Background
Malaria vector control tools such as the long-lasting insecticide-treated nets (LLIN) and indoor residual spraying (IRS) have controlled malaria transmission in Africa by targeting mosquitoes that blood feed on humans (anthrophagy, anthrophily), and rest inside houses (endophagy, endophily) [
1,
2]. Specifically, these indoor interventions reduced feeding frequency, density and survival of
Anopheles gambiae s.s. [
3‐
7] and
Anopheles funestus [
6,
8,
9]. However, transmission risk and burden of malaria in Africa is yet unacceptably high even in areas with high coverage of LLIN and IRS [
10‐
14], because of physiological insecticide resistance [
11,
14,
15], and behaviours of outdoor feeding (exophagy), resting (exophily); biting in the early evening and morning, feeding on animals and resting in cattle shelters (zoophagy, zoophily) in Anopheles mosquitoes [
6,
7,
11,
12,
14,
16‐
19]. Therefore, a novel vector-control strategy capable of reducing the density and survival of outdoor biting, and zoophilic mosquitoes is urgently required to complement LLIN and IRS by controlling residual transmission of malaria.
Endectocides, such as ivermectin, which are broad-spectrum systemic drugs against nematodes and arthropods in public [
20‐
23], and veterinary health importance [
24], they are also potential novel vector-control tools for targeting outdoor biting, and insecticide resistant
Anopheles mosquitoes [
25‐
27]. However, the effects of ivermectin-treated humans against malaria vectors have been extensively investigated relative to treated-cattle. Ivermectin act on glutamate gated chloride channels in parasites including mosquitoes [
28‐
32], which differ from target sites of conventional insecticides used in LLIN and IRS. Therefore, mass drug administration (MDA) of ivermectin to humans may targets malaria vectors regardless of their biting locations [
16,
17,
33], time [
17,
19,
34], and physiological insecticide resistance status [
11,
14,
15]. Laboratory studies demonstrated that ivermectin-treated human blood reduces feeding frequency, blood meal digestion, rate of defecation, survival, fecundity, vector density, and sporozoite rates of mosquitoes after membrane or direct feeding assays [
35‐
40]. Similarly, the MDA of ivermectin to humans may decrease malaria transmission by reducing the survival of wild Anopheles mosquitoes for 1 week that lead to increased young females for 3 weeks, and reduced sporozoite rates for 2 weeks [
41‐
45]. Even if long-lasting formulations of ivermectin administered to humans could sustain reduction in malaria transmission, it may be inefficient to suppress malaria vectors that blood feed on cattle.
Zoophilic vectors contribute to malaria transmission in many parts of the world such as
Anopheles arabiensis in Africa [
46],
An. albimanus in Latin America [
47],
An. sinensis in Asian-Pacific [
48]. For example,
An. arabiensis feed on cattle and continues to transmit malaria outside houses in areas with high coverage of LLIN and IRS across East-Africa [
6,
7,
49‐
51]. Treating cattle with ivermectin could control these mosquitoes [
52‐
54], yet the studies that evaluated this strategy in more realistic environments are scarce. For example, few laboratory and field studies demonstrated that ivermectin-treated cattle reduced the survival and fecundity of Anopheles mosquitoes (e.g.
An. gambiae s.s,
An. arabiensis,
Anopheles coluzzii,
Anopheles culicifacies and
Anopheles stephensi) [
55‐
59]. Although the effects of subcutaneous-treated cattle to mosquitoes were long-lasting than orally- or topically-treated cattle [
55,
57‐
59], most of these studies used laboratory mosquitoes and artificial feeding strategies which have excluded mosquito genetic and phenotypic diversity [
60‐
62], and vertebrate host ecology [
63]. The blood feeding behaviour of
An. arabiensis on cattle has genetic basis [
51,
61], but ivermectin-treated cattle against these mosquitoes has never been evaluated using host-mosquito interactions as in the natural environments. Therefore, several knowledge gaps including the effects of ivermectin-treated cattle against fitness of
An. arabiensis remain to be studied under semi-field conditions before recommending for a large-scale trial.
