Continuing the discussion from Zika virus introduction in the Americas: when and from where?:
It has been speculated that several factor may limit our ability to accurate estimate the evolutionary rate of ZIKV including: potential recombinant sequences, sequence integrity and multiple passage history. Two previous studies also showed that significantly higher rates of nucleotide substitution appeared to occur during urban rather than during enzootic transmission of arboviruses (Sall AA, Faye O, Diallo M, Firth C, Kitchen A, Holmes EC (2010) Yellow fever virus exhibits slower evolutionary dynamics than dengue virus. J Virol 84: 765-772. Volk SM, Chen R, Tsetsarkin KA, Adams AP, Garcia TI, Sall AA, et al. (2010) Genome-scale phylogenetic analyses of chikungunya virus reveal independent emergences of recent epidemics and various evolutionary rates. J Virol 84: 6497-6504).
We have estimated the nucleotide substitution rate of ZIKV E gene by Bayesian MCMC analyses using four different datasets: 1) the ZIKV
COMPLETE that included all ZIKV sequences availables at the time of this study; 2) the ZIKV
NONREC that included all ZIKV sequences selected minus those previously characterized as recombinant in the original Faye et al. (2014) publication 3) the ZIKV
HUMAN that only included non-recombinant ZIKV sequences retrieved from human; and 4) the ZIKV
NON-HUMAN that only included non-recombinant ZIKV sequences retrieved from mosquitos and monkeys.
We found that the median substitution rates estimated from ZIKV
COMPLETE and ZIKV
NONREC were nearly equal (1.0×10-3 susbs./site/year) (Figure), indicating that ZIKV molecular clock estimates were not affected by the inclusion of putative recombinants strains. The median evolutionary rate estimated for the ZIKV
HUMAN data set (4.4×10-3 susbs./site/year), however, was more than 10 times higher than that estimated from ZIKV
NON-HUMAN (2.6×10-4 susbs./site/year), whereas intermediate rates were obtained for those datasets combining ZIKV sequences from both human and non-human hosts (ZIKV
COMPLETE and ZIKV
These results supports the notion that significantly higher rates of nucleotide substitution appeared to occur during urban rather than during enzootic transmission of arboviruses and further indicates that the time-scale of ZIKV outbreaks in human populations should be reconstructed using only sequences from urban epidemics.
Using the ZIKV
HUMAN data set, we estimated the T
MRCA of ZIKV outbreak epidemic in the Americas in 2014 (95% HPD: 2013-2015) (see post http://virological.org/t/phylogeographic-analyses-point-to-a-single-introduction-event-responsible-for-the-2015-zika-virus-outbreak-in-the-americas/211). This estimate is nearly equal to that described by Nuno Faria using a dataset of contemporaneous NS5B gene sequences that was mostly composed (91%) by ZIKV sequences from humans (see post http://virological.org/t/zika-virus-introduction-in-the-americas-when-and-from-where/204). Thus, accurate estimates of the time-scale process of recent ZIKV dissemination in human populations could be obtained from both E and NS5 genes, since datasets were exclusively or mostly composed by viral sequences sampled from urban epidemics.
A manuscript is under preparation by several authors including Daiana Mir
1, Gonzalo Bello
1, Matthew T. Aliota
2 and Edson Delatorre
1. Laboratório de AIDS & Imunologia Molecular, Instituto Oswaldo Cruz, FIOCRUZ, Rio de Janeiro, Brazil.
2. Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin-Madison.