Background
For about two decades, African countries have been intensifying malaria vector control activities. Insecticide-treated nets (ITNs/LLINs) and indoor residual spray (IRS) have been scaled up as major tools for malaria vector control. This has resulted in a substantial reduction in malarial morbidity and mortality [
1,
2]. However, the rapid expansion of insecticide resistance observed in major African malaria vectors,
Anopheles gambiae s.l. and
Anopheles funestus s.l. could jeopardize this achievements [
2,
3]. This has been further accelerated by the use of similar insecticide classes against crop and livestock pests, thus, exerting more selection pressure to the vectors and hence worsening the situation [
4]. Insecticide resistance monitoring in malaria vectors has been carried out in most of the malaria endemic countries [
4]. Many countries in Africa have reported insecticide resistance in the main malaria vectors to at least two of all four major insecticide classes (pyrethroids, organochlorine, carbamates, and organophosphates) recommended for vector control. Of all the classes of insecticides, high levels of resistance have been recorded against pyrethroids, which is the only class of insecticides used in ITNs/LLINs [
2,
3].
In the mainland Tanzania, efficacy and effectiveness trials of ITNs against malaria vectors were conducted between early 1980s and late 1990s [
5], followed by the national wide scaling up of ITNs that started in 1996 [
6,
7]. IRS programmes in the country particularly in the North Western zone around the Lake Victoria, commenced in 2007. Those programmes started with pyrethroid-based insecticide (lambdacyhalothrin) in 2007, followed with bendiocarb (carbamate) in 2011 then pirimiphos methyl (organophosphate), from 2014 to date [
8‐
11]. Since late 1990s there have been some large-scale (having 5–29 study sites) [
6,
11‐
16] and small-scale (having one study site) [
8,
17‐
30] studies on insecticide susceptibility and mechanisms of resistance monitoring in malaria vectors conducted across several sentinel sites in the mainland Tanzania. Although those studies have documented insecticide resistance and/or resistance mechanisms to different insecticides, there is no collective information regarding dynamics and monitoring of insecticide resistance in the country. Therefore, we systematically analysed information from 30 published articles on the susceptibility of malaria vectors to various insecticides and resistance mechanisms across mainland Tanzania. The aim was to elucidate dynamics and monitoring of insecticide resistance in malaria vectors across the country. This information will help revealing the extent of insecticide resistance, and the major challenge in the current malaria vector control strategies. The information will be useful in directing future vector control strategies.
Discussion
This review aimed at demonstrating dynamics and monitoring of insecticide resistance in malaria vectors in mainland Tanzania between 1997 and 2017. A total of 30 published articles reporting results from 35 study sites were reviewed. Most studies were conducted from Eastern, North-Eastern highlands and North-Western parts of the mainland Tanzania. Commencing from late 1990s, number of study sites and types of insecticide classes for controlling disease vectors have been expanding. Similarly, the increase in resistance of malaria vectors to widely studied insecticides (pyrethroids and DDT) was observed. Whereas, most studies on organophosphates and carbamates started after the year 2010. The wide spread of resistance observed between 2011 and 2017 in mainland Tanzania is a result of development of insecticide resistance in malaria vectors in areas which have previously recorded full susceptibility. For instance, 50% (7/14) and 43% (3/7) of study sites, which have recorded An. gambiae s.l. resistance to pyrethroids and DDT respectively by 2017, showed high susceptibility in studies conducted before year 2009.
Moreover, an increasing number of resistance monitoring sites over time has also detected new sites with insecticide resistance in malaria vectors. The mainland Tanzania consists of about 187 districts [
40] and insecticide resistance monitoring studies have so far been conducted in 18% of total districts. Increasing the number of study sites and regular monitoring of susceptibility of mosquitoes to insecticides is needed to give a broader picture of the trend of insecticide resistance status in the country.
By 2017, An. gambiae s.l. resistance to all four recommended classes of insecticides used in malaria vector control had been reported from various parts of the country. Efficacy of pyrethroids and DDT seems to be greatly threatened by the observed resistance. Although DDT is not currently used in vector control, susceptibility status of malaria vectors to this insecticide should be known because of its cross resistance with pyrethroids. The observed DDT resistance might be due to the cross resistance with pyrethroids or historical use of DDT in the country. On the other hand, susceptibility studies on An. funestus s.l. show they are highly resistant to pyrethroids and DDT while highly susceptible to carbamates and organophosphates. This implies that carbamates and organophosphates remain good candidate insecticides for controlling An. funestus s.l. in the country.
