Malaria continues to be a major life-threatening vector-borne disease globally but more so in developing countries. According to the World Health Organization (WHO), over 90% of the 228 million malaria cases and 405,000 deaths observed in 2018 were recorded in Africa [1
]. It is the first cause of mortality and morbidity in West Africa, and the majority of malaria-endemic countries are in sub-Saharan Africa (SSA), which shares 80% of the global malaria burden [1
]. Malaria parasites are transmitted by infected Anopheline mosquito species. Vector control is therefore essential in the fight against malaria and is commonly accomplished through the use of insecticide-treated nets (ITNs) and indoor residual spraying (IRS) [3
]. The latter consists of applying insecticides to interior walls of a house.
IRS is highly effective in controlling malaria vectors and has contributed significantly to the successes obtained in the fight against the disease in recent years [4
]. However, insecticide resistance is now widespread for pyrethroids, one the most used insecticide classes in vector control [6
]. More concerning is that resistance to the other classes of insecticides approved by the WHO (carbamates, organochlorines, and organophosphates) has started to rise and has now been suspected or documented in several countries worldwide [6
]. It is therefore of utmost importance to search for alternatives that not only exhibit new modes of action, but are also safer, cost-effective, sustainable, naturally occurring if possible, and compatible with the synanthropic nature of mosquitoes.
Mechanical insecticides (MIs) are minerals that come into contact with insects and produce a lethal effect. MIs exhibit insecticidal properties through a physical mode of action and have consistently been shown to control numerous agricultural pests, i.e., Coleoptera [12
], Thysanoptera [13
], Lepidoptera [14
], and Hemiptera [15
]. Their potential use in public heath, on the other hand, has received minimal attention and was almost exclusively limited to laboratory trials [16
In this study, we demonstrated the efficacy of a new mineral-based mosquitocide derived from volcanic rock, ImergardTM
WP (expanded perlite, 100%), in commercialization by the company, Imerys. Perlite is an aluminosilicate volcanic glass with a relatively high-water content, typically formed by the hydration of obsidian. It occurs naturally and has the unusual property of greatly expanding in volume under extreme temperature. Because of its low density, perlite is used as insulation or in plant growth media. Perlite is listed in the US Food and Drug Administration (FDA) database as Generally Recognized as Safe (GRAS), is classified as a feed additive by the European Food Safety Authority (EFSA), and is found in products like toothpaste [18
]. It has not been considered before for vector control. The MI described in this paper (applied by spraying in water) provides a new mode of insecticide action for mosquito and malaria control, disrupting insect water balance. The current thinking is that mechanical insecticides disrupt the protective lipid layer of the insect’s cuticle, thus leading to death by desiccation (a non-toxic, i.e., non-systemic mode of action) [16
]. Because the MI is an industrial mineral, its activity does not degrade with time as long as the mechanical insecticide is present and can transfer to the insect [19
]. Since mechanical insecticides have not been used before in mosquito control, it should be effective against pyrethroid-resistant mosquitoes in Africa.
We evaluated the toxicity and residual activity of ImergardTMWP alone and as a mixture with deltamethrin for the control of susceptible and wild-type pyrethroid-resistant mosquitoes under field conditions in Africa. We demonstrated that the MI is a suitable alternative to a chemical insecticide currently in use for mosquito control and offers a long-lasting protection against mosquitoes such as the African malaria mosquito, Anopheles gambiae (s.l.).
Mosquito transmitted malaria parasites kill more people each year worldwide than any other single mortality factor, especially affecting children under the age of five and expecting mothers. The WHO African region accounted for 94% of all malaria deaths in 2018, while 67% (272,000) of all malaria deaths worldwide were children under five years [1
]. Vector control with chemical-based, insecticide-treated bed nets and chemical-based insecticide (residual) wall sprays is critical to mosquito management and the reduction of malaria, the former including insecticide synergists. It is estimated that the use of insecticide-treated nets and the implementation of indoor residual spraying, taken together, averted over 517 million clinical cases of malaria from 2000 to 2015 [28
]. Using a mathematical model to predict the health impact of various IRS products, Sherrard-Smith et al. [29
] reported that Actellic®
300CS (an organophosphate) and SumiShield®
50WG (a neonicotinoid) grouped together averted up to 500 clinical cases per 1000 people per year. These predictions were determined for areas with moderate endemicity, high levels of pyrethroid resistance, low bed net use and 80% IRS coverage. In Benin, two rounds of bendiocarb (a carbamate) spraying in a community trial in 2008–2009 reduced biting rates of pyrethroid-resistant An. gambiae
by over 80% and the parous rate by 70%, while none of the mosquitoes analyzed were infected (entomological inoculation rate = 0) [30
Since the main pesticides currently used for bed nets are pyrethroid insecticides, the use of residual wall sprays with insecticides with a different mode of action like carbamates, organophosphates, and organochlorines are essential to managing mosquito insecticide resistance to bed nets. Despite this effort, insects are increasingly becoming more resistant to pyrethroids and demonstrating cross-resistance to other insecticides. In fact, insecticide resistance in malaria vectors has been documented for pyrethroids [6
], organophosphates and carbamates [9
], and organochlorines [32
]. Currently, there is a significant effort to develop new chemical insecticides with new modes of action for bed nets and residual wall sprays. As of January 2020, 30 IRS products (24 pyrethroids, 1 neonicotinoid, 2 carbamates, 2 organophosphates, and 1 dual-active ingredient neonicotinoid-pyrethroid) as well as 19 ITN kits (18 pyrethroid-based and 1 dual-active ingredient alpha-cypermethrin-chlorfenapyr) have WHO prequalification listings (https://www.who.int/pq-vector-control/prequalified-lists/en
, accessed on 24 April 2020). SumiShield®
50WG, for example, is a new IRS product (WHO prequalification achieved in October 2017) containing clothianidin, a neonicotinoid [22
]. If alternatives are not found, we will lose mosquito control not only from wall sprays but also for bed nets, and the gains achieved in the last ten years in the reduction of malaria cases will be reversed.
