Vector potential of Aedes (Stegomyia) spp. populations and risk of dengue and yellow fever virus transmission in three Kenyan cities

Abstract

Dengue (DENV) and Yellow fever (YFV) viruses are medically important flaviviruses that are transmitted by Aedes mosquito vectors of the subgenus Stegomyia especially Aedes aegypti and Aedes bromeliae, respectively, in East Africa.Urbanization has been identified as a major driver in the emergence of these diseases because of the permissive environment it creates promoting the convergence of susceptible human hosts and local vectors. However, while dengue outbreaks have mainly been reported in urban and semi-urban areas of East Africa, yellow fever outbreaks remain limited to the sylvatic cycle. Urban outbreaks may be facilitated by adaptation of the viruses to sylvatic vectors. In Kenya recurrent dengue outbreaks remain limited to the city of Mombasa and do not occur in the other major cities of Kisumu and Nairobi, despite unplanned urbanization being a problem across Kenya. The risk of the spread of dengue to other major cities as well as the emergence of an urban yellow fever outbreak in Kenya therefore remains unknown and deserves public health attention. This scenario emphasizes the need for risk assessment studies to inform public health decisions on cost effective vaccination for yellow fever and vector control for dengue. With no knowledge of the level of yellow fever risk in the major urban areas of Kenya, or how the dengue risk levels in Mombasa compare to those of other areas of contrasting endemicity (Kisumu and Nairobi), this study initially performed an intensive house-to-house survey of natural and artificial water holding containers to assess the degree of infestation with mosquito immatures (larvae and pupae) using standardized immature sampling tools. Based on estimated Stegomyia indices of the major Aedes (Stegomyia) species encountered, the risk of transmission of DENV/YFV in these three cities of Kenya was established and included seasonal trends (long-rains, short-rains and dry seasons). Based on the container type and the number of immatures collected, the preferred breeding sites of the major vectors were identified and characterized for targeted vector control. Also, following an intensive sampling aided by carbon dioxide baited BG-Sentinel traps, the seasonal abundance and diversity of the host-seeking adult vector populations were compared for the three cities. Using prokopack aspirators, the preferred resting sites of the major vector Ae. aegypti were identified and compared across the different cities. Furthermore, the 12S gene target extensively used for blood meal identification was used to determine the host feeding preference of wild-caught, blood-fed Ae. vaegypti mosquitoes collected in all three cities. The ability of Ae. aegypti populations to transmit strains of DENV/YFV at selected temperatures of 22°C, 28°C and 31°C(representing the minimum/maximum average monthly temperatures), was assessed and the vectorial capacity for DENV transmission estimated and compared for the three cities. Following vector competence studies, the genetic difference of Ae. aegypti specimens found to be susceptible and non-susceptible for YFV were compared by analyzing sequence variation in mitochondrial cytochrome oxidase subunit I (COI) gene, a genome target widely used in molecular evolution studies to differentiate between closely related mosquito species. Findings from this study showed that Ae. aegypti and Ae. bromeliae were the major Stegomyia species in all three cities. While the immatures of Ae. aegypti preferentially occurred in artificial water-holding containers like jerricans, drums, used/discarded containers and tyres, those of Ae. bromeliae utilized more natural sites such as tree holes and leaf axils. These identified breeding sites could be made the focus of targeted vector control. Based on established vector index thresholds, the risk of DENV transmission was high in Kilifi (outskirts of Mombasa) and Kisumu, and low-to-medium in Nairobi, while the risk of YFV transmission was low-to-medium in Kilifi and Kisumu, and low in Nairobi. Ae. aegypti was the most abundant host-seeking Stegomyia species, and found to be significantly more abundant in Kilifi and Kisumu than Nairobi, and during the long-rains (April-June). Low occurrence of Aedes bromeliae was observed and the abundance varied neither by season nor by city. Ae. aegypti displayed differences in resting habits with populations in Kilifi being comparably endo- and exophilic compared to Kisumu and Nairobi (mainly exophilic), suggesting possible genetic difference in the mosquito populations. Further laboratory experiments to assess how dengue risk levels in Kilifi compare to Kisumu and Nairobi revealed that all three populations of Ae. aegypti could transmit the DENV serotype-2 and virus transmission increased with an increase in temperature. Interestingly, blood meal analysis suggested that Ae. aegypti mosquitoes in Kilifi and Nairobi were more anthropophagic compared to Kisumu. Overall, the estimated potential for DENV transmission (vectorial capacity) in Kilifi was 9- and 14-fold higher than in Kisumu and Nairobi, respectively. This pattern was mainly influenced by the low levels of human blood feeding detected in Ae. aegypti mosquitoes in Kisumu, and the low temperatures in Nairobi. Vector competence studies provided evidence that Ae. aegypti mosquitoes from all three cities were susceptible to infection with the East African YFV genotype but unable to disseminate the virus. This was suggestive of a low risk of YFV transmission in all three cities. Two Ae. aegypti lineages, well supported by the maximum likelihood-tree, were observed in Kilifi, Kisumu and Nairobi. However, no unique pattern was observed in the clustering of the YFV susceptible and non-susceptible specimens in the different lineages. It was therefore unlikely that the genetic differences within the Ae. aegypti population affect YFV susceptibility and hence yellow fever epidemic patterns in Kenya. Taken together, this study provides the most in-depth data on entomological risk factors relating to transmission of DENV/YFV and possible emergence of dengue and yellow fever in the major urban areas of Kenya. The improved understanding of the epidemiology of these diseases strongly suggests where to focus control efforts, especially vector control, as well as providing guidance for cost effective vaccination for yellow fever.