Behavioural responses of Anopheles gambiae sensu stricto M and S molecular form larvae to an aquatic predator in Burkina Faso

Background: Predation of aquatic immature stages has been identified as a major evolutionary force driving habitat segregation and niche partitioning in the malaria mosquito Anopheles gambiae sensu stricto in the humid savannahs of Burkina Faso, West Africa. Here, we explored behavioural responses to the presence of a predator in wild populations of the M and S molecular forms of An. gambiae that typically breed in permanent (e.g., rice field paddies) and temporary (e.g., road ruts) water collections. Methods: Larvae used in these experiments were obtained from eggs laid by wild female An. gambiae collected from two localities in south-western Burkina Faso during the 2008 rainy season. Single larvae were observed in an experimental arena, and behavioural traits were recorded and quantified a) in the absence of a predator and b) in the presence of a widespread mosquito predator, the backswimmer Anisops jaczewskii. Differences in the proportion of time allocated to each behaviour were assessed using Principal Component Analysis and Multivariate Analysis of Variance. Results: The behaviour of M and S form larvae was found to differ significantly; although both forms mainly foraged at the water surface, spending 60-90% of their time filtering water at the surface or along the wall of the container, M form larvae spent on average significantly more time browsing at the bottom of the container than S form larvae (4.5 vs. 1.3% of their overall time, respectively; P < 0.05). In the presence of a predator, larvae of both forms modified their behaviour, spending significantly more time resting along the container wall (P < 0.001). This change in behaviour was at least twice as great in the M form (from 38.6 to 66.6% of the time at the wall in the absence and presence of the predator, respectively) than in the S form (from 48.3 to 64.1%). Thrashing at the water surface exposed larvae to a significantly greater risk of predation by the notonectid (P < 0.01), whereas predation occurred significantly less often when larvae were at the container wall (P < 0.05) and might reflect predator vigilance. Conclusions: Behavioural differences between larvae of the M and S form of An. gambiae in response to an acute predation risk is likely to be a reflection of different trade-offs between foraging and predator vigilance that might be of adaptive value in contrasting aquatic ecosystems. Future studies should explore the relevance of these findings under the wide range of natural settings where both forms co-exist in Africa