Novel Clade of Alphaproteobacterial Endosymbionts Associated with Stinkbugs and Other Arthropods

Abstract

Here we report a novel clade of secondary endosymbionts associated with insects and other arthropods. Seed bugs of the genus Nysius (Hemiptera: Lygaeidae) harbor the primary gammaproteobacterial symbiont Schneideria nysicola within a pair of bacteriomes in the abdomen. Our survey of Nysius species for their facultative bacterial associates consistently yielded a novel type of alphaproteobacterial 16S rRNA gene sequence in addition to those of Wolbachia. Diagnostic PCR survey of 343 individuals representing 24 populations of four Nysius species revealed overall detection rates of the alphaproteobacteria at 77.6% in Nysius plebeius, 87.7% in Nysius sp. 1, 81.0% in Nysius sp. 2, and 100% in Nysius expressus. Further survey of diverse stinkbugs representing 24 families, 191 species, and 582 individuals detected the alphaproteobacteria from an additional 12 species representing six families. Molecular phylogenetic analysis showed that the alphaproteobacteria from the stinkbugs form a distinct and coherent monophyletic group in the order Rickettsiales together with several uncharacterized endosymbionts from fleas and ticks. The alphaproteobacterial symbiont clade was allied to bacterial clades such as the endosymbionts of acanthamoebae, the endosymbionts of cnidarians, and Midichloria spp., the mitochondrion-associated endosymbionts of ticks. In situ hybridization and electron microscopy identified small filamentous bacterial cells in various tissues of N. plebeius, including the bacteriome and ovary. The concentrated localization of the symbiont cells at the anterior pole of oocytes indicated its vertical transmission route through host insect generations. The designation “Candidatus Lariskella arthropodarum” is proposed for the endosymbiont clade.

INTRODUCTION

Diverse organisms are in symbiotic association with multiple bacterial species. Many insects are associated with an obligate symbiotic bacterium, which is referred to as the “primary” symbiont. The primary symbiont generally plays important biological roles for growth, survival, and/or reproduction of its host, and therefore its infection is fixed in the host populations. Frequently, these insects also possess one or a few additional facultative bacterial associates, which are referred to as “secondary” symbionts. The secondary symbionts are generally not essential for their host and usually exhibit partial infection frequencies in the host populations (46, 49).

Plant-sucking stinkbugs (Hemiptera: Heteroptera: Pentatomomorpha) are also frequently associated with symbiotic bacteria. While several atypical stinkbugs of the family Lygaeidae are associated with a primary endocellular bacterial symbiont in the bacteriome (39, 40, 41, 43), the majority of stinkbugs are associated with a primary extracellular bacterial symbiont within specialized midgut portions called crypts or ceca (19, 24, 25, 28–30, 32–37, 51–53, 60). When experimentally deprived of the gut symbiont, these stinkbugs suffer retarded growth, sterility, and/or high mortality (1, 19, 24, 27, 33, 35, 37, 51, 60), indicating important biological roles of the symbiont for the host. In contrast, facultative secondary symbionts of stinkbugs have been poorly investigated despite their potential ecological relevance: except for an extensive survey of Wolbachia among diverse stinkbugs (31), only sporadic detections of Wolbachia and Sodalis have been reported (28, 29, 43).

Seed bugs of the genus Nysius, belonging to the family Lygaeidae, comprise an exceptional stinkbug group in association with an endocellular primary symbiont. They possess a pair of bacteriomes in the abdomen, wherein a gammaproteobacterial symbiont Schneideria nysicola is harbored (43). Although biological roles of Schneideria have not been examined experimentally, nutritional supplementation for the host insects is suspected. Meanwhile, their facultative bacterial associates have been poorly characterized, except for sporadic detection of 16S rRNA gene sequences of Wolbachia and other bacteria (43). Here we report a novel clade of alphaproteobacterial secondary symbionts associated with Nysius seed bugs. We demonstrate that this symbiont is not restricted to Nysius spp. but widely found in other stinkbugs and also in some fleas and ticks, identifying a novel arthropod-associated symbiont group belonging to the order Rickettsiales.

MATERIALS AND METHODS

Materials.

