Regular Paper
Highly Reduced Complementary Genomes of Dual Bacterial Symbionts in the Mulberry Psyllid Anomoneura mori
2024 Volume 39 Issue 3 Article ID: ME24041
Details
2024 Volume 39 Issue 3 Article ID: ME24041
The genomes of obligately host-restricted bacteria suffer from accumulating mildly deleterious mutations, resulting in marked size reductions. Psyllids (Hemiptera) are phloem sap-sucking insects with a specialized organ called the bacteriome, which typically harbors two vertically transmitted bacterial symbionts: the primary symbiont “Candidatus Carsonella ruddii” (Gammaproteobacteria) and a secondary symbiont that is phylogenetically diverse among psyllid lineages. The genomes of several Carsonella lineages were revealed to be markedly reduced (158–174 kb), AT-rich (14.0–17.9% GC), and structurally conserved with similar gene inventories devoted to synthesizing essential amino acids that are scarce in the phloem sap. However, limited genomic information is currently available on secondary symbionts. Therefore, the present study investigated the genomes of the bacteriome-associated dual symbionts, Secondary_AM (Gammaproteobacteria) and Carsonella_AM, in the mulberry psyllid Anomoneura mori (Psyllidae). The results obtained revealed that the Secondary_AM genome is as small and AT-rich (229,822 bp, 17.3% GC) as those of Carsonella lineages, including Carsonella_AM (169,120 bp, 16.2% GC), implying that Secondary_AM is an evolutionarily ancient obligate mutualist, as is Carsonella. Phylogenomic analyses showed that Secondary_AM is sister to “Candidatus Psyllophila symbiotica” of Cacopsylla spp. (Psyllidae), the genomes of which were recently reported (221–237 kb, 17.3–18.6% GC). The Secondary_AM and Psyllophila genomes showed highly conserved synteny, sharing all genes for complementing the incomplete tryptophan biosynthetic pathway of Carsonella and those for synthesizing B vitamins. However, sulfur assimilation and carotenoid-synthesizing genes were only retained in Secondary_AM and Psyllophila, respectively, indicating ongoing gene silencing. Average nucleotide identity, gene ortholog similarity, genome-wide synteny, and substitution rates suggest that the Secondary_AM/Psyllophila genomes are more labile than Carsonella genomes.
Animals and microbes have diverse symbiotic relationships, among which the most intimate are bacteriome-associated mutualisms in insects (Moran et al., 2008; McCutcheon et al., 2019; Alarcón et al., 2022). Various insect lineages, particularly those feeding on nutritionally restricted diets, such as plant sap and vertebrate blood, possess the bacteriome, a specialized organ with apparently varied developmental origins (Alarcón et al., 2022), which harbors the ‘primary symbiont,’ providing a nutritional supply to support the survival of holobionts (host-symbiont complexes). Distinct insect lineages harbor phylogenetically diverse primary symbionts, indicating their independent origins from various free-living microbes. They are mostly bacterial and feature organelle-like characteristics, including intracellular localization within the host, complete infection in host populations, host-symbiont cospeciation through strict vertical transmission, and marked genome reductions due to the elevated fixation of deleterious mutations and the resulting loss of non-essential genes (Moran et al., 2008; McCutcheon et al., 2019). Their highly reduced genomes retain many of the genes required to synthesize essential nutrients that fulfill host requirements. However, the contingent acquisition of additional symbionts with a functionally intact genome may further allow the erosion of primary symbiont genomes to a level where the genes essential for the holobiont’s survival are also degraded and complemented by newcomers. In that case, the primary symbionts per se may eventually be replaced by the newly acquired symbionts. Apparent snapshots of these evolutionary processes have been observed in several insect lineages (McCutcheon et al., 2009; Vogel and Moran, 2013; Bennett et al., 2014; Monnin et al., 2020; Michalik et al., 2021).
