Study site
The insecticide susceptibility study in 2018 was conducted at sentinel sites in seven of the nine malaria-endemic regions. The seven regions (sentinel village sites in brackets) that were included in this study were Oshana (Onamutai), Oshikoto (Oniimwandi), Otjozondjupa (Otjituuo), Ohangwena (Okanghudi), Kunene (Otjimuhaka), Omusati (Omiindaba), and Kavango West (Mukekete) (Figure 1). Thereafter, PBO synergist assays were conducted in sentinel sites from six malaria-endemic regions, which included the Oshana region (Onamutai), Oshikoto (Oniimwandi), Otjozondjupa (Otjituuo), Kavango West (Mukekete), Kavango East (Shadikongoro), and Zambezi (Sibbinda) (Figure 1). A sentinel site in each region was a malaria hotspot determined by the NVDCP based on malaria risk maps generated from malaria surveillance data. Each sentinel site included a sampling area of approximately a 10-kilometer radius from the central location of the village.
Collection and rearing of larvae
Larvae and pupae were collected from the study sentinel sites from March to April 2019 for insecticide assays. For PBO synergist assays, larvae and pupae were only collected from sites in the Oshikoto and Oshana regions during the peak malaria season (March to May 2020), and COVID-19 restrictions resulted in larval collections being interrupted from March to April 2020 and collection from sites in the Oshikoto, Otjozondjupa, Kavango West, Kavango East, and Zambezi regions from March to April 2021. Larval sampling was performed by using larval scoops dipped at the edges of breeding sites 17, 18. Specific sampling sites in each region were chosen by the National Vector-borne Diseases Control Programme (NVDCP) based on malaria risk maps generated from malaria surveillance data. Each sentinel site included a sampling area of approximately a 10-kilometer radius from the central location of the village. Anopheles mosquito larvae and pupae were reared to adulthood at the NVDCP insectary in Oshakati. The larvae were fed a mixture of dog biscuits and yeast powder, while the adults were fed on 10% sugar solution ad libitum. Susceptible An. arabiensis mosquitoes (KGB strain) were used for the control assays. The temperature and humidity within the insectary were maintained between 28°C and 29°C and 70% and 80%, respectively. Female Anopheles mosquitoes aged 3–5 days that had never had a blood meal were used for the assays.
WHO insecticide susceptibility bioassays
Using WHO-recommended procedures 13, 17, a complete test included four replicates using 20 adult wild-caught specimens per tube and two control replicates each using 20 susceptible An. arabiensis mosquitoes per tube. In total, 1680 female Anopheles were used for WHO bioassays. Adult wild-caught (as larvae) female mosquitoes, 3–5 days old, were exposed to DDT (4%) and deltamethrin (0.05%) insecticide-impregnated papers in WHO tubes for one hour. Two control tubes were run in parallel at any time during the testing. Following exposure, mosquitoes were provided with a 10% sugar solution and maintained at 28°C to 29°C with 70% to 80% humidity. Mosquito mortality was observed and recorded after 24 hours. The observed mortality of the test sample was calculated by summing the number of dead mosquitoes across all exposure tubes and then expressing this as a percentage of the total number of exposed mosquitoes. Mosquitoes were stored separately on silica gel for molecular analysis.
PBO synergist assays
WHO Whatman papers impregnated with the PBO synergist were used to perform the synergist assay. PBO-synergist assays were conducted using WHO-recommended procedures 13, 17. A complete test included four holding tubes (each with 20 adult female mosquitoes) and five exposure tubes. The sample size was ~60 mosquitoes for sentinel sites and 20 An. arabiensis for the control replicates. A total of 120 female mosquitoes were used for the synergist-insecticide monitoring study from the Oshana (Onamutayi village) and Oshikoto regions (Oniimwandi and Omandongo village) in 2020. In 2021, 300 female mosquitoes were used from Oshikoto (Oniimwandi), Otjozondjupa (Otjituuo), Kavango West (Mukekete), Kavango East (Shadikongoro) and Zambezi (Sibbinda). For each sentinel site, a total of only 60 mosquitoes were exposed to the PBO synergist.
Mosquitoes were exposed in three WHO tubes: i) to a dosage of 4% "PBO only", ii) pre-exposed to 4% "PBO" for one hour before being exposed to 0.05% Deltamethrin for one hour (PBO +Deltamethrin), and iii) exposed to 0.05% "Deltamethrin only". A control assay using a tube lined with non-insecticide-impregnated paper was run in parallel. Test mosquitoes were maintained at 28° to 29°C and a relative humidity range of 70 to 80% during exposure and holding periods. Test mosquitoes were provided with 10% sugar solution ad libitum during the holding periods. Mosquito mortality was observed and recorded after 24 hours. The observed mortality of the test sample was calculated by summing the number of dead mosquitoes across all exposure tubes and then expressing this as a percentage of the total number of exposed mosquitoes.
Molecular identification of Anopheles gambiae s.l.
Dead and live mosquitoes were separated and labelled at the end of each assay. All mosquitoes were morphologically identified using keys 19 as An. gambiae s.l. All samples for 2018 and a subsample from 2020 underwent molecular species identification using polymerase chain reaction (PCR) diagnostic 20.
