Genotype-specific variation in West Nile virus dispersal in California
West Nile virus (WNV) is an arbovirus that is maintained in an enzootic cycle between mosquitoes and birds. Spillover into humans causes annual epidemics of WNV neuroinvasive disease. Since its introduction into the U.S., WNV has caused more than 18,000 cases of WNV neuroinvasive disease and 1600 deaths.
WNV was first reported in North America in 1999 in New York, and, as soon as the first cases of WNV neuroinvasive disease were described, Dr. Bill Reisen at the UC Davis Center for Vectorborne Diseases began surveying mosquitoes in California for the virus. His mosquito surveillance work showed that, within 4 years of its initial detection in New York, WNV spread across the country to the west coast, where it was first detected in southern California. WNV quickly spread to northern California and has remained enzootic across the state. Thanks to his long-term surveillance program, Dr. Reisen has a unique collection of WNV PCR-positive mosquitoes spanning the entire history of WNV in California. In this study, we sequenced over 110 full-length WNV genomes from mosquitoes collected between 2003 and 2011, representing approximately 1% of Dr. Reisen’s collection. One difficulty with this retrospective study was that some of the samples were degraded, and we were unable to sequence a consistent number of isolates from every year.
We used these sequences of WNV isolates from California to perform phylogeographic analyses to visualize the temporal and spatial spread of WNV in California (see figure). Surprisingly, we found evidence for at least 5 introductions of WNV into California occurring between 2003 and 2006. Because several of these introductions occurred nearly simultaneously, the multi-wave front of emerging WNV was masked until this genetic study. Not all of the introduced WNV strains were successful in establishing enzootic maintenance across California; in fact, WNV isolates within the SW03 genotype were restricted to southern California. In addition, we determined that SW03 genotype viruses spread at a faster rate than WN02 genotype viruses. Thus, by sequencing viruses isolated from the mosquito vector during epidemic and non-epidemic years, rather than sequencing isolates collected from humans during outbreaks, we were able to get a more accurate picture of the way a virus spreads – in a series of repeated movements, many of which are unsuccessful, rather than in a single, straightforward pathway – and to identify a viral genetic factor that was associated with WNV dispersal.