Psyllids (Hemiptera: Sternorrhyncha: Psylloidea) are phloem sap-sucking insects encompassing ~4,000 described species worldwide (Burckhardt et al., 2021). They have a large bilobed bacteriome within the abdominal hemocoel (Profft, 1937; Chang and Musgrave, 1969; Waku and Endo, 1987; Nakabachi et al., 2010; Dan et al., 2017; Nakabachi and Suzaki, 2023), which typically harbors two distinct bacterial symbionts (Subandiyah et al., 2000; Spaulding and von Dohlen, 2001; Hall et al., 2016; Morrow et al., 2017; Nakabachi et al., 2022a; Dittmer et al., 2023; Maruyama et al., 2023). The primary symbiont is “Candidatus Carsonella ruddii” (Gammaproteobacteria: Oceanospirillales, hereafter Carsonella) (Thao et al., 2000b; Nakabachi and Moran, 2022), which has been detected in all psyllid species analyzed to date and is, thus, considered to be essential for Psylloidea. Molecular phylogenetic analyses demonstrated cospeciation between Carsonella lineages and their host psyllids, resulting from the single acquisition of an ancestor of Carsonella by a psyllid common ancestor and its subsequent strict vertical transmission (Thao et al., 2000b; Spaulding and von Dohlen, 2001; Hall et al., 2016; Morrow et al., 2017; Maruyama et al., 2023). Carsonella genomes derived from several psyllid lineages were analyzed and revealed to be markedly reduced in size (158–174 kb), AT-rich (14.0–17.9% GC), and structurally conserved. They lack numerous genes that are apparently essential for bacterial life but retain genes to synthesize essential amino acids that are deficient in the phloem sap diet (Nakabachi et al., 2006, 2013b, 2020b; Nakabachi, 2008; Sloan and Moran, 2012; Dittmer et al., 2023). Another bacterial lineage housed in the bacteriome is categorized as a ‘secondary symbiont’, which is phylogenetically diverse among psyllid lineages, suggesting its repeated infection and replacement during the evolution of Psylloidea (Thao et al., 2000a; Spaulding and von Dohlen, 2001; Sloan and Moran, 2012; Hall et al., 2016; Morrow et al., 2017; Nakabachi et al., 2020a, 2022a, 2022b). Although secondary symbionts in diverse insect taxa form various host-symbiont relationships across the mutualism-parasitism continuum (Dale and Maudlin, 1999; Nakabachi et al., 2003; Werren et al., 2008; Oliver et al., 2010; Johnson, 2015), those in the psyllid bacteriome appear to be consistently hosted by all individuals within a particular psyllid species, having obligate mutualistic relationships with the host psyllid. Therefore, they are occasionally called ‘co-primary symbionts,’ as in various auchenorrhynchan insects (McCutcheon et al., 2009; Bennett et al., 2014; Michalik et al., 2021). In contrast to Carsonella, limited genomic information is currently available on secondary symbionts associated with psyllid bacteriomes. Sloan and Moran examined the whole genome sequences of two distantly related secondary symbionts (both Gammaproteobacteria: Enterobacterales) derived from Ctenarytaina eucalypti (Aphalaridae: Spondyliaspidinae) and Heteropsylla cubana (Psyllidae: Ciriacreminae) (Sloan and Moran, 2012). The findings obtained showed that they are larger with less biased nucleotide contents (1,441 kb with 43.3% GC in Ct. eucalypti; 1,122 kb with 28.9% GC in H. cubana) than other insect bacteriome-associated symbionts analyzed thus far, including Carsonella, suggesting their recently formed relationships with the host psyllid. Both genomes encoded genes to complement the incomplete essential amino acid biosynthetic pathways encoded in the genome of co-residing Carsonella (Sloan and Moran, 2012). Subsequent analyses of two lineages of “Candidatus Profftella armatura” (Betaproteobacteria: Burkholderiales, hereafter Profftella) from the Asian citrus psyllid Diaphorina citri and its relative Diaphorina cf. continua (both Psyllidae: Diaphorininae) revealed that their genomes are markedly reduced in size and AT-rich (460 kb with 24.2% GC in D. citri; 470 kb with 24.4% GC in D. cf. continua) (Nakabachi et al., 2013a, 2013b, 2020b), suggesting that Profftella is an ancient symbiont. The highly conserved synteny and comparatively low substitution rates of the genomes implied that Profftella acquired a relatively stable status, which is similar to that of Carsonella. These genomes shared all genes for the biosynthesis of toxins (diaphorin and hemolysin), carotenoids, and B vitamins (riboflavin and biotin), indicating that Profftella is a unique versatile symbiont that plays multiple roles, including nutritional supply and defenses against natural enemies. Subsequent studies demonstrated the distinct activities of diaphorin, such as inhibitory effects against divergent organisms (Nakabachi and Fujikami, 2019; Nakabachi and Okamura, 2019; Yamada et al., 2019; Tanabe et al., 2022; Takasu et al., 2023, 2024). These collective benefits to the host may contribute to the stabilization of symbiotic relationships, leading to the organelle-like status of Profftella. However, these four lineages represent only a small fraction of the diverse secondary symbionts associated with the psyllid bacteriome; therefore, their evolutionary behaviors and the number of versatile symbionts, such as Profftella, in Psylloidea remain unclear.
