Behavioural Responses of Male Aedes albopictus to Different Volatile Chemical Compounds

Simple Summary Many studies have been performed to assess the effects of chemical compounds on mosquito behaviour. These studies almost exclusively involve only female mosquitoes as they can transmit disease pathogens, or at least, cause biting nuisance. Few studies have considered male mosquitoes. The identification of chemical substances that attract males can be very useful for trapping purposes, especially for monitoring the makeup of the male population during control programmes, such as those involving the release of sterile male mosquitoes. Twenty-eight chemical compounds from different chemical classes were evaluated using a dual-port olfactometer assay with at least three serial hexane dilutions against a hexane control. The compounds included known animal, plant and fungal volatiles, and the components of a putative Aedes aegypti pheromone. Many of the compounds were repellent for male mosquitoes, especially at the highest concentration. One compound, decanoic acid, acted as an attractant for males at an intermediate concentration. Decanoic acid did not elicit a significant response from female mosquitoes. Abstract The Asian tiger mosquito, Aedes albopictus, has become one of the most important invasive vectors for disease pathogens such as the viruses that cause chikungunya and dengue. Given the medical importance of this disease vector, a number of control programmes involving the use of the sterile insect technique (SIT) have been proposed. The identification of chemical compounds that attract males can be very useful for trapping purposes, especially for monitoring the makeup of the male population during control programmes, such as those involving the use of the SIT. Twenty-eight chemical compounds from different chemical classes were evaluated using a dual-port olfactometer assay. The compounds included known animal, fungal and plant host volatiles, and components of a putative Aedes aegypti pheromone. Many of the compounds were repellent for male mosquitoes, especially at the highest concentration. One compound, decanoic acid, acted as an attractant for males at an intermediate concentration. Decanoic acid did not elicit a significant response from female mosquitoes.


Introduction
The study of the chemically-mediated behaviour of male mosquitoes has, to a large extent, been ignored in favour of the more exciting animal host odour-related behaviour of their blood-feeding female counterparts [1]. Male mosquitoes, like females, require nectar as a source of carbohydrates and respond to semiochemicals released by suitable host plants. Semiochemicals may also signal the presence of suitable resting sites. As mosquitoes are thought to be important in the pollination of certain plant species [2,3], it is highly likely that coevolution between a mosquito species and a flowering plant species will have resulted in plant-mosquito specialisations. Certain plant volatiles may therefore be expected to be attractive or repellent for some mosquito species but not others. For

Olfactometer Validation
To validate the olfactometer(Dataset 1), a known mosquito attractant, human foot odour [17], was used. A polyamide sock (Ciak 15 Sanpellegrino, CSP International Fashion Group SpA, Ceresara, Italy) was worn by one of the male authors (LMG) for three days and nights to accumulate foot odour [18,19]. The worn sock (treatment) was placed in a plexiglass cylinder and closed at each end with net fabric. An unworn sock (control) was placed in an identical cylinder. The two cylinders were positioned in the distal part of the arms of the olfactometer. The entire olfactometer was covered by a white cotton sheet to avoid visual stimulation of the mosquitoes during the test. The olfactometer was illuminated from above the sheet by a 60 watt halogen light bulb (OSRAM GmbH, Munich, Germany). The illuminance just above the olfactometer was around 2000 lux. Twenty Ae. albopictus individuals (males or females) were released into the main chamber of the olfactometer by inserting the release chamber and rotating it 180 degrees, thus allowing the plexiglass lid to swing open, permitting the mosquitoes to enter the main chamber.
The end caps of the two arms were then fixed in position and the airflow commenced, and maintained, at a constant rate of 3 L/min. After 10 min had elapsed, the number of individuals in the traps of the two arms were counted. The cylinders containing the socks were removed and all of the mosquitoes were removed and eliminated after each run. Clean air was allowed to flow through the olfactometer for 15 min between replicates. The position of the worn and control socks in each arm was alternated for each replicate. Five replicates were performed for each sex. Given that the number of individuals trapped in each arm may not have a normal distribution, and that they could be considered proportions, each number was arcsine transformed [20] and compared using a paired, two tailed t test. This assay was always conducted between 15:00 and 19:00 to limit the influence of circadian factors. The temperature within the olfactometer was 27 ± 0.72 °C with 60.8 ± 4.0 relative humidity (RH), as determined by a thermometer-hygrometer (Tacklife HM01).

