Mitochondrial DNA variation in the malaria vector Anopheles minimus across China, Thailand and Vietnam: evolutionary hypothesis, population structure and population history

Abstract

The effects of Pleistocene environmental fluctuations on the distribution and diversity of organisms in Southeast Asia are much less well known than in Europe and North America. In these regions, the combination of palaeoenvironmental reconstruction and inferences about population history from genetic data has been very powerful. In Southeast Asia, mosquitoes are good candidates for the genetic approach, with the added benefit that understanding the relative contributions of historical and current processes to population structure can inform management of vector species. Genetic variation among populations of Anopheles minimus was examined using 144 mtDNA COII sequences from 23 sites in China, Thailand and Vietnam. Haplotype diversity was high, with two distinct lineages that have a sequence divergence of over 2% and exhibit different geographical distributions. We compare alternative hypotheses concerning the origin of this pattern. The observed data deviate from the expectations based on a single-panmictic population with or without growth, or a stable but spatially structured population. However, they can be readily accommodated by a model of past fragmentation into eastern and western refugia, followed by growth and range expansion. This is consistent with the palaeoenvironmental reconstructions currently available for the region.

Figure 1

Collecting localities for populations of Anopheles minimus in China, Vietnam and Thailand. The 23 populations were divided into nine population groups (circled) based on their geographical distributions.

Figure 2

Two separate haplotype networks for Anopheles minimus. The size of a circle corresponds to the haplotype frequency, and a unit branch represents one mutation. Small ovals indicate unobserved, inferred haplotypes.

Figure 3

Observed mismatch distributions among haplotypes in Lineages A+B (squares) with the results of SIMCOAL runs using the parameters inferred by FLUCTUATE for a single, exponentially expanding population (means, diamonds and 95% confidence intervals from 1000 simulations).

Figure 4

Comparisons between the observed D and F_s estimates (open circle) with those obtained from 1000 SIMCOAL simulated data sets (filled diamonds) under each of three scenarios: an exponentially expanding population (a), a population that has experienced a stepwise expansion (b) and a structured population of constant size (c). See text for further explanation.