Evaluation of honey-baited FTA cards in combination with different mosquito traps in an area of low arbovirus prevalence

Abstract

Background: The threat of mosquito-borne diseases is increasing in continental Europe as demonstrated by several autochthonous chikungunya, dengue and West Nile virus outbreaks. In Switzerland, despite the presence of competent vectors, routine surveillance of arboviruses in mosquitoes is not being carried out, mainly due to the high costs associated with the need of a constant cold chain and laborious processing of thousands of mosquitoes. An alternative approach is using honey-baited nucleic acid preserving cards (FTA cards) to collect mosquito saliva that may be analysed for arboviruses. Here, we evaluate whether FTA cards could be used to detect potentially emerging viruses in an area of low virus prevalence in combination with an effective mosquito trap.

Methods: In a field trial in southern Switzerland we measured side-by-side the efficacy of the BG-Sentinel 2, the BG-GAT and the Box gravid trap to catch Aedes and Culex mosquitoes in combination with honey-baited FTA cards during 80 trapping sessions of 48 hours. We then screened both the mosquitoes and the FTA cards for the presence of arboviruses using reverse-transcription PCR. The efficacy of the compared trap types was evaluated using generalized linear mixed models.

Results: The Box gravid trap collected over 11 times more mosquitoes than the BG-GAT and BG-Sentinel 2 trap. On average 75.9% of the specimens fed on the honey-bait with no significant difference in feeding rates between the three trap types. From the total of 1401 collected mosquitoes, we screened 507 Aedes and 500 Culex females for the presence of arboviruses. A pool of six Cx. pipiens/Cx. torrentium mosquitoes and also the FTA card from the same Box gravid trap were positive for Usutu virus. Remarkably, only two of the six Culex mosquitoes fed on the honey-bait, emphasising the high sensitivity of the method. In addition, two Ae. albopictus collections but no FTA cards were positive for mosquito-only flaviviruses.

Conclusions: Based on our results we conclude that honey-baited FTA cards, in combination with the Box gravid trap, are an effective method for arbovirus surveillance in areas of low prevalence, particularly where resources are limited for preservation and screening of individual mosquitoes.

Keywords: Arbovirus surveillance; Culicidae; Disease control; Mosquito-only flaviviruses; Nucleic acid preservation cards; Usutu virus.

Fig. 1

Mosquito traps used in combination with honey-baited FTA cards. The honey-baited FTA cards are indicated by blue arrows. a BG-Sentinel 2 trap baited with BG-Lure^® and CO₂, image below shows the modified collection bag with the attached FTA card. b BG-GAT on the left wiped with canola oil without FTA card and BG-GAT on the right with an FTA card taped to the translucent chamber, shown below in the open state. c Box gravid trap, mosquitoes approaching the water surface are caught by the up-draft fan and sucked into the collection chamber where the FTA card was attached, shown below in open state

Fig. 2

Comparison of mosquito trap efficacy. The histograms show the frequency of recorded mosquito count per 48-hour trapping session (n = 80 for each trap type) with red bars indicating negative traps. The diamonds and horizontal bars below represent the average mosquito count per 48-hour trapping session with 95% confidence intervals as estimated with the GLMM

Fig. 3

Comparison of sugar-feeding success between trap types. Blue bars represent the cumulative number of mosquitoes that fed on the honey-baited FTA cards in each trap type. On average 75.9% (95% CI: 70.8–80.4%) of the captured females fed on honey-baited FTA cards. There was no significant difference in sugar-feeding rates between the three trap types (χ²= 3.2, df = 2, P = 0.198)

Fig. 4

Map of sampling area in the Canton of Ticino, southern Switzerland. Each filled symbol represents one of the 36 trap positions. Unfilled symbols indicate virus detections at the underlying trap position. Source of base map: Swiss Federal Office of Topography, swisstopo