As the first step to obtaining further insights into the evolution of the dual symbiotic system of the psyllid bacteriome, we analyzed the genomes of the bacteriome-associated secondary symbiont (Gammaproteobacteria: Enterobacterales, hereafter Secondary_AM) and Carsonella (hereafter Carsonella_AM) of a sericultural pest, the mulberry psyllid Anomoneura mori (Psyllidae: Psyllinae), in which the localization of symbionts has already been reported (Fukatsu and Nikoh, 1998). During the analysis, the genomes of 12 Carsonella lineages and 10 lineages of a secondary symbiont “Candidatus Psyllophila symbiotica” (Gammaproteobacteria: Enterobacterales, hereafter Psyllophila) from four Cacopsylla spp. (Psyllidae: Psyllinae) were recently published, revealing that the Psyllophila genomes are markedly reduced in size (221–237 kb) and AT-rich (17.3–18.6% GC), encoding genes complementary to those of Carsonella (Dittmer et al., 2023). We compared the genomes of the bacteriome-associated symbionts of A. mori with the previously sequenced genomes of psyllid symbionts.
The material of A. mori was collected from the mulberry tree Morus sp. (Moraceae) in Tsukuba city, Ibaraki Prefecture, Honshu, Japan (36.048N, 140.101E, 23 m a.s.l.) on May 26, 2015. DNA was extracted from a pool of the bacteriomes of three adult female A. mori using a DNeasy Blood & Tissue Kit (Qiagen) following the manufacturer’s instructions. Whole genome amplification was performed using extracted DNA and a REPLI-g Mini Kit (Qiagen) according to the manufacturer’s instructions.
Sequencing and assemblyAn 800-bp paired-end library and an 8-kbp mate-pair library of A. mori bacteriome DNA were prepared using the TruSeq DNA PCR-Free Sample Preparation kit (Illumina) and Nextera Mate Pair Sample Preparation kit (Illumina), respectively. The libraries were sequenced on the MiSeq instrument (Illumina) with the MiSeq Reagent kit v3 (600-cycles; Illumina). Paired reads in which either of the pair showed similarity (e-value scores <1.0E-5) to the previously reported genomic sequences of eight Carsonella lineages (NC_018417.1, AP009180.1, CP003541.1, CP003542.1, CP003543.1, CP003545.1, CP003467.1, and CP012411.1) were collected from the sequences obtained by a local alignment search with the Nucleotide Basic Local Alignment Search Tool (BLASTN) program (Camacho et al., 2009) using a custom Perl script. Sequencing errors were corrected using ShortReadManager based on 17-mer frequency (Ohtsubo et al., 2022). Collected and refined reads were assembled using Newbler version 2.9 (Roche). Gap sequences between contigs were assessed in silico using GenoFinisher and AceFileViewer (Ohtsubo et al., 2022).
Annotation and structural analysis of genomesInitial gene predictions and annotations were conducted using the DNA Data Bank of Japan (DDBJ) Fast Annotation and Submission Tool (DFAST) pipeline version 1.2.0 (Tanizawa et al., 2018) and the National Center for Biotechnology Information (NCBI) Prokaryotic Genome Annotation Pipeline (PGAP) version 2023-05-17.build6771 (Tatusova et al., 2016), followed by manual corrections with the aid of Rfam version 12.2 (Kalvari et al., 2021), the NCBI ORFfinder (Wheeler et al., 2003), BLAST (Camacho et al., 2009), and eggNOG-mapper version 2.1.12 (Cantalapiedra et al., 2021). The functional categories of Cluster of Orthologous Groups (COG) were assigned to predicted genes using the abovementioned version of eggNOG-mapper (Cantalapiedra et al., 2021). Metabolic pathways were analyzed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa, 2019). Pathway maps were created by examining the presence or absence of genes involved in the biosynthesis of essential amino acids and vitamins. Dinucleotide bias and GC skew were analyzed and circular diagrams were drawn using ArcWithColor version 1.62 (Ohtsubo et al., 2008). The codon adaptation index (CAI) was calculated using the CAIcal server (Puigbo et al., 2008). Pairwise comparisons of genomic structures were performed using GenomeMatcher version 3.06 (Ohtsubo et al., 2008), in which BLASTN of all-against-all bl2seq similarity searches was conducted with the parameter set ‘-F F -W 21 -e 1.0e-10’.
