Background
Over decades, arthropods have been controlled using synthetic pesticides with dramatic reduction on target pest populations and their associated impacts of society [
1,
2]. While effective, their prolonged and widespread use has unintentionally resulted in increased prevalence of pesticide resistance, with agricultural and medical implications [
3,
4]. Pesticide resistance is typically a result of injudicious pesticide use and mounting pressure on population genetics (e.g. through natural selection and evolution) [
5,
6]. Indeed, pesticide resistance is triggered by many genetic, [
7] operational [
8,
9] as well as biological [
10] factors. For instance, selection pressure due to excessive use of insecticides may result in behavioral adaptation of vectors and subsequently, gene mutation expression, leading to temporal and spatial intra-specific heterogeneity [
7]. While well-assessed in certain taxa and regions, pesticide resistance is dynamic in space and time and requires continuous evaluation. There are, however, regions where resistance has not been assessed. Furthermore, without empirical evidence for optimal efficacy, synthetic pesticidal active ingredients continue to be used within same localities.
Malaria in humans is an infectious disease, spread by various mosquito species in the
Anopheles genus, which serve as bridge vector hosts for the
Plasmodium spp. parasites that cause the disease. Given the role of mosquitoes in malarial transmission dynamics, their management is a crucial component of integrated malaria control strategies [
11,
12]. Mitigation of the spread of malaria is typically reliant of vector monitoring and control, which most often involves the use of insecticides [
13], although several other complementary approaches are also widely explored [
14,
15]. With regards to insecticide resistance, there are four common synthetic insecticides that are capable of conferring resistance to insects; namely the organochlorines, organophosphates, carbamates and pyrethroids [
16]. These insecticides are target site specific with organophosphates and carbamates inhibiting activity of the neuro-synaptic enzymes whilst the organochlorines and the pyrethroids target the sodium ion channels [
17,
18]. Mosquitoes have shown alterations in their genetics (e.g. acetylcholinesterase genes), with consequent reduction in the binding efficiency with insecticides and hence reduced efficacy [
19]. According to Williamson et al
. [
20], organochlorines and pyrethroid resistance emanates from point mutations in the voltage-gated sodium channels resulting in knockdown resistance (KDR). Cross resistance as a consequence, may occur when a resistance mechanism, also confers resistance to another insecticide [
21], thus further occurring between pesticides from different chemical classes [
22]. Evidence of multiple insecticide resistance in mosquito vectors, including
Anopheles species, have been reported from many regions [
23,
24]. Furthermore,
Anopheles malaria vectors have also been shown to develop adaptive escape behaviours, through either learning or based on insecticide avoidance and/or repellency, creating further challenges for control and elimination of these vectors and associated infections [
25,
26].
Insecticide resistance is a consistently worsening situation in Africa, requiring urgent intervention for effective control of malaria vector species [
27,
28]. Malaria is the most prevalent mosquito-borne disease in the sub-Saharan Africa and is carried by various species of Anopheline mosquitoes. Botswana, situated in the warm subtropics of southern Africa is no exception, with malaria cases reported annually (~ 0.01% /1000 population) and even spreading to non-endemic parts of the country [
29]. However, regardless of the pronged insecticide use for vector control in Botswana [
30], mosquito susceptibility investigations are scant. For many decades lamda-cyhalothrin (pyrethroid) and dichloro diphenyl trichloroethane (DDT; organochlorine) have been the main insecticides used for mitigation against malaria vectors in Botswana [
30], although recently (2019) pirimiphos-methyl (organophosphate) was deployed for indoor residual spraying (IRS) use across all malaria endemic districts [
13]. Similarly, since the 1940s, DDT has been used in the country for IRS and later complimented by pyrethroids long-lasting insecticide nets (LLINs) and microbial larviciding (
Bacillus thuringiensis serovar
israelensis) [
13,
31]. However, insecticide susceptibility status of malaria vectors is currently unexplored in Botswana. This is regardless of the country having been using these insecticides for > 70 years [
30,
32], a time-scale that will likely have promoted resistance development. This may subsequently regress nationwide or regional planning initiatives on malaria elimination achievement targets by 2023 [
33].
