Background
In Ethiopia, malaria showed a declining trend over the last 15 years mainly due to the high coverage of key control interventions, such as artemisinin-based combination therapy (ACT), use of rapid diagnostic tests (RDTs) at the remote health facilities, wide-scale distribution of long-lasting insecticidal nets (LLINs) and increased coverage of indoor residual spraying (IRS) since 2004/2005 [
1,
2]. As a result, malaria decreased from 4.1% in 2006 to 0.4% in 2007 [
3]. A survey in 2012 showed reduction in malaria admissions by 54% and deaths by 55%, as compared to the rates in 2001–2004 [
4]. Ethiopia also attained a reduction in malaria morbidity from 22 to 10%; malaria case fatality rate from 4.5 to 2% in age groups of 5 years and above, and 5% to 2% in under 5 children [
4].
The vector control interventions targeted
Anopheles arabiensis [
5,
6], which is a member of the
Anopheles gambiae complex [
7] and the main vector of the disease in the country.
Anopheles pharoensis, Anopheles funestus and
Anopheles nili serve as secondary vectors in some areas [
8‐
11].
Although the World Health Organization (WHO) recommends LLINs and IRS for malaria control in many African countries [
12,
13], insecticide resistance appears to be a major challenge for malaria vector control programmes [
14,
15]. The challenge became very serious with the occurrence of pyrethroid resistant
An. funestus in southern Africa [
13] and multiple insecticide resistant populations of
An. arabiensis in some other countries [
16]. Also in Ethiopia, high levels of resistance to DDT and the pyrethroid insecticides [
10,
15,
17] were reported from the South, Gambella, Gorgora and the Ghibe River valley.
Furthermore, additional factors like improper use of LLINs and increased outdoor feeding frequency of malaria vectors may negatively affect the benefits of nets [
18], implicating the need for novel approaches that will supplement the available vector control tools by reducing the use of conventional insecticides. This would reduce the rate of resistance development to the existing insecticides and also target vectors that feed and/or rest outdoors [
19].
Endectocides, the macrocyclic lactones (avermectin/milbemycin), are drugs with endoparasitocidal and ectoparasitocidal activities [
20]. They have been used to treat metazoan parasites from multiple phyla [
21]. Treatment of livestock with endectocides has been reported to be effective against zoophilic malaria vectors [
22], tsetse flies, and ticks in sub-Saharan Africa [
23]. One of the endocticides is ivermectin, which is known to be toxic to mosquitoes [
24] that feed on treated humans and cattle [
25]. Studies conducted by Fritz and her colleagues [
26,
27] showed that mass treatment of livestock with ivermectin could reduce zoophilic vector populations at the onset of the rainy season, precluding epidemics in malaria endemic regions. A therapeutic dose of ivermectin in a mosquito blood meal causes reduced survival, fecundity, and egg hatch rate [
26,
28‐
31]. This has been shown to serve as a supplementary approach to prevent malaria transmission by reducing the population of
Anopheles mosquitoes [
31]. Its effect is enhanced as it affects the age structure of the vector population by changing/pulling it towards the younger age class which does not participate in transmission [
32].
As with ivermectin, LongRange™ (eprinomectin 5%) has a deworming action in cattle and it persists in treated cattle for 150 days [
33]. Hence, the present study investigates the efficacy of LongRange™ eprinomectin (LRE) against insectary-reared
An. arabiensis in Ethiopia.
Methods
Study area
The study was conducted in Edo Kontola (a rural village) which is located about 4 to 5 km north of Batu town (Ziway) from February to May 2016. Batu town is located 167 km South of Addis Ababa, Ethiopia. The area lies at an average altitude of 1653 m above sea level. The total annual rainfall is approximately 700 mm, with peaks during the main rainy season in July (250 mm) and August (220 mm). The mean minimum and maximum annual temperatures are 14.5 °C and 27.7 °C, respectively.