The other consideration with insecticide-treated bed nets and residual wall sprays is the potential long-term use and exposure of people to chemical insecticides in the home. Contradictory results have been reported in the literature regarding the health effects of IRS insecticides, where some studies found adverse associations between exposure to insecticides such as DDT (dichlorodiphenyltrichloroethane, an organochlorine) and some pyrethroids and human health [34
], whereas others found none [36
]. However, the use of chemical insecticides is currently necessary because of the high risk of mortality from contracting malaria, and the insecticides being used have been proven safe. Not only are different modes of action needed for control but effective insecticides from natural sources with more specific action against the mosquito vector would be desirable.
In field trials in Benin, Africa, using WHO-approved huts (Figure 2
) and standard WHO methods to evaluate residual wall sprays, ImG alone and as a mixture with deltamethrin (deltamethrin 62.5 SC-PE) was found to be significantly more effective than the pyrethroid alone and the untreated control. A deterrent effect (reduction in mosquito entry rates compared to the control) was found in the first four months (Table 1
). The deterrent effect with ImG was not expected as the MI is odorless, and there was no obvious airborne MI in or exiting the experimental huts. One possible explanation is that some of the mosquitoes entering the hut made contact with a treated surface close to the hut entrance and became quiescent. When a mosquito moves, because of their proximity to the opening they have a better chance of exiting compared to the control huts. Because the entrance windows were engineered as one-way openings and deterrence levels were highly variable in different months, this explanation is most likely not the mechanism for the calculated ImG deterrence found in some months. Entry reduction rates month to month were highly variable in our study, especially for ImG, and in most cases was less than 10% when compared to the control. It is most likely that entry reductions could simply be an artifact of the random geographical positioning of the control huts used to calculate deterrence compared to the ImG-treated huts and/or could be the result of environmental differences.
In two-choice repellency studies conducted in laboratory settings at North Carolina State University (NCSU), the malaria mosquito was not repelled by an ImG-treated surface [38
]. Results here showed that once mosquitoes were in the hut, ImergardTM
WP did not repel the mosquitoes into the veranda (Table 2
); there was no difference in exophily rates (the proportion of mosquitoes found in the hut veranda) between the control and the MI. This is consistent with the idea that an ImG surface is not repellent. The higher exophily rates observed with deltamethrin alone and the ImG–deltamethrin mixture is expected since pyrethroids are known to demonstrate mosquito spatial repellency [21
No significant blood-feeding inhibition was observed for ImG, the pyrethroid, or the mixture of ImG with the pyrethroid (Table 3
). In fact, blood-feeding rates were high in all treatments with >90% of the mosquitoes entering the huts acquiring a blood meal in both the treated and untreated huts. This is not uncommon for IRS treatments. No significant differences in blood-feeding rates were found between the treated and untreated huts during the evaluation of chlorfenapyr [21
], clothianidin [20
], and bendiocarb [40
] in Benin.
Since there is no physical barrier, host-seeking mosquitoes often feed on the sleeper before landing on the walls to rest and digest their meal [20
]. High percentages of blood-fed An. gambiae
(s.l.) died after resting on the walls treated with ImergardTM
WP (79–82% average mortality) and the mixture (73–81%) during the 1–6-month study (Figure 3
), thus confirming this technology has a significant potential as a new active ingredient for IRS. Furthermore, its residual mean activity as a stand-alone treatment was ≥80% mortality for at least five months for both susceptible (Figure 4
) and resistant (Figure 5
) mosquitoes; activity at six months was 76% and 78%, respectively (although the 95% upper confidence interval exceeds 80%). Although further studies are needed, it is possible that 80% mortality could be achieved past six months using the application rate in this study and/or increased even further by increasing the application rate. Examining application rates versus control duration for ≥80% mortality is needed in the future. The use of ImG at an optimum application rate for maximum control over time could lead to a reduction in the number of applications needed each year, a reduction in IRS application costs per home, and an increase in the number of homes that could be treated with the resources currently available. Furthermore, since ImG is not subject to thermal degradation, metal roofs can be treated, further improving efficacy. Based on the differences in mortality in the huts over six months, the MI killed 2.5-fold more mosquitoes (5517) than the pyrethroid (2180).
