Development and field evaluation of the sentinel mosquito arbovirus capture kit (SMACK)

Abstract

Background: Although sentinel animals are used successfully throughout the world to monitor arbovirus activity, ethical considerations and cross-reactions in serological assays highlight the importance of developing viable alternatives. Here we outline the development of a passive sentinel mosquito arbovirus capture kit (SMACK) that allows for the detection of arboviruses on honey-baited nucleic acid preservation cards (Flinders Technology Associates; FTA®) and has a similar trap efficacy as standard light traps in our trials.

Methods: The trap efficacy of the SMACK was assessed against Centers for Disease Control and Prevention (CDC) miniature light traps (standard and ultraviolet) and the Encephalitis Vector Survey (EVS) trap in a series of Latin square field trials conducted in North Queensland, Australia. The ability of the SMACK to serve as a sentinel arbovirus surveillance tool was assessed in comparison to Passive Box Traps (PBT) during the 2014 wet season in the Cairns, Australia region and individually in the remote Northern Peninsula Area (NPA) of Australia during the 2015 wet season.

Results: The SMACK caught comparable numbers of mosquitoes to both CDC light traps (mean capture ratio 0.86: 1) and consistently outperformed the EVS trap (mean capture ratio 2.28: 1) when CO2 was supplied by either a gas cylinder (500 ml/min) or dry ice (1 kg). During the 2014 arbovirus survey, the SMACK captured significantly (t 6 = 2.1, P = 0.04) more mosquitoes than the PBT, and 2 and 1 FTA® cards were positive for Ross River virus and Barmah Forest virus, respectively, while no arboviruses were detected from PBTs. Arbovirus activity was detected at all three surveillance sites during the NPA survey in 2015 and ca. 27 % of FTA® cards tested positive for either Murray Valley encephalitis virus (2 detections), West Nile virus (Kunjin subtype; 13 detections), or both viruses on two occasions.

Conclusions: These results demonstrate that the SMACK is a versatile, simple, and effective passive arbovirus surveillance tool that may also be used as a traditional overnight mosquito trap and has the potential to become a practical substitute for sentinel animal programs.

Fig. 1

Individual components of the SMACK and field settings with different CO₂ sources (10 kg cylinder and 1 kg dry ice). a The individual components of the SMACK: (a) 20 L clear plastic tub with lid, (b) 5 mm gas tubing with air stone, (c) 500 ml water reservoir and chamois sponge, (d) 70 ml FTA® card holders with exposed FTA® cards visible on top, (e) 18 cm sieve entry, (f) eye bolt from which the trap is hung, and (g) a fully assembled trap. b The SMACK set in the field with CO₂ supplied by a 10 kg cylinder with a timer-regulator setup (a) and when CO₂ supplied from 1 kg of dry ice released from an insulated (b) cooler

Fig. 2

Examples of recently ingested sugar meals and daily mosquito sugar-feeding rates on honey-baited FTA® cards. a Recent sugar meals easily observable in the abdomens of field-collected mosquitoes exposed to differently dyed honey-soaked FTA® cards. The figure depicts individual sugar-feedings on either blue (day 1) or yellow (day 2) dyed honey and the generation of a green colour if a second sugar-feeding occurred during day 2. b Daily sugar-feed rates (mean ± SE) of females on honey-baited FTA® cards when housed in SMACKs for a period of 3 d

Fig. 3

Mean number of mosquitoes collected per entry type per night and collection composition summarized by genera. a Mean (±SE) number of mosquitoes collected by each entry type (PVC spigot, 10 cm and18 cm diameter mesh sieve with a 5 cm diameter entry hole) per trap night and b the mean proportion of each nightly collection total comprised of mosquitoes species belonging to four different mosquito genera (Aedes, Anopheles, Culex, and Verrallina). Different letters indicate a significant (P-value < 0.05) difference (ANOVA, Tukey HSD post-hoc analysis)

Fig. 4

Mean nightly mosquito collections, trap composition, and non-target organisms collected per trap per night, a Nightly collection totals (mean ± SE) by trap type across the two CO₂ sources (gas cylinder at 500 ml/min and 1 kg dry ice). b Proportion of each collection total comprised of mosquitoes species belonging to four different mosquito genera (Aedes, Anopheles, Culex, and Verrallina). c Mean (±SE) of non-target insects collected per trap night for each trap type across the two CO₂ sources (gas cylinder at 500 ml/min and 1 kg dry ice). d Mean number of non-target insects collected per trap night for each trap averaged across both CO₂ treatments and summarized by order. Different letters indicate a significant (P-value < 0.05) difference (ANOVA, Tukey HSD post-hoc analysis)

Fig. 5

Daily survival of field-collected mosquitoes housed in a SMACK and unmodified PBT over 14 days. Daily survival probability (mean and 95 % confidence intervals) of field-collected mosquitoes housed in a SMACK and unmodified PBT for a period of 14 days