Study design
A randomized control trial was conducted on six calves—three control and three treatment. Epirnomectin was subcutaneously injected by a veterinarian and the efficacy of the drug on An. arabiensis was assessed for 85 days.
Mosquitoes
A colony of
An. arabiensis that was maintained at Aklilu Lemma Institute of Pathobiology, Addis Ababa University since 2001 was used in the study. The generation of these mosquitoes was F-165 (where F is for filial) and they were reared at 28–31 °C, 70–80% relative humidity and a 12:12 lightdark cycle. It was originally collected from Bishoftu (Debrezeit), which is located at 45 km East of Addis Ababa [
10]. The colony is susceptible (100% mortality) to DDT, permethrin and deltamethrin [
10]. Larvae were raised on a diet of ground Tetramin
® fish meal and adults maintained on 10% sterile sugar solution. A high bovine blood index of
An. arabiensis in the village was known from the study of Gari et al. [
34].
Experimental drug
The LongRange™ (eprinomectin 5%), “LongRange™”, a trademark of Merial (registration pending in the USA) [
33], was kindly provided by Dr. Seth Irish, research entomologist at CDC in Atlanta and stored as recommended by the manufacturer and used in the study. It is a semi-synthetic compound of the ivermectin family drugs and was originally selected as a nematocide, insecticide, and miticide [
35,
36].
Treatment of calves with eprinomectin
The experiment was performed following the method used by Naz et al. [
22] with some modifications. Six calves (1–3 year old) of local breed having approximately equal weights were divided into two groups (three calves per group). Calves were kept together with the herd in the field grazing around Lake Ziway, 5 km North East of Batu town. The body mass of the calves was measured by tightly stretching the Heart Girth (mass measuring metre) [
37] behind the fore limbs from the ventral to dorsal (hump) area. Prior to conducting the study, information on the history of previous drug exposure was checked for each calf, to avoid possible interference from other drugs that may affect mosquitoes 1 week ahead and a day before drug injection. The drug dose for each calf was calculated (1 mg/kg which is equivalent to 1 ml/50 kg) according to the instruction of the manufacturing company. The drug was injected to the calves by a veterinarian subcutaneously under the loose skin in front of the shoulder by using 18 gauge, ½ in. needle after sanitizing the area with denatured alcohol.
Bioassay of eprinomectin against Anopheles arabiensis that fed on drug dosed calves
The bioassay was done according to Poche et al. [
38], with little modification. Briefly, paper cups were 24 h after drug injection to the calves, 20 unfed
An. arabiensis females 3–5 days old were transferred into paper cups covered with nylon netting by using mouth aspirator. A total of 40
An. arabiensis in two cups of 20 individuals were allowed to feed on the neck of each calf for one hour in the evening (7:00–8:00 p.m.) by wrapping the cups on the left and right sides of their neck with a stripe of cotton cloth. After the feeding session, fully engorged mosquitoes were carefully selected and transferred into small Barraud cages that were covered with a large polyethylene bag by using glass tube mouth aspirator. In the cages, the mosquitoes were provided with 10% sterile sugar solution soaked in cotton pads. Cotton pads soaked in water were also placed in the polythene bags to maintain 70–80% humidity. The controls and all blood-fed mosquitoes were maintained inside a thatched hut field laboratory, under natural temperature condition, and followed for mortality for four consecutive days. Mortality was recorded from each cage every 24 h. A correction for treatment response was made from control response by using the Abbott’s formula [
39]. However, since the control response values in the present study were < 10% and had only small effect on the value of mean experimental treatment responses, the application of Abbott’s formula was not found applicable. The drug is claimed to provide a second peak concentration in the plasma after 70 days of a single dose treatment on calves [
35]. Therefore, exposure of mosquitoes was conducted every week for about 85 days to treated and untreated calves.