In this review, high frequencies of both west and east African
kdr mutations in
An. gambiae s.l. were observed in Eastern and North-Western parts of the country. Similarly, pyrethroids and DDT resistance had been detected in the same areas. This is due to the cross resistance pattern existing among these compounds that share the common target site that is voltage gated sodium channels (VGSC) of insect nerve cell [
41]. However,
kdr mutations are strongly associated with DDT and less associated with pyrethroids resistance [
42]. Different studies in the country have recorded resistance to either pyrethroids alone, DDT alone or resistance to more than one compound with unrelated target sites (e.g. pyrethroids or DDT with organophosphates or carbamates). This situation might be triggered by metabolic resistance of which cross resistance is difficult to generalize [
43]. For instance, P450 s enzymes in some cases have shown specificity by acting differently in compounds within the same insecticide class [
43‐
45]. In other cases P450s enzymes have metabolized compounds between different classes of insecticides [
44,
45]. Moreover, a metabolic resistance study conducted by Kisinza et al. [
11] across the country has shown elevation of detoxification enzymes associated with different classes of insecticides.
Areas, which have recorded high
kdr frequencies in this review, are dominated by
An. gambiae s.s., while other areas with high phenotypic resistance and low or no
kdr mutations are being dominated by
An. arabiensis [
8,
28,
30,
32,
36]. This suggests that
An. arabiensis resistance to insecticides mostly depend on metabolic path-way. According to Rusell et al. [
46],
An. arabiensis, is the predominant species of the
An. gambiae s.l. over
An. gambiae s.s. in most areas of the country. Their exophilic and exophagic behavior have reduced their mortality from exposure of the indoor insecticide applications of LLINs and IRS, thus, increase in their dominance. This increase dominance of
An. arabiensis over
An. gambiae s.s. in the country should be coupled with deployment of new tools for reducing residual malaria transmission.
There are limited studies on molecular mechanisms of metabolic insecticide resistance in
An. gambiae s.l. in Tanzania. A study by Nkya et al. [
15] revealed over transcription of several genes in
An. arabiensis as compared to the susceptible laboratory colony, mostly in agricultural areas. The most significantly over transcribed genes in
An. arabiensis were P450s: CYP6P3 and CYP9J5 associating with pyrethroid and DDT resistance in Dar es Salaam (urban); and CYP9J4 and CYP6P1 genes in associating with pyrethroid resistance in Hai district (agricultural areas) [
15]. Conversely, another study by Matowo et al. [
37] conducted in Moshi district (agricultural areas) near Hai district has recorded over transcription of different genes in
An. arabiensis, which include over transcription of CYP4G16 and ABC transporter genes in
An. arabiensis associating with pyrethroids resistance [
37]. Nevertheless, over transcription of CYP4J10 and CY6P3 genes in
An. gambiae s.s. associating with DDT resistance was also observed in urban areas [
39]. These findings are contrary to studies conducted elsewhere in Africa where over transcription of CY6P3, CYP6M2, CYP6Z1, CYP6Z2 and Cyp9K1 genes associating with pyrethroid or DDT resistance was observed [
47‐
52]. This indicates that different genes, depending on the environment in which mosquitoes are exposed, might mediate metabolic resistance in
An. gambiae s.l. In another study, over transcription of CPAP3-A1b genes encoding cuticle protein was also recorded in agricultural areas in the country and were associated with pyrethroid resistance [
15]. Over transcription of these genes was also recorded elsewhere in West Africa [
53]. There is a need to explore further involvement of other metabolic and cuticular resistance genes associated with pyrethroid resistance in Tanzania where pyrethroid LLINs is scaled up as the main malaria vector control tool. So far, no studies in the country that have reported mechanisms of resistance in the
An. funestus s.l.
There are several factors, which might have contributed to the intensified development, spread and variations in malaria vectors resistance to various insecticides across different sites in the country. These factors include historical use of pesticides, contemporary use of pesticides in agriculture, non-pesticides pollutants and insecticides use in vector control.
