Dual Effects of Light on Regulating Aedes aegypti Heat-Seeking Behavior

Dear Editor,

Mosquitoes, as important arthropod disease vectors, efficiently transmit pathogens such as dengue virus, Zika virus, yellow fever virus, and malaria. To find their host, female mosquitoes use multiple sensory systems to detect host-associated cues such as CO₂, volatile odorants, visual signals, and temperature. Recent genetic studies have identified several sensory molecules that are required for detecting those host-associated cues. Gustatory receptor (Gr3) is required for Aedes aegypti to sense CO₂ produced by the host and to activate flight [1]. Olfactory receptor (Or4) is responsible for sensing sulcatone from the human body and might be required for the host preference of Ae. aegypti [2]. Female mosquitoes are also attracted to host body heat, and recent studies have found that the cooling receptor ionotropic receptor (Ir21a) is required for heat-seeking [3]. Mosquitoes are attracted to high-contrast visual cues after sensing CO_2, which indicates that visual signals also play an important role in host localization [4]. A recent study demonstrated that vision-guided target attraction requires opsin1 (Op1)and opsin2 (Op2) in Ae. aegypti [5].

Mosquitoes are distributed worldwide, and their vectorial capacity is dramatically affected by different environmental conditions. Despite that many studies have revealed the importance of host-associated cues during mosquito host-seeking, fewer investigations have investigated other environmental factors such as light, humidity, and temperature in regulating host-seeking. Light modulates a wide range of mosquito behaviors such as flight activity, mating, oviposition, and biting time [6]. Light also serves as a strong zeitgeber to entrain the circadian clock, and mosquito light preference and blood-sucking behaviors have been reported to be under circadian control [7]. A recent study in Anopheles gambiae showed that brief light exposure during night-time decreases the biting activity [8]. These studies strongly suggest that ambient light signals play important roles in regulating mosquito host-seeking behavior and might impact their vectorial capacity. Thus, examining how light regulates mosquito host-seeking behavior might lead to new measures to control mosquito-borne diseases.

In this study, we designed experiments to explore the effects of ambient light on Ae. aegypti host-seeking behavior as well as the underlying molecular mechanisms. Heat-seeking behavior serves as an excellent model to study mosquito host-seeking in the laboratory [1]. We established a heat-seeking behavior assay (Figs 1A, S1, and Supplementary Methods) and asked whether heat-seeking behavior is modulated by different ambient light conditions. Female mosquito heat-seeking activity was recorded under either ambient infrared light or ambient high-intensity white light conditions (~650 lux). Since Ae. aegypti is a diurnal species that might display changing heat-seeking activity with the time of day, we measured the heat-seeking activity in both day-time (zeitgeber time ZT3 and ZT9) and night-time (ZT15 and ZT21). We found that heat-seeking activity under the white light condition was significantly weaker than that under the infrared light condition, suggesting that light acutely suppresses Ae. aegypti heat-seeking (Fig. 1B). On the other hand, consistent with previous studies on Ae. aegypti biting [9], mosquitoes displayed significantly stronger heat-seeking during the day (Fig. 1B). Since Ae. aegypti displays strong host-seeking during dawn and dusk, we hypothesized that mosquito heat-seeking might only be suppressed by strong light. Therefore, we further studied the heat-seeking behavior under different light conditions: infrared, red, and white light at different intensities from 100 lux to 650 lux. We found that the heat-seeking was greatly suppressed when the light intensity was stronger than 300 lux (Fig. 1C). To investigate whether the decrease in heat-seeking is attributable to the suppression of locomotion by light, the CO₂ activation rate and locomotor behavior at ZT9 under 650 lux white or infrared light were measured, and no difference was found (Fig. 1D, E), indicating that the heat-seeking behavior suppression by light is not caused by a decrease of locomotion or the CO₂ activation rate.

Fig. 1.

Dual effects of light on mosquito heat-seeking. A Schematic of the behavioral assay for heat-seeking. Each chamber has 20 female mosquitoes. B Heat-seeking is stronger during the day and is suppressed by strong ambient light. “% on Peltier” means the the number of mosquitoes that land on zone 37°C minus the number of mosquitoes on zone 25°C then divided by the total number of mosquitoes (mean ± SD, n = 16–28 trials per experiment; letters denote distinct categories; between a and b, P <0.05, between b and c, P <0.0001; Kruskal-Wallis and post hoc Mann-Whitney U tests). C Heat-seeking is suppressed by white light >300 lux (mean ± SD, n = 7–15 trials for each experiment; letters denote distinct categories; P <0.05; Kruskal-Wallis and post hoc Mann-Whitney U tests). D, E Light does not decrease locomotion and CO₂ activation rate. The locomotion of wild-type mosquitoes was activated by CO₂ under 650 lux white light or infrared light. The number of mosquitoes that took off in response to the CO₂ and the speed at which they flew was recorded (mean ± SD, n = 21–23 trials for each experiment; n.s., no significant; Student’s t-test). F Schematic of ZT time and light-dark switching. G Heat-seeking behavior at ZT9, ZT21, and the same times after 1 h of dark or light treatment in the incubator. The index increases after light treatment and decreases after dark treatment under the ambient infrared light condition (mean ± SD, n = 10–14 trials for each experiment; **P <0.01, ****P <0.0001; Kruskal-Wallis and post hoc Mann-Whitney U tests).