Table S1 in the supplemental material lists the stinkbug samples examined in this study. Some of the samples were used in previous studies (31, 43). Some samples were freshly dissected and subjected to DNA extraction, while other samples were immediately put in glass vials filled with acetone and preserved at room temperature until use (18). Several strains of Nysius plebeius and Kleidocerys resedae were maintained in the laboratory at 25°C under a long-day regime (16 h of light and 8 h of dark) on sunflower seeds, whole wheat, and distilled water supplemented with 0.05% ascorbic acid, and these insects were subjected to histological analyses.

Dissection and DNA extraction.

The insects were dissected in phosphate-buffered saline (PBS) (137 mM NaCl, 8.1 mM Na₂HPO₄, 2.7 mM KCl, 1.5 mM KH₂PO₄ [pH 7.4]) with fine forceps under a dissection microscope; the symbiotic organs (either the bacteriome or the midgut fourth section with crypts), the reproductive organs, and the alimentary tracts (excluding the symbiotic section if present) of the insects were isolated for DNA extraction. For small insects whose body length is less than 1 cm, the whole body or the whole abdomen was used instead. Each of the tissue samples was homogenized in a 1.5-ml plastic tube and digested with lysis buffer (10 mM Tris-HCl [pH 8.0], 1 mM EDTA, 0.1 M NaCl, 0.5% sodium dodecyl sulfate, 0.2 mg/ml proteinase K) at 56°C overnight. The lysate was extracted with phenol-chloroform and subjected to ethanol precipitation, and all the precipitated nucleic acid was dried and dissolved in TE buffer (10 mM Tris-HCl [pH 8.0], 1 mM EDTA).

PCR, cloning, and DNA sequencing.

A 1.5-kb fragment of bacterial 16S rRNA gene was amplified with primers 16SA1 and 16SB1 (see Table S2 in the supplemental material). The PCR was conducted with AmpliTaqGold DNA polymerase (Applied Biosystems) and its supplemented buffer system under the following temperature profile: (i) 95°C for 10 min; (ii) 35 cycles, with 1 cycle consisting of 95°C for 30 s, 52°C for 30 s, and 72°C for 2 min; and (iii) a final extension step at 72°C for 10 min. The PCR products were subjected to cloning, restriction fragment length polymorphism (RFLP) genotyping, and DNA sequencing with BigDye terminator v3.1 cycle sequencing kit and 3130/3130xl genetic analyzer (Applied Biosystems) as described previously (35). A 1.2-kb fragment of the 16S rRNA gene of the alphaproteobacterial symbiont was amplified with the primers 16SA1 and Alp1203R (Table S2), and the PCR products were cleaned using exonuclease I (New England BioLabs) and alkaline phosphatase (shrimp) (TaKaRa) and subjected to direct cycle sequencing. Artifactual chimeric sequences were detected by Mallard software (3).

Diagnostic PCR.

A 0.25-kb region of the 16S rRNA gene of the alphaproteobacterial symbiont was amplified with the primers Alp976F (Alp stands for alphaproteobacterial, and F stands for forward) and Alp1203R (R stands for reverse) (see Table S2 in the supplemental material). A 0.6-kb fragment of the wsp gene of the Wolbachia symbiont was amplified with the primers wspF and wspR (Table S2). The PCRs were performed under the following temperature profile: 95°C for 10 min, followed by 35 cycles, with 1 cycle consisting of 95°C for 30 s, 55°C for 45 s, and 72°C for 1 min. Negative- and positive-control samples were included. To confirm the quality of the DNA samples, a 0.65-kb fragment of insect 18S rRNA gene was amplified with primers 2880 and B (Table S2) under the following temperature profile: 95°C for 10 min, followed by 30 cycles, with 1 cycle consisting of 95°C for 15 s, 55°C for 15 s, and 72°C for 1 min.

Molecular phylogenetic analysis.

Multiple alignments of nucleotide sequences were generated by the program MUSCLE (10) within a software package of MEGA5.05 (61). Ambiguously aligned nucleotide sites were corrected manually or removed. Phylogenetic analyses were conducted by three methods: maximum likelihood (ML), maximum parsimony (MP), and neighbor joining (NJ). ML trees were created by the PhyML 3.0 program (20), while MP and NJ trees were constructed by MEGA 5.0.5. In the ML analyses, the GTR + I + G (general time reversible with invariable sites allowed and the rate variation among sites) model (for Fig. 1 and for Fig. S2 in the supplemental material) and the HKY (Hasegawa-Kishino-Yano) + I + G model (for Fig. S1) were chosen for the 16S rRNA gene data set on the basis of the Akaike information criterion using the program Modeltest 3.7 (50) in conjunction with the program PAUP* (59). Bootstrap tests were conducted with 100 replications in the ML analyses and with 1,000 replications in the MP and NJ analyses.

