Comparative Genomics of Wolbachia- Cardinium Dual Endosymbiosis in a Plant-Parasitic Nematode

Abstract

Wolbachia and Cardinium are among the most important and widespread of all endosymbionts, occurring in nematodes and more than half of insect and arachnid species, sometimes as coinfections. These symbionts are of significant interest as potential biocontrol agents due to their abilities to cause major effects on host biology and reproduction through cytoplasmic incompatibility, sex ratio distortion, or obligate mutualism. The ecological and metabolic effects of coinfections are not well understood. This study examined a Wolbachia-Cardinium coinfection in the plant-parasitic nematode (PPN), Pratylenchus penetrans, producing the first detailed study of such a coinfection using fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR), and comparative genomic analysis. Results from FISH and single-nematode PCR showed 123/127 individuals in a focal population carried Cardinium (denoted strain cPpe), and 48% were coinfected with Wolbachia strain wPpe. Both endosymbionts showed dispersed tissue distribution with highest densities in the anterior intestinal walls and gonads. Phylogenomic analyses confirmed an early place of cPpe and long distance from a sister strain in another PPN, Heterodera glycines, supporting a long history of both Cardinium and Wolbachia in PPNs. The genome of cPpe was 1.36 Mbp with 35.8% GC content, 1,131 predicted genes, 41% having no known function, and missing biotin and lipoate synthetic capacity and a plasmid present in other strains, despite having a slightly larger genome compared to other sequenced Cardinium. The larger genome revealed expansions of gene families likely involved in host-cellular interactions. More than 2% of the genes of cPpe and wPpe were identified as candidate horizontally transferred genes, with some of these from eukaryotes, including nematodes. A model of the possible Wolbachia-Cardinium interaction is proposed with possible complementation in function for pathways such as methionine and fatty acid biosynthesis and biotin transport.

Keywords: Cardinium; Pratylenchus penetrans; Wolbachia; endosymbiont; genomics; horizontal gene transfer; plant-parasitic nematode; symbiosis.

FIGURE 1

Localization of coinfecting Wolbachia wPpe (red) and Cardinium cPpe (yellow) in the anterior body of Pratylenchus penetrans by fluorescence in situ hybridization (FISH) and confocal microscopy. (A,B) Split panels showing confocal z-stacks combining signals from brightfield DIC, blue (DAPI), and red, and yellow FISH probes (top); wPpe (red FISH probe) alone (middle); and cPpe (yellow FISH probe) alone (bottom). Arrows point to excretory ducts. Yellow autofluorescence can be seen in lip region and excretory duct (arrows). Scale bars = 20 μm.

FIGURE 2

Coinfecting Wolbachia wPpe (red) and Cardinium cPpe (yellow) in P. penetrans in esophageal region and ovaries by FISH and confocal microscopy. (A,B) Confocal sections of esophageal regions combining brightfield DIC, blue DAPI, and red and yellow FISH probes, localizing Wolbachia wPpe (red) around subventral glands (subv. g) and elsewhere, with cPpe (yellow) in intestinal wall (int). Yellow autofluorescence can be seen in lip region. (C,D) Split panels of ovaries showing DAPI alone (top), yellow cPpe FISH probe alone (second panel), red wPpe FISH probe alone (third) panel, and all three signals (bottom), with arrows in (C) showing low levels of Wolbachia (red) and (D) low levels of cPpe (yellow). Scale bars = 20 μm.

FIGURE 3

Localization of Cardinium cPpe in whole body and ovaries of P. penetrans by FISH and confocal microscopy. (A–C) Split panels showing nematode nuclei (blue DAPI stain) alone (top), cPpe (yellow FISH probe) alone (middle), and combined images (bottom). (A) Whole female with o = ovaries, and yellow autofluorescence of lip and spermathecal regions. (B) Female with combined brightfield DIC image (bottom), v = vulva. (C) High magnification of (A) showing ovaries and developing oocytes (arrows), v = vulva. Scale bars = 20 μm.

FIGURE 4

Genome features of Cardinium strains and outgroups (number of proteins, %GC, proportion of the genome that is coding, and sum of ortholog lengths for 37 genes in phylogenomic analyses). Colors: orange = Cardinium cPpe from P. penetrans, yellow = Cardinium strains cEper1 and cBtQ1, brown = Candidatus Amoebophilus asiaticus, blue = outgroup Bacteroidetes, and gray = plasmids from the above bacteria. (Full data are given in Supplementary Table 4.)

