Widespread insecticide resistance in Aedes aegypti L. from New Mexico, U.S.A

Abstract

Background: Aedes aegypti mosquitoes are vectors of a variety of emerging viral pathogens, including yellow fever, dengue, chikungunya, and Zika virus. This species has established endemic populations in all cities across southern New Mexico sampled to date. Presently, control of Aedes-borne viruses relies on deployment of insecticides to suppress mosquito populations, but the evolution of insecticide resistance threatens the success of vector control programs. While insecticide resistance is quite common in Ae. aegypti field populations across much of the U.S., the resistance status of this species in populations from New Mexico has not previously been assessed.

Results: First, we collected information on pesticide use in cities in southern New Mexico and found that the most commonly used active ingredients were pyrethroids. The use of insecticides with the same mode-of-action over multiple years is likely to promote the evolution of resistance. To determine if there was evidence of resistance in some cities in southern New Mexico, we collected Ae. aegypti from the same cities and established laboratory strains to assess resistance to pyrethroid insecticides and, for a subset of populations, to organophosphate insecticides. F2 or F4 generation mosquitoes were assessed for insecticide resistance using bottle test bioassays. The majority of the populations from New Mexico that we analyzed were resistant to the pyrethroids permethrin and deltamethrin. A notable exception to this trend were mosquitoes from Alamogordo, a city that did not report using pyrethroid insecticides for vector control. We screened individuals from each population for known knock down resistance (kdr) mutations via PCR and found a strong association between the presences of the F1534C kdr mutation in the para gene of Ae. aegypti (homologue to F1534C in Musca domestica L.) and pyrethroid resistance.

Conclusion: High-level pyrethroid resistance is common in Ae. aegypti from New Mexico and geographic variation in such resistance is likely associated with variation in usage of pyrethroids for vector control. Resistance monitoring and management is recommended in light of the potential for arbovirus outbreaks in this state. Also, alternative approaches to mosquito control that do not involve insecticides should be explored.

Fig 1. Permethrin-resistance levels in Aedes aegypti from New Mexico.

Shown are mortality curves of a permethrin-susceptible control strain (Rock, dotted line), a resistant control strain (Puerto Rico, dashed line), susceptible mosquitoes in acetone-treated bottles (Rock, dotted/dashed line), and strains established from field collections (red line). Please note that the X-axis is not a linear time scale but has five minute intervals until 15 min and then switches to 15 min intervals. In all cases (A-H), the control (Acetone) mortality curves were significantly different (P<0.0001) from the curves for all other treatments and the curves plotted for the Puerto Rico strain and the Rock strain were also significantly different from each other (P < .0001). The curves plotted for the permethrin-exposed Rock strain were also significantly different from the other strains (P < .0001). A. The curve plotted for the Puerto Rico strain (a known resistant strain) is significantly different than that plotted for the Alamogordo strain (P < .0001). B. The curve plotted for the Puerto Rico strain is significantly different than that plotted for the Carlsbad strain (P < .0001). C. The curve plotted for the Puerto Rico strain is not significantly different than that plotted for the Carlsbad strain (P = 0.597). D. The curve plotted for the Puerto Rico strain is significantly different than that plotted for the Las Cruces strain (P < .0001). E. The curve plotted for the Puerto Rico strain is not significantly different than that plotted for the Lovington strain (P < .0001). F. The curve plotted for the Puerto Rico strain is not significantly different than that plotted for the Roswell strain (P = .206). G. The curve plotted for the Puerto Rico strain is not significantly different than that plotted for the Sunland strain (P < .0001). H. There was a significant difference between the curves plotted for the permethrin-exposed mixed strain (50% Rock, 50% Puerto Rico) and the curves plotted for the pure strains (P < .0001). The curve plotted for the Puerto Rico strain is not significantly different than that plotted for the mixed strains (P < .0001).

Fig 2. Insecticide resistance in Aedes aegypti from New Mexico.

The resistance status of individual populations was determined via bottle testing (see Fig 1 and Figs A-J in S1 File). Populations were classified as resistant when their mortality curves were significantly shifted towards longer survival compared to the mortality curves of the sensitive Rock strain. Resistant populations are marked with red squares, sensitive populations are marked with yellow triangles. The grey circles represent populations that have not yet been tested for this particular resistance. ALA–Alamogordo, CRL- Carlsbad, DEM-Deming, LC-Las Cruces, LOV-Lovington, ROS-Roswell, SOC-Socorro, SUN-Sunland Park. A. resistance map for permethrin, B. resistance map for deltamethrin, C. resistance map for etofenprox, D. resistance map for chlorpyrifos.

Fig 3. Genotype frequency map.

The pie charts represent the percentage of the three specific kdr F1534C genotypes in mosquitoes collected at different locations. Wildtype, susceptible mosquitoes are shown in blue. Mutant, resistant mosquitoes are shown in yellow. Heterozygotes are shown in green. The genotype frequencies of the laboratory strains, Rockefeller and Puerto Rico, are shown to the upper right of the map. The location of New Mexico within the United States is shown in the lower right of the figure.