Olfactometer Validation
To validate the olfactometer (Dataset S1), a known mosquito attractant, human foot odour [17], was used. A polyamide sock (Ciak 15 Sanpellegrino, CSP International Fashion Group SpA, Ceresara, Italy) was worn by one of the male authors (LMG) for three days and nights to accumulate foot odour [18,19]. The worn sock (treatment) was placed in a plexiglass cylinder and closed at each end with net fabric. An unworn sock (control) was placed in an identical cylinder. The two cylinders were positioned in the distal part of the arms of the olfactometer. The entire olfactometer was covered by a white cotton sheet to avoid visual stimulation of the mosquitoes during the test. The olfactometer was illuminated from above the sheet by a 60 watt halogen light bulb (OSRAM GmbH, Munich, Germany). The illuminance just above the olfactometer was around 2000 lux. Twenty Ae. albopictus individuals (males or females) were released into the main chamber of the olfactometer by inserting the release chamber and rotating it 180 degrees, thus allowing the plexiglass lid to swing open, permitting the mosquitoes to enter the main chamber.
The end caps of the two arms were then fixed in position and the airflow commenced, and maintained, at a constant rate of 3 L/min. After 10 min had elapsed, the number of individuals in the traps of the two arms were counted. The cylinders containing the socks were removed and all of the mosquitoes were removed and eliminated after each run. Clean air was allowed to flow through the olfactometer for 15 min between replicates. The position of the worn and control socks in each arm was alternated for each replicate. Five replicates were performed for each sex. Given that the number of individuals trapped in each arm may not have a normal distribution, and that they could be considered proportions, each number was arcsine transformed [20] and compared using a paired, two tailed t test. This assay was always conducted between 15:00 and 19:00 to limit the influence of circadian factors. The temperature within the olfactometer was 27 ± 0.72 • C with 60.8 ± 4.0 relative humidity (RH), as determined by a thermometer-hygrometer (Tacklife HM01).
A 400 µL aliquot of the hexane dilution of the test compound, or 400 µL of hexane (control), was spotted on a 20 mm × 35 mm Whatman 3MM filter paper rectangle supported by a 24 mm× 40 mm glass microscope coverslip. The hexane was allowed to evaporate for 2 min (under a benchtop extractor arm) and then the control and treatment filters/coverslips were positioned in each arm of the olfactometer on an inverted 2 cm thick solid watch glass ( Figure 1). The olfactometer was covered by a white cotton sheet and illuminated from above as described in the previous section. Twenty Ae. albopictus males were released into the main chamber of the olfactometer, and after 10 min had elapsed, the number of individuals in the trap from each arm was counted. The filter papers with supporting microscope coverslips, and all the mosquitoes, were removed and eliminated after each run. Clean air was allowed to flow through the olfactometer for 15 min between replicates. Compounds were tested with at least six replicate runs for each concentration. The position of the test stimuli and control in each arm was alternated for each replicate. Each compound was initially tested using the low concentration followed by the intermediate and high concentrations. After each set of runs, the apparatus was thoroughly wiped with 70% ethanol using paper towels and air was pumped through the apparatus for 24 h to aid the evaporation and elimination of any contaminating compound. This assay was conducted between 15:00 and 19:00.
We described the effects of the compounds on the mosquitoes in terms of attraction and repulsion: attraction, when a significantly higher number of individuals accumulated in the trap from the arm with the test compound; repulsion, when a significantly higher number accumulated in the control arm. Some compounds might, of course, elicit alternative responses, for example, as an excitant or an arrestant, without attracting or repelling the mosquitoes. We were not able to assess these responses.

