1 NaCl, 10 mmol l−1 Tris pH 8.0, 1 mmol l−1 EDTA, (5′-CAATGGCTTACCCCATCTGGC). The 20 µl reaction pH 8.0) containing proteinase K at 0.4 mg ml−1. After mixture consisted of 10 % (v/v) LightCycler DNA master SYBR −1 −1 incubation at 55 °C for 90 min, DNA was purified by Green I (Roche Diagnostics), 3 mmol l MgCl2, 0.2 µmol l of phenol/chloroform extraction followed by ethanol precipitation. each primer, 2 µl of template DNA and 0.16 µl of TaqStart Antibody (Clontech). For the amplification, 50 cycles consisting Phylogenetic analysis of 0 s at 95 °C, 5 s at 57 °C and 30 s at 72 °C were performed. The wsp gene was amplified from the DNA solution The fluorimetric intensity of SYBR Green I, a dye specific for prepared from testis by PCR using the general wsp primers double-stranded DNA, was measured at the end of each (Braig et al., 1998) wsp81F (5′-TTGTCCAATAAGTGAA- elongation phase, and the original concentration of the VirD4 Wolbachia infection in cricket 2505 gene in the template DNA solution was determined by analysis Table 1. Crossing experiments in Teleogryllus taiwanemma of the amplification kinetics. Standard reactions were performed Number Number Number Percentage with the template solutions containing VirD4 PCR products at Cross of of eggs of eggs of eggs known concentrations. The primer set used amplified the VirD4 male × female crosses collected hatched hatched gene without production of primer dimers or non-specific Infected × infected 9 1497 1282 82.0±10.9 background, which was confirmed by a melting-curve analysis Infected × uninfected 15 2712 861 20.3±18.7* as recommended by the manufacturer. On the basis of the Uninfected × infected 5 667 553 82.1±10.4 observation that the VirD4 gene of wTai is a single copy gene Uninfected × uninfected 8 1067 920 86.7±7.1 (S. Masui, unpublished data), the number of Wolbachia cells in each sample was calculated from the concentration of the gene. *Mann-Whitney U-test (compared with all other crosses): P<0.001. Crossing experiments Values are means ± S.D. Crossing experiments were performed using single pairs of virgin individuals. Females and males were separated at the late bipuntata (wBipa, wBipb). These Wolbachia strains constituted nymphal stage; they were paired on adult emergence and kept in a monophyletic group supported by the bootstrap probability of a cage with a 50 ml tube containing moistened tissue paper. 98.9 %, although the branches within this clade were not strongly Approximately 1 week after they had been paired, the female laid supported except for the branch consisting of wBipa and wBipb eggs on the tissue paper. The eggs were washed off the tissue (Fig. 1). In terms of sequence divergence, wTai differed by at paper and placed onto moistened filter papers in a plastic dish. least 7 % from all other Wolbachia strains. According to the More than 100 eggs were collected from each pair to count the grouping criterion of a 2.5 % sequence difference as proposed number of hatched nymphs. After the eggs had been collected, by Zhou et al. (1998), wTai represents a new group. all individuals used in the experiment were diagnosed by the PCR using the ftsZB primers to confirm the infection status. The DNA Crossing experiments samples were prepared from the testes or ovaries. The expression of cytoplasmic incompatibility, a possible reproductive alteration induced by the Wolbachia, was Electrophoresis of sperm proteins examined by crossing experiments between the infected insects Spermatozoa were dissolved in 240 µl of the lysis buffer and those from which the Wolbachia had been removed by [9.8 mol l−1 urea, 0.5 % (v/v) Triton-X, 2 % Ampholine, tetracycline treatment. When uninfected females were mated (pH 3.5–10), 0.075 % (w/v) dithiothreitol and a small amount with infected males, the egg hatch rates were 0–48.6 %, with of Orange G], then applied to gels (Immobiline DryStrip, a mean rate of 20.3 % (Table 1). In each of the other three pH 3–10, 11 cm; Amersham Pharmacia Biotech). The gels crosses, more than 80 % of the eggs hatched. The egg hatch were rehydrated for 12 h and then subjected to electrophoresis rate in the crosses between infected males and uninfected in a Multiphor II electrophoresis unit (Amersham Pharmacia females was statistically different from those in the other Biotech). Isoelectric focusing was performed at 15 °C with combinations (Mann–Whitney U-test, P<0.001). The duration voltage limits of 300 V for 1 h, a gradient from 300 V to 3500 V of embryonic development was approximately 3 weeks at for 3 h, and 3500 V for 17 h. The current limit was 0.5 mA per 25 °C. In the compatible crosses, the eggs swelled within a few strip throughout the electrophoresis. In the second dimension, days of oviposition. Most eggs from the incompatible cross did SDS–PAGE was performed as described by Laemmli (1970) not swell, suggesting that embryonic death had occurred at a using slab separation gels of 10 % polyacrylamide. The relatively early stage of the development. proteins were visualized by silver staining using two- dimensional silver stain II ‘DAIICHI’ (Daiichi Pure Distribution of Wolbachia Chemicals). The protein profiles were compared between To examine the distribution of Wolbachia in T. infected and uninfected males by superimposing the two gels taiwanemma, DNA was prepared from the antennae, heads, on a light box. forewings, hindwings, testes, ovaries, Malpighian tubules, foot muscles, fat bodies, haemolymph and spermatozoa. In the PCR assay using ftsZB primers, we detected Wolbachia in Results all samples examined, except in the haemolymph and Phylogenetic relationship spermatozoa (Fig. 2A). The success of the DNA extraction was To estimate the relative phylogenetic position of Wolbachia confirmed by PCR using pgi primers (Fig. 2B). infecting T. taiwanemma (wTai), the wsp gene was sequenced We also performed quantitative PCR to determine the and aligned to those previously reported from other host species. densities of Wolbachia in the tissues in which the bacteria had The phylogenetic analysis suggested that wTai is closely related been detected. The total number of bacteria per tissue was to Wolbachia strains infecting the small brown planthopper highest in the ovaries (480×106 bacteria) and relatively high in Laodelphax striatellus (wStri), the flower beetle Tribolium the fat bodies, testes and heads. The bacterial density, confusum (wCon), the butterfly Acraea encedon (wEnc), the estimated by normalising the total bacterial number by the wasp Trichogramma deion (wDei) and the ladybird Adalia wet mass for each tissue, was highest in the fat bodies 2506 S. KAMODA AND OTHERS