Phylogenomic analysisSingle-copy orthologous proteins shared among Secondary_AM, ten lineages of Psyllophila from four Cacopsylla spp. (six from C. melanoneura, two from C. picta, and one each from C. pyricola and C. pyri) (Dittmer et al., 2023), 33 other insect nutritional endosymbionts belonging to the order Enterobacterales, and Pseudomonas oryziphila and two strains of Pseudomonas entomophila (both Gammaproteobacteria: Pseudomonadales) as outgroups were identified using OrthoFinder version 2.5.5 (Emms and Kelly, 2019). The amino acid sequences of the identified proteins were aligned with MAFFT 7.452 (Katoh and Standley, 2013) using the E-INS-i algorithm with the aid of Perl script MultiMafft.pl (https://fish-evol.org/MAFFT.html). After removing amino acid sites corresponding to alignment gaps, multiple fasta alignments were concatenated into a partitioned supermatrix using catsequences version 1.4 (https://github.com/ChrisCreevey/catsequences.git). Phylogenetic trees were inferred by the Maximum Likelihood (ML) method using RAxML-NG version 1.2.1 (Kozlov et al., 2019) with 1,000 replicates with the MTREV+I+G4+F model, which was selected using ModelTest-NG version 0.1.7 (Darriba et al., 2020). Trees were visualized using Interactive Tree of Life (iTOL) version 5 (Letunic and Bork, 2021).
Average nucleotide identity (ANI) analysisTo measure the nucleotide-level genomic similarity of A. mori symbionts to their counterparts in four Cacopsylla spp. (ten Psyllophila lineages as described above and eight Carsonella lineages from C. melanoneura, two from C. picta, and one each from C. pyricola and C. pyri), ANI was calculated using the ANI calculator (Rodriguez-R and Konstantinidis, 2016). The calculation was performed using reciprocal best hits (two-way ANI) between two genomic datasets with the default parameters. Window and step sizes were set to 1,000 and 200 bp, respectively.
Substitution rate analysisAmino acid sequences were deduced from the protein-coding genes (CDSs) shared between A. mori symbionts and their counterparts in the melET isolate of C. melanoneura, which were then aligned with MAFFT version 7.452 (Katoh and Standley, 2013) using the E-INS-i algorithm with default parameters. The resulting protein alignments were converted to nucleotide alignments using PAL2NAL version 13.0 (Suyama et al., 2006). Non-synonymous (dN) and synonymous (dS) substitution rates and dN/dS ratios between orthologous pairs were calculated using the KaKs_Calculator 1.2 package (Zhang et al., 2006) with the YN model (Yang and Nielsen, 2000). All statistical analyses were performed using R software version 4.2.1 (R Core Team 2022, https://www.r-project.org).
MiSeq sequencing of A. mori bacteriome DNA libraries yielded 1.36 million paired-end reads (754 Mbp) and 1.03 million mate-pair reads (447 Mbp). We collected sequence reads that showed detectable similarity (BLASTN e-value scores <1.0E-5) to published Carsonella genomes, aiming to elucidate the Carsonella_AM genome. The assemblage of the 126,000 paired-end reads (72.6 Mbp) and 25,000 mate pair reads (5.58 Mbp) that were obtained and refined yielded five scaffolds and 220 large contigs (>0.5 kbp). After filling gap sequences in silico, we unexpectedly succeeded in obtaining circular complete genomes with a coverage of ca. 100× of not only Carsonella_AM (169,120 bp, 16.2% GC), but also Secondary_AM (229,822 bp, 17.3% GC) (Table 1, S1, and S2, Fig. 1, S1), which were identified by the presence of known sequences of 16S rRNA genes (Fukatsu and Nikoh, 1998; Nakabachi et al., 2022b). Notably, the size and GC content of the Secondary_AM genome were similar to those of Carsonella lineages, including Carsonella_AM, implying an evolutionarily ancient and obligate mutualistic relationship of Secondary_AM with the host insect. These features were similar to those of the recently published genomes of Psyllophila (221,413–237,114 bp, 17.3–18.6% GC), the bacteriome-associated secondary symbiont of Cacopsylla spp. (Psyllidae: Psyllinae) (Dittmer et al., 2023), which was in contrast to previously sequenced markedly larger genomes with less nucleotide composition bias derived from the secondary symbionts of two psyllid species, Ct. eucalypti (Aphalaridae: Spondyliaspidinae) (1.4 Mb, 43.3% GC) and H. cubana (Psyllidae: Ciriacreminae) (1.1 Mb, 28.9% GC) (Sloan and Moran, 2012).
