Altitudinal population structure and microevolution of the malaria vector Anopheles cruzii (Diptera: Culicidae)

BACKGROUND: In Brazil, the autochthonous transmission of extra-Amazonian malaria occurs mainly in areas of the southeastern coastal Atlantic Forest, where Anopheles cruzii is the primary vector. In these locations, the population density of the mosquito varies with altitude (5–263 m above sea level), prompting us to hypothesise that gene flow is also unevenly distributed. Describing the micro-geographical and temporal biological variability of this species may be a key to understanding the dispersion of malaria in the region. We explored the homogeneity of the An. cruzii population across its altitudinal range of distribution using wing shape and mtDNA gene analysis. We also assessed the stability of wing geometry over time. METHODS: Larvae were sampled from lowland (5–20 m) and hilltop (81–263 m) areas in a primary Atlantic Forest region, in the municipality of Cananéia (State of São Paulo, Brazil). The right wings of males and females were analysed by standard geometric morphometrics. Eighteen landmarks were digitised for each individual and a discriminant analysis was used to compare samples from the hilltop and lowland. A 400-bp DNA fragment of the mitochondrial cytochrome oxidase gene subunit I (CO-I) was PCR-amplified and sequenced. RESULTS: Wing shapes were distinct between lowland and hilltop population samples. Results of cross-validated tests based on Mahalanobis distances showed that the individuals from both micro-environments were correctly reclassified in a range of 54–96%. The wings of hilltop individuals were larger. The CO-I gene was highly polymorphic (haplotypic diversity = 0.98) and altitudinally structured (Фst = 0.085 and Jaccard = 0.033). We found 60 different haplotypes but only two were shared by the lowland and hilltop populations. Wing shape changed over the brief study period (2009–2013). CONCLUSIONS: Wing geometry and CO-I gene analysis indicated that An. cruzii is vertically structured. Wing shape varied rapidly, but altitude structure was maintained. Future investigations should identify the biotic/abiotic causes of these patterns and their implications in the local epidemiology of malaria. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13071-014-0581-8) contains supplementary material, which is available to authorized users.