NEW YORK (GenomeWeb) – Researchers have reconstructed the spread of the Zika virus in Central America and Mexico using the genomes of viruses collected from patients in the region.
Zika virus, a Flavivirus, was first discovered in Uganda in 1947 and is transmitted primarily by mosquitos. It was first reported in the Americas in 2015, where it is expected to infect about 100 million people by the end of the epidemic.
In a new study appearing in Cell Host & Microbe today, researchers sequenced 61 Zika virus samples. In combining phylogenetic, epidemiological, and environmental data, they found that there were multiple introductions of the virus into Central America and Mexico, and that there have been peaks in transmission that coincide with variations in mosquito abundance.
"In this study, we were able to reconstruct in detail how Zika was introduced and spread throughout this region," senior author Oliver Pybus from the University of Oxford said in a statement. "By looking at the genomes of Zika viruses obtained from people who had been infected, we discovered a surprising amount of information about the epidemic's transmission history."
Pybus and his colleagues collected serum and urine samples from patients who lived in or traveled to Central America or Mexico and who showed symptoms of Zika infection. The 95 samples they collected — 59 from Mexico, 16 from Nicaragua, nine from Honduras, eight from Guatemala, and three from El Salvador — were positive via qRT-PCR testing for the virus.
For 81 of those samples, the researchers took a combined metagenomic next-generation sequencing and spiked primer enrichment strategy to tease out Zika viral genomes from their samples. From this, they identified sequencing reads that matched the Zika virus in 71 samples, with an average consensus coverage of 64 percent. For the remaining samples, all of which were from Nicaragua, the researchers used a different metagenomic approach that relied on bait probe capture.
In all, they generated 61 complete and partial viral genomes with an average genome coverage of nearly 83 percent, after leaving out samples with low coverage.
Pybus and his colleagues constructed a phylogenetic tree using these and 298 published Zika virus genomes. Their tree placed 102 of the 107 Zika virus genomes from Central America and Mexico into a single monophyletic clade, dubbed clade B, which was closely related to viruses from Brazil. Four other Zika virus genomes from Panama and Mexico formed clade C, suggesting multiple introductions of the virus to the region, though only one has dominated.
The researchers traced the most recent common ancestor of clade B back to December 2014 and found that it diverged from Brazilian strains around July 2014. Based on this, the researchers suggested that the clade B lineage was exported from Brazil to Central America between July and December 2014, about three months earlier than other estimates.
They further reported that the most recent common ancestor of the clade B lineage likely lived in Honduras. They timed the introduction of the Zika virus to Honduras to between July and September 2014, and said the virus spread from there to Guatemala, Nicaragua, and southern Mexico in late 2014 and early 2015 before then dispersing within Mexico in mid-2015.
By also tying in epidemiological data, the researchers noticed that there were peaks in transmission in winter and summer in Belize, Honduras, and Guatemala that occurred during the wet and dry seasons. Using environmental data, the researchers also noted that these timeframes coincided with environmental suitability for mosquito growth.
"What we found was not what we expected," said first author Julien Thézé from Oxford. "There is usually only one breeding season per year, but we found two separate outbreaks in a relatively small country like Honduras."
This, he and his colleagues noted in their paper, indicates there are complex annual trends in Zika virus transmission in the region that could complicate viral control efforts.