Observations on the effect of eprinomectin on the fecundity of Anopheles arabiensis
In addition to mortality, observations were carried out on the effect of eprinomectin on fecundity and fertility of
An. arabiensis. Using the method described by Pooda et al. [
40], mosquitoes surviving 4 days post-blood feeding on drug-dosed calves were grouped into two; half were dissected for fecundity (egg counting) and the remaining half were observed for egg hatchability. Ovarial dissection was done under a 40× dissecting microscope and the number of eggs from each mosquito counted and recorded.
Observations on the effect of eprinomectin on Anopheles arabiensis egg hatchability
The fertility of mosquitoes was determined based on the method of Derua et al. [
41]. This was done for each group (treatment and control) of mosquitoes. On weeks 2, 4 and 5, treatment group mosquitoes were placed in one cage and the control group mosquitoes in another cage and followed for egg oviposition for 4 days by providing egg laying substrates (wetted filter papers on petri dishes). Eggs were counted and then transferred to rearing pans for incubation and hatching. Upon hatching, all first instar larvae were counted and removed daily and this was done for 3 days [
26]. Fertility of individual mosquitoes was monitored from 6 to 12 weeks post drug exposure, by placing them in a plastic cup covered with nylon cloth mesh, with a lining of wet filter paper for egg laying. Five mosquitoes from each cage (12 cages; a total of 60 mosquitoes) were transferred singly into 60 plastic cups. Eggs laid from each cup were counted and transferred into individual pans to maintain and follow larval development up to adult emergence during which the number of larvae, pupae and adults were counted and recorded.
Data analysis
Data were entered into Microsoft Excel and analysed using IBM, SPSS version 20. Mortality data was compared between treatment and control arms using Kaplan–Meier estimator. The percent of fed mosquitoes were obtained by calculating the ratio of total fed to total exposed. The fecundity, fertility and larval development data was compared between the two groups using independent t-test. The comparison was based on means of the target variables, i.e., mortality, fecundity and fertility rates obtained from both treated and control groups. Statistical significance was assumed whenever P-values are < 0.05.
Discussion
This research evaluated the endectocidal activity of eprinomectin in calves against
An. arabiensis in a semi-field environment. The effects of the drug with varying degrees of efficacy were observed on the later feeding weeks from treatment. The finding that mortality of all the mosquitoes that fed on treated calves within a week of treatment shows that eprinomectin can reduce adult zoophagic
An. arabiensis populations. A previous study in Kenya also reported that eprinomectin mixed with bovine blood and provided to laboratory-reared
An. arabiensis in a membrane feeder killed all mosquitoes at low concentrations [
27]. In the present study, the drug effect was significantly higher up to 7 weeks post-treatment. However, the rate of mortality of
An. arabiensis after a week was significantly reduced. This finding is in agreement with previous studies by Poché et al. [
36] in Western Kenya where eprinomectin, fipronil and ivermectin were found effective killing
An. arabiensis for at least 7 days at the lower doses of 0.25 and 0.5 mg/kg and at a concentration of 0.2 mg/kg for up to 21 days post-treatment.
A similar study by Derua et al. [
41] using ivermectin also showed that
An. gambiae that fed on treated human volunteers were killed within 3 days post-treatment. The studies present evidences on the potential impact of endectocides including eprinomectin, on vector mosquitoes, and hence, it can be suggested that where the vectors are zoophagic and outdoor feeders combining cattle treatment with IRS and LLINs could be taken alternative approaches in vector control interventions particularly during epidemic seasons. The approach can also be a component of integrated vector management. In areas where there is continuous transmission, there could be a need for repeated treatment of cattle and humans with endectocides.
Eprinomectin was observed to be very effective in protecting cattle from a variety of nematode parasites [
33,
35] which is an added advantage in countries like Ethiopia where malaria and other parasites of humans and animals are most common. In contrast with ivermectin, eprinomectin has a better pharmacokinetic profile with possibly longer effect and has no withdrawal period for milking i.e. a treated cow can be milked right after treatment and the milk for human consumption is safe as the drug has limited distribution in milk [
33]. In addition, eprinomectin can provide high levels of parasite control against a range of nematodes of cattle for up to 5 months following a single treatment [
35].