wStri (Laodelphax striatellus) 0.01 564 wCon (Tribolium confusum) Fig. 1. A phylogenetic tree 655 wEnc (Acraea encedon) based on wsp sequences. The wsp sequence from wTai was 522 wDei (Trichogramma deion) aligned with wsp sequence data of B-group Wolbachia, with 998 wTai (Teleogryllus taiwanemma) A-group wRi and wCauA as outgroups, and analyzed wBipa (Adalia bipunctata A) B using the neighbour-joining 579 algorithm. Bootstrap values out 1000 wBipb (Adalia bipunctata B) of 1000 replications are indicated above the branches. wCauB (Ephestia cautella) The scale bar indicates genetic 998 1000 distance in units of nucleotide wOri (Tagosedes orizicolus) substitutions per site. The right- wAlbB (Aedes albopictus) hand vertical lines show A- and 712 B-groups identified by Werren et al. (1995a) and detailed wNou (Drosophila simulans Noumea) grouping based on the wsp 1000 wPip (Culex pipiens) sequence (Zhou et al., 1998). The GenBank accession wCauA (Ephestia cautella) numbers for wsp sequences are A AF020070 for wRi, AF020075 wRi (Drosophila simulans Riverside) for wCauA, AF020076 for wCauB, AF020083 for wCon, AF020084 for wDei, AF020085 for wOri, AF020080 for wStri, AF020061 for wPip, AF020074 for wNou, AF020059 for wAlbB, AJ130714 for wEnc, AJ130715 for wBipa and AJ130716 for wBipb.

Table 2. Infection level of Wolbachia in the tissues of Teleogryllus taiwanemma A Total number of Bacterial density × 10−3 Tissue bacteria × 10−3 (µg−1 tissue)