Fig 1.

Phylogenetic placement of the alpha-symbionts, the Wolbachia symbionts, and the Rickettsia symbionts from Nysius spp., Kleidocerys resedae, and other stinkbugs on the basis of 16S rRNA gene sequences. A maximum likelihood phylogeny inferred from 1,118 aligned nucleotide sites is shown, while maximum parsimony and neighbor-joining analyses gave substantially the same results. Bootstrap support values of 50% or higher are shown at the nodes in the following order: maximum likelihood/maximum parsimony/neighbor joining. Asterisks indicate support values lower than 50%. Sequences identified from stinkbugs are indicated in boldface type, while the collection localities in Japan are indicated in parentheses. Sequence accession numbers are shown in brackets. str., strain. The scale bar indicates the branch length measured by the number of substitutions per site. Accession numbers starting with AB were from DDBJ, those beginning with FM and FN were from EMBL, those beginning with JQ, FJ, AF, GU, EF, CP, GQ, AE, DQ, AY, and U were from Genbank, and those beginning with NC were from NCBI RefSeq.

Fluorescent in situ hybridization.

Table S2 in the supplemental material lists the fluorochrome-labeled oligonucleotide probes used for 16S rRNA-targeted fluorescent in situ hybridization. The insects were dissected, thoroughly washed in PBS, and fixed in Carnoy's solution (ethanol-chloroform-acetic acid [6:3:1]). After the tissues were fixed overnight, they were treated with 6% hydrogen peroxide in 80% ethanol for several weeks for quenching autofluorescence of the tissues (38). After thorough washing with absolute ethanol and phosphate-buffered saline with Tween 20 (PBST) buffer (137 mM NaCl, 8.1 mM Na₂HPO₄, 2.7 mM KCl, 1.5 mM KH₂PO₄ [pH 7.4], 0.2% Tween 20), the samples were incubated with hybridization buffer (20 mM Tris-HCl [pH 8.0], 0.9 M NaCl, 0.01% sodium dodecyl sulfate, 30% formamide) three times for 5 min each time. Then, the samples were hybridized with hybridization buffer containing the probes (100 nM each) and SYTOX green (0.25 μM) overnight. After thorough washing with PBST, the samples were mounted with Slowfade antifade solution (Invitrogen) and observed with a laser confocal microscope (Pascal 5; Carl Zeiss).

Electron microscopy.

Insects were dissected with fine forceps in 0.1 M sodium phosphate buffer (pH 7.4) containing 2.5% glutaraldehyde, and isolated bacteriomes and ovaries were prefixed in the fixative at 4°C overnight and postfixed in 2% osmium tetroxide at 4°C for 60 min. After dehydration through an ethanol series, the materials were embedded in Spurr resin. Ultrathin sections were made on an ultramicrotome (Ultracat-N; Leichert-Nissei), mounted on copper meshes, stained with uranyl acetate and lead citrate, and observed with a transmission electron microscope (model H-7600; Hitachi).

Nucleotide sequence accession numbers.

The nucleotide sequences determined in this study have been deposited in the DDBJ/EMBL/GenBank nucleotide sequence database under the accession numbers JQ726711 to JQ726831 (see Tables S1 and S3 and Fig. S1 and S2 in the supplemental material).

RESULTS

Bacterial 16S rRNA gene sequences from Nysius plebeius.

