UC Davis

Abstract Background Genetic diversity is a key factor that enables adaptation and persistence of natural populations towards environmental conditions. It is influenced by the interaction of a natural population's dynamics and the environment it inhabits. Anopheles gambiae s.s. and Anopheles arabiensis are the two major and widespread malaria vectors in sub-Saharan Africa. Several studies have examined the ecology and population dynamics of these vectors. Ecological conditions along the Kilombero valley in Tanzania influence the distribution and population density of these two vector species. It remains unclear whether the ecological diversity within the Kilombero valley has affected the population structure of An. gambiae s.l. populations. The goal of this study was to characterise the genetic structure of sympatric An. gambiae s.s and An. arabiensis populations along the Kilombero valley. Methodology Mosquitoes were collected from seven locations in Tanzania: six from the Kilombero valley and one outside the valley (~700 km away) as an out-group. To archive a genome-wide coverage, 13 microsatellite markers from chromosomes X, 2 and 3 were used. Results High levels of genetic differentiation among An. arabiensis populations was observed, as opposed to An. gambiae s.s., which was genetically undifferentiated across the 6,650 km2 of the Kilombero valley landscape. It appears that genetic differentiation is not attributed to physical barriers or distance, but possibly by ecological diversification within the Kilombero valley. Genetic divergence among An. arabiensis populations (F ST = 0.066) was higher than that of the well-known M and S forms of An. gambiae s. s. in West and Central Africa (F ST = 0.035), suggesting that these populations are maintained by some level of reproductive isolation. Conclusion It was hypothesized that ecological diversification across the valley may be a driving force for observed An. arabiensis genetic divergence. The impact of the observed An. arabiensis substructure to the prospects for new vector control approaches is discussed.