As part of a larger project on mosquito control in the region, here we conducted a baseline assessment on
Anopheles arabiensis insecticide susceptibility for World Health Organization (WHO) recommended and currently used pesticides across three malaria endemic regions in Botswana [
34].
Anopheles arabiensis is the biggest contributor to malaria in the region [
35] and is widely distributed across malaria endemic and even non-endemic parts of the country [
36‐
38]. Specifically,
An. arabiensis susceptibility status to eleven registered insecticide products was assessed, comprising four classes of pesticides and determined their knockdown times (KDT
50) and differences in susceptibility patterns across malaria endemic districts. It was hypothesized that (i) locally used insecticides and those with a similar mode of action, had differed efficacy on
An. arabiensis, with (ii) knockdown times differed across recommended insecticides based on their mode of action and intensity of use, and (iii) that
An. arabiensis susceptibility status will differ in space owing to differences in insecticide use.
Discussion
The study highlighted the general reduced efficacy of organochlorines and pyrethroids on
An. arabiensis across malaria endemic areas. Results showed high resistance development for lambda-cyhalothrin (0.05%), permethrin (0.75%), deltamethrin (0.5%) and cyfluthrin (0.15%) in Okavango and for only lambda-cyhalothrin (0.05%) and permethrin (0.75%) in Ngamiland. It further showed that resistance dynamics were variable in space and likely as a result of differing insecticide intensities and nature of application regimes across districts [
31]. In Botswana, a combination of vector control strategy using LLINs (pyrethroid based) and IRS (pyrethroid and DDT) have been practiced continuously over decades with first intervention (DDT) rolled out in Okavango, Ngamiland and Chobe districts in mid-1940s [
13,
30]. Okumu and Moore [
8] suggested that the two intervention strategies used together, may promote the evolution of insecticide resistance in mosquito populations due to increased pesticide selection pressure. Little information is, however, available regarding extent and difference among national vector control and even domestic insecticide use, across the three districts and its role in resistance development. What is known is that the Okavango and Ngamiland districts are formally considered more problematic for malaria and likely receive more intensive and frequent use of intervention strategies [
45].
Botswana has been able to effectively reduce malaria cases over the years through chemical based intervention strategies [
13,
31]. Parallel to this achievement, this study demonstrated compromised vector efficacy to insecticides used for IRS and LLINs, similar to observations in other parts of the world [
46]. The reduced mosquito sensitivity to insecticides observed in this study may be due to the prolonged (> 70 years) use of IRS (DDT and lambda-cyhalothrin) and massive area-wide distribution of LLINs (pyrethroid impregnated) [
32]. Moreover, lambda-cyhalothrin (pyrethroid) and DDT (organochlorine) are insecticides of similar mode of action [
47]. Insensitivity to pyrethroids not currently registered for vector control in the country
vis permethrin, deltamethrin, cyfluthrin and etofenprox was recorded. One of the reasons for this could be the prolonged use of lambda-cyhalothrin [
13], with the same mode of action as these pesticides. Okavango showed the most prominent insecticide resistance which may be associated with extensive pesticidal usage since 1940s from the national vector control programme and household interventions, as an area of targeted malaria ‘hotspot’ compared to Ngamiland and Bobirwa [
31,
32]. Indeed, the results showed that Bobirwa only had cases of suspected resistance. This may be associated with (i) the vector insecticidal intervention intensification strategy post 2012 in Bobirwa and its categorization as a malaria ‘hotspot’ area [
31] and (ii) the interventions are less frequent/intense than other malaria endemic areas (e.g. Okavango and Ngamiland) [
48]. Moreover, Simon et al
. [
31] reported public defiance in Bobirwa toward national intervention strategies (e.g. IRS), likely ‘delaying’
An. arabiensis resistance in the area. However, the data are only based on samples from one location (village) per district. As such, future work should consider monitoring susceptibility status of malaria vectors in more exhaustive human settlements receiving unique chemical intervention to establish other bio-physical factors contributing to insecticide resistance.
