Draft genome sequence of the male-killing Wolbachia strain wBol1 reveals recent horizontal gene transfers from diverse sources

Abstract

Background: The endosymbiont Wolbachia pipientis causes diverse and sometimes dramatic phenotypes in its invertebrate hosts. Four Wolbachia strains sequenced to date indicate that the constitution of the genome is dynamic, but these strains are quite divergent and do not allow resolution of genome diversification over shorter time periods. We have sequenced the genome of the strain wBol1-b, found in the butterfly Hypolimnas bolina, which kills the male offspring of infected hosts during embyronic development and is closely related to the non-male-killing strain wPip from Culex pipiens.

Results: The genomes of wBol1-b and wPip are similar in genomic organisation, sequence and gene content, but show substantial differences at some rapidly evolving regions of the genome, primarily associated with prophage and repetitive elements. We identified 44 genes in wBol1-b that do not have homologs in any previously sequenced strains, indicating that Wolbachia's non-core genome diversifies rapidly. These wBol1-b specific genes include a number that have been recently horizontally transferred from phylogenetically distant bacterial taxa. We further report a second possible case of horizontal gene transfer from a eukaryote into Wolbachia.

Conclusions: Our analyses support the developing view that many endosymbiotic genomes are highly dynamic, and are exposed and receptive to exogenous genetic material from a wide range of sources. These data also suggest either that this bacterial species is particularly permissive for eukaryote-to-prokaryote gene transfers, or that these transfers may be more common than previously believed. The wBol1-b-specific genes we have identified provide candidates for further investigations of the genomic bases of phenotypic differences between closely-related Wolbachia strains.

Figure 1

(a)Maximum likelihood phylogeny, based on concatenated MLST genes,showing the relationships between the five Wolbachia strains for which genomes are available (wMel, wRi, wPip, wBm and wBol1-b)and the two other strains known from H.bolina,wBol1-a and wBol2. (b) A Neighbor-Net phylogenetic network based on a concatenated alignment of 654 core genes from the five complete genomes. Conflicting phylogenetic signals in the data are represented as boxes or parallelograms in the network. Very narrow boxes are present along the internal axes of the network, indicating that conflicting signal is present but minor in these genes. The large grey circle contains an expanded representation of the region of the network contained within the smaller grey circle, to make the narrow boxes visible.

Figure 2

Dot plots showing differences in syntenic conservation between Wolbachia genomes. Dots and lines represent unique genomic sequence matches, red for a forward match and blue for a reverse match (inversions). Comparisons are shown between the genomes of (a) wBol1-b and wPip, the two complete B-group genomes, (b) wMel and wRi, the two complete A-group genomes, and (c) wBol1-b and wMel, a representative comparison between A- and B-group genomes. Numbers along the axes represent genome coordinates.

Figure 3

Location of horizontally transferred secA genes, wBol1_1091 and wBol1_1092,in the wBol1-b genome. Arrows represent CDSs, are drawn to scale, and are labelled with abbreviated locus names (e.g. wBol1_1091 is 1091). Core Wolbachia genes are shown in green, non-core genes with identifiable Wolbachia homologs are in blue.

Figure 4

Maximum likelihood phylogenetic tree based on the amino acid sequences of Wolbachia secA gene wBol1_1092 and its homologs. Prokaryote and eukaryote secA genes cluster into two separate, well-supported clades. There are two apparent transfers from eukaryote to prokaryote species, marked with arrowheads. Bootstrap values over 50 are shown; for clarity they have been removed from short internal branches within the prokaryote clade and the large Daphnia and Harpegnathos clades. Each tip is labelled with the sequence’s Genbank GI number and the species name.