ИСТИНА

Introduction: Enteroviruses are amongst the most rapidly changing viruses. Poliovirus is known to accumulate 1% of substitutions in VP1 per year. Enteroviruses are also an important source of emerging infections. We did a parallel analysis of 30 EV types to study common trends and variation in evolution of VP1. Methods: All enterovirus types represented by at least 50 partial VP1 sequences (3’ terminal 300 nt) were analyzed by Bayesian phylogenetic approach using BEAST software. Results: The inferred nucleotide substitution rates in non-polio enteroviruses were on average 0.9% per year, and varied between 0.6-1.2% in most of the types. The rate estimates for a type were reproducible and almost not affected by a sample bias. In all types the age of the most recent common ancestor of a type was between 55-200 years. This age was significantly affected by presence of a prototype strain in a sample. Without a prototype strain, common ancestors of types appeared up to 40 years younger. Moreover, the topology of phylogenetic trees suggested that the global populations of VP1 region sequences of roughly half of the most common types underwent global population bottlenecks. This resulted in very recent common ancestor ages of contemporary isolates and extinction of most lineages that hypothetically existed in mid-20th century. On the other hand, diversification of the known types was inferred to have occurred within the last 500 years. These data, which should be treated with caution, suggest that the current enterovirus diversity emerged rather recently. Enteroviruses of the same type differ by no more than 25% nucleotide sequence in VP1. At their fast substitution rates, they could accumulate enough substitutions to drift beyond these boundaries in a matter of years. Analysis of 2200 full VP1 sequences of 15 types found that the type criteria suggested in 1999 remain generally valid. Therefore, enteroviruses change fast on a scale of years, but accumulate mostly synonymous substitutions and remain surprisingly stable on a scale of decades. The hypothetical global-scale bottlenecks also resulted in distinct virus groups within types. Distribution of genetic distances suggests that genotypes within types can be identified based on VP1 sequence distances in roughly a half of types; however, the distance criteria of such genotypes would be different, ranging between 11-20%. Conclusion: VP1 genes of a type exist as global populations that undergo global expansion-extinction cycles. At the same time, none of the types that existed in 1950s were extinct to date, therefore there must be long-term reservoirs of enterovirus genes. Generally poor surveillance and common detection of rare genetic variants of VP1 in Africa suggest that rare enterovirus variants can be preserved on this continent. Reports of known and novel enterovirus types in non-human primates call for further studies of this potential reservoir of enterovirus genes. The stability of types despite rapid accumulation of substitutions supports a hypothesis of types as fitness peaks, where viruses drift around, but cannot normally jump to another fitness peak because of unacceptable fitness penalty for going off the current state.