Aedes aegypti (l.) (Diptera: culicidae), a potential vector of dengue viruses in South Africa: taxonomy, ecology and vector competence.

Abstract

The potential of Aedes aegypti to act as a vector of dengue (DEN) viruses was assessed in the tropical and subtropical regions of South Africa. The prevalence in bamboo pot ovitraps, utilization of artificial larval habitats and antihropophflism of Ae. aegypti was measured in fifteen populations from coastal Natal and five from the Transvaal. Pot indices generally ranged from 40-100 per cent (x±s;c= 60,3±9,8), with a mean abundance index of 0,43±0,15 (sx) mosquitoes/pot/day. Artificial container indices ranged from 11-83 per cent (X±si = 56,8±5,6). Biting rates ranged from 0,2-29,0 per manhour, in direct relationship to the level of urbanization. The feral Nduntupopulation was least anthropophilic, although pot and abundance indices were high. To investigate the presence of subspp. aegypti and formosus, the pale scales on tergite-1 were counted in ten or more siblings from each of 196 families, representing eighteen populations. At least 118 of these families were heterogeneous, each containing some siblings with no pale scales and others with pale scales on tergite-1, thus invalidating this character for distinguishing between the subspp. Isozyme electrophoresis 'did not provide diagnostic electromorphs for distingllishing between different populations. Allelomorph frequencies and the numberof pale scales on tergite-l and tergite-2 differed significantly in individual populations but not between anthropophilic and non-anthropophilic populations. As there was no correlation between tergal morphology, isozymes and

anthropophilism, the populations could not be resolved into the two subspp. Based on the only morphological character that appears to be reliable, viz. the blackness of the background scales, all of the populations are probably Ae. aegypti formosus. The vector competence of five populations for DEN-1 and DEN-2 viruses was tested in the laboratory after allowing mosquitoes to feed on an infective blood-virus mixture. Viral antigen was detected by indirect fluorescent antibody test. Head-squash infection rates (HSIRs)ranged from 11-54 per cent for DEN-land from 19- 46 per cent for DEN...2. Transmission rates (TRs) were determined by in vitro capillary method and ranged from 67-100 per cent for DEN-1 and from 11-86 pier cent for DEN-2.Meanvector competence indices (calculated from HSIRsand TRS) ware 0,13-0,41 for DEN-l and 0,,18-0,34 for DEN-2.It is concluded that, should DEN be reintroduced to South Africa via the shipping or tourist industries, Ae. aegypti would be an efficient urban vector. The Durban population is of particular epidemiological importance because it was highly anthropophilic and was the most vectorially competent of the South African populations.