WP showed a time-dependent increase in insecticidal activity against mosquitoes from 24 to 72 h. When a mosquito lands on a treated surface, a few particles of the product are statically transferred onto the insect’s body. The insect’s epicuticle contains lipids to prevent water loss [41
]. Current understanding is that the lipid layer is disrupted by the mechanical insecticide. This increases water loss, disrupts the osmolality of the hemolymph, and results in death. The disruption of the lipid layer of the insect’s cuticle is the presumed mechanism of action for MIs [16
], and this process can take a few hours to a couple of days depending on the MI, the insect species, dose, environmental conditions (temperature and humidity), and the time since the last consumption of water or a blood meal. This delayed effect was also noticed with non-pyrethroid insecticides in IRS trials, e.g., with clothianidin (a neonicotinoid) [20
] and chlorfenapyr (a pyrrole) [21
]. It is therefore important that the WHO IRS insecticide evaluation guidelines, originally conceived for fast-acting insecticides, be reviewed or adjusted to account for the varying modes of action of new IRS alternatives currently being investigated.
Mosquitoes exposed to Imergard in the residual activity tests died at a faster and higher rate than the mosquitoes collected inside the huts. In the latter, mosquitoes entered freely and, once inside the huts, had the possibility of moving to different surfaces of the huts and into the exit traps. Some of these surfaces were untreated. In the residual efficacy tests, mosquitoes were confined to a small space (WHO cones) presumably increasing their exposure to the treated surface for 30 min, and the mosquitoes may have acquired a higher amount of the insecticide in a shorter period. In laboratory studies, when An. gambiae
were continuously exposed to Imergard in cones, 50% of the mosquitoes were dead in 5 h [38
During the study, mosquitoes were not provided access to a 10% honey-in-water solution during the holding period. The sugar solution fed to mosquitoes used in the residual activity tests prior to their use in this study and the high blood-feeding rates (>90%) in mosquitoes collected inside the huts allowed them to survive without dying from starvation for the time course of the bioassays [43
]. Blood feeding may have also decreased the rate of death during holding for the hut-collected mosquitoes. The MI could easily be washed away or become wet during prolonged interactions of the mosquito with a water source in their holding container. The close housing of mosquitoes within a few cm of a water source, comprised of a large wet surface area optimized for mosquito availability and feeding, creates artificial conditions not found under field conditions. For mechanical insecticides, water provisioning in confinement artificially blocks their mode of action. This is an important example where having rigid testing guidelines for evaluating new IRS technology requiring water for the mosquito might prevent the discovery and implementation of a new mode of action.
A legitimate concern could be that, in houses, access to wet environments might be possible and could reduce or prevent ImergardTM
WP action. However, this is not de facto true. After blood feeding, endophilic/anthropophilic mosquitoes seek a protected resting site inside the house (or sometimes outside) where they can digest the blood meal and develop eggs. At least 50% of An. gambiae
females stay inside the home after a blood meal and invade interior walls, crevices, furniture, etc., for that purpose [46
]. This stage of the gonotrophic cycle takes 1.8 days up to several days, where the duration varies depending on the mosquito species, the amount of blood consumed, and the environmental conditions [46
]. Because blood digestion and egg maturation occur at rest with little to no interaction with a water source, it is likely that under natural conditions in a home the MI would have produced its lethal effect before the mosquito finished blood meal digestion and began to seek an oviposition site.
One concern about the use of a mechanical insecticide on the walls of a home is that human contact with treated surfaces might prematurely reduce the duration of control. A possible solution could be to treat only the upper part of the walls or target places where mosquitoes rest, such as crevices. Djenontin et al. [50
] showed that covering only the upper one-third of the wall of huts with a carbamate-treated plastic sheeting (CTPS) provided equal or better efficacy compared to a traditional IRS application method using bendiocarb.
Improved toxicity and residual activity were obtained by mixing deltamethrin with ImG compared to deltamethrin alone especially for wild-type insecticide-resistant An. gambiae
, Figure 4
and Figure 5
). Mixing two insecticides with different modes of actions to improve vector control even in areas of resistance was suggested before and recommended [6
]; using two different modes of action could delay the evolution of resistance or reverse current levels of resistance. The WHO recommends that the two insecticides be co-formulated into a single product [6
The poor performance observed with deltamethrin in our study (even though not surprising) is worrisome. Deltamethrin 62.5 SC-PE (a.i. 62.5 g/L deltamethrin; Bayer, WHOPES recommendations achieved in September 2013) is a polymer-enhanced suspension concentrate that came on the market only a few years ago and has been shown in several studies to be more effective than the standard deltamethrin WG25 against both susceptible and wild resistant mosquitoes [51
]. The work here is suggesting that improving current active ingredients by formulation alone will not be sufficient in resolving resistance issues for the long term and that new active ingredients with a different mode of action are needed.