Mosquito feeding patterns in the context of West Nile and eastern equine encephalitis viruses in eastern Ontario, Canada.

West Nile virus (WNV) and eastern equine encephalitis virus (EEEV) are typically maintained via enzootic cycles between birds and mosquitoes (Diptera: Culicidae). In North America, the blood hosts of mosquitoes involved in pathogen circulation are poorly described for many regions, limiting our ability to predict disease outbreaks. Our objectives were to (i) identify local (Eastern Ontario, Canada) bloodmeal hosts fed on by mosquitoes using DNA barcoding and (ii) investigate if the bloodmeal hosts identified were representative of the entire host community using species accumulation curves and Chao richness estimates. We successfully identified 313 bloodmeal hosts from 292 mosquitoes belonging to the species Aedes cinereus, Aedes vexans, Coquillettidia perturbans, Culex pipiens/restuans, Ochlerotatus stimulans and Ochlerotatus trivittatus, with 257 and 56 bloodmeal hosts identified at the species and genus levels, respectively. Concurrent with the literature, Culex species almost exclusively fed on birds, C. perturbans fed on a mixture of birds and mammals, and the remaining mosquito species fed almost exclusively on mammals. Eastern cottontail rabbits (Sylvilagus floridanus) were the most common mammal species fed on, driven by A. vexans and O. trivittatus. With respect to WNV and EEEV enzootic cycles, Culex species, the main vectors for WNV transmission, were observed to forage on American robins (Turdus migratorius), an amplification host for WNV, highlighting a pathway by which WNV could be maintained in our study region. A. vexans, a potential bridge vector, was also observed to forage on a relatively high frequency of humans (Homo sapiens), suggesting a pathway by which humans could be exposed to WNV. C. perturbans, a potential bridge vector for EEEV transmission, was also observed to forage on both an American Robin (also an amplifying host for EEEV) as well as on humans in a relatively high frequency, illustrating a potential pathway by which EEEV exposure in humans could occur. Our species accumulation curves and Chao estimates of richness with respect to bloodmeal hosts, indicated that the bloodmeal host community was not completely characterized. Together, our results contribute to a better understanding of mosquito bloodmeal hosts in eastern North America and suggest a larger or more directed sampling effort would be needed to characterize the entire bloodmeal host community in our study region.