Assessing Host-Virus Codivergence for Close Relatives of Merkel Cell Polyomavirus Infecting African Great Apes

Abstract

It has long been hypothesized that polyomaviruses (PyV; family Polyomaviridae) codiverged with their animal hosts. In contrast, recent analyses suggested that codivergence may only marginally influence the evolution of PyV. We reassess this question by focusing on a single lineage of PyV infecting hominine hosts, the Merkel cell polyomavirus (MCPyV) lineage. By characterizing the genetic diversity of these viruses in seven African great ape taxa, we show that they exhibit very strong host specificity. Reconciliation analyses identify more codivergence than noncodivergence events. In addition, we find that a number of host and PyV divergence events are synchronous. Collectively, our results support codivergence as the dominant process at play during the evolution of the MCPyV lineage. More generally, our results add to the growing body of evidence suggesting an ancient and stable association of PyV and their animal hosts.

Importance: The processes involved in viral evolution and the interaction of viruses with their hosts are of great scientific interest and public health relevance. It has long been thought that the genetic diversity of double-stranded DNA viruses was generated over long periods of time, similar to typical host evolutionary timescales. This was also hypothesized for polyomaviruses (family Polyomaviridae), a group comprising several human pathogens, but this remains a point of controversy. Here, we investigate this question by focusing on a single lineage of polyomaviruses that infect both humans and their closest relatives, the African great apes. We show that these viruses exhibit considerable host specificity and that their evolution largely mirrors that of their hosts, suggesting that codivergence with their hosts played a major role in their diversification. Our results provide statistical evidence in favor of an association of polyomaviruses and their hosts over millions of years.

FIG 1

Maximum likelihood tree derived from an alignment of partial VP1 sequences. This tree was rooted at its center. The six gray circles stand for the main nodes whose date estimates are given in full in Tables 5 and 6; the black circle indicates the node that was used to calibrate the analyses. Note that these circles coincide with putative codivergence events. This tree was rooted using TempEst. Bp, bootstrap; pp, posterior probability. G. g. gorilla, G. gorilla gorilla; G. b. graueri, G. beringei graueri; G. b. beringei, G. beringei beringei; P. t. verus, P. troglodytes verus; P. t. troglodytes, P. troglodytes troglodytes; P. t. schweinfurthii, P. troglodytes schweinfurthii; H. sapiens, Homo sapiens.

FIG 2

Chronogram derived from an alignment of partial VP1 sequences. This chronogram was obtained through BMCMC analyses run under a multispecies coalescent model (the clades corresponding to entities considered species are highlighted in blue). Other BMCMC analyses run under different tree priors and ML analyses gave similar results. The root of the tree was the most frequently observed in all posterior samples of trees (posterior probability of ca. 0.60) and was also retrieved by rooting the ML tree at its center. The six gray circles stand for the main nodes whose date estimates are given in full in Tables 5 and 6; the black circle indicates the node that was used to calibrate the analyses. Note that these circles coincide with putative codivergence events. Bp, bootstrap; pp, posterior probability.