Antenna 135±73 415±168 a h fw hw t o mt fm fb hl s nc m Head 37 878±266 695±91 Forewing 218±159 53±30 B Hindwing 76±30 27±13 Testis 33 733±3 717 1 030±91 Ovary 480 000±85 440 3 279±584 Malpighian tubules 5 960±2 546 969±425 Foot muscle 4 313±2 990 122±79 Fig. 2. Polymerase chain reaction (PCR) assay using ftsZB primers Fat body 51 034±16 724 5 103±2 128 for detection of Wolbachia in various tissues of Teleogryllus taiwanemma (A). Successful DNA extraction was confirmed by PCR Samples were collected from adult individuals less than 1 week with primers for the pgi gene coding for phosphoglucose isomerase after the final ecdysis. (B). Lanes: a, antenna; h, head; fw, forewing; hw, hindwing; t, testis; Values are means ± S.D.(N=4). o, ovary; mt, Malpighian tubules; fm, foot muscle; fb, fat body; hl, haemolymph; s, spermatozoa; nc, negative control; m, 1 kb DNA ladder (GIBCO BRL). (5105 µg−1), followed by the ovaries, testes and Malpighian tubules (Table 2). In an attempt to find such a difference, we analyzed the profiles of sperm proteins using two-dimensional gel electrophoresis. Analysis of sperm proteins However, no repeatable difference in protein profile was The absence of Wolbachia in the spermatozoa of the infected detected between infected and uninfected individuals (Fig. 3). T. taiwanemma suggested that the Wolbachia modified the spermatozoa to induce cytoplasmic incompatibility during sperm development. For this reason, any difference between Discussion the spermatozoa of infected and uninfected males should have Reproductive incompatibilities have been reported in been related to the expression of cytoplasmic incompatibility. allopatric field populations of crickets (Harrison, 1983; Wolbachia infection in cricket 2507

kDa A B

97

68

43

29

14 Fig. 3. Silver-stained two-dimensional gel electrophoresis of proteins from mature spermatozoa of infected (A) and uninfected (B) Teleogryllus taiwanemma.