Olfactometer Validation
Female mosquitoes showed a significant preference for the arm of the olfactometer containing the foot odour/worn sock (two tailed paired t test, p = 0.010). Male mosquitoes did not show any preference and were not attracted to either the foot odour/worn sock or the control sock ( Figure 2).  [21] and TGSC Information System [22].

Olfactometer Validation
Female mosquitoes showed a significant preference for the arm of the olfactometer containing the foot odour/worn sock (two tailed paired t test, p = 0.010). Male mosquitoes did not show any preference and were not attracted to either the foot odour/worn sock or the control sock ( Figure 2).

Response of Male Mosquitoes to Chemical Compounds
Seven of the ten ketones tested showed repellency at one or more concentrations (Figure 3b-e,g-i). For 2-nonanone, acetophenone, 6-methyl-5-hepten-2-one, 2,6,6-trimethyl-2-cyclohexanene-1,4-dione, 3,7-dimethyl-oct-6-en-1-yn-3-ol and 4 -ethyl-acetophenone, there was significant repellency at only the high concentration (10 −2 ), whereas 4-propylbenzaldehyde showed repellency at both the intermediate and high concentrations, with greater repellency at the high concentration. For the majority of the assayed ketones, there was an evident, though not always significant, correspondence between the responses of the males and the concentration of the compound. For 4-propyl-benzaldehyde (Figure 3d), there appeared to be a trend of decreasing repellency and increasing attraction for each   Two of the five terpenes tested showed repellency at the 10 −2 concentration (geraniol, p = 0.034; linalool oxide, p = 0.030; Figure 3m,o). The other terpenes did not produce a significant response from the males at any concentration.
Two of the three aldehydes tested showed repellency. Nonanal showed significant repellency only at the high concentration (Figure 3w, p = 0.025), whereas phenylacetaldehyde showed significant repellency only at the low concentration (Figure 3x, p = 0.039).
The only acid tested, decanoic acid, had no significant effect at the low concentration (10 −6 ), but at the intermediate concentration (10 −4 ), there was significant attraction of males (p = 0.025). At the high concentration, the compound acted as a repellent (10 −2 , p = 0.027) (Figure 3ab). Given this result, the same serial dilutions of decanoic acid were assayed using female mosquitoes. No significant response to any dilution was recorded (Figure 3ac, Dataset S2).

Discussion
We tested 28 compounds belonging to six chemical classes originating from plants, fungi, or animals, and we provide evidence that 13 acted as repellents at one or more concentrations, usually the highest concentration. Fourteen compounds elicited no significant response, while one compound, decanoic acid, acted as an attractant for males at the intermediate concentration, and as a repellent at the high concentration. Female Ae. albopictus showed no significant response to decanoic acid. We discuss our results in relation to those of previous studies, considering for the most part only the mosquito species. This is not a simple task due to a high number of variables, such as different treatment concentrations, test assay conditions, and species and sex of the mosquitoes tested, to mention a few. We have divided the discussion into sections based on the origin of each volatile.