The results showed entire susceptibility to the organophosphate (malathion) and carbamates (propoxur, bendiocarb), irrespective of region. The country’s national vector control strategies are based on insecticides (pyrethroids and DDT) that target one site (voltage-gated sodium channel proteins), which may facilitate selection pressure for possible mutation. Therefore, it may be logical, from the perspective above to use insecticides with different modes of action (e.g. organophosphates and carbamates) on rotational/alternation to improve efficacy while simultaneously managing insecticide resistance [
49]. To err on the side of caution regarding the deployment of insecticides in microhabitats (e.g. human habitation structures), abiotic factors shown to influence mosquitocides’ efficacy [
50,
51] should also be considered during application. For example, temperature can interact with mosquito chemical intervention approaches (LLINs and IRS) within structures of ‘unstandardized’ thermal condition [
50]. As such, assessing temperature coefficient (TC) of pesticides prior to regional use is recommended, as insecticides with positive TC are likely less efficacious at elevated temperatures [
51,
52]. Future work on monitoring and evaluation of other
Anopheles vectors [
35] and mechanisms of resistance is warranted (although see Kgoroebutswe et al
. [
38]).
Assessment and selection of pesticides based on their time of action for vector control is an essential component that has a bearing in management of insecticide resistance. The KDT
50 determines the time that enables 50% of mosquito population to be knocked down by an insecticide. Although it may be necessary to opt for insecticides that are fast in action (shorter KDT), this can be overruled if induced insecticide resistance is observed. The results demonstrated that pyrethroids generally reported shorter KDT
50 than other classes tested. This is in keeping with Wakeling et al
. [
53] that this group of insecticide is fast in action. Regardless of their ability to knockdown mosquitoes within a short period of time, this group appeared not to be efficacious for
An. arabiensis, at least in Okavango and Ngamiland, indicating potential development of pyrethroid resistance. In contrast, malathion, propoxur and bendiocarb were generally observed to be slow to action (high KDT
50) across the study sites with mosquito vector susceptible to their discriminating dozes. Therefore, it may follow that, if mosquitoes do not show resistance, a fast-acting insecticide may be given priority of choice but with pyrethroid-resistance areas (as in Okavango), slow-acting insecticides (e.g. organophosphates and carbamates) may serve as alternatives. Hence, it may be important that vector response assessment on both the KDT
50 and insecticide susceptibility status be carefully considered and merged appropriately for future insecticide selection and subsequently managing resistance. Furthermore, alternative vector control strategies [
54], applied in an integrated and area-wide approach may help bridge resistance development.
Though pyrethroids were observed to be fast in action,
An. arabiensis displayed a compromised sensitivity to the insecticides, which has implications for future vector control strategies using this pesticidal group. This baseline assessment work advocates for continuous monitoring of insecticide resistance to all potential mosquito vectors in the country before conclusive recommendations on susceptibility status are made. Moreover, malaria elimination in the country and region is a priority that necessitates efforts in vector management to monitor insecticide resistance. This may be achieved using integrated approaches that complement the current vector management strategies in minimizing resistance and at the same time delivering environmental benefits [
55].
Conclusion
While requiring further investigation, the results suggest that
An. arabiensis, one of the important malaria vectors in the country [
35], may be showing a genetic drift towards resistance as reported in other southern African neighbouring countries [
56,
57]. The current study adds to other reports on insecticide resistance in Africa and may be extended to other disease vectors on fine-to large-scale susceptibility to insecticides for both the endemic and non-endemic districts across diverse landscapes [
43,
58]. With the resistance reported here, Botswana should integrate the current national intervention strategies with other approaches of vector management in minimizing resistance and simultaneously considering environmental benefits. This may include novel complimentary non-chemical ‘bio-friendly’ approaches targeting both immature and adult vector life stages [
59,
60]. Furthermore, public records of governmental, industrial and private insecticidal use and availability should be considered to aid delimit drivers of resistance development. This may help integrated vector management, frameworks for pest decision-making, continued insecticide resistance monitoring, and the implementation of insecticide resistance management strategies while maintaining biodiversity and essential ecosystem services [
61]. The country needs to be more conservative with the continuous use of pyrethroids especially in the Okavango delta, where insecticide resistance was evident and biodiversity sustenance is key for sustainable livelihoods and the tourism economy [
62].
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.