16S rRNA Gene Sequencing of Six Psyllid Species of the Family Carsidaridae Identified Various Bacteria Including Symbiopectobacterium

Psyllids (Hemiptera: Sternorrhyncha: Psylloidea) are plant sap-sucking insects that are closely associated with various microbes. To obtain a more detailed understanding of the ecological and evolutionary behaviors of microbes in Psylloidea, the bacterial populations of six psyllid species, belonging to the family Carsidaridae, were analyzed using high-throughput amplicon sequencing of the 16S rRNA gene. The majority of the secondary symbionts identified in the present study were gammaproteobacteria, particularly those of the order Enterobacterales, including Arsenophonus and Sodalis, which are lineages found in a wide variety of insect hosts. Additionally, Symbiopectobacterium, another Enterobacterales lineage, which has recently been recognized and increasingly shown to be vertically transmitted and mutualistic in various invertebrates, was identified for the first time in Psylloidea. This lineage is closely related to Pectobacterium spp., which are plant pathogens, but forms a distinct clade exhibiting no pathogenicity to plants. Non-Enterobacterales gammaproteobacteria found in the present study were Acinetobacter, Pseudomonas (both Pseudomonadales), Delftia, Comamonas (both Burkholderiales), and Xanthomonas (Xanthomonadales), a putative plant pathogen. Regarding alphaproteobacteria, three Wolbachia (Rickettsiales) lineages belonging to supergroup B, the major group in insect lineages, were detected in four psyllid species. In addition, a Wolbachia lineage of supergroup O, a minor group recently found for the first time in Psylloidea, was detected in one psyllid species. These results suggest the pervasive transfer of bacterial symbionts among animals and plants, providing deeper insights into the evolution of the interactions among these organisms.

According to a definition revised by Burckhardt et al., psyllids are classified into seven extant families: Aphalaridae, Calophyidae, Carsidaridae, Liviidae, Mastigimatidae, Psyllidae, and Triozidae (Burckhardt et al., 2021).Carsidaridae is a relatively small family consisting of three subfamilies, Carsidarinae, Homotominae, and Pachypsyllinae, which comprise 148 species of 23 genera (Ouvrard, 2023).This family encompasses agricultural pests, including Mesohomotoma tessmanni (Carsidarinae), Allocarsidara malayensis (Carsidarinae), and Homotoma ficus (Homotominae), which feed on cacao (Yana et al., 2015), durian (Ketsa et al., 2020), and fig (Tasi et al., 2021) trees, respectively.A recent study performed a relatively comprehensive analysis of 16S rRNA gene amplicons using 44 psyllid species of five families, including four Carsidaridae species: H. ficus, Macrohomotoma gladiata (Homotominae), Mesohomotoma hibisci (Carsidarinae), and Protyora sterculiae (Carsidarinae) (Kwak et al., 2021).The analysis revealed various bacteria, including two lineages of "Ca.Liberibacter" that are potential plant pathogens.One of them was a close relative of "Ca.Liberibacter asiaticus", which is a notorious pathogen of the most devastating citrus disease, Huanglongbing or greening disease, and the other was "Ca.Liberibacter capsica", a novel species that may potentially cause diseases in solanaceous crops.Although this study provided important insights into the evolution of microbiomes in Psylloidea, the analysis was performed using (1) only one individual or a pool of two individuals for each psyllid species, making the analysis less reliable; (2) primers unsuitable for detecting bacteria with AT-rich 16S rRNA genes, including Carsonella; and (3) the analytical method that clusters sequence reads with a similarity threshold of 97%, resulting in a lower resolution of the analysis.Moreover, (4) none of the representative sequences were deposited in public databases, making phylogenetic analyses difficult to reproduce.
In the present study, amplicon analyses were performed based on Illumina sequencing of the V3 and V4 regions of 16S rRNA genes to assess the microbiomes of six Carsidaridae species collected in Japan (Table 1) using (1) ten (five adult males and five adult females) pooled individuals for each species, (2) primers optimized to detect 16S rRNA genes with a wider variety of G+C contents, and (3) the method to remove sequencing errors and resolve sequence variants (SVs) down to the level of single-nucleotide differences.Furthermore, (4) all of the main SVs (>1% of the total reads) have been deposited in public databases, providing the research community with a basis for the further investigation and reexamination of the analytical results.