The historical role of pesticides usage in the development of insecticide resistance in malaria vectors
Between 1940s and 1960s malaria vector control by IRS and aerial spray of organochlorines (DDT, dieldrin and lindane) were conducted in Dar es Salaam [
5]. As part of global malaria eradication campaign between 1950 and 1960, pilot projects of IRS with DDT and dieldrin were conducted in Pare-Taveta border. Some other low scale IRS activities were conducted in selected areas from the five region of mainland Tanzania [
5]. From 1970s to 1980s organochlorines pesticides were also vastly donated to the country for agricultural pest control [
54]. Areas in the country with the known history of massive utilization of these pesticides include: Dar es Salaam, Lake Victoria basin and Moshi [
55]. Use of DDT and other persistent organochlorine insecticides were banned in the country for public health use in 1980s and for agricultural use in 1997 [
56]. This was due their hazardous effect on human beings, biodiversity and environmental concerns [
56]. Following banning, there were stockpiles of improperly kept pesticides which had contaminated the environment around storage sites [
54]. Some of the nine major unattended storage sites are located in the six regions of the main-land Tanzania [
56]. Several studies conducted around some storage sites and other areas which have largely utilized these pesticides, have detected their residuals in water, soil and plants [
24,
55‐
61]. These insecticide residuals may also be exposed to mosquitoes by leaching to their breeding sites [
24]. Currently, there has been a rapid development of DDT and pyrethroids resistance in some parts of the Lake Victoria zone after deployment of pyrethroids intervention (LLINs and IRS). This might be due to the existence of small population of resistant mosquitoes from DDT residuals exposures followed by pyrethroid exposures. DDT and pyrethroids share the same target site, hence cross resistance between these insecticides [
24]. However, the effect of the DDT contamination around former storage sites on malaria vector insecticide resistance has not been established. Therefore, investigations of pyrethroids and DDT resistance around those former pesticide storage sites are needed, to evaluate their contribution to the current DDT and pyrethroids resistance in malaria vectors.
The role of contemporary usage of pesticide in agriculture in the development of insecticide resistance in malaria vectors
Importation of pesticides increased enormously in Tanzania after trade liberalization of agricultural inputs in 1990s [
62]. Pesticides imports increased from about 500 to 12,000 metric tons between 2000 and 2014, respectively [
63]. More than 80% of pesticides have been used for agricultural and veterinary purposes [
54]. Large amount of pesticides have been used in the country to control pests in horticulture, floriculture, cotton, cashew nuts, coffee, sugar cane, legumes and rice cultivation [
64]. Pesticides frequently used in the country include: pyrethroids (alpha cypermethrin, deltamethrin, lambdacyhalothrin and permethrin), organophosphates (chlorpyrifos, dimethoate, profenofos, pirimiphos methyl, diazinon, malathion, fenitrothion and glyphosate) organochlorines (endosulfan, chlorothalonil) carbamates (carbofuran, carbaryl, aldicarb and mancozeb) [
20,
62‐
71]. These insecticide classes are the same as ones used in public health vector control, hence accelerating selection of resistance in malaria vectors. Accordingly, several studies in Africa have shown an association between agricultural practices and insecticides resistance in malaria vectors, reviewed in [
72]. Additionally, similar observations have been reported in Tanzania [
15,
20,
24,
37,
39]. Studies on molecular mechanisms of resistance conducted in the areas with intensive agricultural practice in Lower Moshi and Hai have reported over transcription of several genes associated to pyrethroid resistance compared to non-agricultural areas [
15,
37]. Moreover, a study by Kisinza et al. [
11] conducted in Arumeru, has shown the high resistance of
An. gambiae s.l. to pyrethroids and pirimiphos methyl (organophosphate). This
An. gambiae s.l. resistance to pyrethroids and pirimiphos methyl might be a result of the long time large scale use of pyrethroids and organophosphates used in agricultural pest control in Arumeru [
67,
69,
73‐
76].
The role of non-pesticide pollutants in the development of insecticide resistance in malaria vectors
Mosquitoes breeding sites are exposed to various pollutants from different human activities.
