Abstract
Anopheles punctimacula s.l. is a regional malaria vector in parts of Central America, but its role in transmission is controversial due to its unresolved taxonomic status. Two cryptic species, An. malefactor and An. calderoni, have been previously confused with this taxon, and evidence for further genetic differentiation has been proposed. In the present study we collected and morphologically identified adult female mosquitoes of An. punctimacula s.l. from 10 localities across Panama and one in Costa Rica. DNA sequences from three molecular regions, the three prime end of the mitochondrial cytochrome c oxidase I gene (3' COI), the Barcode region in the five prime end of the COI (5' COI), and the rDNA second internal transcribed spacer (ITS2) were used to test the hypothesis of new molecular lineages within An. punctimacula s.l. Phylogenetic analyses using the 3' COI depicted six highly supported molecular lineages (A-F), none of which was An. malefactor. In contrast, phylogenetic inference with the 5' COI demonstrated paraphyly. Tree topologies based on the combined COI regions and ITS2 sequence data supported the same six lineages as the 3' COI alone. As a whole this evidence suggests that An. punctimacula s.l. comprises two geographically isolated lineages, but it is not clear whether these are true species. The phylogenetic structure of the An. punctimacula cluster as well as that of other unknown lineages (C type I vs C type II; D vs E) appears to be driven by geographic partition, because members of these assemblages did not overlap spatially. We report An. malefactor for the first time in Costa Rica, but our data do not support the presence of An. calderoni in Panama.