Insect sampling and DNA extraction
Adults of six psyllid species belonging to the family Carsidaridae were collected from their host plants at various locations in Japan (Table 1).Insect samples were stored in acetone (Carsidara limbata, Homotoma radiata, and Celtisaspis japonica) or 99.5% ethanol (the other species) at -20°C until DNA extraction.DNA was extracted from the whole bodies of pooled individuals of five adult males and five adult females of each psyllid species using the DNeasy Blood & Tissue Kit (Qiagen).The quality of extracted DNA was assessed using a NanoDrop 2000c spectrophotometer Maruyama et al.

/ 16
Article ME23045 (Thermo Fisher Scientific).Its quantity was assessed using a Qubit 2.0 fluorometer with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific).

Construction and sequencing of amplicon libraries
Bacterial populations in psyllids were analyzed in accordance with the instructions provided by Illumina (Illumina, 2013) with some modifications (Nakabachi et al., 2020a(Nakabachi et al., , 2022a(Nakabachi et al., , 2022b).An amplicon polymerase chain reaction (PCR) was performed using DNA extracted from psyllids, KAPA HiFi HotStart ReadyMix (KAPA Biosystems), and the primer set 16S_341Fmod (5′-TCGTC GGCAGCGTCAGATGTGTATAAGAGACAGYYTAMGGRNGG CWGCAG-3′) and 16S_805R (5′-GTCTCGTGGGCTCGGAGAT GTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3′), targeting the V3 and V4 regions of the 16S rRNA gene.Underlined areas indicate regions corresponding to target 16S rRNA gene sequences.Other nucleotides were the overhangs required for preparing and sequencing Illumina libraries.Although both primers were based on the instructions by Illumina (Illumina, 2013), 16S_341F was modified (bold italics), where the original CC, C, and G were replaced with the mixed bases YY (C or T), M (A or C), and R (A or G), respectively.This modification was based on an alignment of 16S rRNA genes encoded in divergent bacterial genomes with various G+C contents.Referenced sequences included those of 44 diverse lineages of Carsonella derived from 38 psyllid species from 23 genera, belonging to five families, which were available in the nucleotide database of the National Center for Biotechnology Information (NCBI) as of August 2014.The modification achieved a 100% match of the primers to most Carsonella sequences, retaining a single mismatch each in four lineages (Supplementary Fig. S1).In our reanalysis in May 2023, we identified 11 more Carsonella sequences deposited in the NCBI nucleotide database, nine of which showed no mismatch to the primers.One sequence had a single mismatch and another showed two mismatches (single each for 16S_341Fmod and 16S_805R) (Supplementary Fig. S1).This modification increased sensitivity to detect symbionts with AT-rich genomes, including Carsonella, without reducing sensitivity to those with GC-rich genomes (Nakabachi et al., 2020a(Nakabachi et al., , 2022a(Nakabachi et al., , 2022b)).Dual indices and Illumina sequencing adapters were attached to the amplicons with index PCR using Nextera XT Index Kit v2 (Illumina).The libraries were combined with PhiX Control v3 (Illumina), and 250 bp of both ends were sequenced on the MiSeq platform (Illumina) with MiSeq Reagent Kit v2 (500 cycles; Illumina).

Computational analysis of bacterial populations
Output sequences were processed using the QIIME2 platform (version 2022.8)(Bolyen et al., 2019).Primer sequences were removed from the demultiplexed sequence reads using the cutadapt plugin (Martin, 2011).The denoising and joining of paired-end reads and removal of low-quality or chimeric reads were subsequently performed using the DADA2 plugin (Callahan et al., 2016).In this process, parameters were set to --p-trunc-len-f 230 and --p-trunc-len-r 225 to remove 3′-end nucleotides with quality scores below the Phred score 30, referring to the quality histograms drawn with fastQC (version 0.11.9)(Andrews, 2023) and multiQC (version1.12)(Ewels et al., 2016).Dereplicated amplicon reads were classified, and taxonomic information was assigned using the q2-feature-classifier (Bokulich et al., 2018), which was trained with the V3 and V4 regions of the 16S rRNA gene (Silva 138 SSURef NR99) (Glöckner et al., 2017).The SVs obtained were manually checked by BLASTN searches against the NCBI non-redundant database (Camacho et al., 2009).After SVs were aligned to related sequences using SINA (version 1.2.11) (Pruesse et al., 2012), phylogenetic trees were inferred by the maximum likelihood (ML) method using RAxML (version 8.2.12) (Stamatakis, 2014).The GTR+Γ model was used with no partitioning of the data matrix, with 1,000 bootstrap iterations (options -f a -m GTRGAMMA -# 1000).