Genomic features of bacteriome-associated symbionts in A. mori
| Carsonella_AM | Secondary_AM | |
|---|---|---|
| Chromosome (bp) | 169,120 | 229,822 |
| G+C content (%) | 16.2 | 17.3 |
| CDS | 196 | 215 |
| rRNA | 3 | 3 |
| tRNA | 28 | 26 |
| Pseudogene | 0 | 2 |
Circular representation of Secondary_AM and Carsonella_AM genomes. Concentric rings denote the following features (from the outside): (i) the scale in kilobases, (ii) forward strand genes, (iii) reverse strand genes, (iv) the dinucleotide bias, (v) GC skew, and (vi) G+C content. To calculate (iv), (v), and (vi), sliding windows of 100 bp and a step size of 10 bp were used.
Previous microbiome and phylogenetic analyses using 16S rRNA genes suggested that Secondary_AM and Psyllophila are related and comprise a lineage widely distributed in Cacopsylla spp. and also in some species of other genera, including Anomoneura and Cyamophila (all Psyllidae: Psyllinae) (Nakabachi et al., 2022b). This implies that the lineage is ancient and was acquired before the divergence of these psyllid genera. In this context, we performed a more detailed phylogenetic analysis using the sequences for orthologous proteins encoded in these genomes. A total of 14,766 amino acid positions derived from 60 single-copy orthologous proteins that were found to be shared among analyzed symbionts (Table S2) were used in an ML analysis (Fig. 2). The results obtained revealed that Secondary_AM formed a robustly supported clade (bootstrap: 100%) with a clade formed by 10 Psyllophila lineages derived from Cacopsylla spp., which was also robustly supported (bootstrap: 100%) in the ML tree. This branching pattern indicates that Secondary_AM is a sister lineage of Psyllophila, sharing a recent common ancestor. The clade of Secondary_AM and Psyllophila further formed a robustly supported clade (bootstrap: 100%) with “Candidatus Annandia” spp., the bacteriome-associated symbionts of adelgids (Hemiptera: Sternorrhyncha: Phylloxeroidea: Adelgidae), and “Candidatus Nardonella” spp., the bacteriome-associated symbionts of weevils (Coleoptera: Curculionoidea). The previously sequenced secondary symbionts of Ct. eucalypti and H. cubana were shown to be distantly related to this clade encompassing Secondary_AM and Psyllophila (Fig. 2).
Maximum likelihood phylogram showing phylogenetic relationships among Secondary_AM, 10 Psyllophila lineages from four Cacopsylla spp., and 33 other insect endosymbionts belonging to the order Enterobacterales. For simplicity, species names are presented without “Candidatus.” A total of 14,766 unambiguously aligned amino acid positions derived from 60 single-copy orthologous proteins shared among these bacteria were subjected to the analysis. On each branch, bootstrap support values 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 oryziphila and two strains of Pseudomonas entomophila (all Gammaproteobacteria: Pseudomonadales) were used as an outgroup.
ANI was calculated to further assess genomic similarities between Secondary_AM and 10 Psyllophila lineages (Table 2). As a comparison, ANI was also calculated between Carsonella_AM and 12 Carsonella lineages from Cacopsylla spp. The results obtained showed that 1) the mean ANI is higher in Carsonella (86.915±0.005%) (mean±standard deviation [SD]; n=12) than in Psyllophila (79.571±0.004%; n=10) (P<0.001, Welch’s t-test), and 2) ANI values are similar within Carsonella or Psyllophila (see SD of 1 and Table 2). However, since the ANI of Secondary_AM was the highest (79.78%) with the Psyllophila lineage derived from the melET isolate of C. melanoneura (PSmelET) (Table 2), we compared their sequences in more detail featuring this lineage.
