Evaluating synthetic odours and trap designs for monitoring Anopheles farauti in Queensland, Australia

Vector surveillance is an important component of the endeavour to eliminate malaria. An extensive and reliable vector surveillance network requires efficient, standardized methods for sampling adult vector populations, for monitoring of behavioural and physiological resistance to insecticides, to characterize biting behaviour, and to define the receptivity of different regions to local transmission of malaria.

Malaria control has recently stalled, with increases in malaria seen globally, including the Southwest Pacific [1]. There is a need for more effective monitoring of malaria vectors for changes in biting habits that can threaten the effectiveness of vector control strategies. The main vector species in the region belong to the Anopheles punctulatus complex, including Anopheles farauti sensu stricto (s.s.), which is the dominant vector in coastal areas from eastern Indonesia to Vanuatu [2, 3]. In the 1960s and 1970s, in response to indoor residual insecticide spraying (IRS) with DDT, An. farauti shifted its biting behaviour from all-night biting both indoors and outdoors to earlier evening and more outdoor biting [4‐6]. Despite these behavioural adaptations, indoor interventions such as long-lasting insecticidal bed nets (LLINs) and IRS contribute significantly to malaria control in the Southwest Pacific because An. farauti is a species that typically feeds frequently and will enter houses late at night to blood feed [7] and will be likely to encounter LLINs or IRS.

Monitoring vector populations to determine endophagy, peak biting time, seasonality, and sporozoite infections rates requires a standardized, effective collection method. In the Southwest Pacific, the human landing catch (HLC) is still the ‘gold standard’, the most efficient (and often the only) surveillance method for collecting blood-seeking An. farauti sensu lato (s.l.). During HLCs, mosquitoes landing on a trained collector are captured as they begin to probe. The HLC targets anthropophagic anophelines, but is expensive, requiring collectors to work under close supervision with a risk of exposure to infectious mosquito bites [8]. Although mosquito collectors have been shown to not be at an increased risk of malaria (when provided with prophylaxis) [9], there is a possibility of exposure to arboviruses by biting culicines. To overcome these disadvantages, many mechanical trapping methods have been developed to replace the HLC. The Centers for Disease Control (CDC) Light Trap attracts mosquitoes with either an incandescent or ultraviolet light bulb and CO₂ [10]. Where a source of CO₂ is not available, the CDC light trap can be placed close to a (human) host, protected by a bed net, who provides CO₂ and odorant lures [10]. Traps with an odour lure and CO₂ that mimic a human host have been developed with an ultimate aim of replacing the HLC with a safer, cheaper, more standardized approach [11‐14].

In these odour-baited traps, volatiles from the odour lure within the trap are emitted and spread by a constant airflow generated by a fan, while attracted mosquitoes are sucked into the trap. This counter-current concept has been applied in several traps, including the Mosquito Magnet traps (American Biophysics Corporation, USA) and the BG-Sentinel trap (Biogents AG, Germany). The inclusion of a synthetic odour lure would remove the need to place a trap close to a human host, thus enabling the trap to be used outdoors as well as indoors.

This first part of this study was conducted to evaluate the attraction of An. farauti to two synthetic lures, previously shown to be effective in attracting anthropophagic African anophelines, and a natural human odour sample. The most effective lure from the first part of the study was subsequently used to bait three different fan-powered traps and a passive trap to determine which trap was most effective in capturing An. farauti in tropical North Queensland, Australia.

All field experiments were conducted in a rural area 15 km north of Cairns, Queensland, Australia, in June and July 2017 (16.8221°S, 145.6972°E). The site is situated in a swamp forest dominated by paperback (Melaleuca) trees, with human activity (suburbs, industrial areas, sugarcane farms) within a 1-km radius [15]. The climate is tropical, with hot, humid summers (November–March) and milder, dry winters (April–October). The min/max temperatures range from 23 °C/31 °C in summer to 18 °C/26 °C in winter, with an annual average rainfall of approximately 2000 mm. The prevailing wind direction in summer is north-eastern, dictated by the monsoon, and is dominated by the south-eastern trade winds during winter (data obtained from the Bureau of Meteorology, Australia).