FIGURE 5

Maximum-likelihood phylogenies of Cardinium strains and outgroup species. (A) Tree generated from 37 protein coding genes comprising 72,194 nucleotide alignment positions generated with RAxML under the GTR model, showing bootstrap values (1,000 replicates) and posterior probabilities from MrBayes. (B) Tree generated from 16S rRNA and gyrB genes comprising 2,426 nucleotide positions calculated as described for (A). Shading indicates major Cardinium groups. For both (A,B), similar analyses with varied alignment filtering stringency in Gblocks, different outgroups, different models, or using amino acid sequences, produced similar supported clades. (List of 37 genes in Supplementary Table 2; accession numbers in Supplementary Table 4; additional phylogenies in Supplementary Figures 1, 2.)

FIGURE 6

Genome content (ortholog) comparisons among Cardinium strains and outgroups. (A) Genes shared between the three available Cardinium strains, cEper1 from Encarsia wasps, cBtQ1 from Bemisia tabaci whiteflies, and cPpe from P. penetrans plant-parasitic nematodes (PPN) (gene list in Supplementary Table 7). (B) Gene content overlap for completely shared “core” genes among groups of strains for Bacteroidetes in Supplementary Table 4 (gene list in Supplementary Table 8). (C) Genetic repertoire similarity for all genes (“pangenome”) between different groups, including plasmids (gene list in Supplementary Table 9).

FIGURE 7

Gene copy number expansions as measured by orthoMCL orthologous gene sets for Candidatus Amoebophilus asiaticus and Cardinium strains from P. penetrans (cPpe), whitefly (cBtQ1), and Encarsia wasps (cEper1). (A) Number of homologous sets of genes by protein category, normalized for differences in genome size. (B) Gene number among all homologous sets, normalized for genome size, for sets with three or more homologs. (C) Genes names and numbers in homolog sets for predicted protein-coding genes of known function (“unique genes”). In all comparisons (A–C), only sets with three or more homologs were used.

FIGURE 8

Comparison of antifeeding prophage-like protein secretion system synteny among Amoebophilaceae showing flanking genes for transport and other relevant functions of interest. Color block arrows show genes homologous to Serratia entomophila pADAP “Afp” antifeeding prophage and other universally conserved Amoebophilaceae genes, whereas dark gray blocks show flanking genes of interest. Synteny regions are shown with light gray connectors. Proteins in black font are conserved in this clade; proteins in gray font are conserved among only Cardinium strains; proteins in red font are uniquely located in specific Cardinium strains; proteins in blue font appear to be lost in Cardinium from P. penetrans (cPpe).

FIGURE 9

Relative ranges of blast bit score for candidate HGT genes, by phylogenetic clade of blast hits. Relative divergence (in blast bit score) of candidate HGTs compared with ordinary flanking genes in Cardinium cPpe from P. penetrans, Cardinium cBtQ1/cEper1 from insects (yellow boxes), and Wolbachia wPpe from P. penetrans (red boxes). Black boxes indicate top blast hits outside the clades Cardinium or Wolbachia, while other colors indicate the top hit matches Cardinium or Wolbachia, but all remaining hits match distant taxa (blue = eukaryotes, green = other bacterial phyla, pink = Wolbachia/Rickettsiales). Dotted lines serve as a visual guide to indicate the range in blast bit scores relative to the upper quartile of cPpe and wPpe genes for which the gene tree matches the species tree. (Box and whisker plot: box represents interquartile range, whisker represents remaining quartiles, with midline and ‘x’ denoting the median and mean, respectively, and outliers as dots beyond 1.5 times the interquartile range.)

FIGURE 10

Candidate horizontally transferred genes (HGTs) in Cardinium strains from P. penetrans (cPpe), whitefly (cBtQ1), and Encarsia wasps (cEper1) based on absence of blast similarity to proteins from outgroup genomes and presence of high blast similarity to other organisms. This list emerged from a filtered subset of blastx hits comprising the outliers in hit length and similarity shown in Supplementary Figure 3 and Supplementary Table 10, and thus is a fairly conservative subset of candidate HTGs. Source taxa for these genes are shown to the left and place of entry into the Cardinium lineage is shown at the bottom.

FIGURE 11

A model of possible interactions between Cardinium cPpe and Wolbachia wPpe in the PPN P. penetrans.