Smith and Cade, 1987; Cade and Tyshenko, 1990). Recently, if these genes are located on a transposable element that can some of them were found to be infected with Wolbachia move between strains. It has been demonstrated that the (Werren et al., 1995a; Giordano et al., 1997). In the present genome of wTai contains transposon-like sequences (Masui et study with T. taiwanemma, we generated a Wolbachia-free al., 1999) and phages (S. Masui, unpublished data). strain from the infected strain to perform crossing experiments, In the PCR assay using ftsZB primers, Wolbachia were and observed that the cross between the infected males and detected in all the samples examined, except in the uninfected females was incompatible, giving an embryonic spermatozoa and haemolymph (Fig. 2). Since Wolbachia are mortality at 79.7 %. To our knowledge, this is the first clear- maternally transmitted through the egg cytoplasm and induce cut evidence for the association between Wolbachia infection reproductive alterations, it has been assumed that they infect and unidirectional cytoplasmic incompatibility in an the reproductive tissues of the hosts. Contrary to this orthopteran insect. It is known that the strength of cytoplasmic assumption, we have shown here that Wolbachia are widely incompatibility varies depending on the Wolbachia strain and distributed in the tissues of T. taiwanemma, a finding consistent host species. For example, Wolbachia induce almost complete with the previous reports on Wolbachia infections in incompatibility in the almond moth Ephestia cautella (Sasaki Drosophila simulans, the mosquitoes Aedes albopictus and and Ishikawa, 1999), in Laodelphax stritellus (Noda, 1984) Culex pipiens, the moth Ephestia cautella (Dobson et al., 1999) and in the Tribolium (Stevens and Wade, 1990), while and the isopod Armadillidium vulgare (Martin et al., 1973). In Drosophila melanogaster shows weak expression of the present study, we further examined the somatic infection cytoplasmic incompatibility with an embryonic mortality of the Wolbachia in terms of bacterial density. The quantitative lower than 30 % (Hoffmann, 1988; Hoffmann et al., 1994). In PCR assays (Table 2) showed that the density in fat bodies was comparison with other host species, the strength of cytoplasmic higher even than that in the ovaries or testes. The bacterial incompatibility observed in T. taiwanemma may be considered density differed greatly among tissues, suggesting that some as intermediate or relatively high. tissues, such as the muscle and brain, are not suitable for the Phylogenetic analysis based on the wsp gene (Fig. 1) proliferation of wTai. The distribution and density of suggested that wTai is phylogenetically close to wStri, wCon, Wolbachia in the tissues of adult insects may also be related wEnc, wDei, wBipa and wBipb. Interestingly, these Wolbachia to the bacterial localization in the host egg, which has not yet strains induce different reproductive phenotypes: wStri and been examined for T. taiwanemma. wCon induce cytoplasmic incompatibility, wDei causes To explain the mechanism of cytoplasmic incompatibility, thelytokous parthenogenesis (Stouthamer et al., 1993) and the it has often been postulated that Wolbachia secrete a protein others are male-killers (Hurst et al., 1999). Thus, closely that remains in mature spermatozoa (Hurst, 1991; Lassy and related Wolbachia have different effects on their host, as was Karr, 1996). When Sasaki et al. (1998) analyzed sperm also described in a recent study by Van Meer et al. (1999). One proteins from Drosophila simulans, no protein specific to the explanation for the lack of congruence between Wolbachia infected flies was detected. Compared with Drosophila spp., T. phylogeny and their phenotypes is that the host, rather than taiwanemma has some features suitable for sperm analysis: Wolbachia, plays a major role in the determination of the (i) mature spermatozoa can be obtained easily from phenotype. It is also possible that Wolbachia easily acquire the spermatophores without contaminating the spermatids; (ii) ability to induce various effects. Different phenotypes may spermatophores can be collected from a male repeatedly evolve rapidly if the genes responsible for them are similar or without killing it; and (iii) spermatophores contain sufficient 2508 S. KAMODA AND OTHERS volumes of spermatozoa for biochemical and molecular Hurst, L. D. (1991). The evolution of cytoplasmic incompatibility or biological analyses. In this study, we took advantage of these when spite can be successful. J. Theor. Biol. 148, 269–277. features of T. taiwanemma to analyze larger numbers of Jost, E. (1971). Meiosis in the male of Culex pipiens and Aedes spermatozoa and performed isoelectric focusing over a wider albopictus and fertilization in the Culex pipiens-complex. Can. J. range than in the previous study with D. simulans. In spite of Genet. Cytol. 13, 237–250. these improvements in the methodology, we were still unable Katz, L. A. and Harrison, R. G. (1997). Balancing selection on electrophoretic variation of phosphoglucose isomerase in two to detect any repeatable difference in the protein profiles species of field cricket: Gryllus veletis and G. offnsylvanicus. between the infected and uninfected individuals. This could Genetics 147, 609–621. mean that the amount of the protein involved in cytoplasmic Kimura, M. (1980). A simple method for estimating evolutionary incompatibility expression is extremely small or that the pI rates of base substitutions through comparative studies of and/or molecular size of the protein are out of range of the two- nucleotide sequences. J. Mol. Evol. 16, 111–120. dimensional gel system employed. It is also possible that Laemmli, U. K. (1970). Cleavage of structural proteins during cytoplasmic incompatibility is caused by such a small the assembly of the head of bacteriophage T4. Nature 227, modification in protein structure that it could not be detected. 680–685. Alternatively, the molecule responsible for cytoplasmic Lassy, C. W. and Karr, T. L. (1996). Cytological analysis of incompatibility may not be a protein, but some other molecule, fertilization and early embryonic development in incompatible such as a nucleic acid or a compound of low molecular mass. crosses of Drosophila simulans. Mech. Dev. 57, 47–58. Martin, G., Juchault, P. and Legrand, J. J. (1973). Mise en évidence To examine these possibilities, further experiments are needed. d’un micro-organisme intracytoplasmique symbiote de l’oniscoïde Armadillidium vulgare L., dont la présence accompagne l’intersexualité ou la féminisation totale des males génétiques de la References lignée thélyghne. C.R. Acad. Sci. Ser. D 276, 2313–2316. Binnington, K. C. and Hoffman, A. A. (1989). Wolbachia-like Masui, S., Kamoda, S., Sasaki, T. and Ishikawa, H. (1999). The organisms and cytoplasmic incompatibility in Drosophila first detection of the insertion sequence ISW1 in the intracellular simulans. J. Invertebr. Path. 54, 344–352. reproductive parasite Wolbachia. Plasmid 42, 13–19. Braig, H. R., Zhou, W., Dobson, S. L. and O’Neill, S. L. (1998). Masui, S., Sasaki, T. and Ishikawa, H. (1997). groE-homologous Cloning and characterization of a gene encoding the major surface operon of Wolbachia, an intracellular symbiont of arthropods: a protein of the bacterial endosymbiont Wolbachia pipientis. J. new approach for their phylogeny. Zool. Sci. 14, 701–706. Bacteriol. 180, 2373–2378. Noda, H. (1984). Cytoplasmic incompatibility in allopatric field Cade, W. H. and Tyshenko, M. G. (1990). Geographic variation in populations of the small brown planthopper, Laodelphax hybrid fertility in the field crickets Gryllus integer, G. rubens and striatellus, in Japan. Ent. Exp. Appl. 35, 263–267. Gryllus sp. Can. J. Zool. 68, 2697–2700. O’Neill, S. L. (1989). Cytoplasmic symbionts in Tribolium confusum. Callaini, G., Dallai, R. and Riparbelli, M. G. (1997). Wolbachia- J. Invertebr. Path. 53, 132–134. induced delay of paternal chromatin condensation does not prevent O’Neill, S. L. and Karr, T. L. (1990). Bidirectional incompatibility maternal chromosomes from entering anaphase in incompatible between conspecific populations of Drosophila simulans. Nature crosses of Drosophila simulans. J. Cell Sci. 110, 271–280. 348, 178–180. Dobson, S. L., Bourtzis, K., Braig, H. R., Jones, B. F., Zhou, W., Reed, K. M. and Werren, J. H. (1995). Induction of paternal genome Rousset, F. and O’Neill, S. L. (1999). Wolbachia infections are loss by the paternal-sex-ratio chromosome and cytoplasmic distributed throughout insect somatic and germ line tissues. Insect incompatibility bacteria (Wolbachia): a comparative study of early Biochem. Mol. Biol. 29, 153–160. embryonic events. Mol. Reprod. Dev. 40, 408–418. Felsenstein, J. (1981). Evolutionary trees from DNA sequences: a Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new maximum likelihood approach. J. Mol. Evol. 17, 368–376. method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, Giordano, R., Jackson, J. J. and Robertson, H. M. (1997). The role 406–425. of Wolbachia bacteria in reproductive incompatibilities and hybrid Sasaki, T., Braig, H. R. and O’Neill, S. L. (1998). Analysis of zones of Diabrotica beetles and Gryllus crickets. Proc. Natl. Acad. Wolbachia protein synthesis in Drosophila in vivo. Insect Mol. Sci. USA 94, 11439–11444. Biol. 7, 101–105. Harrison, R. G. (1983). Barriers to gene exchange between closely Sasaki, T. and Ishikawa, H. (1999). Wolbachia infections and related cricket species. I. Laboratory hybridization studies Gryllus cytoplasmic incompatibility in the almond moth and Mediterranean pennsylvanicus, Gryllus firmus. Evolution 37, 245–251. flour moth. Zool. Sci. 16, 739–744. Hoffmann, A. A. (1988). Partial cytoplasmic incompatibility between Smith, C. J. and Cade, W. H. (1987). Relative fertility in two Australian populations of Drosophila melanogaster. Ent. Exp. hybridization experiments using three song types of the field Appl. 48, 61–67. crickets Gryllus integer and Gryllus rubens. Can. J. Zool. 65, Hoffmann, A. A., Clancy, D. J. and Merton, E. (1994). Cytoplasmic 2390–2394. incompatibility in Australian populations of Drosophila Stevens, L. and Wade, M. J. (1990). Cytoplasmically inherited melanogaster. Genetics 136, 993–999. reproductive incompatibility in Tribolium flour beetles: the rate of Hurst, G. D. D., Jiggins, F. M., von-der-Schulenburg, J. H. G., spread and effect on population size. Genetics 124, 367–372. Bertrand, D., West, S. A., Goriacheva, I. I., Zakharov, I. A., Stouthamer, R., Breeuwer, J. A. J. and Hurst, G. D. D. (1999). Werren, J. H., Stouthamer, R. and Majerus, M. E. N. (1999). Wolbachia pipientis: microbial manipulation of Male-killing Wolbachia in two species of insect. Proc. R. Soc. reproduction. Annu. Rev. Microbiol. 53, 71–102. Lond. B 266, 735–740. Stouthamer, R., Breeuwer, J. A. J., Luck, R. F. and Werren, J. Wolbachia infection in cricket 2509