We evaluated the effects of ivermectin-treated cattle against free-flying population of An. arabiensis within the semi-field system that is closely related to the natural environments in south-eastern Tanzania. The specific objectives were: (1) to demonstrate that ivermectin-treated cattle reduce blood meal digestion and the subsequent egg production, and survival in An. arabiensis, and (2) to assess the duration of these effects (residual activity) against these mosquitoes. This information will be useful in assessing the appropriateness of ivermectin-treated cattle as complementary to LLIN and IRS for controlling outdoor-biting, and insecticide-resistant An. arabiensis.
Discussion
This study demonstrates that the ivermectin-treated cattle reduce blood meal digestion, and subsequent egg production, and survival of free-living population of An. arabiensis in South-Eastern Tanzania. The ivermectin-treated cattle decreased the efficiency of blood meal digestion in An. arabiensis, and their subsequent egg production for up to 2 weeks. The proportion surviving and their median survival times of these mosquitoes were reduced after blood feeding on ivermectin-treated cattle than control cattle for up to 3 weeks. Additionally, ivermectin-treated cattle increased daily mortality rates of An. arabiensis by five folds than control cattle, but it gradually declined for up to 3 weeks post-treatment. These results imply that the effects of ivermectin-treated cattle on efficiency of blood meal digestion, egg production, and survival of population of An. arabiensis may suppress their vector density and reduce outdoor transmission of malaria.
The digestion of blood proteins (haemoglobin) in the mosquito stomach generates two key products: heme that is detoxified and defecated as haematin, and essential nutrients for egg production [
77‐
79]. The present study found that ivermectin-treated cattle reduced both oviposition rates and fecundity of
An. arabiensis than untreated cattle for 15 days post-treatments. This reduction in mosquito egg production may be explained by the decreased mass of haematin defecated by these mosquitoes at the end of digestion suggesting that small amount of blood proteins in their stomach was converted to less nutrients required for egg development. Perhaps ivermectin changes digestive responses to blood meal in mosquito stomach (e.g. malformation of peritrophic matrix) that lead to reduced efficiency of blood digestion, defecated haematin, and nutrients for egg production. For instance, previous laboratory studies confirmed that ivermectin, chitinase, and silencing disrupted peritrophic matrix in blood fed mosquitoes or sand flies that reduced their blood meal digestion, haematin defecation, and their subsequent egg production [
36,
82‐
84]. Additionally, our finding is consistent with previous studies that observed ivermectin-treated cattle decreased egg production in
An. coluzzii [
57],
An. gambiae s.s [
55], and
An. arabiensis [
55,
56], for up to 10 days after subcutaneous injection [
55,
57]. The present study suggests that ivermectin-treated cattle may reduce digestion of blood proteins in mosquito stomach resulting to small amount of defecated haematin, and essential nutrients absorbed for egg production.
This study also demonstrated that the negative effects of ivermectin-treated cattle on the long-term survival of
An. arabiensis was strong, and declined with post-treatment time. The ivermectin-treated cattle reduced the proportion of surviving
An. arabiensis, and their median survival times for up to 3 weeks after treatment. Similarly, the ivermectin treated cattle increased daily mortality rates by five-folds than control cattle, but the risk of death gradually declined until 3 weeks post-treatments. The possible explanation could be that ivermectin act on nervous system leading to flaccid paralysis and death of mosquitoes [
29,
31]. Another possibility could be that ivermectin may inhibit or delay heme detoxification to haematin in the mosquito abdomen that results to small amounts of defecated haematin, and increased heme toxicity that reduce mosquito survival; therefore, further investigations are required to confirm this possibility. Our results are similar to previous studies which found that ivermectin-treated cattle significantly reduced the survival of
Anopheles mosquitoes by >80% than control cattle in the first week, and the effect gradually declined up until 4 weeks after subcutaneous injection [
55‐
59]. In contrast, Poche et al. [
58] reported that oral administration of ivermectin to cattle significantly reduced the survival of
An. arabiensis than control cattle for up to 1 week post-treatment. Like orally-treated cattle, many previous studies found that orally-treated humans reduced the survival of
Anopheles mosquitoes for at least 1 week [
35,
40,
42,
44,
85].
