Aestivation of the African malaria mosquito, Anopheles gambiae in the Sahel

Abstract

The African malaria mosquito, Anopheles gambiae, inhabits diverse environments including dry savannas, where surface waters required for larval development are absent for 4-8 months per year. Under such conditions, An. gambiae virtually disappears. Whether populations survive the long dry season by aestivation (a dormant state promoting extended longevity during the summer) or are reestablished by migrants from distant locations where larval sites persist has remained an enigma for over 60 years. Resolving this question is important, because fragile dry season populations may be more susceptible to control. Here, we show unequivocally that An. gambiae aestivates based on a demographic study and a mark release-recapture experiment spanning the period from the end of one wet season to the beginning of the next. During the dry season, An. gambiae was barely detectable in Sahelian villages of Mali. Five days after the first rain, before a new generation of adults could be produced, mosquito abundance surged 10-fold, implying that most mosquitoes were concealed locally until the rain. Four days after the first rain, a marked female An. gambiae s.s. was recaptured. Initially captured, marked, and released at the end of the previous wet season, she has survived the 7-month-long dry season. These results provide evidence that An. gambiae persists throughout the dry season by aestivation and open new questions for mosquito and parasite research. Improved malaria control by targeting aestivating mosquitoes using existing or novel strategies may be possible.

Figure 1.

Abundance (Bottom) and composition (Top) of An. gambiae from the late wet season (2008) to the early subsequent wet season (2009). Density is measured by daily live indoor density per house (Bottom). The left axis (black) corresponds to the data from the wet season, and the right axis (red) corresponds to the data from the dry season (note difference in scale). Vertical gray lines depict one standard error of the mean density per house. Composition of An. gambiae is shown in bars based on pooled collections representing 1- to 18-day intervals centered on the date shown above the bars. Sample size used in calculation of composition is shown beneath each bar, and percentage of each population is given inside the bars. Stars indicate day of desiccation of the last larval site in the village (November 8) and 1.5 km away (November 25), which was the last larval site in a radius of 6 km or more from Thierola. Blue ellipse indicates day of scattered rains (March 22) during the dry season.

Figure 2.

House density of An. gambiae, An. Funestus, and An. rufipes before and after the first rain (May 24, 2009) measured by the total daily live collections indoors in all 120 houses of Thierola. In a box-whisker plot, the box extends between the 25th and 75th percentiles [i.e., across one interquartile range (IQR)], and the whiskers extend up to the most extreme value but not beyond 1.5 times the IQR. Values located over 2.5 IQR from the median are shown. Non-overlapping notched belts indicate significant difference between means (P < 0.05). This figure appears in color at www.ajtmh.org.

Figure 3.

Photographs of the aestivating female recaptured after the first rains nearly 7 months (212 days) after being marked. The painted dot on the dorsal side of the thorax (Top) and the two dots on the ventral side of the abdomen (Bottom) are visible. Thick arrows point to light-blue paint dots that are clearly visible on the image. The narrow arrow points to the white paint dot that is faintly visible on the image (contrast and brightness of the images were optimized using GIMP2 by Dr. Nick Manoukis).