Functional genomics of a Spiroplasma associated with the carmine cochineals Dactylopius coccus and Dactylopius opuntiae

Abstract

Background: Spiroplasma is a widely distributed endosymbiont of insects, arthropods, and plants. In insects, Spiroplasma colonizes the gut, hemolymph, and reproductive organs of the host. Previous metagenomic surveys of the domesticated carmine cochineal Dactylopius coccus and the wild cochineal D. opuntiae reported sequences of Spiroplasma associated with these insects. However, there is no analysis of the genomic capabilities and the interaction of this Spiroplasma with Dactylopius.

Results: Here we present three Spiroplasma genomes independently recovered from metagenomes of adult males and females of D. coccus, from two different populations, as well as from adult females of D. opuntiae. Single-copy gene analysis showed that these genomes were > 92% complete. Phylogenomic analyses classified these genomes as new members of Spiroplasma ixodetis. Comparative genome analysis indicated that they exhibit fewer genes involved in amino acid and carbon catabolism compared to other spiroplasmas. Moreover, virulence factor-encoding genes (i.e., glpO, spaid and rip2) were found incomplete in these S. ixodetis genomes. We also detected an enrichment of genes encoding the type IV secretion system (T4SS) in S. ixodetis genomes of Dactylopius. A metratranscriptomic analysis of D. coccus showed that some of these T4SS genes (i.e., traG, virB4 and virD4) in addition to the superoxide dismutase sodA of S. ixodetis were overexpressed in the ovaries.

Conclusion: The symbiont S. ixodetis is a new member of the bacterial community of D. coccus and D. opuntiae. The recovery of incomplete virulence factor-encoding genes in S. ixodetis of Dactylopius suggests that this bacterium is a non-pathogenic symbiont. A high number of genes encoding the T4SS, in the S. ixodetis genomes and the overexpression of these genes in the ovary and hemolymph of the host suggest that S. ixodetis use the T4SS to interact with the Dactylopius cells. Moreover, the transcriptional differences of S. ixodetis among the gut, hemolymph and ovary tissues of D. coccus indicate that this bacterium can respond and adapt to the different conditions (e.g., oxidative stress) present within the host. All this evidence proposes that there is a strong interaction and molecular signaling in the symbiosis between S. ixodetis and the carmine cochineal Dactylopius.

Fig. 1

Phylogenetic position of Spiroplasma symbionts associated with Dactylopius cochineals. Maximum-likelihood trees of 16S rRNA genes (a) and 168 concatenated single-copy gene markers (b). Scale bars indicate 10 and 20% estimated sequence divergence, respectively. See Additional file 2 Data set 1 for the accession numbers of the sequences used in these analyses

Fig. 2

Comparative genomics of Spiroplasma associated with Dactylopius and other Spiroplasma. COG profiles of S. ixodetis DCF, DCM and DO (blue bars) compared to other Spiroplasma genomes (red bars). Blue arrows indicate enrichment of genes in the COG categories on S. ixodetis DCF, DCM and DO genomes. The mean ± SEM proportion of genes belonging to each COG category is shown

Fig. 3

Gene structure of the glycerol catabolic genes (glpF, glpO and glpK) present in different Spiroplasma spp. genomes. Color-empty blocks represent pseudogenes. Gray blocks show genes annotated as hypothetical (hyp) protein. Maximum-likelihood phylogenetic tree in the left shows the cladogenesis of Spiroplasma spp. genomes using the Glycerol-3-phosphate oxidase encoding gene glpO. MAFFT was used to align all glpO sequences from each genome and the phylogenetic tree, based on the LG + G4 substitution model obtained by ModelFinder, was calculated by IQtree with 1000 Bootstrap replicates for internal branch support. Scale bar indicates 5% estimated sequence divergence

Fig. 4

Differences in genes encoding secretion systems among Spiroplasma spp. genomes. Heatmaps showing the number of genes associated with (a) Intracellular trafficking, secretion, and vesicular transport COG category. (b) Heatmap showing the number of genes associated with the type IV secretion system (T4SS) among all the Spiroplasma spp. genomes. Colors at the top of the heatmaps indicate the Spiroplasma phylogenetic position of each genome

Fig. 5

Variation on gene expression of S. ixodetis DCF in the gut, ovary, and hemolymph of D. coccus. (a) Heatmap showing all the differentially expressed genes in each tissue comparison (i.e. gut vs hemolymph, gut vs ovary and hemolymph vs ovary). Color range in heatmaps indicates variation in log2 fold-change. Gray boxes indicate no differential expression observed after DESeq2 analysis (absolute log₂ fold-change ≥0.58 and p-adjust value ≤0.05). hyp means genes annotated as hypothetical proteins. (b) Volcano plot of differential gene expression of S. ixodetis DCF in gut vs hemolymph, gut vs ovary and hemolymph vs ovary comparisons. Red dots show genes considered as differentially expressed after DESeq2 analysis absolute log₂fold change > 0.58 (outer broken vertical lines) and p-adjust value ≤0.05 (−log₁₀ p-value ≥1.3, broken horizontal line). Green dots show non-differentially expressed genes with an absolute log₂ fold-change > 0.58 and p-adjust value > 0.05. Gray dots show non-differentially expressed genes with an absolute log₂fold change < 0.58 and p-adjust value > 0.05 (non-differentially expressed). All annotated genes are represented, red dots without annotation legend represent hypothetical protein coding genes differentially expressed