Assessment of population genetic structure in the arbovirus vector midge, Culicoides brevitarsis (Diptera: Ceratopogonidae), using multi-locus DNA microsatellites

Abstract

Bluetongue virus (BTV) is a major pathogen of ruminants that is transmitted by biting midges (Culicoides spp.). Australian BTV serotypes have origins in Asia and are distributed across the continent into two distinct episystems, one in the north and another in the east. Culicoides brevitarsis is the major vector of BTV in Australia and is distributed across the entire geographic range of the virus. Here, we describe the isolation and use of DNA microsatellites and gauge their ability to determine population genetic connectivity of C. brevitarsis within Australia and with countries to the north. Eleven DNA microsatellite markers were isolated using a novel genomic enrichment method and identified as useful for genetic analyses of sampled populations in Australia, northern Papua New Guinea (PNG) and Timor-Leste. Significant (P < 0.05) population genetic subdivision was observed between all paired regions, though the highest levels of genetic sub-division involved pair-wise tests with PNG (PNG vs. Australia (FST = 0.120) and PNG vs. Timor-Leste (FST = 0.095)). Analysis of multi-locus allelic distributions using STRUCTURE identified a most probable two-cluster population model, which separated PNG specimens from a cluster containing specimens from Timor-Leste and Australia. The source of incursions of this species in Australia is more likely to be Timor-Leste than PNG. Future incursions of BTV positive C. brevitarsis into Australia may be genetically identified to their source populations using these microsatellite loci. The vector's panmictic genetic structure within Australia cannot explain the differential geographic distribution of BTV serotypes.

Figure 1

Sites of collections of C. brevitarsis and plot of the genetic structure in this study. A STRUCTURE plot results (K = 2) integrated into a map showing the locations of the study area. B The Q matrix derived from STRUCTURE clustering analysis show the inferred ancestry membership proportions of each individual in each cluster (K = 2). Each individual is represented by a single vertical line, partitioned into K colored segments that represent the individual’s estimated membership fraction in each of the K inferred clusters. The X-axis corresponds to the pre-defined populations (TL, PNG, QLD, NT and NSW) and the Y-axis represents the proportional estimates of the estimated membership in clusters, which add up to one.

Figure 2

The distribution pattern of primers across the different microsatellite repeats in Culicoides brevitarsis genome. Considerable variation was observed among the 12 validated microsatellite loci. Exact tests for linkage disequilibrium identified significant association between locus G7B17 and GO2AH. Subsequently locus G7B17, which was less polymorphic than GO2AH was excluded from downstream analysis. The number of alleles per locus ranged from 3–13 and 78 alleles were scored across the 11 loci.