Analysis
WHO guidelines 13 were used to interpret the susceptibility status of An. gambiae s.l. mosquitoes after 24 hours of exposure to insecticides. A mean mortality of >98% indicated susceptibility to the insecticide, while a mean mortality between 90% and 98% indicated reduced susceptibility to the insecticide, and a mean mortality of <90% indicated resistance to the insecticide. The Kruskal‒Wallis test was used to compare mosquito mortality between the sentinel sites. Pearson's chi-square test was used to compare the Anopheles mosquito species composition between the regions. WHO guidelines 13 were used to interpret the effect of the PBO synergist on mosquitoes. "If the mean mortality in the "insecticide only" samples is ≥ 90%, the effect of PBO cannot be reliably assessed. If the mean mortality in the "insecticide only" samples is < 90%, the effect of PBO can be interpreted according to the following criteria:
- Complete restoration of susceptibility (mitigation of resistance) by pre-exposure to PBO (i.e., ≥ 98% mean mortality in the “PBO followed by insecticide” samples) implies that a mono oxygenase-based resistance mechanism fully accounts for the expression of the resistant phenotype in the test population.
- Partial restoration of susceptibility by pre-exposure to PBO (i.e., mean mortality in the "PBO followed by insecticide" samples is greater than mean mortality in the "insecticide only" samples but < 98%) implies that a mono oxygenase-based resistance mechanism only partially accounts for expression of the resistant phenotype and that other resistance mechanisms are likely to be present in the test population.
All insecticide resistance analyses are presented for An. gambiae s.l., with the exclusion of non-amplifiers.
Anopheles species compositions
Morphologically, all species in the bioassays were identified as An. gambiae s.l. Of the mosquitoes tested for insecticide susceptibility (n= 1680), 1582 (94%) were successfully amplified using the An. gambiae s.l. diagnostic assay 20. An. gambiae s.s. made up the majority of amplified specimens (62%), and An. arabiensis came in second (38%). Non-amplified samples (n = 98) were not included in downstream analyses.
Due to capacity and funding limitations, only 72% (n = 309) of the PBO synergist assay samples from 2020 were subjected to molecular analysis. Approximately 84.3% (n = 200) of the samples molecularly identified from the Oshana region (n = 237) were An. arabiensis, and 15.2% (n = 36) did not amplify with PCR (non-amplifiers), with a single An. quadriannulatus specimen. From the Oshikoto region, 73.7% (n = 53) were An. arabiensis, and 26.4% (n = 19) were non-amplifiers. Samples from 2021 were not molecularly analysed. All results are presented for An. gambiae s.l. Overall, An. arabiensis (n = 253, 81.8%) was the dominant vector recorded.
A higher proportion of An. gambiae s.s. mosquitoes were sampled at most sites: 60.9% (n = 214) in Otjozondjupa, 80.3% (n = 196) in Oshana, 59.8% (n = 183) in Omusati, and 100% (n =113) in Kunene. Approximately equivalent proportions of An. gambiae s.s. and An. arabiensis were observed in Oshikoto and Ohangwena (50.3% and 49.4% An. gambiae s.s., respectively), while Kavango West was the only site with more An. arabiensis (56.8%; n = 54) (Figure 2). There was a significant difference in the species composition of the An. gambiae s.l. members between the regions (Pearson χ2, DF = 6, P = 0.001).
WHO Susceptibility bioassays
After conducting the WHO tube bioassays, molecular analyses were performed on a total of 1582 female An. gambiae s.l. mosquitoes. Of those, 980 were An. gambiae s.s. and 602 were An. arabiensis mosquitoes.
Table 1. An. gambiae s.s. and An. arabiensis insecticide susceptibility results using the WHO tubes in 2018.
The results demonstrate that An. gambiae s.s. was resistant to deltamethrin 0.05% in Oshikoto, Kunene, and Kavango West (79%, 86%, and 67%, respectively) and that An. gambiae s.s. was less sensitive to deltamethrin in Omusati (97%) and Ohangwena (94%), as shown in Table 1. An. gambiae s.s. showed reduced susceptibility to DDT in Kavango West (91%), while they showed full susceptibility (100%) in the rest of the regions. An. arabiensis showed resistance to deltamethrin 0.05% in the Oshikoto region (82%), reduced susceptibility in Kavango West (96%), and full susceptibility (100%) in the rest of the regions. There is a reduced susceptibility of An. arabiensis to DDT 4% in the Kavango West (96%) region, whereas An. arabiensis vectors from the other regions showed full susceptibility (100%). A Kruskal‒Wallis H test was performed for these data from 2018 and suggests that there is a difference in mean mortality by DDT and deltamethrin between the regions (P = 0.008; DF = 8).
PBO-Synergist Assay
In 2020, deltamethrin (0.05%) alone induced 93.3% and 95.0% mortality in Oshana and Oshikoto, respectively. After pre-exposure to the PBO synergist (4%) followed by exposure to deltamethrin (0.05%), the resistance status of mosquitoes in the Oshana and Oshikoto regions was completely restored to 100% susceptibility (Figure 3). Mortality after pre-exposure to the PBO synergist increased by 6% in Oshana (from 93.3% to 100%) and by 5% in Oshikoto (from 95% to 100%). In 2021, deltamethrin (0.05%) alone induced 95.0%, 90.0%, 70.0%, 80.0%, and 75.0% mortality in Otjozondjupa, Oshikoto, Kavango East, Kavango West, and Zambezi, respectively. After pre-exposure to the PBO synergist (4%) followed by exposure to deltamethrin (0.05%), the resistance status of mosquitoes was completely restored to 100% susceptibility in all five regions (Otjozondjupa, Oshikoto, Kavango East, Kavango West, and Zambezi). The results also indicated that deltamethrin resistance in the Oshikoto region increased by 5% from 2020 to 2021 (from 95.0% to 90.0% mortality) (Figure 3).