Field relevant variation in ambient temperature modifies the density-dependent establishment of Plasmodium falciparum in mosquitoes: implications for the infectious reservoir and beyond?

The relationship between Plasmodium falciparum gametocyte density and infections in mosquitoes is central to understanding the rates of transmission with important implications for control. Here, we determined whether field relevant variation in environmental temperature could also modulate this relationship. Anopheles stephensi were challenged with three densities of P. falciparum gametocytes spanning a [~]10-fold gradient, and housed under diurnal/daily temperature range (\"DTR\") of 9{degrees}C around means of 20{degrees}C, 24{degrees}C and 28{degrees}C. At the peak stages of infection for each temperature, the proportion of mosquitoes infected with oocysts in the midguts or infectious with sporozoites in the salivary glands were measured (referred to collectively as vector competence hereon), in addition to oocyst intensities from infected midguts. While vector competence was similar at 20 DTR 9{degrees}C and 24 DTR 9{degrees}C, the proportion of mosquitoes infected and subsequently infectious were also comparable, with evidence, surprisingly, for higher vector competence in mosquitoes challenged with intermediate gametocyte densities. For the same gametocyte densities however, severe reduction in the proportion of infectious mosquitoes was accompanied by a significant decline in vector competence at 28 DTR 9{degrees}C, although density per se showed a positive and linear effect at this temperature. Unlike vector competence, oocyst intensities decreased with increasing temperatures with a predominantly positive and linear association with gametocyte density, especially at 28 DTR 9{degrees}C. Oocyst intensities across individual infected midguts suggested temperature-specific differences in mosquito susceptibility/resistance: at 20 DTR 9{degrees}C and 24 DTR 9{degrees}C, dispersion (aggregation) increased in a density-dependent manner but not at 28 DTR 9{degrees}C where the distributions were consistently random. Limitations notwithstanding, our results have manifold implications in, for instance, how variation in temperature could modify seasonal dynamics of infectious reservoirs and transmission and the contribution of high-/patent- and low-density/sub-patent carriers, to suggestions for design and deployment of transmission-blocking vaccines/drugs, but with a cautionary note suggesting how low efficacy could lead to transmission enhancement in certain environments.