Background
Malaria vectors require a blood meal to develop their eggs [
1] and the process of finding blood hosts is primarily mediated by host odour [
2‐
5]. Carbon dioxide (CO
2) is one of the important components of human host odour affecting mosquito host-seeking behaviour [
6]. It is thought that this gas activates mosquitoes by eliciting take-off behaviour. The presence of CO
2 then sustains the mosquitoes in host-seeking flight [
6,
7], guiding them towards their blood meal hosts [
3]. It is not surprising, therefore, that CO
2 is a key ingredient of synthetic mosquito attractants for host-seeking mosquitoes [
8]. The application of this gas from pressurized cylinders, fermenting sugar (i.e., sucrose) or molasses and/or the use of dry ice present major challenges to the use of CO
2 -based mosquito attractants under field conditions. The gas cylinders are heavy, bulky, expensive and prone to leakages [
9] and dry ice can be difficult to obtain, transport and store [
9‐
12]. Whilst CO
2 produced by fermenting refined sugar or molasses can offer a solution to these problems [
13] this method of CO
2 production is also expensive and presents logistical challenges when used on a large scale because the gas is only produced over one trapping night (ca ten hours) and must be replenished daily.
In a study by Turner et al. 2-butanone was identified as a potential replacement for CO
2 in a synthetic blend of mosquito attractants [
14]. These authors demonstrated the capacity of 2-butanone to induce a dose-dependent activation of the cleavage product A (cpA) CO
2 receptor neuron in the maxillary palps of
Anopheles gambiae, Aedes aegypti and
Culex quinquefasciatus. This simulated the activity of CO
2. In related studies, acetone and cyclopentanone have also been tested as substitutes for CO
2 [
15‐
17] but with little success under field conditions [
16,
17]. The current study sought to: (a) evaluate the synergistic importance of CO
2 as a mosquito attractant in counter-flow MM-X traps; (b) assess the attraction of mosquitoes to different concentrations of 2-butanone; (c) determine the optimal concentration of 2-butanone for attracting mosquitoes; and, (d) evaluate the attraction of mosquitoes to odour baits containing 2-butanone in the field.
Discussion
In this study, it was observed that the responses of laboratory-reared
An. gambiae s.s. mosquitoes to the MB5 reference attractant blend with CO
2 were significantly higher compared to MB5 alone, CO
2 alone or a trap without a bait. In all semi-field investigations MB5 + CO
2 attracted a significantly higher number of mosquitoes than its variants containing the different dilutions of 2-butanone used to replace CO
2. When using the blends of MB5 + 2-butanone, the highest catches were associated with the 99.5% concentration of 2-butanone and the 1.0% concentration of 2-butanone. Overall catches of
An. arabiensis were far much lower than those of
An. gambiae under semi-field conditions, probably because the human-mimicking attractant blends were developed and customized using
An. gambiae as the test organism [
25,
26] and that
An. arabiensis has a more opportunistic host preference [
1]. In the field study,
An. gambiae s.l.,
An. funestus and
Culex species were attracted to both MB5 + CO2 and MB5 + 99.5% 2-butanone with no significant difference in catch size between the blends. These results demonstrate that 2-butanone can be used as a replacement for CO2 under field conditions.
The finding that significantly more laboratory-reared mosquitoes were attracted to MB5 + CO
2 than to MB5 alone (P < 0.001), CO
2 alone (P < 0.001) or a trap without a bait (P < 0.001) underscores the action of CO
2 as a synergist in mosquito attractants [
21,
29]. The gas is known to activate mosquitoes by eliciting take-off behaviour and sustaining them in host-seeking flight [
6,
7,
30]. These findings are in line with the results of studies which demonstrated that compounds are more attractive to host-seeking mosquitoes when blended than when applied alone [
31].
It was observed that the pure (99.5%) form of 2-butanone is a potential replacement for CO
2 in mosquito attractants. Under field conditions there were no differences between the numbers of
An. gambiae s.l. and
An. funestus mosquitoes attracted to MB5 + CO
2 compared with MB5 + 99.5% 2-butanone. 2-butanone is a natural product identified in the emanations of various vertebrates and arthropods [
32,
33] and several insects express a behavioural response upon exposure to this compound. Two separate studies [
34,
35] have reported that the olfactory receptor cell of the fruit fly
Bactrocera tyoni that responds to CO
2 also responds to 2-butanone. And more recently, Turner et al. [
14] demonstrated the capacity of 2-butanone to induce a dose-dependent activation of the CO
2 receptor neuron in the maxillary palps of
An. gambiae, Aedes aegypti and
Culex quinquefasciatus. Failure to observe this effect under semi-field conditions may imply that 2-butanone acts as a long range rather than a short or medium-range cue or that the proximity of a more attractive alternative (MB5 + CO
2) was preferred when mosquitoes were presented with a direct choice. Furthermore, there were no statistical differences in the numbers of
An. funestus attracted to MB5 + CO
2 or MB5 + 2-butanone (99.5%) under field conditions, but because a colony of this mosquito species has not been established at the research station in Mbita the response of this species under semi-field conditions could not be tested.
The relatively high number of wild male
An. gambiae s.l. and
An. funestus mosquitoes in traps baited with synthetic odour blends is contrary to expectations because males are phytophagous and are thought unlikely to respond to host-seeking odour blends compared to female mosquitoes [
3,
37]. It may be assumed that the males were pursuing the females for mating, if this life history trait ever occurs indoors without swarming. Currently, there is an urgent need for potent synthetic odour blends for sampling and control of male mosquitoes. Such blends could be deployed to reduce mating success, and also to reduce the number of gravid female mosquitoes and prevalence of mosquito-borne diseases. Because the prospects of eliminating malaria are largely threatened by rapid development of drug-resistant
Plasmodium parasites and insecticide-resistant malaria vectors, novel tools are needed. Odour-baited trapping technology has been used successfully in western Kenya to reduce mosquito bites and malaria prevalence [
38]. The number of mosquitoes around houses was reduced by mass deployment of outdoor traps that were baited with MB5 augmented with 2-butanone instead of CO
2 [
36,
38]. The baited traps were effectively and repeatedly used for removal trapping of outdoor-biting mosquito vectors. Although residual attraction of mosquitoes to synthetic compounds impregnated on nylon strips has been reported [
23], similar studies are needed for mosquito attractants that are augmented with 2-butanone. An odour bait with a residual activity of long duration without the need for frequent replenishment would be convenient for both monitoring as well as removal trapping of mosquitoes in remote areas of sub-Saharan Africa.
Human landing catches, light traps, bed nets occupied by humans, pyrethrum spray catches, and man-landing catches are commonly used for sampling of malaria vectors and estimation of malaria transmission intensity [
39]. The methods vary in terms of reliability and efficacy, hence the need for standardized tools that are sensitive, specific, reliable and ethically acceptable for trapping and sampling malaria vectors. Recent studies indicate that synthetic odour baits dispensed by MM-X traps can be used reliably to collect live and species specific samples of both indoor and outdoor biting malaria and other mosquito vectors, particularly those which are host-seeking [
13,
18,
29]. Thus, it is important to compare the odour blend, putative CO
2 replacement and odour baited traps that were used in the current study with the common trapping tools outlined above.
Authors’ contributions
MMM, CKM, WT, and WRM designed the study. MMM, AH and WRM analysed the data. All authors read and approved the final manuscript.