However, the subcutaneous implants containing ivermectin may extend the negative effects of ivermectin-treated blood on mosquito survivorship for up to 24 weeks after treatment [
86]. The long lasting effects of ivermectin on the survival of mosquitoes may be linked with the fact that subcutaneous injection distribute large amount of ivermectin to adipose tissues than oral route, where it slowly released into the peripheral blood circulations which are available to mosquitoes [
38‐
40,
86]. These results suggest that treating cattle with long-lasting subcutaneous of ivermectin may sustain strong reduction in survival of
An. arabiensis.
Our findings suggest that ivermectin-treated cattle has great potential of controlling residual transmission of malaria by reducing vector density, survival, and vector competence of
An. arabiensis. The present study revealed that the ivermectin-treated cattle could reduce egg production in
An. arabiensis for at least 2 weeks. This suggests that mosquitoes would produce fewer eggs after feeding on ivermectin-treated cattle leading to reduced density of adult mosquitoes in the subsequent generations. Additionally, this study also revealed that ivermectin-treated cattle reduced probability of survival and median survival times of
An. arabiensis for up to 3 weeks. For example, ivermectin treated cattle killed >80% of mosquitoes within 2–4 days post-feeding. These findings indicate that majority of mosquitoes will die before completing egg productions [
71,
72], and
Plasmodium falciparum development to infective sporozoites (10–14 days) [
87,
88]. The effects of ivermectin on digestion of blood meal in the stomach of surviving mosquitoes suggests that it may also inhibit establishment of parasite development in mosquitoes [
37]. In contrast, the MDA to humans contributed to decreased malaria transmission by reducing survival of wild Anopheles mosquitoes for 1 week that consequently shifted age structure to young females (less infectious mosquitoes), and reduced sporozoite rates for at least 2 weeks [
39,
44,
45]. Therefore, the long lasting effects of ivermectin-treated cattle could similarly reduce the vectorial capacity of
An. arabiensis in the field, and further investigations are required. This study also suggests that treating non-lactating cattle (i.e. calves, heifers and bulls) with long-lasting formulations of ivermectin for large-scale malaria vector control may be the best alternative because it allows milk and meat consumption by the communities in rural settings.
The potential limitation of the experimental design was that the population of
An. arabiensis within one chamber of the SFS was exposed to untreated or ivermectin-treated cattle in alternating nights, and sample of blood fed mosquitoes were collected in the morning (Table
1). Under this system, some of mosquitoes missed in prior collection may be mixed to those of new night leading to systemic bias (carrying-over effects) in mosquito sample between treatments. Ideally, these experiments were to be conducted in two different chambers (untreated and ivermectin-treated) to avoid mixing samples between treatments, but it was logistically impossible to establish several chambers with
An. arabiensis populations during dry seasons. Nevertheless, visual observation of mosquito blood meal digestion status in their abdomen was used as a marker to separate between fresh blood-fed mosquitoes from those fed prior nights. The fresh blood fed mosquitoes (i.e. within 12 h) have full reddish/dark red abdomen, but those with two-third black to no blood contents abdomen (semi-gravid and gravid) were considered fed prior nights (i.e. >30 h) [
74,
75]. Therefore, these semi-gravid and gravid females were excluded from the sample of fresh blood fed mosquitoes. Besides, our experimental design considered experimental nights, and individual cows as random effects to control for the variations of mosquito catches between nights and individual hosts.
This study confirms that ivermectin-treated cattle reduce blood meal digestion, and subsequent egg production, and survival of An. arabiensis for up to 3 weeks in South-eastern Tanzania. These results suggest long lasting effects of ivermectin-treated cattle treated may sustainably suppress An. arabiensis, and reduce outdoor transmission of malaria. To ensure continued milk and meat consumption in communities, this study recommends that non-lactating cattle (i.e. calves, heifers and bulls) could be treated with ivermectin for large-scale malaria vector control in villages. Furthermore, the simulation model is under-way to predict the impact of ivermectin-treated cattle alone, or in combination with LLIN/IRS on reducing residual transmission of malaria.
Authors’ contributions
INL and LLM designed, and supervised execution of experiments. INL analysed the data and drafted the manuscript. IL and LLM critically reviewed the manuscript. STK, KM, AAD, and DDM performed experiments. All authors read and approved the final manuscript.