"Mosquito and Arboviral Landscape in Akagera National Park, Rwanda"

Abstract

Mosquito-borne viruses are an ongoing public and veterinary health concern in East Africa. In protected areas such as Akagera National Park (ANP), high wildlife diversity, varied mosquito breeding habitats, and close contact zones with livestock and people create conditions in which arboviruses could circulate and, in some situations, spill over which is transmission from an enzootic cycle (often involving wildlife and mosquitoes) into livestock or humans. From a One Health perspective, this kind of setting is important because the same ecological drivers can influence risks for wildlife health, livestock production, and human exposure. However, ANP has limited baseline data on mosquito species composition and on whether arboviruses are detectable in local mosquito populations. In parallel, for many arboviruses it remains unclear how viruses persist between recognised outbreaks. They continue at low levels in wildlife or livestock, occur within poorly characterised reservoir cycles, or persist through vector-associated processes. Consequently, site-specific ecological data are needed to interpret potential transmission pathways. Against this background, the aim of this thesis was to generate a baseline for ANP by (i) characterising mosquito species diversity using cytochrome c oxidase subunit I COI DNA barcoding and (ii) screening female mosquito pools for arboviruses using generic PCR assays designed for broad detection. COI barcoding improved mosquito identification compared with morphology alone and clarified the taxa represented in the ANP collections, although species-level assignment was sometimes limited by the availability of reference sequences for regional mosquitoes. In addition, arbovirus screening detected an insect-specific flavivirus (ISF), Culex flavivirus (CXFV), and Bunyamwera virus, indicating that mosquitoes in ANP carry both mosquito-associated viruses and viruses with relevance to vertebrate infection. Overall, these results provide baseline evidence that arbovirus related signals are detectable in mosquitoes from ANP and offer a clearer description of the mosquito community that could support transmission. Building on this baseline, the findings help guide future work to link virus detections to specific mosquito taxa, habitats, and seasons, and to connect mosquito data with host contact patterns, which is central to a One Health approach at wildlife–livestock–human interfaces. Finally, they provide a starting point for addressing inter-outbreak questions in this system by evaluating whether detections are most consistent with low-level circulation in vertebrate hosts, local reservoir cycles, or other persistence mechanisms that may contribute to re-emergence.