Possible Roles of New Mutations Shared by Asian and American Zika Viruses

Abstract

Originating in Africa, the Zika virus (ZIKV) has spread to Asia, Pacific Islands and now to the Americas and beyond. Since the first isolation in 1947, ZIKV strains have been sampled at various times in the last 69 years, but this history has not been reflected in studying the patterns of mutation accumulation in their genomes. Implementing the viral history, we show that the ZIKV ancestor appeared sometime in 1930-1945 and, at that point, its mutation rate was probably less than 0.2 × 10-3/nucleotide site/year and subsequently increased significantly in most of its descendants. Sustaining a high mutation rate of 4 × 10-3/site/year throughout its evolution, the Ancestral Asian strain, which was sampled from a mosquito in Malaysia, accumulated 13 mutations in the 3'-untranslated region of RNA stem-loops prior to 1963, seven of which generate more stable stem-loop structures and are likely to inhibit cellular antiviral activities, including immune and RNA interference (RNAi) pathways. The seven mutations have been maintained in all Asian and American strains and may be responsible for serious medical problems we are facing today and offer testable hypotheses to examine their roles in molecular interactions during brain development.

Keywords: RNA structures.; Zika viruses; evolutionary rates; evolutionary tree.

Fig. 1.

Phylogenetic trees of 20 representative ZIKV strains. Numbers in front of ZIKV strains indicate the year of isolation. The root was determined using Spondweni virus (SM-6V-1) as the outgroup. (A) The NJ tree, which was obtained by applying the Neighbour-Joining (NJ)-method directly to the sequence data (Tamura et al. 2013). Numbers at various nodes indicate clustering percent supports generated by 1000 bootstrap analyses. Blue and orange rectangles indicate groups 1–3 of African strains and Asian strains which were distinguished using the NJ method, respectively. (B) The PE tree, where the tree topology was determined first and then branch lengths were determined using the ML method (Yang 2007).

Fig. 2.

Estimation of divergence times and evolutionary rates of ZIKV strains. (A) A modified PE tree with divergence times (T₁–T₉) and evolutionary rates of nucleotide substitution per site per year (a₁–a₁₉). Three groups of strains are distinguished by blue, black and red branches with evolutionary rates of 0.05, 0.2–1.5 and 2.0–5.1 × 10 ^− 3/site/year, respectively. (B) A phylogenetic relationship among MR766 (strain 0), Aae-Malaysia (strain 1), Micronesia (strain 3), and Aaf1 (strain 12). d_i,js are the branch lengths between pairs of strains i and j (i, j = strains 0, 1, 3, and 12). Evolutionary rates a₁, a₃, and a₁₂ are followed by divergence times (in parentheses).

Fig. 3.

mfold RNA structure models of MR 766 and Asian strains. (A) The SL-I structures. (B) The SL-DB structures. (C) The 3′-SL structures. The comparable segments of the African and Asian strains are shown by blue strips, whereas the mutated sites are highlighted by yellow rectangles. The SL-I, SL-DB, and 3′-SL structures were modeled considering nucleotide sites 1–60, 237–322, and 332–428, respectively, and minor loops for the latter two are not shown.