Background:  Simian immunodeficiency virus (SIV) has had considerable success at crossing species barriers:  both HIV-1 and HIV-2 have been transmitted on multiple occasions from SIV-infected natural host species. Temporal phylogenetic analysis of SIV_sm envelope V1/V2 variants sampled from experimentally inoculated sooty mangabeys and rhesus macaques revealed a rhesus macaque-specific restriction in viral diversity during the acute phase of infection. Phylogenetic and population genetic analyses were employed to understand the selective forces operating after acute infection and to identify specific codons involved in adaptation to the new rhesus macaque host.

Methods:  To elucidate the overall temporal dynamics of selection in SIV_sm-infected sooty mangabeys and rhesus macaques, nucleotide sequence alignments of each time point from each animal were compared to the SIV_sm inoculum to obtain frequencies of synonymous and non-synonymous polymorphism. These data were used to estimate the number of adaptive events. A similar site-by-site analysis of non-synonymous variation was also performed. Rates of non-synonymous and synonymous substitution at each codon were estimated on maximum-likelihood trees of individual time points using a likelihood model of codon substitution. Amino acid changes were mapped onto bayesian phylogenetic trees of all samples from individual animals to estimate numbers of amino acid substitutions at each site.

Results:  Adaptive events in rhesus macaque viral populations increased linearly from 0 to 9 between days 14 and 70, while adaptive events in sooty mangabeys remained around 0. Of 124 amino acid sites analyzed, only 27 were under positive selection. Of these 3 sites, 2 were evolving in both species and occur in V1. Of the 9 codons evolving under positive selection in the rhesus macaques only, 8 fall in V2. These included the loss of an N-glyc site at day 578, a 2-amino acid shift in an N-glyc site at days 40 and 70, and a substitution in V2 at days 40 and 70. Temporal analysis of polymorphisms at each site revealed that most selected mutations were already present in the SIV_sm inoculum.

Conclusions:  Strong species-specific positive selection in the V2 loop at days 40 and 70 underscores its importance in adaptation of SIV_sm to rhesus macaques. Since neutralizing antibody responses are low or not yet present, these changes may be adaptations to a divergent CD4 receptor or co-receptor. Examining how SIV_sm adapts to non-natural hosts may help us to understand and prevent cross-species transmission of other deadly RNA viruses.