Animal-Related Volatiles
Decanoic acid is a fatty acid that is a surface component of human skin [23] and is present in many plants. It is also a component of mosquito cuticular and internal lipids [24]. In our assay, decanoic acid was found to be attractive at the intermediate concentration and repellent at the high concentration for male Ae. albopictus. No effect was observed against females. Decanoic acid acts as a biting deterrent for Ae. aegypti with an efficacy similar to that of N,N-diethyl-m toluamide (DEET), but decanoic acid remains effective for a longer period of time [25].
Decanoic acid-treated pools initially became repellent and then subsequently highly attractive for ovipositing females of Cx. restuans [26]. The attraction of decanoic acid-treated pools for ovipositing mosquitoes was confirmed in another study (this time for Cx. pipiens molestus and Ae. aegypti), apparently as a consequence of the attractiveness of the bacterial breakdown products from the decanoic acid substrate [27].
Nonanal, a major odour component of birds and human skin, was repellent at the highest concentration in our tests. At low concentrations, it acts as an attractant for hostseeking mosquitoes (Ae. aegypti, An. gambiae and Cx. quinquefasciatus), while in gravid females it acts as a cue for a suitable ovipositioning site (An. arabensis, Cx. quinquefasciatus, Cx. tarsalis). The same chemical also contributes to the recognition of plant hosts for Ae. aegypti [2,10,[28][29][30]. The Plasmodium parasite induces an increase in the production of certain volatiles by infected individuals, including the aldehydes heptanal, octanal and nonanal. The emission of these volatiles makes humans infected with malaria more attractive to Anopheles vectors, resulting in greater transmission of the parasite [31,32].
The alcohol, 1-octen-3-ol, a component of cattle breath and human sweat [33,34], was repellent for Ae. albopictus males at the highest concentration we tested. As a lure in traps, it Insects 2022, 13, 290 9 of 13 was attractive for females of An. gambiae, Ae. aegypti and Ae. albopictus, but failed to attract Cx. quinquefasciatus [35,36] and may be a repellent for the latter species [37].
Finally, 2-nonanone and 6-methyl-5-hepten-2-one (sulcatone) are components of human odour that have been shown to be attractive for Ae. albopictus females [36]. In our assays, at the highest concentration, sulcatone acted as a repellent. Indeed, sulcatone appears to act as a "masking", or even a repellent, odorant for host-seeking Ae. aegypti [30,38]. The emission of 6-methyl-5-hepten-2-one from overcrowded or pre-occupied larval sites appears to act as an ovipositioning deterrent for gravid An. coluzzii females to reduce the risk of intraspecific competition and cannibalism [39].

