BackgroundVector control is a crucial element of anti-malaria campaigns and works best when there is a thorough knowledge of the biology and behaviour of the Anopheles vector species responsible for transmitting malaria within a given locale. With the push to eradicate malaria stronger than ever, there is a growing need to develop and deploy control strategies that exploit the behavioural attributes of local vector species. This is especially true in regions where the vectors are exophagic (i.e., prefer to bite outdoors), exophilic (i.e., prefer to remain outdoors), and zoophagic (i.e., as likely to feed on non-humans as humans). One promising strategy targeting vectors with these behavioural traits is the administration of avermectin-based endectocides, such as ivermectin, to humans and livestock. When ingested in a blood meal, ivermectin has been shown to reduce mosquito survivorship and fecundity in a number of Anopheles species. In this study, the relative toxicity of ivermectin was compared between two zoophagic, exophilic malaria vectors—Anopheles albimanus and Anopheles stephensi.ResultsToxicity of ivermectin was assessed using membrane feedings, intrathoracic injections, and mosquito feedings on treated mice. When ingested in a blood meal, ivermectin was much less toxic to An. albimanus (4-day oral LC50 = 1468 ng/ml) than to An. stephensi (4-day oral LC50 = 7 ng/ml). However when injected into the haemocoel of An. albimanus, ivermectin was much more toxic (3-day parenteral LC50 = 188 ng/ml). Because the molecular targets of ivermectin (i.e., glutamate-gated chloride channels) reside outside the midgut in nerves and muscles, this suggests that ingested ivermectin was not readily absorbed across the midgut of An. albimanus. In contrast, ivermectin was considerably more toxic to An. stephensi when ingested (4-day oral LC50 = 7 ng/ml) than when injected (3-day parenteral LC50 = 49 ng/ml). This suggests that metabolic by-products from the digestion of ivermectin may play a role in the oral toxicity of ivermectin to An. stephensi. Blood meal digestion and subsequent oviposition rates were significantly hindered in both species by ingested ivermectin but only at concentrations at or above their respective oral LC50 concentrations. To test mosquitocidal activity of ivermectin in a live host system, two groups of three mice each received subcutaneous injections of either ivermectin (600 µg/kg BW) or saline (control). One day after injection, the ivermectin-treated mice (n = 3) exhibited significant mosquitocidal activity against both An. stephensi (85% mortality vs 0% in control-fed) and, to a lesser degree, An. albimanus (44% mortality vs 11% in control-fed). At 3 days, the mosquitocidal activity of ivermectin-treated mice waned and was effective only against An. stephensi (31% mortality vs 3% in control-fed).ConclusionsIvermectin was not uniformly toxic to both Anopheles species. Previous studies indicate that ivermectin is a good choice of endectocide to use against malaria vectors in southeast Asia and Africa. However, these data suggest that ivermectin may not be the optimal endectocide to use in Central America or the Caribbean where An. albimanus is a major malaria vector species. If endectocides are to be used to help eradicate malaria, then additional efficacy data will be needed to define the activity of specific endectocides against the major malaria vector species of the world.