H. (1993). Molecular identification of microorganisms associated Werren, J. H., Zhang, W. and Guo, L. R. (1995a). Evolution and with parthenogenesis. Nature 361, 66–68. phylogeny of Wolbachia: reproductive parasites of arthropods. Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). Proc. R. Soc. Lond. B 261, 55–63. CLUSTAL W: improving the sensitivity of progressive multiple Werren, J. H., Windsor, D. and Guo, L. R. (1995b). Distribution sequence alignment through sequence weighting, position-specific of Wolbachia among neotropical arthropods. Proc. R. Soc. Lond. B gap penalties and weight matrix choice. Nucleic Acids Res. 22, 262, 197–204. 4673–4680. Wright, J. D. and Barr, A. R. (1980). The ultrastructure and Van Meer, M. M. M., Witteveldt, J. and Stouthamer, R. (1999). symbiotic relationships of Wolbachia of mosquitoes of the Aedes Phylogeny of the arthropod endosymbiont Wolbachia based on the scutellaris group. J. Ultrastruct. Res. 72, 52–64. wsp gene. Insect Mol. Biol. 8, 399–408. Zhou, W., Rousset, F. and O’Neil, S. L. (1998). Phylogeny and Werren, J. H. (1997). Biology of Wolbachia. Annu. Rev. Ent. 42, PCR-based classification of Wolbachia strains using wsp gene 587–609. sequences. Proc. R. Soc. Lond. B 265, 509–515.