The bacteriomes, ovaries, and midguts were dissected from five adult females of N. plebeius collected in Tsukuba, Japan, and subjected to DNA extraction. The DNA samples were subjected to PCR, cloning, and RFLP genotyping of a 1.5-kb region of the bacterial 16S rRNA gene. A total of 173 clones were analyzed, and almost all of them, except for three clones with unique RFLP patterns, were classified into three RFLP genotypes, tentatively named type A (98 clones), type B (43 clones), and type C (29 clones) (see Table S3 in the supplemental material). The type A clones, 1,479 bp in size, were predominant in the bacteriome (48/55 clones) and the ovary (38/59 clones) but minor in the midgut (12/57 clones) and represented Schneideria nysicola, the bacteriome-associated gammaproteobacterial primary symbiont of Nysius spp. (43). The type B clones, 1,427 bp in size, were obtained from all the tissues at lower frequencies (4/55 clones from the bacteriome, 12/59 clones from the ovary, and 27/57 clones from the midgut), which represented an alphaproteobacterial Wolbachia symbiont of N. plebeius (43). The type C clones, 1,420 bp in size, were identified from all the tissues at the lowest frequencies (3/55 clones from the bacteriome, 9/59 clones from the ovary, and 17/57 clones from the midgut), with the highest BLAST hit to an alphaproteobacterial 16S rRNA gene sequence belonging to the Rickettsiales from the tick Ixodes ovatus (99.1% [1,147/1,157] identity; GenBank accession number AB297807) (17). Hereafter, we refer to the type C bacterium as the alpha-symbiont of N. plebeius. Of three exceptional clones, two turned out to be chimeras of the different symbiont sequences, whereas a clone obtained from the midgut exhibited the highest BLAST hit to a 16S rRNA gene sequence of Phyllobacterium myrsinacearum (99.9% [1,335/1,337] identity; GenBank accession number AB681132).

Ubiquitous occurrences of the alpha-symbiont and Wolbachia symbiont in Nysius spp. and Kleidocerys resedae.

Similarly, we analyzed the bacterial 16S rRNA sequences obtained from eight additional populations of N. plebeius, four populations of Nysius sp. 1, a population of Nysius sp. 2, and a population of N. expressus. In addition to the primary symbiont Schneideria, the alpha-symbiont was consistently detected from all the species and populations. The 16S rRNA gene sequences of the alpha-symbiont, 1,421 bp in size, were completely or nearly identical (99.9 to 100% identities) to the sequence from N. plebeius. The Wolbachia symbiont was also detected from all the species and populations but the population of N. plebeius from Iriomote-jima, Japan. From two populations of Nysius sp. 1, a population of Nysius sp. 2, and a population of N. expressus, a different type of alphaproteobacterial 16S rRNA gene sequences was also detected, which exhibited the highest BLAST hit to the sequence of Rickettsia bellii from the tick Dermacentor variabilis (99.6% [1,418 or 1,419/1,424] identity; GenBank accession number CP000087). Furthermore, a number of minor bacterial 16S rRNA gene sequences were identified from the midgut and whole-body DNA samples, most of which are probably derived from gut microbes (see Table S3 in the supplemental material).

We also analyzed bacterial 16S rRNA sequences from Kleidocerys resedae, a lygaeid species related to but distinct from Nysius spp., which possesses a bacteriome-associated primary symbiont Kleidoceria schneideri (39). Of two insect populations examined, the alpha-symbiont and a Rickettsia symbiont were detected at low frequencies in the insect population from Okunikko, Japan, whereas Wolbachia symbionts were found in both populations (see Table S3 in the supplemental material). The 16S rRNA gene sequences of the alpha-symbiont, 1,421 bp in size, were almost identical (99.4 to 99.5% identities) to the sequences from Nysius spp.

Next we performed a wider diagnostic PCR survey of the alpha-symbiont and the Wolbachia symbiont across many populations of Nysius spp. and K. resedae. A total of 183 individuals from 13 populations of Nysius plebeius, 106 individuals from seven populations of Nysius sp. 1, 21 individuals from four populations of Nysius sp. 2, 33 individuals from a population of N. expressus, and 46 individuals from two populations of K. resedae were analyzed. Both the alpha-symbiont and the Wolbachia symbiont were detected from almost all the insect species and populations at high frequencies. The overall infection frequencies were 77.6% (142/183) in N. plebeius, 87.7% (93/106) in Nysius sp. 1, 81.0% (17/21) in Nysius sp. 2, 100% (33/33) in N. expressus, and 87.0% (40/46) in K. resedae for the alpha-symbiont and 86.9% (159/183) in N. plebeius, 88.7% (94/106) in Nysius sp. 1, 95.2% (20/21) in Nysius sp. 2, 100% (33/33) in N. expressus, and 100% (46/46) in K. resedae for the Wolbachia symbiont (see Table S1 in the supplemental material).

Occurrences of the alpha-symbiont in diverse stinkbugs.