Data availability
Nucleotide sequence data are available in the DDBJ/EMBL/ GenBank databases under the accession numbers DRR420941-DRR420946 (raw fastq files) and TAAE01000001-TAAE01000035 (dereplicated SVs).

Overview
Raw libraries after demultiplexing yielded 48,874-83,223 pairs of forward and reverse reads for the six psyllid species.The denoising and joining of paired-end reads and removal of low-quality or chimeric reads resulted in 42,709-64,067 non-chimeric high-quality reads (Supplementary Table S1).The dereplication of these reads resulted in 342 independent SVs, 33 of which accounted for >1% of the total reads (Supplementary Table S2 and Fig. S2).The present study focused on the 33 main SVs unless otherwise noted because (1) filtering with a threshold of 1% was previously demonstrated to be among the most accurate and effective methods to eliminate potential contaminants (Karstens et al., 2019) and ( 2) the targets of the present study were relatively abundant symbionts with close associations with the host psyllids.SVs with a relative abundance of <1% were collectively categorized as 'others' (Fig. 1) and accounted for 0.1 (H.radiata) -13.6% (C.japonica) of the total reads in each psyllid species (Supplementary Table S2).

Identification of Carsonella
The taxonomic classification by QIIME2 (Supplementary Table S2) followed by independent BLAST searches and phylogenetic analyses showed that all psyllid species, except for Mesohomotoma camphorae, possess distinct lineages of Carsonella (Fig. 1).The ML tree placed these Carsonella sequences at positions that are largely consistent with the host psyllid phylogeny inferred by mitochondrial and nuclear gene analyses with the aid of morphological analyses (Percy et al., 2018;Cho et al., 2019;Burckhardt et al., 2021) (Fig. 2).SV4 and SV5, derived from the congeneric species Homotoma unifasciata (27.9% of the reads) and H. radiata (38.2% of the reads; both Homotominae), respectively, formed a robustly supported clade (bootstrap: 98%).This clade further formed a well-supported clade (bootstrap: 88%) with other psyllid species belonging to the same subfamily Homotominae (Fig. 2).SV12 and SV28, derived from C. limbata (14.8% of the reads) and Tyora ornata (4.1% of the reads; both Carsidarinae), respectively, also formed a moderately supported clade (bootstrap: 69%) with a sequence from P. sterculiae (Carsidaridae: Carsidarinae, KY427941).SV18, which accounted for 8.4% of C. japonica (Pachypsyllinae) reads, was placed at the basal position in the Carsonella tree (Fig. 2).Although SV18 did not form a clade with sequences derived from other Pachypsyllinae species of the New World endemic genus Pachypsylla, the phylogenetic position was proximal to the moderately supported clade consisting of these sequences (bootstrap: 67%) (Fig. 2).These results were consistent with previous findings showing that host psyllids and Carsonella cospeciated due to the stable vertical transmission of Carsonella (Thao et al., 2000b;Spaulding and von Dohlen, 2001;Hall et al., 2016;Nakabachi et al., 2020aNakabachi et al., , 2022aNakabachi et al., , 2022b)).
In contrast, no Carsonella-like SVs with a relative abundance of ≥1% were detected in M. camphorae (Supplementary Table S2).This was unexpected because Carsonella was considered to be essential in Psylloidea, and the primer set (Supplementary Fig. S1) used in the present study was shown to be suitable for detecting Carsonella (Nakabachi et al., 2020a(Nakabachi et al., , 2022a(Nakabachi et al., , 2022b)).Therefore, the search was extended to include SVs with a relative abundance of <1% in M. camphorae.SV73, the most abundant SV in M. camphorae showing similarity to Carsonella sequences, accounted for only 0.16% of M. camphorae reads (Supplementary Table S2).The sequence was only 93.3% identical to the most similar sequences in the database, which were derived from Carsonella of Diaphorina cf.continua (AP023214, TAAA01000009) and Diaphorina lycii (TAAA01000012).These psyllid species belong to the family Psyllidae and are distantly related to M. camphorae.Moreover, the ML tree placed this SV at a position apart from Carsonella sequences found in other Carsidaridae species (Fig. 2).No sequence reads analyzed in the same MiSeq flow cell showed higher similarity to SV73 than the sequences found in Diaphorina spp.These results imply that SV73 was derived from a chimeric PCR artifact that QIIME2 failed to remove or a minor contaminant of an unknown source, not a true symbiont in M. camphorae.Similarly, all other M. camphorae SVs showing similarity to Carsonella, namely, SV125 (0.06% of the reads), SV213 (0.02%), SV219 (0.02%), and SV293 (0.01%), appeared to be derived from artifacts or contaminants.These types of minor SVs that showed similarity to Carsonella sequences, but appeared to correspond to PCR artifacts or contaminants of unknown sources were also detected in C. limbata, T. ornata, and C. japonica (Supplementary Table S2).

