Comparative Characterization of the Microbiota of the Western Kenyan Malaria Vector, Anopheles Gambiae S.S, with Differing Insecticide Resistance Profiles

Abstract

Anopheles mosquitoes are known to transmit malaria, the major cause of morbidity and mortality in sub-Saharan Africa. The persistent reliance on insecticide-based vector control interventions has inevitably led to advancement of insecticide resistance. This resistance to insecticides has been reported in Anopheles gambiae complex from Western Kenya of which the mechanisms behind it have not been fully understood. Following recent reports of relationships between insecticide resistance and mosquito microbiota, this study focused on characterizing the microbial composition of Anopheles gambiae s.s. in relation to their insecticide resistance status and kdr mutation profiles in two locations (Kitabisi and Tulukuyi) within Bungoma County. F1 non-blood fed female A. gambiae s.s. obtained from wild-caught mosquitoes were exposed to 1X (21.5μg/ml) and 5X (107.5μg/ml) the discriminating dose of permethrin using CDC bottle bioassay. RT-PCR was then conducted to determine the kdr status. Microbiota of susceptible samples at 1X (n=14) and microbiota of both resistant (n=61) and susceptible (n=100) samples at 5X were characterized by high throughput sequencing which targeted the universal bacterial and archaeal 16S rRNA gene. Of all the mosquito samples from Kitabisi (n= 8) and Tulukuyi (n=20) at 1X, 50% were resistant in both sites while at 5X in Kitabisi (n= 90) and Tulukuyi (n= 133), 12% and 47% of the mosquitoes were resistant. The frequency of Vgsc_1014S mutation in Kitabisi (n = 71) and Tulukuyi (n = 116) were 98.59% and 99.14%, respectively. This indicated that microbial composition did not have an influence on kdr gene mutation or vice versa and, therefore, no further analyses for associations were conducted. Downstream analysis for the comparison of microbial composition was carried out on susceptible (n= 50) and resistant (n= 50) samples at 5X from Tulukuyi. Samples from Kitabisi and those exposed to 1X were excluded due to small and unequal sizes of the resistant and susceptible mosquitoes, however, taxonomic annotation of the microbiota was conducted on samples from both sites. The phylum of, Proteobacteria was the most abundant taxa in both sites. The genera Asaia (38.33%) and Comamonas (19.38%) were the most abundant taxa in Tulukuyi and Kitabisi respectively. The bacterial composition between resistant and susceptible individuals from Tulukuyi were significantly different (PERMANOVA, F=2.33, P=0.001), with Lysinibacillus, Sphingobacterium and Streptococcus comprising of known pyrethroid-degrading species. The radio-tolerant Rubrobacter was at least three folds more abundant in the resistant mosquitoes compared to the susceptible mosquitoes (P<0.05). This first report of association between the microbiota and pyrethroid resistance in A. gambiae s.s. corroborates results of previous studies conducted on Anopheles albimanus from Peru and Guatemala. High 1014S frequency along with resistance phenotype-driven differences in the microbiota suggests no relationship between that specific gene mutation and microbial composition, and posits that instead, other microbial mechanisms e.g., metabolic, may be involved in resistance. These findings will be used to guide future research into the function of mosquito microbiota in insecticide resistance development and the identification of novel microbial indicators of insecticide resistance in mosquito populations.