Plant-Related Volatiles
Mosquitoes visit flowers for nectar and may, in turn, act as pollinators for the plants [2,40,41]. The interaction between the mosquito and the plant is determined by the composition of the inflorescence odour, which may contain a mixture of attractive and repellant compounds [2]. One such case is provided by the Spanish catchfly, Silene otites, whose inflorescences emit a strong odour at night that is attractive for Culex pipiens molestus [41]. Individual components of the S. otites odour were demonstrated to elicit antennal responses in both male and female Cx. pipiens and Ae. aegypti [6]. The floral emissions of S. otites includes six of the compounds that we tested against male Ae. albopictus, namely: 2-phenyl ethanol, phenylacetaldehyde, (Z)-3-hexenyl acetate, linalool oxide, acetophenone and methyl salicylate. These compounds were not attractive in our behavioural test; moreover, three of them-phenylacetaldehyde, linalool oxide and acetophenone-were repellent. It is noteworthy that 2-phenyl ethanol, phenylacetaldehyde, linalool oxide and acetophenone were attractive in two choice bioassays for Cx. p. molestus females (males were not tested but were equally attracted by the odour of Silene otites inflorescences) [6].
A given chemical cue induces behavioural responses by activating some receptors and inhibiting, or having no effect, on others, and this may differ between species [42]. In addition, the concentration of a compound, the assay type and scale of the assay can influence the response. For example, the floral components phenylacetaldehyde and acetophenone, which had previously been shown to be attractive for Ae. aegypti in small scale experiments, were not attractive in a larger setting [43]. Moreover, Von Oppen and colleagues [44] found no response from Ae. aegypti to 2-phenyl ethanol (at a concentration equivalent to our 10 −4 dilution) in a Y-tube olfactometer. However, when undiluted 2-phenyl ethanol is applied directly to the skin or clothing, it becomes an extremely effective repellent against female Ae. aegypti [45]. Compounds may act as attractants for insects at low concentrations but as repellents at higher concentrations [46]. This is what we observed for decanoic acid.
Methyl cinnamate and methyl salicylate have been identified as the active components of Ocimum forskolei, a plant traditionally used in Eritrea as a repellent against mosquitoes, black flies and ticks [47][48][49]. At a concentration of 10 −3 in hexane, both methyl cinnamate and methyl salicylate significantly reduced landing of Ae. aegypti females on human skin odour baits. Ocimum forskolei, and other plants in the genus, have been shown to be repellent for other mosquito species such as An. arabiensis and An. stephensi, suggesting a similar modality of repulsion across mosquito taxa [47]. In our tests, methyl cinnamate was repellent at the lowest concentration whereas methyl salicylate elicited no response at any concentration.
The malarial vector, An. Gambiae, has been shown to respond to three of the compounds tested here: R-(+)-limonene, linalool oxide and (E)-β-farnesene [50]. These were among the components isolated from three plants favoured by nectar-feeding An. Gambiae: Santa Maria feverfew, Parthenium hysterophorus; the castor oil plant Ricinus communis; Cobbler's pegs, Bidens Pilosa. We did not record attraction of male Ae. albopictus to R-(+)limonene or farnesene, although we used a mixture of isomers rather than (E)-β-farnesene, and all of our concentrations were higher than the concentrations used against An. gambiae. The potential of linalool oxide as a single-component plant-based lure has been investi-gated for trapping Aedes species in unlit bait traps at field sites in Kenya [51]. Linalool oxide-baited traps performed comparably with commercial BioGent (BG) Lure-baited traps for trapping female Ae. aegypti, but significantly more males were collected in the linalool oxide traps. When CO 2 was added, linalool oxide was significantly better than the BG Lure with a 2.8-fold increase in male Ae. aegypti captures [51]. In our tests, however, male Ae. albopictus were repelled by linalool oxide at our highest concentration.
Hao and colleagues [13] tested several plant volatiles, including geraniol, eugenol and anisaldehyde against Ae. albopictus females using a two-port olfactometer. Geraniol and eugenol did not induce a significant response, regardless of concentration, while anisaldehyde elicited significant attraction at a 6% concentration. In our assays, we found no response of male Ae. albopictus to eugenol or anisaldehyde. Geraniol did, however, act as a repellent for males at our highest concentration. Geraniol was also shown to be highly repellent for An. gambiae females when tested at concentrations between 10 −2 and 10 −5 g/mL [52]. Afify and colleagues employed a close proximity assay to demonstrate that 60% eugenol was a repellent for Ae. aegypti and Cx. quinquefasciatus, but not An. coluzzii. They argued that the response of mosquitoes to different repellents, and presumably attractants, is species-specific [4]. We would add that each sex of a mosquito species might also respond differently to odour stimuli, as observed, for example, with foot odour.
We found that geranyl acetate elicited no response at any concentration, however, repellent activity assays with female Ae. aegypti found that higher concentrations (10 and 25%) offered 97-100% protection for more than 60 min [53].
The compound, 3,7-dimethyl-oct-6-en-1-yn-3-ol (dehydrolinalool), which is present in the floral odours of many plant species, elicited no response from male Ae. albopictus, except at the highest concentration, when it acted as a repellent. For Ae. aegypti, however, it resulted in activation and/or orientation towards the chemical source [54].

Conclusions
Some considerations emerge from these preliminary data. Firstly, compounds that are emitted by plants and animals may elicit different responses depending on its concentration and the species and sex of the mosquito. Moreover, the effect of a single compound may be enhanced or diminished, depending on the other compounds that are present. Further investigations should evaluate different blends of these compounds, as many studies have shown that blends are more likely to elicit a response than individual volatiles [55,56]. Wide-scale tests should be performed with individual candidate compounds, and mixtures of compounds, in the environment where the mosquito is found and where trapping is planned, testing different concentrations, formulations and methods of emission. These wide-scale tests could be performed in association with an evironmental evaluation of the presence of plants, animals and fungi to help explain the absence of mosquitoes and frequent negative collections at certain trapping points.
One of the compounds tested in this study, decanoic acid, appears to be a promising candidate for male trapping. Further tests in the field, perhaps in combination with other compounds or sound stimuli [57], will be necessary to identify the optimal dose for its potential use in monitoring during SIT and population-modification programmes aimed at Ae. albopictus populations.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/insects13030290/s1, Dataset S1: Data for olfactometer validation using human foot odour, Dataset S2: Data for response of male mosquitoes to chemical compounds (and female response to decanoic acid).