Furthermore, we performed a diagnostic PCR survey of the alpha-symbiont across diverse heteroptaran bugs representing 24 families, 190 species, and 536 individuals. Among these bugs, the alpha-symbiont was detected from 11 species representing six families, including Arocatus melanostomus (2/2) (Lygaeidae), Dimorphopterus pallipes (1/1) (Blissidae), Horridipamera inconspicua (1/2) and Paromius exguus (3/3) (Rhyparochromidae), Physopelta gutta (2/2) and Physopelta cincticollis (1/3) (Largidae), Neuroctenus castaneus (1/1) (Aradidae), and Dolycoris baccarum (1/15), Eysarcoris guttiger (1/8), Eysarcoris annamita (12/12), and Piezodorus hybneri (1/7) (Pentatomidae) (see Table S1 in the supplemental material).

For each of the alpha-symbiont-positive species, a 1.2-kb region of the 16S rRNA gene was sequenced. All the sequences, 1,129 to 1,150 bp in size, exhibited very high sequence identities (99.6 to 100%) to the alpha-symbiont sequences from Nysius spp.

Phylogenetic placement of alpha-symbionts from diverse stinkbugs.

Molecular phylogenetic analyses based on the 16S rRNA gene sequences revealed that the alpha-symbionts of Nysius spp., K. resedae, and other stinkbugs form a very compact monophyletic group, wherein only a few nucleotide variations are present among them. Notably, the alpha-symbiont clade also contained 16S rRNA gene sequences of uncultured bacteria derived from fleas (13) and ticks (17, 44). In the Alphaproteobacteria, the alpha-symbiont clade was placed within the Rickettsiales, whose members are endocellular symbionts/parasites/pathogens such as Rickettsia, Wolbachia, Ehrlichia, Anaplasma, etc. (6, 9). The alpha-symbiont clade was allied to the following bacterial groups: the endosymbionts of acanthamoebae (15), the endosymbionts of cnidarians (14, 58), and Midichloria spp., the mitochondrion-associated endosymbionts of ticks (12, 54) (Fig. 1).

Phylogenetic placement of Wolbachia symbionts and Rickettsia symbionts of Nysius spp. and Kleidecerys resedae.

Most of the 16S rRNA gene sequences of the Wolbachia symbionts from Nysius spp. and K. resedae were placed in the Wolbachia supergroup B, except for a sequence from K. resedae in the Wolbachia supergroup A (Fig. 1; see also Fig. S1 in the supplemental material). In the supergroup B, the Wolbachia sequences from different populations of the same Nysius species were almost identical to each other, except for the two distinct lineages of the Wolbachia sequences from Nysius sp. 1. The 16S rRNA gene sequences of the Rickettsia symbionts from N. expressus, Nysius sp. 1, and Nysius sp. 2 were all allied to the sequence of Rickettsia bellii, while the Rickettsia sequence from K. resedae clustered with Rickettsia felis and R. akari (Fig. 1; see also Fig. S2 in the supplemental material).

Localization and fine structure of the alpha-symbiont in Nysius plebeius and Kleidocerys resedae.

The following laboratory-maintained strains of N. plebeius were subjected to fluorescent in situ hybridization of the symbionts. The NpFky002 (Np stands for N. plebeius, and Fky stands for Fukuyama) strain was established from an adult female collected at Fukuyama, Hiroshima, Japan, and infected with the alpha-symbiont only in addition to the Schneideria symbiont. The NpTsk002 (Tsk stands for Tsukuba) strain was derived from an adult female collected in Tsukuba, Ibaraki, Japan, and infected with both the alpha-symbiont and the Wolbachia symbiont in addition to the Scineideria symbiont. In the disymbiotic strain NpFky002, the alpha-symbiont signals were detected in various tissues, including bacteriomes, ovaries, and midguts at low densities (Fig. 2A to D). In the bacteriomes, the alpha-symbiont signals were sporadically found in the cytoplasm of bacteriocytes together with intense signals of the Schneideria symbiont (Fig. 2A). In the ovaries, the alpha-symbiont signals were sparsely detected in nurse cells, follicular cells, and Schneideria-harboring ovarial bacteriocytes (Fig. 2B). At the anterior pole of oocytes, the alpha-symbiont signals were seen as filamentous bacterial cells smaller than the Schneideria symbiont cells, where the alpha-symbiont cells formed a cluster adjacent to but distinct from a larger cluster of tubular cells of the Schneideria symbiont (Fig. 2C). Midgut epithelial cells also exhibited sparse signals of the alpha-symbiont (Fig. 2D). In the trisymbiotic strain NpTsk002, the alpha-symbiont signals generally exhibited localization patterns similar to those in strain NpFky002, together with signals of the Wolbachia symbiont. The alpha-symbiont signals cooccurred in bacteriocytes, nurse cells, follicular cells, and ovarial bacteriocytes and at the anterior poles of oocytes, with signals of the Wolbachia symbiont, but usually at lower densities (Fig. 2E to G). Transmission electron microscopy revealed small filamentous cells of the alpha-symbiont between numerous tubular cells of the Schneideria symbiont within the bacteriocytes constituting the bacteriome (Fig. 3A). The alpha-symbiont cells, containing characteristic vacuoles in their cytoplasm, were also found in nutritive cord, nurse cells, and follicular cells of the ovarioles (Fig. 3B).