Average nucleotide identity between symbionts of A. mori and Cacopsylla spp.
| Host species | Cacopsylla melanoneura | C. picta | C. pyricola | C. pyri | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Strains | melAO1 | melAO2 | melAO3-1 | melAO3-2 | melET | melST4 | melST17 | melST21 | pic1 | pic2 | pyc | pyr |
| Carsonella | 87.13 | 87.08 | 87.08 | 87.08 | 87.05 | 87.06 | 87.05 | 87.07 | 86.78 | 86.64 | 86.65 | 86.31 |
| Psyllophila | 79.75 | — | 79.53 | 79.75 | 79.78 | — | 79.60 | 79.63 | 79.70 | 79.75 | 79.08 | 79.14 |
Genome-wide alignments revealed highly conserved synteny between the genomes of Secondary_AM and 10 Psyllophila lineages (alignment with the PSmelET genome is shown as a representative in Fig. 3), indicating that most genes are shared between these symbionts and essentially no genome rearrangements have occurred since they diverged. Between the genomes of Secondary_AM and PSmelET, 230 pairs of orthologous genes were 79.8% identical at the nucleotide level and the amino acid sequences of 200 pairs of orthologous proteins were 69.0% identical on average (Table S2). They shared all genes involved in synthesizing biotin and riboflavin or complementing essential amino acid biosynthesis by Carsonella, which will be further discussed later.
Comparison of genomic structures of Secondary_AM and Psyllophila derived from Cacopsylla melanoneura (PSmelET). The genomes of Secondary_AM and PSmelET are represented by the x and y axes, respectively. The thick line indicates shared synteny between the two genomes. The color of the line indicates percentage similarity between the nucleotide sequences. The genes found in Secondary_AM but not in PSmelET are presented below the line plot; the genes present in PSmelET but not in Secondary_AM are shown above the line plot.
Despite this high level of conservation between the genomes, random gene silencing appeared to be ongoing in Secondary_AM and Psyllophila. The genes found in one of the lineages, but not in the other, are shown in Fig. 3. Neither the G+C content nor CAI of these genes was significantly different (P>0.05, Welch’s t-test) from those of genes shared between Secondary_AM and Psyllophila (Table S3), showing no signs of recent horizontal acquisition, which strongly suggested that the different gene inventories reflect gene silencing on either genome. cysCDHIJN encoding enzymes for sulfur assimilation were retained in Secondary_AM but were missing in PSmelET and all other Psyllophila lineages (Fig. 3 and S2, Table S2). These genes potentially contribute to the synthesis of the sulfur-containing amino acids, cysteine and methionine, the latter of which is an essential amino acid that the host insects cannot synthesize. However, no other genes related to their synthesis were retained in the Secondary_AM genome (Fig. S2 and Table S2). Moreover, Carsonella_AM lacked genes for the biosynthesis of cysteine/methionine other than metE, which converts homocysteine into methionine (Fig. 4 and S2). Therefore, the cysteine/methionine synthesis pathway appeared to be incomplete, making the role of the conservation of cysCDHIJN genes in Secondary_AM unclear.
Pathways for synthesizing essential amino acids and B vitamins reconstructed from genes encoded in Carsonella_AM and Secondary_AM genomes. Genes denoted with green and magenta arrows are found in the genomes of Carsonella_AM and Secondary_AM, respectively, while those shown with white arrows appear to be absent in these genomes. ribC shown with a yellow arrow is suspected to be encoded in the host psyllid genome, which was horizontally acquired from an unknown bacterium. PEP, phosphoenolpyruvate; E4P, erythrose 4-phosphate; 2MM, 2-methylmalate; PRPP, phosphoribosyl diphosphate; PME, pimelyl-(acyl-carrier protein) methyl ester.
On the other hand, Secondary_AM lacked genes for the synthesis of carotenoids (crtB, crtI, and crtY) (Fig. 3), all of which were retained in all Psyllophila lineages (Dittmer et al., 2023). These genes were also retained in Profftella, a distantly related symbiont whose primary role appears to be the protection of the holobiont from natural enemies (Nakabachi et al., 2013b, 2020b), presenting an example of intriguing convergence. Carotenoids are organic pigments found in diverse organisms (Mussagy et al., 2019). In animals, including insects, they are supposed to play important roles, including as antioxidants and pigments for photoprotection, ornamentation, or camouflage. Although various microbes and plants produce carotenoids, metazoa cannot generally synthesize carotenoids and must acquire them through diet (Mussagy et al., 2019). Since all Carsonella lineages sequenced to date lack genes for synthesizing carotenoids (Nakabachi et al., 2006, 2013b, 2020b; Sloan and Moran, 2012; Dittmer et al., 2023), the carotenoid biosynthetic genes of Psyllophila and Profftella presumably supplement host requirements. The lack of these genes in Secondary_AM may reflect the availability of carotenoids in the phloem sap of the host plant.
