The biting midge Culicoides imicola Kieffer (Diptera: Ceratopogonidae) is the most important Old World vector of African horse sickness (AHS) and bluetongue (BT). Recent increases of BT incidence in the Mediterranean basin are attributed to its increased abundance and distribution. The phylogenetic status and genetic structure of C. imicola in this region are unknown, despite the importance of these aspects for BT epidemiology in the North American BT vector. In this study, analyses of partial mitochondrial cytochrome oxidase subunit I gene (COI) sequences were used to infer phylogenetic relationships among 50 C. imicola from Portugal, Rhodes, Israel, and South Africa and four other species of the Imicola Complex from southern Africa, and to estimate levels of matrilineal subdivision in C. imicola between Portugal and Israel. Eleven haplotypes were detected in C. imicola, and these formed one well-supported clade in maximum likelihood and Bayesian trees implying that the C. imicola samples comprise one phylogenetic species. Molecular variance was distributed mainly between Portugal and Israel, with no haplotypes shared between these countries, suggesting that female-mediated gene flow at this scale has been either limited or non-existent. Our results provide phylogenetic evidence that C. imicola in the study areas are potentially competent AHS and BT vectors. The geographical structure of the C. imicola COI haplotypes was concordant with that of BT virus serotypes in recent BT outbreaks in the Mediterranean basin, suggesting that population subdivision in its vector can impose spatial constraints on BT virus transmission.

No abstract

Collections of biting midges were made over 24 months from sixty sites spread across Iberia. Information on the distribution of the vector of African horse sickness virus, Culicoides imicola, from these 3119 samples showed that this species was annually present across south-western Spain as far as 3 degrees 53'W and throughout most of Portugal, up to 41 degrees 5'N. C. imicola was found in all areas where African horse sickness epizootics had occurred in 1987-90 and also in areas outside the epizootic zones. Seasonal patterns of capture success of C. imicola, from seventeen frequently sampled sites where the vector was present, usually showed a late summer-early autumn peak. At the sites furthest south there was a discrete peak, mostly in September or October, before and after which the numbers captured increased or decreased steadily. At higher latitudes peak abundances occurred as early as May or as late as November, population build up was less uniform and numbers often declined rapidly after the peak was reached. Both the distribution and seasonal abundance patterns closely matched transmission patterns of African horse sickness virus, which rose during late summer and caused most cases during the autumn months.

No abstract