LARVAL COMPETITION DIFFERENTIALLY AFFECTS ARBOVIRUS INFECTION IN AEDES MOSQUITOES

Abstract

Both density-mediated and trait-mediated indirect biotic interactions may be important in structuring communities. Indirect interactions in many study systems remain unexplored; in part, because they are often difficult to detect, and in many instances, have been identified empirically only when unexpected results arise. Indirect effects induced by competition may be particularly important among organisms with complex life cycles, wherein competitive effects experienced in one life stage influence species interactions in one or more subsequent stages. We determined whether species-specific effects of larval competition in the mosquitoes Aedes albopictus and Aedes aegypti have indirect effects at the adult stage, specifically testing for effects on arboviral infection with Sindbis virus (SINV). For A. albopictus, but not for A. aegypti, competition resulted in greater infection, body titer, and dissemination rates compared to low-competition conditions. Whole body titers of virus increased with adult size irrespective of competition. However, between competitive treatments, mosquitoes from low-competition conditions had greater mean size, with lower infection rates and lower whole body titers than the smaller mosquitoes from high-competition conditions. These results suggest that larval competition, common in natural mosquito populations, has important indirect effects on adults by altering mosquito-virus interactions. Such indirect effects may change transmission parameters of pathogens.

Fig. 1

Aedes albopictus least-squares means (±se) for female survivorship and size at emergence. Different lowercase letters indicate significant differences between bivariate means. Competition treatments consisted of species density ratios of A. albopictus : A. aegypti—160:0, 320:0, and 160:160.

Fig. 2

Aedes aegypti least-squares means (±se) for female survivorship and time to emergence. Different lowercase letters indicate significant differences between bivariate means. Competition treatments consisted of species density ratios of A. albopictus : A. aegypti—0:160, 0:320, and 160:160.

Fig. 3

Least-squares means (±se) for estimated finite rate of increase, λ′, for Aedes albopictus and A. aegypti. Points without bars have standard errors too small to appear on the graph. Different lowercase and uppercase letters indicate significant differences between means for A. albopictus and A. aegypti, respectively.

Fig. 4

Bivariate plots of least-squares means (±se) for three dependent variables for Aedes albopictus females fed on a Sindbis virus blood meal. (A) Proportion of infected females vs. proportion with disseminated infection. (B) Proportion of infected females vs. body titer. In both graphs, the dashed ellipse indicates multivariate means that are not significantly different. Numbers in the figure key represent the ratio of A. albopictus to A. aegypti.

Fig. 5

Least-squares means for body titer and size of adult Aedes albopictus females with disseminated (i.e., infection spread beyond the midgut, infecting secondary target organs such as body, legs) and isolated (i.e., infection limited to the midgut) Sindbis virus (SINV) infections. The size effect on females with disseminated infections gives a slope of 5.48 (se = 1.28). Solid and dashed lines drawn through bivariate means show the best fit for A. albopictus with disseminated infections in three competitive treatment conditions. Numbers in the figure key represent the ratio of A. albopictus to A. aegypti.