Carsonella_AM and Secondary_AM are metabolically interdependentAmong the COG functional categories assigned to the protein-coding genes of Carsonella_AM and Secondary_AM, “translation” (category J) exhibited the highest percentage in both symbionts (33.2% in Carsonella_AM, 45.5% in Secondary_AM), while those for “transcription” (category K, 2.5% in Carsonella_AM, 3.6% in Secondary_AM) and “DNA replication, recombination and repair” (category L, 2.5% in Carsonella_AM, 6.4% in Secondary_AM) were low (Fig. S1). This result implies that translation is an especially important information-processing process for retaining the autonomy of symbionts within the host cell. In Carsonella_AM, the second most prominently represented COG category was E “amino acid transport and metabolism” (21.3%) (Fig. S1). Similar to other Carsonella strains previously analyzed (Nakabachi et al., 2006, 2013b, 2020b; Nakabachi, 2008; Sloan and Moran, 2012; Dittmer et al., 2023), Carsonella_AM retained most of the genes required for the biosynthesis of essential amino acids; however, some appeared to be missing (Fig. 4 and S2, Table S1). Regarding the tryptophan biosynthesis pathway, only trpE and trpG, genes for anthranilate synthase components involved in the first catalytic step after diverging from that for phenylalanine, were retained in the Carsonella_AM genome. All other genes (trpD, trpCF, trpA, and trpB) required for the remainder of the pathway were missing. Notably, these genes lost in Carsonella_AM were entirely retained in the Secondary_AM genome (Fig. 4 and S2). Moreover, these were the only Secondary_AM genes directly involved in the synthesis of essential amino acids, underlining the elaborate metabolic complementarity encoded in the genomes of Carsonella_AM and Secondary_AM. The interdependent biosynthetic pathway of tryptophan, involving Carsonella encoding only trpE and trpG, complemented by a secondary symbiont, was observed not only in Cacopsylla spp., but also in Ct. eucalypti (Aphalaridae: Spondyliaspidinae) and H. cubana (Psyllidae: Ciriacreminae) with distantly related and possibly more recently acquired Enterobacterales symbionts with larger genomes (Sloan and Moran, 2012), exemplifying other cases of convergence.
Although only 5.5% of genes encoded in the Secondary_AM genome were assigned COG category H “coenzyme transport and metabolism,” they were bioA, bioD, and bioB, which are required for the synthesis of the B vitamin, biotin, and ribA, ribD, ribB, and ribE, genes for the synthesis of another B vitamin, riboflavin (Fig. 4). This result suggests that supplementation of B vitamins, which are also scarce in the phloem sap diet (Ziegler et al., 1975), is the pivotal role of Secondary_AM. All of these genes were retained not only in Psyllophila, but also in Profftella, further adding an example of intriguing convergence (Fig. S2). Although all the lineages of Secondary_AM, Psyllophila, and Profftella lacked ribC, which is required for the final step of riboflavin biosynthesis, previous screening of the genomic/transcriptomic data of several divergent psyllid lineages demonstrated that host psyllids horizontally acquired ribC from an uncertain bacterial lineage before the radiation of major psyllid lineages (Sloan et al., 2014; Nakabachi, 2015), which is expected to complement its absence in bacterial symbionts (Fig. 4). The involvement of host genes horizontally acquired from bacteria (mostly lineages not leading to extant bacteriome residents) in metabolic pathways with symbionts has been demonstrated in the bacteriome symbioses of various hemipterans (Nakabachi et al., 2005, 2014; Nikoh et al., 2010; Shigenobu et al., 2010; Husnik et al., 2013; Luan et al., 2015; Ren et al., 2020; Smith et al., 2021), implying its importance in the evolution of this type of intimate symbiosis. Moreover, newly acquired symbionts repeatedly add or replace the ability to synthesize B vitamins in diverse plant-sap feeding insects (McCutcheon et al., 2009; Nakabachi et al., 2013b, 2020b; Monnin et al., 2020; Michalik et al., 2021), showing that microbial supplementation with B vitamins is important, if not essential, for these insects (Nakabachi and Ishikawa, 1999).
While metabolic complementarity between Carsonella_AM and Secondary_AM was shown as described above, the vulnerability of these symbionts was simultaneously indicated. Regarding the COG categories for energy production and the metabolism of carbohydrates, nucleotides, and lipids (C, G, F, and I), a limited number of genes were retained, apparently constituting incomplete pathways (Fig. S1, Table S1 and S2). Furthermore, these symbionts lacked genes for many functional categories, including A, “RNA processing and modification,” M, “Cell envelope biogenesis,” D, “Cell division and chromosome partitioning,” V, “Defense mechanisms,” and Q “Secondary metabolites biosynthesis, transport and catabolism,” indicating their limited ability. The absence of Secondary_AM/Psyllophila genes involved in defense and secondary metabolism further emphasizes Profftella’s uniqueness in having both nutritional and defensive roles (Fig. S1, Table S1 and S2).