Fig 2.

In vivo localization of the alpha-symbiont visualized by fluorescent in situ hybridization. (A to D) Disymbiotic strain NpFky002 of Nysius plebeius infected with the alpha-symbiont in addition to the primary Schneideria symbiont. Red, green, and blue signals indicate the alpha-symbiont, the Schneideria symbiont, and host insect DNA, respectively. (A) Bacteriocytes making up a bacteriome. (B) Tip region of an ovariole. (C) Anterior pole of an oocyte. (D) Midgut epithelium. (E to G) Trisymbiotic strain NpTsk002 of N. plebeius infected with the alpha-symbiont, the Wolbachia symbiont, and the Schneideria symbiont, Red, green, and blue signals indicate the alpha-symbiont, the Wolbachia symbiont and host insect DNA, respectively, while the Schneideria symbiont is not visualized except for the DNA staining. (E) Bacteriocytes constituting a bacteriome. (F) Tip region of an ovariole. Signals of the Wolbachia symbiont are particularly concentrated in the germarium region, which mainly consists of nurse cells and also contains somatic and germ line stem cells (16). (G) Developing oocytes, wherein the Wolbachia symbiont and the alpha-symbiont colocalize and form a cluster at the anterior pole. (H and I) Trisymbiotic strain KrOkn001 of Kleidocerys resedae infected with the alpha-symbiont, the Wolbachia symbiont, and the primary Kleidoceria symbiont. Red, green, and blue signals indicate the alpha-symbiont, the Wolbachia symbiont, and host insect DNA, respectively, while the Kleidoceria symbiont is not visualized except for the DNA staining. (H) Bacteriocytes constituting a bacteriome. (H) Midgut epithelium. Abbreviations: a, signal of the alpha-symbiont; f, follicle cell; g, germarium mainly consisting of nurse cells; iz, infection zone in germarium (43); n, nucleus of bacteriocyte; o, oocyte; ob, ovarial bacteriocyte in infection zone (43); s, signal of the Schneideria symbiont; w, signal of the Wolbachia symbiont. The arrows pointing up and down indicate the anterior poles and posterior poles of oocytes, respectively.

Fig 3.

Transmission electron microscopy of the alpha-symbiont in Nysius plebeius. (A) In the bacteriocyte, filamentous alpha-symbiont cells are seen between numerous Schneideria cells of larger size. (B) An enlarged image of alpha-symbiont cells in the ovariole. Abbreviations: a, alpha-symbiont cell; s, Schneideria symbiont cell.

We also histologically inspected a laboratory-maintained strain KrOkn001 (Kr stands for K. resedae, and Okn stands for Okunikko) of K. resedae collected in Okunikko, Tochigi, Japan, which was infected with both the alpha-symbiont and the Wolbachia symbiont in addition to the primary Kleidocerya symbiont, and observed similar localization patterns of the alpha-symbiont and the Wolbachia symbiont (Fig. 2H and I).

DISCUSSION

In this study, we identified a novel clade of alphaproteobacterial 16S rRNA gene sequences from 16 stinkbug species representing six families (Fig. 1; see also Table S1 in the supplemental material). The alpha-symbiont clade also contained endosymbiont gene sequences from fleas and ticks (Fig. 1). In Nysius seed bugs, the alpha-symbionts were at high infection frequencies ranging from 76% to 100% within the species (Table S1) and observed as filamentous bacteria in various cells and tissues (Fig. 2 and 3). The alpha-symbiont cells were particularly concentrated at the anterior poles of oocytes in the ovaries of adult females, indicating the vertical transmission route of the symbiont through the host generations (Fig. 2 and 3). On the basis of these results, we conclude that the alpha-symbionts comprise a facultative symbiont clade belonging to the Rickettsiales, whose hosts are stinkbugs and other arthropods.

