The sugar meal of the African malaria mosquito Anopheles gambiae and how deterrent compounds interfere with it: a behavioural and neurophysiological study.

In this study, we show that female African malaria mosquitoes Anopheles gambiae starved for 3-5 h start to engorge on sucrose at concentrations between 50 and 75 mmol l(-1). Half of the feeding response (ED50) is reached at 111 mmol l(-1) and the maximum response (0.4 mg) occurs at 250 mmol l(-1). Two receptor cells in a trichoid sensillum of the labellum, called the 'sucrose' and 'water' neurones, are activated by sucrose and water, respectively. The electrophysiological response of the sucrose receptor cell starts well below the level of sugar necessary to induce engorgement. The sugar receptor cell is most sensitive to small increments in sucrose concentration up to 10 mmol l(-1) with a response plateau from 25 mmol l(-1). Fructose has a mild phagostimulatory effect on A. gambiae, whereas no significant differences in meal sizes between water and glucose were found. However, when 146 mmol l(-1) fructose plus glucose are mixed, the same engorgement as on 146 mmol l(-1) sucrose is observed. Likewise, even though the sucrose receptor cell is not activated by either fructose or glucose alone, equimolar solutions of fructose plus glucose activate the neurone. We conclude that there is a behavioural and neurophysiological synergism between fructose and glucose, the two hexose sugars of sucrose. We show that some bitter-tasting products for humans have a deterrent effect on feeding in A. gambiae. When 1 mmol l(-1) quinidine, quinine or denatonium benzoate is added to 146 mmol l(-1) sucrose, feeding is almost totally inhibited. The effect of berberine is lower and no significant inhibition on engorgement occurs for caffeine. The deterrent effect depends on the concentration for both quinine and quinidine. Capillary feeding experiments show that contact chemosensilla on the mouthparts are sufficient for the detection of sucrose and bitter products. The feeding assay findings with deterrents correlate with the neurophysiological responses of the sucrose and water labellar neurones, which are both inhibited by the bitter compounds denatonium benzoate, quinine and berberine between 0.01 and 1 mmol l(-1), but not by the same concentrations of caffeine. In conclusion, sucrose stimulates feeding and activates the labellar sucrose neurone, whereas feeding deterrents inhibit both the sucrose and water neurones. This study provides an initial understanding of the physiological mechanisms involved in sugar feeding in A. gambiae and shows how some bitter products interfere with it.