Enterobacterales symbionts in Carsidaridae
Of the 33 main SVs obtained in the present study, 28 corresponded to gammaproteobacteria, 14 of which belonged to the order Enterobacterales (Supplementary Table S2).

First report on a Symbiopectobacterium symbiont in Psylloidea
SV3, which accounted for as much as 41.0% of M. camphorae reads (Supplementary Table S2), was 97.4% identical to the sequence of Symbiopectobacterium purcellii, an endosymbiont found in the leafhopper Empoasca decipiens (CP081864) (Nadal-Jimenez et al., 2022b).These sequences formed a robustly supported clade (bootstrap: 99%) in the ML tree of the order Enterobacterales (Fig. 3).Since the genus Symbiopectobacterium is closely related to the genus Pectobacterium, which includes important plant pathogens (Rossmann et al., 2018;Oulghazi et al., 2021), we performed a more comprehensive analysis that focused on Symbiopectobacterium and related bacterial genera.The results obtained placed SV3 in a well-supported clade (bootstrap: 71%) of vertically transmitted Symbiopectobacterium symbionts recently recognized and identified in various invertebrate lineages (Fig. 4) (Hosokawa et al., 2010;Kuechler et al., 2011Kuechler et al., , 2012;;da Mota et al., 2012;Koga et al., 2013;Lopez-Madrigal et al., 2014;Husnik and McCutcheon, 2016;Martinson et al., 2020;Nadal-Jimenez et al., 2022b).This clade is distinct from that of Pectobacterium spp., which are plant pathogens that cause soft rot disease in various economically important crops (Rossmann et al., 2018;Oulghazi et al., 2021) (Fig. 4).Bacteria belonging to this newly emerging Symbiopectobacterium clade have mainly been discovered in hemipteran insects, including plant-sap feeders and vertebrate-blood feeders.However, to the best of our knowledge, this is the first study to report Symbiopectobacterium in Psylloidea.
Other Enterobacterales symbionts SV2 (84.7% of C. limbata reads), SV6 (24.4% of H. unifasciata reads), SV8 (30.7% of H. radiata reads), SV10 (26.2% of T. ornata reads), and SV13 (16.6% of T. ornata reads) were placed at ambiguous positions in the ML tree (Fig. 1 and 3, Supplementary Table S2).Since their branching patterns were mostly poorly supported (bootstrap: <50%) and their sequence identities with those of bacteria with a genus name were low (<94.5%),they were collectively referred to as "Enterobacterales symbionts" (Fig. 1  and 3).Among them, SV6 and SV8, which were derived from congeneric psyllid species, formed a robustly supported clade (bootstrap: 100%) in the ML tree (Fig. 3), suggesting that the corresponding symbionts are sister lineages that share a common ancestor in the common ancestral host.This phylogenetic relationship implies the important and conserved roles of these symbionts in host psyllids.SV31, which accounted for 1.7% of C. japonica reads, was 100% identical to the sequence of Enterobacter spp.(Fig. 1 and 3, Supplementary Table S2).This low relative abundance and 100% identity with free-living bacteria implied that the corresponding bacterium was a transient associate, not a stable symbiont.

Bacteria of Pseudomonadales, Burkholderiales, and Xanthomonadales
Six SVs detected in C. japonica were placed in the clade of the order Pseudomonadales (Fig. 1 and 5, Supplementary Table S2).SV16, SV23, SV27, and SV33, which accounted for 9.2, 3.2, 2.1, and 1.1%, respectively, of C. japonica reads, were resolved as distinct lineages in the robustly supported clade (bootstrap: 99%) of the genus Acinetobacter (Fig. 5).SV22 and SV26, which accounted for 3.7 and 2.5%, respectively, of C. japonica reads, were shown as distinct lineages in the robustly supported clade (bootstrap: 100%) of the genus Pseudomonas (Fig. 5).SV19 and SV25, which accounted for 5.1 and 2.5%, respectively, of C. japonica reads, were inferred as distinct lineages in the robustly supported clade (bootstrap: 100%) within the order Burkholderiales (Fig. 1 and 6, Supplementary Table S2).Although Profftella, a unique organelle-like defensive symbiont (Nakabachi et al., 2013b), also belongs to this order, the ML tree showed that these SVs were distantly related to Profftella.The SVs placed in Pseudomonadales or Burkholderiales were 100% identical to the sequences of free-living species, and their abundance was relatively low.Therefore, these SVs may be derived from transient associates, not stable symbionts of C. japonica.