The alphaproteobacterial order Rickettsiales contains well-known bacterial groups, such as Wolbachia, Rickettsia, Ehrlichia, and Anaplasma, whose members are associated with a diverse set of host/vector species with various relationships ranging from parasitic/pathogenic through commensal to mutualistic (6, 9). The alpha-symbiont clade was allied not only to an endosymbiont clade associated with ticks (Midichloria spp.) but also to endosymbiont clades associated with cnidarians, acanthamoebae, and ciliates (Fig. 1), highlighting the diversity of host-symbiont associations in the Rickettsiales.

Currently, infections with the alpha-symbiont have been detected only from stinkbugs (Hemiptera), fleas (Siphonaptera), and ticks (Acari). Biological or ecological connections between the disparate host taxa are obscure, but their nutritionally unbalanced food sources, such as plant sap deficient in essential amino acids for stinkbugs and vertebrate blood lacking in B vitamins for fleas and ticks (8), might be of some relevance. Meanwhile, it seems likely that further surveys across diverse insects and other arthropods would reveal a much wider host range of the alpha-symbiont.

The biological effects of the alpha-symbiont infection on the host stinkbugs and other arthropods deserve future study. Considering the high infection frequencies in Nysius spp. and K. resedae (see Table S1 in the supplemental material), it seems plausible that the alpha-symbiont is conditionally beneficial for the hosts, as facultative symbionts Serratia, Hamiltonella, and Regiella are for their host aphids (45, 48, 49, 56, 63). Alternatively, the possibility that the alpha-symbiont may induce cytoplasmic incompatibility or other reproductive phenotypes in the hosts, thereby spreading in the host populations, as known for facultative symbionts Wolbachia and Cardinium (11, 66), should be taken into account. Since no remarkable bias in the sex ratio is observed in natural populations of Nysius spp. and K. resedae (Y. Matsuura, personal observation), it seems unlikely that the alpha-symbiont is causing either parthenogenesis, feminization, or the male killing. Comparative genomic analyses of the alpha-symbiont with other symbiotic bacteria in the Alphaproteobacteria will shed further light on its functional and evolutionary aspects (55).

Among facultative microbial associates of insects and other arthropods, Wolbachia is the most prevalent one. Many strains induce cytoplasmic incompatibility or other reproductive phenotypes, thereby driving their own infections into the host populations (11, 66). Some strains conditionally benefit the fitness of their hosts (4, 5, 21, 62, 64). Some strains are indispensable for their hosts (7, 23, 26, 42). Wolbachia infections are estimated to be prevalent in more than 60% of insect species (22). Similarly, Rickettsia (65), Spiroplasma (2), Arsenophonus (47), Cardinium (67), and Sodalis (57) have been recognized as comprising ecologically widespread and phylogenetically coherent facultative symbiont groups. Although our current data are still limited, we expect that the alpha-symbiont clade identified in this study would constitute a new group of such widespread symbionts associated with insects and other arthropods.

On account of the distinct and coherent phylogenetic and microbiological traits described in this study, we propose the designation “Candidatus Lariskella arthropodarum” for the novel Rickettsiales clade associated with stinkbugs, fleas, and ticks. The generic name is after a Russian animation character Lariska, a long, black, and rat-like creature in the classic movie series Cheburashka, reflecting the first molecular identification of the bacterium from Russia (44) and the morphology of the bacterium reminiscent of the character. The specific epithet refers to its association with arthropod hosts.

The order Rickettsiales is known for a number of animal and human pathogens, including Rickettsia prowazekii for epidemic typhus, Rickettsia typhi for murine typhus, Rickettsia rickettsii for Rocky Mountain spotted fever, Rickettsia felis for flea-borne spotted fever, Orientia tsutsugamushi for scrub typhus, Ehrlichia chaffeensis for monocytic ehrlichiosis, Ehrlichia ruminantium for cattle heartwater, etc. (6). Notably, Mediannikov et al. (44) detected a 16S rRNA gene sequence of “Candidatus Lariskella arthropodarum” from blood and biopsy samples of patients with acute fevers who had been bitten by Ixodes ticks. Hence, potential medical relevance of the bacterial group should be a focal issue in future studies.