Meanwhile, together with previous studies [5, 10], our finding that Ae. aegypti mosquitoes showed much stronger heat-seeking during the day (Fig. 1B) strongly suggested that light might regulate mosquito heat-seeking behavior through light entrainment of the biological clock. So, we switched the light condition 1 h ahead of the behavioral test to investigate whether prior light experience could entrain mosquito behavior (Fig. 1F, G). We found that turning on the light for 1 h during the night significantly facilitated heat-seeking, while turning off the light during the day for 1 h significantly decreased heat-seeking (Fig. 1G), indicating that light might also act as an entraining signal for heat-seeking behavior in Ae. aegypti. Thus, these results suggested that light plays dual roles in heat-seeking behavior: while light serves as an entraining signal to promote heat-seeking during the day, strong light also inhibits heat-seeking in Ae. aegypti.

Light is primarily detected through the visual system [11], and histamine is an essential neurotransmitter for visual signal transduction in Drosophila melanogaster [12]. To determine whether the visual system is required for light to regulate mosquito heat-seeking behavior, we generated blind mosquitoes by mutating histidine decarboxylase (Hdc) that is required for histamine synthesis. Two Hdc (AAEL009593) mutant alleles of Ae. aegypti were generated via the CRISPR/Cas9 (clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9) system (Fig. S2A). Electroretinogram (ERG) recording showed that the depolarization voltages in the Hdc mutant mosquitoes were significantly weaker than in the wild-type, and this was be rescued by feeding Hdc mutant mosquitoes with histamine (Fig. S2C–E). These results demonstrated that histamine is required for a normal ERG potential, suggesting that the Hdc mutant mosquitoes were blind. To test if these mutants are truly blind, we further performed a flight simulator experiment (Fig. S2B and Supplementary Methods). The yaw of the mutant’s flight indicated that they could not see moving objects. In contrast, the yaw of the wild-type and mutant fed with histamine was normal (Fig. S2F–K). These experiments indicate that Hdc mutant mosquitoes are indeed blind.

Using these Hdc mutant mosquitoes, we investigated whether white light suppresses mosquito heat-seeking through the visual system. Compared to wild-type mosquitoes whose heat-seeking was strongly suppressed by white light (650 lux) (Fig. 2A), the Hdc mutant mosquitoes showed no light-induced suppression such that they performed similar heat-seeking behavior under both white and red light conditions (Figs 2B, S3). The light-suppression defect of Hdc mutant mosquitoes was rescued by feeding 2‰ histamine (Figs 2C, S3). No locomotor defect was found in Hdc mutant mosquitoes (Fig. S4). We also used a pharmacological strategy to inhibit histamine synthesis and found that the heat-seeking of mosquitoes fed with histamine inhibitors was no longer suppressed by strong light, which is consistent with the findings in Hdc mutant mosquitoes (Fig. S5). These results demonstrated that Hdc-dependent visual transduction is required for light to suppress Ae. aegypti heat-seeking behavior.

Fig. 2.

Function of Hdc and Cry in mediating light effects on heat-seeking and host-seeking. A–C Heat-seeking behavior under 650 lux white light and red light. D Heat-seeking is no longer suppressed by white light in Hdc mutants. Wild-type and Hdc mutant mosquitoes fed histamine show reduced heat-seeking under white light (mean ± SD, n = 12–39 trials for each experiment). E Schematic of host-seeking behavior using human hands. F, G Host-seeking behavior under red light and 650 lux white light conditions. H Host-seeking is no longer suppressed by white light in Hdc mutants (mean ± SD, n = 17–22 trials for each experiment; n.s., not significant). I Similar to wild-type (WT) mosquitoes, Hdc mutant mosquitoes show strong heat-seeking during the day and weak heat-seeking at night under the ambient infrared light condition (mean ± SD, n = 11–17 trials per experiment; letters denote distinct categories; between a and b, P <0.01, between b and c, P <0.001). J Heat-seeking of Hdc mutant mosquitoes with light-dark switching (mean ± SD, n = 11–15 trials per experiment). K Heat-seeking behavior of mosquitoes after injection of Cry dsRNA is significantly enhanced (mean ± SD, n = 18–29 trials per experiment; letters denote distinct categories; between a and b, P <0.01, between b and c, P <0.05). L Heat-seeking of Cry knockdown mosquitoes with light-dark switching (mean ± SD, n = 6–12 trials per experiment; **P <0.01, ***P <0.001, ****P <0.0001; Kruskal-Wallis and post hoc Mann-Whitney U tests).

