Biting by Anopheles funestus in broad daylight after use of long-lasting insecticidal nets: a new challenge to malaria elimination

During the past decade, the control of malaria has made slow but steady progress, the overall mortality rate dropping by more than 25% since 2000 [1]. Behind the statistics and graphs, however, lies a great, needless tragedy, with the life of an African child taken every minute [2]. The strategic approaches to control malaria rely on antivectorial programmes, which include mainly indoor residual spraying and insecticide-treated bed nets. The effectiveness of these strategies depends on the susceptibility of the vector species to insecticides and their behaviour, ecology and population genetics.

In Africa, these front-line control tools are very efficient against the main vectors of malaria, including Anopheles gambiae and Anopheles funestus, which prefer to bite and rest indoors at night when people are in bed [3, 4]. Both species have high capacity to transmit malaria parasites, because of their anthropophagic and endophilic characteristics, their longevity and their abundance [3]. Therefore, indoor interventions may protect people against infectious bites. These include reducing vector populations by mass killing, leading to a significant reduction in the lifespan, human contact and malaria sporozoite rate of Anopheles mosquitoes [5], and use of excito-repellent insecticides that cause mosquitoes to leave rooms for outdoors [6].

Many studies show that the effectiveness of these tools has been compromised by the emergence of kdr insecticide resistance, physiological and metabolic resistance of the vectors to insecticides [7–9]. Pyrethroid-treated bed nets are widely deployed in Africa, the estimated percentage of households with one impregnated net having increased from 3% in 2000 to 53% to 2012 [2]. The dramatic increase in Anopheles resistance may, however, reverse the effects of vector control programmes, leading to a resurgence of malaria morbidity in several parts of Africa [10, 11].

The emergence of insecticide-resistant Anopheles mosquitoes is also associated with behavioural changes that lead the vector to avoid intra-domiciliary vector control tools. Formerly, the maximum of anophelines aggressiveness was typically observed in the middle of night but since the introduction of LLINs significant changes were observed. In 1990, Fontenille et al.[12] showed that the An. funestus biting peak occurred from 01:00 to 03:00 indoors and from 02:00 to 05:00 outdoors. Furthermore, in Tanzania, Russell and al [13] in 2011 reported a nocturnal activity of An. funestus before insecticide-treated use became widespread. No studies reported behavioural activities of An. funestus prior to implementation of strategies control. Indoors interventions, such as LLINs and IRS, have become an ideal tool to protect people from biting mosquitoes and the correct use of these tools decrease the burden of malaria transmission [14, 15]. However, to ensure their survival mosquitoes adopt behavioural change and the best time to find host is the early hours of the night or at dawn. This form of resistance is an adaptive strategy, which allows the mosquitoes to avoid or circumvent control strategies [16–18].

Charlwood et al.[19] in Papua New Guinea reported that although many anophelines were killed after the introduction of impregnated nets, a portion was diverted outside. Females that have been diverted have a longer oviposition cycle and therefore tend to bite earlier the following evening. In Burkina Faso, Riehle et al. [20] identified a new subgroup, An. gambiae Goundry, which presumably bites only outside and is more susceptible to Plasmodium infection. The authors reported that this new subgroup would probably be selectively favored by indoor vector control measures.

The emergence of the new phenotypes of mosquitoes was also reported elsewhere, probably natural selected by changing environmental conditions that create new selection pressures [21] Moiroux et al. [22] reported that, in Benin, three years after universal coverage with LLINs, An. funestus showed substantial diurnal and early biting activity and more frequent outdoor biting. The behavioural effect on An. funestus may, therefore, be genetic or phenotypic adaptation to widespread use of LLINs and IRS. This change in mosquito biting habits could jeopardize the success of control operations. In addition to insecticide resistance, behavioural adaptation of mosquitoes that feed exclusively outdoors at early hours of the night or morning may affect the epidemiology of malaria transmission. The aim of this study was to investigate the population dynamics of mosquitoes in relation to their biting activity after prolonged use of insecticide-treated bed nets.

In Africa, An. funestus plays a major role in malaria transmission in several areas [31, 32]. The An. funestus group is composed of many species, which can be differentiated only from discrete traits of both larvae and adults [33]; specific PCR has been developed to differentiate the species in the group [30]. Nine species have been described, with widely different biology and vectorial capacity. Apart from An. funestus, the other species appear to be mainly zoophilic [34]. Human Plasmodium has been found mainly in An. funestus, which is an excellent vector, with high vectorial capacity, and only rarely in Anopheles rivulorum[35]. Anopheles vaneedeni has been experimentally infested [36].

