Since its discovery in Djibouti in 2012, the Asian Anopheles stephensi mosquito has spread throughout the Horn of Africa. This invasive vector continues to spread across the continent, posing a serious threat to malaria control programmes. Vector control methods, including insecticide-treated bed nets and indoor residual spraying, have significantly reduced the malaria burden. However, the increasing prevalence of insecticide-resistant mosquitoes, including Anopheles stephensi populations, is hampering ongoing malaria elimination efforts. Understanding population structure, gene flow between populations, and the distribution of insecticide resistance mutations is essential to guide effective malaria control strategies.
Improving our understanding of how An. stephensi became so established in the HOA is critical to predicting its potential spread to new areas. Population genetics has been used extensively to study vector species to gain insight into population structure, ongoing selection, and gene flow18,19. For An. stephensi, studying population structure and genome structure can help elucidate its invasion route and any adaptive evolution that may have occurred since its emergence. In addition to gene flow, selection is particularly important because it can identify alleles associated with insecticide resistance and shed light on how these alleles are spreading through the population20.
To date, testing of insecticide resistance markers and population genetics in the invasive species Anopheles stephensi has been limited to a few candidate genes. The species’ emergence in Africa is not fully understood, but one hypothesis is that it was introduced by humans or livestock. Other theories include long-distance migration by wind. The Ethiopian isolates used in this study were collected in Awash Sebat Kilo, a town located 200 km east of Addis Ababa and on the main transport corridor from Addis Ababa to Djibouti. Awash Sebat Kilo is an area with high malaria transmission and has a large population of Anopheles stephensi, which is reported to be resistant to insecticides, making it an important site for studying the population genetics of Anopheles stephensi8.
The insecticide resistance mutation kdr L1014F was detected at low frequency in the Ethiopian population and was not detected in the Indian field samples. This kdr mutation confers resistance to pyrethroids and DDT and was previously detected in An. stephensi populations collected in India in 2016 and Afghanistan in 2018.31,32 Despite evidence of widespread pyrethroid resistance in both cities, the kdr L1014F mutation was not detected in the Mangalore and Bangalore populations analyzed here. The low proportion of Ethiopian isolates carrying this SNP that were heterozygous suggests that the mutation arose recently in this population. This is supported by a previous study in Awash that found no evidence of the kdr mutation in samples collected in the year prior to those analyzed here.18 We previously identified this kdr L1014F mutation at low frequency in a set of samples from the same region/year using an amplicon detection approach.28 Given the phenotypic resistance at the sampling sites, the low allele frequency of this resistance marker suggests that mechanisms other than target site modification are responsible for this observed phenotype.
A limitation of this study is the lack of phenotypic data on insecticide response. Further studies combining whole genome sequencing (WGS) or targeted amplicon sequencing in combination with susceptibility bioassays are needed to investigate the impact of these mutations on insecticide response. These novel missense SNPs that may be associated with resistance should be targeted for high-throughput molecular assays to support monitoring and facilitate functional work to understand and validate potential mechanisms associated with resistance phenotypes.
In summary, this study provides a deeper understanding of Anopheles mosquito population genetics across continents. Application of whole genome sequencing (WGS) analysis to larger cohorts of samples in different geographic regions will be key to understanding gene flow and identifying markers of insecticide resistance. This knowledge will enable public health authorities to make informed choices in vector surveillance and insecticide use.
We used two approaches to detect copy number variation in this dataset. First, we used a coverage-based approach that focused on identified CYP gene clusters in the genome (Supplementary Table S5). Sample coverage was averaged across collection locations and divided into four groups: Ethiopia, Indian fields, Indian colonies, and Pakistani colonies. Coverage for each group was normalized using kernel smoothing and then plotted according to the median genome coverage depth for that group.
Post time: Jun-23-2025