Optimization of formulation and delivery technology of entomopathogenic fungi for malaria vector control

Vector control is one of the most effective means of controlling mosquito-borne diseases such as malaria. The broad goal of this strategy is to protect individuals against infective mosquito bites and, at the community level, to reduce the intensity of disease transmission. With the deployment of mainly insecticide-treated nets (ITN) and indoor residual spraying (IRS), aided by effective drug treatment, certain countries particularly those within the low endemic zones have documented more than 50% reduction in malaria cases over the past decade. To keep up the pace and expand effective malaria control, in line with the global effort to eliminate malaria, IRS and ITN need to be complemented with alternative control methods. Indeed, neither long lasting insecticide nets (LLINs) nor IRS alone will be sufficient to achieve and maintain interruption of transmission in malaria holoendemic and hyperendemic areas. Besides, the sustainability of both methods is inescapably threatened by mosquito resistance to insecticides. Scientific evidence indicates that biological control based on entomopathogenic fungi has the potential to complement existing vector control methods. Two species of entomopathogenic fungi, Metarhizium anisopliae and Beauveria bassiana, have demonstrated ability to infect and kill adult malaria vectors. This thesis describes the results of a series of laboratory investigations followed by small scale field trials in Tanzania in an area of high malaria endemicity, with abundant populations of the malaria vector Anopheles gambiae sensu lato. The overall aim was to optimize fungal formulations, develop delivery techniques that maximize fungus infection rates in wild malaria populations, evaluate impact on survival of these mosquitoes and asses the impact on malaria transmission levels. A series of variables that we hypothesized affect the efficacy and persistence of the fungal isolates Metarhizium anisopliae ICIPE-30, M. anisopliae IP 46 and Beauveria bassiana I93-825 against adult An. gambiae were assessed. These included a) conidia concentration (1×107- 4×1010 conidia m-2), b) exposure time (15 min - 6 h), c) delivery substrates (netting, cotton cloth & mud wall), d) mosquito age (2 - 12 d), e) time since blood meal (3 - 72 h) as well as f) mosquito behaviour (repellency by conidial formulations). Co-formulations of M. anisopliae ICIPE-30 and B. bassiana I93-825 in ratios of 4:1, 2:1 & 1:1 were also tested. Metarhizium anisopliae IP 46 was exposed to An. gambiae and An. arabiensis to determine its pathogenicity on these mosquito species before being used for the field trials. Mosquitoes were exposed to fungal formulations applied on paper inside holding tubes, except when different delivery substrates were assessed. For the delivery substrates, sections of netting and black cotton cloth were joined using Velcro strips to fit over 20 × 20 × 20 cm wire frame cages; and mud-lined plywood panels were similarly assembled into 20 × 20 × 20 cm cages. Laboratory experiments were performed using laboratory reared mosquitoes at the Ifakara Health Institute, Ifakara, Tanzania. Following the laboratory experiments, fungal formulations were assayed in experimental hut trials in a field setting at Lupiro village (Ulanga District, Tanzania), a rural hamlet 30 km south of Ifakara. Five different techniques that each exploited the behaviour of mosquitoes when entering (eave netting, eave curtains, eave baffles), host-seeking (cloth strips hung next to bed nets) or resting (cloth panels) were assessed. The degree at which mosquito survival was reduced varied with conidia concentration; 2×1010 conidia m-2 was the optimum concentration above which no further reductions in survival were detectable. Co-formulations exerted neither synergistic nor additive effect in reducing mosquito survival. The exposure of mosquitoes to fungal formulations for time periods as short as 15 and 30 min was adequate to achieve 100% mortality of mosquitoes within 14 d post exposure. Longer exposure times did not result in a more rapid killing effect. Conidia impregnated on papers remained infective up to 28 d post application, and such trait did not seem to be influenced by the conidia concentration. Mosquitoes of the age between 2-12 d equally succumbed to fungus infection, with them, however, being relative more susceptible when non-blood fed. Oil-formulations of the fungi did not exhibit any repellency to mosquitoes. Metarhizium anisopliae IP 46 was pathogenic to both An. gambiae and An. arabiensis. Conidia were more effective when applied on mud panels and cotton cloth compared with polyester netting. Cotton cloth and mud, therefore, represent potential surfaces for delivering fungi to mosquitoes in the field. Two delivery techniques, cotton cloth eave baffles and strips hung next to the bed net were successful in exploiting the behaviour of wild anopheline mosquitoes. Up to 75% of house-entering mosquitoes became infected with fungus applied with either technique. By contrast, eave netting, eave curtains and cotton panels placed next to the bed net were ineffective in infecting mosquitoes with sufficiently high doses of fungi to affect their survival. Based on the survival data of the mosquitoes infected with fungus by means of eave baffles, model estimates indicated that fungus alone can reduce EIR by more than 75%. In conclusion, these findings indicate that with well-optimized fungal formulations and correctly-designed delivery techniques, a high proportion of house-entering wild malaria mosquitoes can be infected with entomopathogenic fungi to achieve considerable reduction in their survival and possibly malaria transmission. More importantly, these findings provide baseline information that is highly relevant for designing and conducting large-scale field trials to validate the projected impact of fungal infection under realistic field situations.