Most of the An. funestus in this study were captured between 07:00 and 11:00 in broad daylight. This new phenotype might have resulted from a shift in sibling species composition, but molecular analyses showed the same population of An. funestus, with a difference in biting activity after prolonged use of LLINs. Anopheles funestus populations captured after dawn are highly anthropophilic and endophilic, when most of the active population is occupied with household tasks, rustic employment and commerce. This is, therefore, a potential danger for vulnerable populations (children, pregnant women, patients and elderly individuals) who spend most of their time indoors.

Anopheles funestus is a major vector of malaria transmission in Dielmo [23, 37]. The studies in the 1990s showed entomological inoculation rates that reached up 180 infected bites per person per year. In contrast, after introduction of LLINs, An. funestus almost completely disappeared, and the annual aggressiveness dropped from 17.2 bites per person per night to fewer than 1.2 (unpublished data). This may be due to the susceptibility of An. funestus to insecticides, an ineffective sampling method to estimate human exposure to mosquito bites or a behavioural change to avoid intra-domicilary control. In Africa, resistance of An. funestus to insecticides is widespread, although no kdr alleles has been found; nevertheless, the presence of biochemical resistance of this species to all classes of insecticide, pyrethroids, carbamates and DDT, has been demonstrated [38, 39]. This phenomenon has not been seen in the village of Dielmo (unpublished data). However, inadequate capture methods may be the reason why this vector is not encountered. In a recent study in Zambia [40], it was observed that an Ifakara Tent Trap (model-C) placed outdoors was thirty four times more sensitive than HLC for sampling An. funestus after deployment of LLINs. In the case of An. funestus behavioural change, many studies have shown that new mosquito behaviour is an adaptive response to control strategies. Indeed the excito-repellents effects of treated bed nets repel mosquitoes outdoor, it follow a behavioural change strategy which is marked by changes in the biting time and place if the vector is highly endophilic and sometimes shift in host feeding [16]. In the current study, diurnal activity was seen in An. funestus, with a peak of aggressiveness between 08:00 and 9:00. These results are similar to those of studies in southern Benin after scaling up of universal coverage with LLINs, where 26.4% of An. funestus were caught after 06:00 [22]. The authors concluded that this vector has exophilic behaviour, because the villagers wake up early to work on crops. Exophilic behaviour was also reported in Tanzania by Russel et al. [13] after massive deployment of LLINs, although early biting activity of An. funestus was demonstrated. The authors showed that this behavioural change in An. funestus was responsible for residual outdoor transmission of malaria. Previous studies reported that behavioural changes to others Anopheles vectors. In Dielmo, Trape et al.[10] reported an early biting of An. gambiae after prolonged use of LLINs and in Benin Corbel et al.[11] showed greater exophagy rates of An. gambiae and An. funestus was recorded with a massive presence coupled with a shift from endophagic to exophagic behaviour. Another sibling species of An. gambiae, Anopheles arabiensis showed early biting activities in Ethiopia after use of LLINs both indoors and outdoors; and 80 per cent of this vector were captured before 22:00 with a peak activity between 19:00 to 20:00 [17]. All these studies showed the emergence of the behavioural adaptation of mosquitoes in response to indoor interventions, which can thus jeopardize the efficacy of these tools and also constitute a risk to people who are so accessible.

In the present study, one mosquito was found infected, which was captured during the day, while since distribution of impregnated nets in this village, no An. funestus had been found positive by ELISA. This constitutes a real risk for malaria transmission in this locality. The new phenotypes of mosquito have a human blood index and high parity, showing that a close relationship has been maintained between Anopheles and humans, despite the distribution of LLINs. The concept of phenotypic plasticity or the selection of specific genetic traits for adapting to environmental conditions appears to confirm Pates et al. [21]. This behavioural change represents a characteristic to respond to new environmental conditions represented by the use of LLINs. Russell et al.[41] showed in Tanzania that An. funestus was strongly exophilic and made a trophic deviation to cattle after universal coverage with LLINs, while in Benin [22] and Senegal, an exceptional, remarkable adaptation has been seen. The fundamental concept of biology appears to be respected: in the face of stressful situations, organisms adapt or disappear [42].

Control strategies against malaria, such as indoor residual spraying and LLINs, represent powerful, appropriate tools, as African malaria vectors were considered to be entirely endophilic and anthropophilic. This study suggests An. funestus has adapted its biting time to the new situation (use of impregnated nets). Behavioural change in response to insecticides may make control tools ineffective, thus negatively affecting malaria control strategies. This remarkable behavioural adaptation of mosquitoes to insecticide-based vector control interventions requires heightened, close awareness in the context of pre-elimination of malaria.