We further investigated whether light suppresses host-seeking using human hands as targets (Fig. 2E). We found that, similar to heat-seeking behavior, bright light (650 lux) strongly suppressed host-seeking behavior in wild-type mosquitoes but not in Hdc mutant mosquitoes, and feeding the Hdc mutants with 2‰ histamine rescued the light-suppression effect (Figs 2F–H, S3). Thus, light suppresses both heat-seeking and host-seeking behaviors through Hdc-dependent visual transduction in Ae. aegypti.

Animals synchronize their behavior to light/dark cycles through photoentrainment of the circadian clock [13, 14]. The visual system has been reported to mediate the photoentrainment in D. melanogaster, and we asked whether the mosquito visual system also contributes to the light/dark cycle of heat-seeking activity in Ae. aegypti. Using Hdc mutant mosquitoes in heat-seeking experiments, we found that these mutants still showed stronger heat-seeking during the day than they did during the night, which is similar to wild-type mosquitoes (Fig. 2I). When we turned off the light for 1 h during the day, the heat-seeking of Hdc mutant mosquitoes decreased significantly, which is also similar to wild-type mosquitoes. However, light treatment during the night was less effective in facilitating heat-seeking in Hdc mutant mosquitoes (Figs 2J, S3H). These results suggested that Hdc-dependent visual transduction only partially mediates the circadian-dependent heat-seeking behavior (Fig. 2I, J), indicating that other light transduction pathways might be involved. Cryptochrome (Cry) is a blue light-sensitive protein that mediates light-dependent degradation of the clock protein Timeless and plays an important role in circadian light entrainment in D. melanogaster [15]. We hypothesized that exposure to light might lead to the degradation of Cry to facilitate heat-seeking behavior in Ae. aegypti. So, we knocked down the Cry gene (AAEL004146) using RNA interference (RNAi) and conducted heat-seeking experiments at different ZTs under infrared light conditions. The expression of Cry significantly decreased after double-stranded RNA (dsRNA) injection into Ae. aegypti (Fig. S6). After Cry knockdown, mosquitoes showed stronger heat-seeking at all ZTs tested, while still maintaining the difference in heat-seeking activity between day and night (Fig. 2K), suggesting that Cry functions as a heat-seeking suppressor but does not significantly affect the rhythm of heat-seeking behavior. The light switch experiments also showed that mosquitoes were still able to respond to the light switch after the knockdown of Cry (Fig. 2L).

Taken together, our results revealed dual effects of light on heat- and host-seeking in Ae. aegypti. On one hand, strong light inhibits heat-seeking and host-seeking behaviors, which requires Hdc-dependent visual transduction. On the other hand, light serves as an entraining signal to promote heat-seeking during the day. The latter effect does not only depend on the visual system or the light-dependent circadian regulator CRY and possibly requires both or additional pathways.

Our results also indicated that light is able to regulate host-seeking behavior by mosquitos through CRY. In Drosophila, Cry is expressed in a subset of clock neurons [15], which might be similar in mosquitoes. Further study of the Cry-expressing neural circuit in host-seeking behavior might provide new insights to understand the central neural mechanism of host-seeking behavior in mosquitoes.

Ae. aegypti has been found to be a diurnal mosquito that sucks blood almost exclusively during the day [9]. This might be controlled by the circadian clock. Here, we found that the heat-seeking of Ae. aegypti was stronger during the day, and light exposure during the night could facilitate heat-seeking, which might be a result of light entrainment of the mosquito's circadian clock. We studied the function of Hdc and Cry in mediating the light entrainment effects and found that loss-of-function of either pathway did not disrupt the diurnal heat-seeking activity in Ae. aegypti, suggesting the involvement of both pathways and potentially other unidentified pathways. Studies in Drosophila have also found that the light signal is transduced through multiple pathways to entrain the circadian clock, and it would be of great interest to identify those pathways in both flies and mosquitoes, which might help to understand the evolutionary mechanisms of how different mosquito species adapted to be either diurnal or nocturnal and lead to new measures for mosquito control as well.

Supplementary Information

Below is the link to the electronic supplementary material.

Conflict of interest

The authors have no financial conflicts of interest.