Thus, treatment of cattle with eprinomectin can be implemented as a supplementary control of An. arabiensis in conditions where the mosquito is zoophagic and exophagic, crepuscular and resistant to the available insecticide-based control methods, such as LLINs and IRS in particular and outdoor malaria transmission in general.
Anopheles arabiensis that fed on treated calves were observed to lay relatively lesser number of eggs (not statistically significant) but their reproduction capacity was not affected. This is similar with a study by Fritz et al. [
27] which showed no difference in the fecundity of
An. arabiensis fed on eprinomectin mixed bovine blood and DMSO-treated blood (control) in a membrane feeder system. The delay in hatching time, larval development, pupation and adult emergence detected in the present study in mosquitoes fed on treated calves than the non-treated is similar to that reported by Pooda and his colleagues, in which it was shown that a therapeutic dose of ivermectin delayed the first larviposition of
Glossina palpalis gambiensis [
30]. According to previous reports, factors like delay in the ovulation process, an increase in the duration of maturation, and/or a disruption of pupation reflect the effect of endectocides on the fertility of flies [
42].
In the present study, the temperature fluctuation in the area might have affected the larval development in the field laboratory. This is consistent to the observation made by Beck-Johnson et al. [
43] and others [
44] who reported that the fall in mean temperature below 18 °C inhibits egg development. The data showed that such temperature changes affected larval development and adult emergence on weeks 4, 5, 8, 9 and 10. Yang et al. [
45] also observed that, the increase in ambient temperature is a beneficial factor for the maintenance of vital physiological activities of mosquitoes leading to maturation of fertilized eggs.
The effective weekly test of mosquitoes, maintenance of fed laboratory, rearing and controlling the daily temperature and humidity at the field laboratory were the strengths of the present study. The limitations of the study were that, the exposed mosquitoes should have been wild-caught and the study should have run for up to 120 days to generate a more robust data, which unfortunately was not financially and technically possible.
Conclusions
Treating calves with a therapeutic dose of eprinomectin killed An. arabiensis that fed on calves within 7 days post-treatment. However, mortality of An. arabiensis was reduced gradually after day seven post-treatment. The drug showed no significant effect on fecundity and fertility of An. arabiensis. Based on the present study, eprinomectin has the potential for use as a supplementary vector control tool together with IRS and LLINs, against An. arabiensis, if weekly mass drug administration of cattle is used for controlling zoophilic mosquitoes and reduce outdoor malaria transmission. This would fill the gap in mosquito control where the country is mainly using control measures targeting the human–vector contact, which would fail as mosquitoes shift their feeding behaviour from humans to domestic cattle. However, the present findings must be refined through future studies that are based on large sample size, use of tent traps and allowing wild caught mosquitoes to feed and by conducting the study under different eco-epidemiological conditions.
Acknowledgements
We are grateful to the East Shewa Zone Health Department and the Adami Tullu District Health Office for their supporting facilitating the implementation of this study. A special thanks go to the Veterinary Drug and Animal Feed Administration and Control Authority of Ethiopia for allowing the first trial of this drug and the study participants who willingly took part in this study. Our thanks also goes to the Microbial, Cellular and Molecular Biology Department, College of Natural Sciences, and the Aklilu Lemma Institute of Pathobiology (ALIPB) at Addis Ababa University for helping to facilitate the field work. A special thanks to the driver Alemayehu Assefa, entomology technicians Selamawit Yaregal and Wossen Sesay at Vector Biology Control and Research Unit, ALIPB. We sent our gratitude to Dr. Seth Irish, Entomologist at Center for Disease Control and Prevention, Atlanta, Georgia.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.