Anopheles gambiae s.l., which normally breed in clean water, have currently been found in polluted breeding sites of the urban and rural areas in Tanzania [
77,
78]. Breeding sites of mosquitoes have been found to be contaminated with several pollutants including organic pollutants (sewage, rotting plants, domestic and animal wastes and other industrial organic chemicals); also inorganic pollutants such as heavy metals (Zn, Au, Mn, Hg, Pb, Cu, Fe, Cd and Co) [
79,
80] and others (industrial and agrochemicals). Pollutants affect the physical–chemical parameters of breeding sites, consequently may affect mosquitoes’ genetic makeup [
81‐
84]. These physical–chemical parameters include pH, conductivity, turbidity, total dissolved solids (TDS) and total suspended solids (TSS), Na
+, Ca
2+, K
+, NH
4+, SO
4
2−,
PO
42−, NO
3−, NO2
−, H
2O
2, Cl
− and F
− [
78,
81]. Limited studies exist on the influence of those pollutants to the development of insecticide resistance in malaria vectors [
79‐
81,
84,
85]. Some pollutants or the change in the levels of physical–chemical parameters in mosquitoes’ breeding sites have shown toxic effect to larval stages [
80,
86]. Mosquito larvae surviving such environment have shown tolerance against the effect of those toxic pollutants and subsequently develop insecticide resistance in adult stages [
84,
87].
Vividly, a study by Tene Fossog et al. [
86] has demonstrated that
Anopheles coluzzi larvae developed tolerance after being exposed to the lethal dose of NH
3. Moreover, other studies have shown that Anopheles larvae exposed to heavy metals or disinfectants (H
2O
2 and/or soap) developed insecticide tolerance in their adult stages [
80,
87]. Some pollutants and levels of physical–chemical parameters in
An. gambiae s.l. breeding sites have shown to be positively associated with their insecticide resistance [
81,
83,
84,
88]. These pollutants and physical–chemical parameters include greases, oils, TDS, NH
3, NH
4+, SO
42−, PO4
2−, NO
3−, NO
2− and F
−. A study by Emidi et al. [
78] in Muheza district, Tanzania, correlating physico-chemical parameters in mosquitoes breeding sites and larval density, have reported for the first time the occurrence of
An. gambiae s.l. in polluted habitat in rural areas. However, the influence of these pollutants to the insecticide resistance across the different ecological zones in the country is yet to be investigated.
The role of public health insecticides use in the development of resistance in malaria vectors
The use of LLINs and IRS are major vector control strategies currently utilizing substantial amount of insecticides [
89]. Scaling up of ITNs in mainland Tanzania started in the late 1990s after their efficacy and effectiveness trials between 1980s and early 1990s. Scaling up of ITNs has been going on through different programmes including: Kilombero Net Project (KINET) in 1996, Social Marketing for ITNs Project (SMITN)/Strategic Social Marketing for Expanding the Commercial Market (SMARTNET) in 1998 [
6]. Others are Tanzania National Voucher Scheme (TNVS) in 2004, Under Five Catch-Up Campaign (U5CC) in 2008 and Universal Catch-Up Campaign (UCC) since 2010 [
7]. Finally, Keep-Up projects started from 2013 to date [
90]. From those projects ITNs coverage of one net per household raised from about 23% in 2004 to more than 91% in 2011. On the other hand, IRS scaling up in the lake zone started in 2007 with the successive replacement of insecticide classes. Insecticides used on IRS include lambdacyhalothrin since 2007, bendiocarb since 2011 and pirimiphos methyl since 2014 to date. It is estimated that, the quantity of insecticides imported in Tanzania for ITNs and IRS had approximately increased from 10 metric tons in 2007 [
91‐
94] to 311 metric tons in 2012 [
7,
91,
94]. This amount of insecticides used in both ITN and IRS is about 2 to 3% of total imported pesticides, which is relatively very small compared to 80% of total insecticide used agriculture and veterinary pest control [
63]. However, mosquitoes’ contact with insecticides in ITNs and IRS might be high because these strategies target their indoor resting and blood feeding behaviour. Evidently, scaling up of ITNs and IRS activities, as well as the use of aerosols, fumigations and coils, particularly among urban populations, are strongly associated with rapid expansion of insecticide resistance in malaria vectors since 2004. The insecticide-based approaches might have exerted selection pressure on Tanzanian populations of both
An. gambiae s.l. and
An. funestus s.l. Indeed,
An. gambiae s.l. has currently developed resistance to at least one compound in all four insecticides classes in Muleba. This area has a long history of utilizing both ITNs and IRS to reduce existing high malaria burden [
11]. Therefore, mitigation of resistance situation in the country needs involvement of different authorities from agricultural, public health and environmental sectors in order to preserve effectiveness of insecticides for vector control.
Authors’ contributions
DJM: conceived, designed, did literature search; participated in data analysis and interpretation; wrote the first draft of the manuscript. JJM, APB, and WK critically revised the manuscript. All authors read and approved the final manuscript.