Evaluation and optimization of membrane feeding compared to direct feeding as an assay for infectivity

Background: Malaria parasite infectivity to mosquitoes has been measured in a variety of ways and setting, includind direct feeds of and/or membrane feeding blood collected from randomly selected or gametocytemic volunteers. Anopheles gambiae s.l is the main vector responsible of Plasmodium falciparum transmission in Bancoumana and represents about 90% of the laboratory findings, whereas Plasmodium malariae and Plasmodium ovale together represent only 10%. Materials and methods: Between August 1996 and December 1998, direct and membrane feeding methods were compared for the infectivity of children and adolescent gametocyte carriers to anopheline mosquitoes in the village of Bancoumana in Mali. Gametocyte carriers were recruited twice a month through a screening of members of 30 families using Giemsa-stained thick blood smears. FI generation mosquitoes issued from individual female wild mosquitoes from Bancoumana were reared in a controlled insectary conditions and fed 5% sugar solution in the laboratory in Bamako, until the feeding day when they are starved 12 hours before the feeding experiment. These FI generation mosquitoes were divided in two groups, one group fed directly on gametocyte carriers and the other fed using membrane feeding method. Results: Results from 372 Plasmodium falciparum gametocyte carriers showed that children aged 4-9 years were more infectious than adolescents (p = 0.039), especially during the rainy season. Data from 35 carriers showed that mosquitoes which were used for direct feeding were about 1.5 times more likely to feed (p < 0.001) and two times more likely to become infected, if they fed (p < 0.001), than were those which were used for membrane feeding. Overall, infectivity was about three-times higher for direct feeding than for membrane feeding (p < 0.001). Conclusion: Although intensity of infectivity was lower for membrane feeding, it could be a surrogate to direct feeding for evaluating transmission-blocking activity of candidate malaria vaccines. An optimization of the method for future trials would involve using about three-times more mosquitoes than would be used for direct feeding.