How Did Zika Virus Emerge in the Pacific Islands and Latin America?

Abstract

The unexpected emergence of Zika virus (ZIKV) in the Pacific Islands and Latin America and its association with congenital Zika virus syndrome (CZVS) (which includes microcephaly) and Guillain-Barré syndrome (GBS) have stimulated wide-ranging research. High densities of susceptible Aedes spp., immunologically naive human populations, global population growth with increased urbanization, and escalation of global transportation of humans and commercial goods carrying vectors and ZIKV undoubtedly enhanced the emergence of ZIKV. However, flavivirus mutations accumulate with time, increasing the likelihood that genetic viral differences are determinants of change in viral phenotype. Based on comparative ZIKV complete genome phylogenetic analyses and temporal estimates, we identify amino acid substitutions that may be associated with increased viral epidemicity, CZVS, and GBS. Reverse genetics, vector competence, and seroepidemiological studies will test our hypothesis that these amino acid substitutions are determinants of epidemic and neurotropic ZIKV emergence.

FIG 1

ZIKV phylogeny of African-Asian/Pacific and Latin American virus isolates, including mapping of amino acid substitutions. Dashed boxes show positions of amino acid substitutions relative to the open reading frame of ZIKV NC_012532.1, and dashed lines indicate the branch positions in the tree at which amino acid substitutions occurred. Amino acid substitutions of specific interest are highlighted in the boxes. The 5′- and 3′-UTR positions refer to the number of nucleotide positions before the start codon and after the stop codon, respectively. For clarity, the tree displays all Pacific virus branches in blue and all Latin American virus branches in red. Red circles highlight virus sequences that were retrospectively associated with CZVS. Posterior probability support values were identical except for node F1, which were 0.92 and 0.98, respectively, for the analyses that included or excluded the African lineages (Fig. 1, nodes 1 to 2 and A to G). Accession numbers for African genomes used in one of the analyses but not shown in the figure are as follows: HQ234498.1, DQ859059.1, KU720415.1, LC002520.1, NC_012532.1, AY632535.2, KU963573.1, KF268949.1, KF268948.1, KF268950.1, KU963574.1, HQ234501.1, KU955595.1, KU955592.1, KU955591.1, KF383116.1, and KX198134.1.