Secondary_AM/Psyllophila genomes are more labile than Carsonella genomesAlthough diverse lineages of insect primary symbionts have been shown to have highly stable genomic structures (Tamas et al., 2002; Rio et al., 2012; Sloan and Moran, 2012; Mao et al., 2017), this is generally not the case for more recently acquired secondary symbionts (Bennett et al., 2014; Campbell et al., 2015; McCutcheon et al., 2019; Nakabachi et al., 2020b). In comparisons between symbionts of A. mori and their counterparts in the melET isolate of C. melanoneura, for which the same divergence time is assumed to be applicable, ANI (Table 2), genome-wide synteny (Fig. 3 and S3), and the similarity of orthologous genes (87.3±9.9% [mean±SD; n=215] for Carsonella vs. 79.8±9.5% for Secondary_AM/Psyllophila [n=230]) (Table S1 and S2) suggested that Secondary_AM/Psyllophila genomes are more labile than Carsonella genomes. To further assess genomic stability, we analyzed the genome-wide rates of synonymous (dS) and non-synonymous (dN) substitutions for these symbionts (Fig. 5, Table S1 and S2). Orthologous protein-coding genes, namely, 185 pairs of Carsonella genes and 200 pairs of Secondary_AM/Psyllophila genes, were used in the analysis. The mean values for both dN (0.076±0.046 for Carsonella vs. 0.148±0.053 for Secondary_AM/Psyllophila) and dS (2.84±12.34 for Carsonella vs. 13.16±30.36 for Secondary_AM/Psyllophila) were higher (P<0.001, Brunner-Munzel test) in the Secondary_AM/Psyllophila lineage than in Carsonella. Synonymous divergence appeared to be saturated (dS>3.0) in 15 (8.1%) Carsonella genes and 95 (47.5%) Secondary_AM/Psyllophila genes (Table S1 and S2). To be conservative, only genes with dS<3.0 were used to calculate dN/dS, which was also higher (P<0.001, Brunner-Munzel test) in the Secondary_AM/Psyllophila lineages (0.134±0.111, n=105) than in Carsonella (0.121±0.151, n=170) (Fig. 5, Table S1 and S2). However, no gene was estimated to have dN/dS>1, indicating purifying selection for all genes analyzed in these symbionts. These results further indicate that the Secondary_AM/Psyllophila genomes are more labile, though still functional, than those of Carsonella genomes. The tendency that genomes of younger symbionts are less stable than those of ancient symbionts may lead to the preferential replacement of secondary symbionts, retaining the primary symbionts. Indeed, while the Secondary_AM/Psyllophila lineage appears to be widely distributed in Cacopsylla and other psyllid genera, several Cacopsylla spp. lack the lineage and harbor other secondary symbionts (Nakabachi et al., 2022b), suggesting relatively recent replacements.
Synonymous (dS) and non-synonymous (dN) substitution rates inferred from pairwise comparisons of orthologous CDSs in Secondary_AM/Psyllophila lineages (magenta dots) and Carsonella (green dots). The values of genes with dS<3.0 (104 and 172 orthologous pairs in Secondary_AM/Psyllophila and Carsonella, respectively), which were used to calculate dN/dS, are shown. Box plots (Secondary_AM/Psyllophila, magenta; Carsonella, green) on the x- and y-axes indicate the distributions (median, quartiles, minimum, maximum, and outliers) of dS and dN values, respectively.
Yasuda, Y., Inoue, H., Hirose, Y., and Nakabachi, A. (2024) Highly Reduced Complementary Genomes of Dual Bacterial Symbionts in the Mulberry Psyllid Anomoneura mori. Microbes Environ 39: ME24041.
https://doi.org/10.1264/jsme2.ME24041
We thank Nami Uechi at the Institute for Plant Protection for kindly collecting and providing us with A. mori specimens. This work was supported by the Japan Society for the Promotion of Science (https://www.jsps.go.jp) KAKENHI (grant numbers 21687020, 26292174, and 20H02998). The funders had no role in the study design, data collection and analysis, publication decisions, or manuscript preparation.
Data availabilityThe genomes of Secondary_AM and Carsonella_AM have been deposited in the DDBJ/EMBL/GenBank databases under the accession numbers AP031380 (Carsonella_AM) and AP031381 (Secondary_AM).