Anopheles farauti s.l. were captured using three different, commercially available, fan-powered traps baited with CO₂ and an odour lure. Although all three odour lures attracted female mosquitoes, CO₂ combined with the BG-Lure or a natural human odour, or CO₂ alone attracted three times more females than the MB5 lure combined with CO₂. The MB5 odour lure is a blend, which was developed to attract Anopheles females [19]. However, the MB5 blend was designed for and tested on African Anopheles species, specifically Anopheles gambiae s.s. and Anopheles coluzzii [20]. These species are both highly anthropophilic [21], whereas tests on Anopheles arabiensis, which shows opportunistic feeding behaviour, resulted in much more variation in catch numbers when the MB5 lure was used [20]. Similar to An. arabiensis, An. farauti displays opportunistic feeding behaviour [22]. This may explain the lower attractiveness of the MB5 blend in this experiment.

Anopheles farauti opportunistic blood feeding may have evolved because potential hosts vary greatly by location and relative abundance, hence females attracted by odorous compounds that are produced by many mammalian blood host species are more likely to be successful in finding a blood meal. These findings show that traps baited with CO₂ and a natural host odour, CO₂ and the synthetic BG-Lure or CO₂ alone all yielded high catch sizes. However, it is unlikely that host-seeking females follow CO₂ alone when searching for a blood meal.

All of the fan-powered traps tested (BG-Sentinel, BG-Suna and BG-Bowl) captured high numbers of An. farauti females over the course of the experiment. The traps were all equipped with a BG-Lure cartridge and CO₂, as this combination yielded the highest catch size in the first experiments. The results suggest that fan-powered trapping of Australasian malaria mosquitoes could have the potential to replace human landing catches, but studies to compare traps with HLCs are required to validate this conclusion.

Catch sizes in the fan-powered traps were significantly higher than numbers captured by the Passive Box trap and it was noted that samples from the BG-Sentinel 2 and BG-Suna trap were in better condition than samples from the BG-Bowl trap. To enable easy and accurate identification in the field, sampled mosquitoes need to remain relatively undamaged. A well-known issue for fan-powered traps is that they can seriously damage mosquitoes during collections [23]. In the BG-Sentinel 2 trap, this is prevented by positioning of the mesh catch bag directly behind the inlet funnel, so mosquitoes are captured before they pass through the fan (see Fig. 1b). Mosquitoes pass through the fan in both the BG-Suna and the BG-Bowl trap (see Fig. 1c, d), making morphological identification more challenging.

The BG-Suna trap is a weather-resistant trap hanging from a rope, whereas the BG-Bowl trap is essentially a simplified version of the BG-Sentinel 2 trap, with a perforated cover which makes these traps less rain-resistant than the BG-Suna trap. In addition, the BG-Bowl trap does not possess a catch bag and mosquitoes are collected in the main body of the trap where water can accumulate and cause morphological damage. Indeed, it was observed that the condition of mosquito samples from the BG-Bowl trap was better after a sampling night without precipitation, whereas this difference was not observed in the BG-Suna trap.

Despite their ready availability and advantages, fan-powered traps are used rarely in the Southwest Pacific region. Limited usage of traps may be a function of trap cost, as well as requirements for a power source (usually a battery that needs regular replacement) and CO₂. As the results of this study have shown, a source of CO₂ is an essential component in a trap designed to lure host-seeking mosquitoes in the Southwest Pacific, as has been shown previously for An. coluzzii [24]. Traditional sources of CO₂, such as compressed gas or dry ice, are expensive, cumbersome and not readily available in places such as the Solomon Islands where tools for monitoring malaria vectors are urgently needed. CO₂ produced by yeast fermentation of sugar or molasses has been an effective solution in Kenya where traps are used to monitor malaria vectors in the field where dry ice and compressed gas are difficult to use [25].

All three of the fan-powered traps tested in this study attracted large numbers of An. farauti as well as other mosquito species, including Aedes vigilax, Verrallina carmenti and Verrallina funerea. These three species are important vectors of Australian vector-borne diseases such as Ross River Virus [26]. The high sampling yield of fan-powered traps shows their potential for vector monitoring in the Southwest Pacific. Targeting multiple key vector species using one trap could make future surveillance studies much more efficient, since it enables accurate, simultaneous population monitoring of a range of important disease vectors. Future studies should be conducted to compare capture rates of traps against HLCs for An. farauti in the field, in areas with both low and high background densities of An. farauti. Comparisons between traps and HLCs would allow confirmation of whether traps could replace traditionally used HLCs.

The use of fan-powered traps such as the BG-Suna trap or BG-Sentinel 2 trap, baited with CO₂ and an appropriate odour lure, can be a promising solution to increase vector monitoring in the Southwest Pacific. BG-Suna traps equipped with CO₂ alone, CO₂ plus a BG-Lure cartridge, or CO₂ plus a natural host odour perform equally well in the field with respect to An. farauti catch sizes in the field.