Discussion
The present study identified various bacterial populations in six psyllid species of the family Carsidaridae collected in Japan.Although distinct lineages of Carsonella were identi-fied in five psyllid species, essentially no sequence for Carsonella was detected in M. camphorae (Fig. 1 and 2, Supplementary Table S2).The universal primers suggested by Illumina (Illumina, 2013) tend to fail to detect AT-rich 16S rRNA genes of Carsonella (Fromont et al., 2017;Morrow et al., 2017Morrow et al., , 2020;;Kwak et al., 2021).However, the primer set used in the present study was modified to detect sequences with a wider variety of G+C contents (Supplementary Fig. S1) and successfully identified Carsonella in diverse psyllid lineages (Nakabachi et al., 2020a(Nakabachi et al., , 2022a(Nakabachi et al., , 2022b)).Therefore, the present results may imply the absence of Carsonella at least in the M. camphorae individuals analyzed in the present study.How- ever, the failure of PCR detection may also be attributed to other reasons, including exceptionally diverged nucleotide sequences at primer annealing sites.Further studies, including comprehensive histological analyses of a large set of specimens from several sampling points, are required to establish whether M. camphorae truly lacks Carsonella.
Regarding Wolbachia (Rickettsiales), the present study identified three lineages belonging to supergroup B, the major group in insect lineages, in four psyllid species.Moreover, supergroup O, another Wolbachia lineage, which is a relatively minor taxon and was recently found for the first time in Psylloidea (Nakabachi et al., 2022a), was identified in M. camphorae.This implies that supergroup O is widespread in Psylloidea, although it has only recently been recognized.Wolbachia is among the most widely distributed symbionts worldwide, estimated to infect 40% of terrestrial arthropods (Werren et al., 2008;Brinker et al., 2019).Although limited lineages are recognized to be obligate mutualists (Hosokawa et al., 2010), they are mostly parasites in insects, in which they manipulate host reproduction to drive their dissemination (Werren et al., 2008;Brinker et al., 2019).They induce resistance to various microbes, including viruses and protozoans, in some insect taxa (Brinker et al., 2019;Wang et al., 2021).Moreover, a genome editing technique using the CRISPR-Cas9 system to transform Wolbachia has recently been developed (Pelz-Stelinski, 13 May 2021, United States Patent and Trademark Office).Therefore, they are regarded as promising agents in the control of pest insects and the microbes that they transmit (Brinker et al., 2019;Pelz-Stelinski, 13 May 2021, United States Patent and Trademark Office;Wang et al., 2021).A high incidence of Wolbachia in pest psyllids worldwide (Spaulding and von Dohlen, 2001;Sloan and Moran, 2012;Arp et al., 2014;Chu et al., 2016Chu et al., , 2019;;Morrow et al., 2017;Nakabachi et al., 2020aNakabachi et al., , 2022b) ) and the interactions observed between Wolbachia and other symbionts (Chu et al., 2016(Chu et al., , 2019;;Jain et al., 2017;Kruse et al., 2017;Killiny, 2022) have led researchers to anticipate using Wolbachia to manipulate pest psyllids (e.g.providing them with resistance to plant pathogens) and/or plant pathogens (e.g.preventing them from infecting plants) (Chu et al., 2016(Chu et al., , 2019;;Kruse et al., 2017;Pelz-Stelinski, 13 May 2021, United States Patent and Trademark Office).The present study suggested the pervasive horizontal transmission of various Wolbachia strains among various insects, including psyllids (Fig. 7).This implies that Wolbachia may be artificially infected into psyllids, forming the basis for the use of this bacterial group for application purposes.
Furthermore, the present study detected potential plant pathogens, including Xanthomonas sp., in H. unifasciata.Since H. unifasciata feeds on Ficus spp.(Moraceae), further studies are warranted to establish whether the host plants are infected with Xanthomonas and if the infection causes disease symptoms.

