Differential attraction of malaria mosquitoes to volatile blends produced by human skin bacteria

Abstract

The malaria mosquito Anopheles gambiae sensu stricto is mainly guided by human odour components to find its blood host. Skin bacteria play an important role in the production of human body odour and when grown in vitro, skin bacteria produce volatiles that are attractive to A. gambiae. The role of single skin bacterial species in the production of volatiles that mediate the host-seeking behaviour of mosquitoes has remained largely unknown and is the subject of the present study. Headspace samples were taken to identify volatiles that mediate this behaviour. These volatiles could be used as mosquito attractants or repellents. Five commonly occurring species of skin bacteria were tested in an olfactometer for the production of volatiles that attract A. gambiae. Odour blends produced by some bacterial species were more attractive than blends produced by other species. In contrast to odours from the other bacterial species tested, odours produced by Pseudomonas aeruginosa were not attractive to A. gambiae. Headspace analysis of bacterial volatiles in combination with behavioural assays led to the identification of six compounds that elicited a behavioural effect in A. gambiae. Our results provide, to our knowledge, the first evidence for a role of selected bacterial species, common on the human skin, in determining the attractiveness of humans to malaria mosquitoes. This information will be used in the further development of a blend of semiochemicals for the manipulation of mosquito behaviour.

Figure 1. Schematic drawing of the dual-choice olfactometer used to examine the response of Anopheles gambiae to VOCs.

The olfactometer contains three test chambers in a stacked configuration.

Figure 2. Response of A. gambiae to VOCs released from broths of five species of skin bacteria.

Bacterial species were tested at the time of exponential growth (Eg) and during the stationary phase (St). The percentage of mosquitoes caught in the trapping device baited with the bacterial broth (B) and in the trapping device baited with medium alone (M) are given. N =  number of mosquitoes released. Error bars represent standard errors of the mean; ***: χ²-test P<0.001; **: χ²-test P<0.01: *: χ²-test P<0.05.

Figure 3. Ranking different species of skin bacteria for attractiveness to A. gambiae.

Results of dual-choice olfactometer experiments in which VOCs of different skin bacterial species in liquid medium were tested for their attractiveness to A. gambiae against ammonia released from an LDPE sachet. Bars show the mean relative attractiveness (i.e. the number of mosquitoes entering the trapping device baited with bacteria as a proportion of the total number entering either this trapping device or the trapping device baited with ammonia). N = total number of mosquitoes released. Error bars represent standard errors of the mean. Means not sharing the same superscript letter differ significantly at P<0.05 (GLM).

Figure 4. Responses of A. gambiae to individual compounds added to a basic blend in an olfactometer.

Compounds were added in three concentrations (dilution 1∶100, 1∶1,000 or 1∶10,000) to the basic blend (ammonia, lactic acid, tetradecanoic acid) (basic blend + compound  =  Treatment) and tested against the basic blend (Control) alone in two series (A and B). N = number of mosquitoes released. Error bars represent standard errors of the mean; ***: χ²-test P<0.001; **: χ²-test P<0.01: *: χ²-test P<0.05.