Host-symbiont conflicts: positive selection on an outer membrane protein of parasitic but not mutualistic Rickettsiaceae.

Abstract

The Rickettsiaceae is a family of intracellular bacterial symbionts that includes both vertically transmitted parasites that spread by manipulating the reproduction of their host (Wolbachia in arthropods) and horizontally transmitted parasites (represented by Cowdria ruminantium), and mutualists (Wolbachia pipientis in nematode worms). We have investigated the nature of natural selection acting on an outer membrane protein, the wsp gene in Wolbachia and its homologue map1 in Cowdria, thought likely to be involved in host-parasite interactions in these bacteria. The ratio of nonsynonymous to synonymous substitution rates (d(N)/d(S)) at individual amino acid sites or at lineages within the gene's phylogeny was estimated using maximum likelihood models of codon substitution. The first hypothesis we tested was that this protein is under positive selection in the parasitic but not in the mutualistic Rickettsiaceae. This hypothesis was supported as positive selection and was detected in Cowdria and arthropod Wolbachia sequence evolution but not in the evolution of Wolbachia sequences from nematodes. Furthermore, this selection was concentrated outside the transmembrane region of the protein and, therefore, in the regions of the protein that may interact with the host. The second hypothesis tested was that positive selection would be stronger in the strains of arthropod Wolbachia that distort the host sex ratio than in those that induce cytoplasmic incompatibility. However, we found no support for this hypothesis. In conclusion, our results are consistent with the hypothesis that antagonistic coevolution causes faster evolution of surface protein sequences in parasites than in mutualists. Confirmation of this conclusion awaits the replication of these results both in additional genes and across more bacterial taxa. The regions of the wsp and map1 genes we identified as likely to be involved in host-parasite arms races should be examined in future studies of parasite virulence and host immune responses, and during the design of vaccines.