Conclusions
The present study identified various bacterial symbionts in six psyllid species of the family Carsidaridae.The majority of the secondary symbionts were gammaproteobacteria, particularly those of the order Enterobacterales, including Arsenophonus and Sodalis.In addition, Symbiopectobacterium, another lineage belonging to Enterobacterales, was detected for the first time in Psylloidea.Regarding non-Enterobacterales gammaproteobacteria, Acinetobacter, Pseudomonas (both Pseudomonadales), Delftia, Comamonas (both Burkholderiales), and Xanthomonas (Xanthomonadales), a putative plant pathogen, were identified.Regarding alphaproteobacteria, three Wolbachia (Rickettsiales) lineages belonging to supergroup B, the major group in insect lineages, were detected in four psyllid species.Moreover, a Wolbachia lineage of supergroup O, a minor group that was recently found for the first time in Psylloidea, was detected in M. camphorae, suggesting that this supergroup is widespread in Psylloidea.These results provide deeper insights into the interactions among insects, bacteria, and plants, which will help establish a basis for the better control of pest species.

Fig. 1 .Fig. 2 .
Fig. 1.Composition of bacterial populations in psyllids of the family Carsidaridae.The relative abundance of Illumina reads belonging to assigned bacterial taxa are shown.

SV10Fig. 3 .
Fig. 3. Maximum likelihood phylogram of Enterobacterales.A total of 427 unambiguously aligned nucleotide sites of 16S rRNA genes were subjected to the analysis.On each branch, bootstrap support values of >50% are shown.The scale bar indicates substitutions per site.Regarding symbiotic bacteria, host organisms are shown in brackets.Symbionts of animals other than psyllids are shown in blue, while symbionts of psyllids are shown in red.Sequences from the present study are shown in bold.DDBJ/EMBL/GenBank accession numbers are provided in parentheses.Carsonella was used as an outgroup.

Fig. 4 .
Fig. 4. Maximum likelihood phylogram of Symbiopectobacterium and related genera.A total of 427 unambiguously aligned nucleotide sites of 16S rRNA genes were subjected to the analysis.On each branch, bootstrap support values of >50% are shown.The scale bar indicates substitutions per site.Regarding symbiotic bacteria, host organisms are shown in brackets.Symbionts of animals other than psyllids are shown in blue, while symbionts of psyllids are shown in red.The sequence from the present study is shown in bold.DDBJ/EMBL/GenBank accession numbers are provided in parentheses.Pseudomonas aeruginosa and Pseudomonas entomophila were used as an outgroup.

Fig. 5 .
Fig. 5. Maximum likelihood phylogram of Pseudomonadales.A total of 427 unambiguously aligned nucleotide sites of 16S rRNA genes were subjected to the analysis.On each branch, bootstrap support values of >50% are shown.The scale bar indicates substitutions per site.Regarding symbiotic bacteria, host organisms are shown in brackets.Symbionts of animals other than psyllids are shown in blue, while symbionts of psyllids are shown in red.The sequence from the present study is shown in bold.DDBJ/EMBL/GenBank accession numbers are provided in parentheses.Escherichia coli was used as an outgroup.

Fig. 6 .
Fig. 6.Maximum likelihood phylogram of Burkholderiales.A total of 427 unambiguously aligned nucleotide sites of 16S rRNA genes were subjected to the analysis.On each branch, bootstrap support values of >50% are shown.The scale bar indicates substitutions per site.Regarding symbiotic bacteria, host organisms are shown in brackets.Symbionts of animals other than psyllids are shown in blue, while symbionts of psyllids are shown in red.The sequence from the present study is shown in bold.DDBJ/EMBL/GenBank accession numbers are provided in parentheses.Escherichia coli was used as an outgroup.

Ca.Fig. 7 .
Fig. 7. Maximum likelihood phylogram of Wolbachia.A total of 402 unambiguously aligned nucleotide sites of 16S rRNA genes were subjected to the analysis.On each branch, bootstrap support values of >50% are shown.Host organisms are shown in brackets.Symbionts of animals other than psyllids are shown in blue, while symbionts of psyllids are shown in red.The sequence from this study is shown in bold.DDBJ/EMBL/ GenBank accession numbers for sequences are provided in parentheses.Supergroups of Wolbachia are shown in angle brackets.The scale bar represents nucleotide substitutions per position.Liberibacter was used as an outgroup.

Table 1 .
Psyllid species used in the present study