First report of AChE1 (G119S) mutation and multiple resistance mechanisms in Anopheles gambiae s.s. in Nigeria

Susceptibility and PBO synergist bioassays were done using 3–5 days old female Anopheles mosquito collected from Lagos State, Nigeria with WHO test papers DDT (4%), permethrin (0.75%), Bendiocarb (1%) and PBO (4%) according to standard procedures. The activities of cytochrome P450s, glutathione S-transferase and carboxylesterases were determined using biochemical assays. The presence of kdr-w, kdr-e and Ace-1R mutations were examined using molecular assays. Resistance to DDT and permethrin in An gambiae s.s from the four Local Government Areas (LGAs) was recorded while suspected resistance to bendiocarb was recorded in mosquitoes from Alimosho and Kosofe LGAs. PBO synergist reduced the knockdown time and also recorded significantly (P < 0.05) higher 24 hrs percentage mortality compared to non-synergized bioassays. Increased activities of detoxifying enzymes was recorded in wild mosquito compared to the insecticides susceptible laboratory strain and this was significant (P < 0.05) in P450s, esterase α and β. Kdr-w was detected in An. gambiae s.s from all the LGAs, kdr-e (L1014S) was detected in Alimosho, Kosofe and Ibeju-Lekki, while the Ace-1R gene was detected in Alimosho and Kosofe. Results from this study provide evidence for resistance of An. gambiae from Lagos State to multiple classes of neurotoxic insecticides with multiple resistance mechanisms to these insecticides.

Efficacy of PBO synergist on DDT and permethrin resistant An. gambiae s.s. Pre-exposure to PBO synergist significantly (P < 0.05) increases the 24hrs percentage mortality of An. gambiae s.l. in all the LGAs to DDT and permethrin. In addition, the estimated knockdown time for 50% and 95% of mosquitoes was also reduced when compared to that of the non-synergized bioassay, although only PBO + permethrin from Badagry was able to attain full susceptibility ( Table 1). The percentage knockdown at various time intervals also shows that Location  Detection of point mutation associated neurotoxic insecticides resistance in An. gambiae s.s. population from lagos state. Table 3 shows the allele frequency at the kdr (L1014S and L104F) and Ace-1 R   www.nature.com/scientificreports www.nature.com/scientificreports/ (G119S) loci of An. gambiae in Lagos State. Homozygotes and heterozygotes resistance kdr west (L1014F) genotype were detected in all the sampled LGAs, with allele frequency ranging from 0.37 to 0.5. The distribution of L1014F mutation within the surveyed location did not significantly depart from the Hardy-Weinberg equilibrium (P > 0.05). Kdr east (L1014S) was also detected in Alimosho, Kosofe and Ibeju-Lekki LGAs with allele frequencies of 37%, 29% and 17% respectively and only heterozygotes resistance genotype identified. AChE1 mutation was detected in An. gambiae from Alimosho (32%) and Kosofe (36%) LGAs, meanwhile no Ace-1 R mutation was detected in Badagry and Ibeju-Lekki LGAs. The observed genotypic frequencies from Ace-1 R mutation in Kosofe and Alimosho LGAs were significantly different from Hardy-Weinberg expectations (P < 0.05).

Discussion
The WHO and SDGs aim to reduce the global malaria prevalence by 90% in the year 2030, therefore, the regular monitoring of the limited available malaria control options including the use of insecticides needs to be harnessed strategically. IRS and ITNs are the two main insecticides-based strategies in malaria vector control 2 . The efficacy of the few Pyrethriods, DDT and carbamates insecticides currently approved for use in public health is critical to the sustainability of these strategies. In this study, high level of resistance to DDT and permethrin was recorded in An. gambiae s.l. from the evaluated LGAs of Lagos State. Several previous studies in different parts of Nigeria have recorded similar resistant level to different Pyrethriods and DDT 7,9,11,12 . Pyrethriods resistance recorded in these study areas could be detrimental to the effectiveness of LLINs as they are the only class of insecticide currently approved for use in LLINs. Though, the level of Pyrethriods resistance in this study area should not be allowed to affect the utilization of LLINs in these areas because it also offers partial protections to users by serving as barrier from mosquito bites thereby reducing the risk of infection 30

. Suspected resistance to bendiocarb was recorded in Alimosho and Kosofe
LGAs in this study. Another previous study has also reported resistance to propoxur in several parts of Lagos State 31 . Resistance to carbamates could have a detrimental effect on malaria vector being one of the few alternative available to widely used Pyrethriods especially with the widespread resistance to DDT and Pyrethriods.
Result from this study also highlights the relevance of PBO based control measures in DDT and Pyrethriods resistance management in malaria vector in Lagos State, Nigeria. PBO synergized bioassays did not only achieve faster knockdown rate and time in DDT and permethrin resistant An. gambiae s.s, it also recorded significantly higher 24hrs percentage mortality. Similarly, studies in several parts of Africa have also proven the efficacy of PBO synergist plus insecticides in resistant malaria vector management 24,28,29,[31][32][33] .
Elevated level of detoxifying enzymes (cytochrome P450, glutathione S-transferase, esterase α and esterase β) activities was observed in field population of An. gambiae s.s. when compared to that of the laboratory susceptible Kisumu strain in this study. Also, studies have linked the increased activities of detoxifying enzymes (P450s and GSTs) and/or mutation in some P450s and GSTs gene to Pyrethriods and other classes of neurotoxic insecticides resistance in malaria vector 7,16,[34][35][36][37][38] . Likewise increased activities of esterase α and esterase β have been implicated in resistance to different classes of insecticides 39,40 .
The kdr-w (L1014F) mutation was detected in all the study LGAs in this study. Previous studies in Nigeria 8,9,15,31 and several parts of West and Central Africa 16,41-43 have described similar mutation to Pyrethriods and DDT which confers resistance to the vectors. Finding from this study also showed the presence of kdr-e www.nature.com/scientificreports www.nature.com/scientificreports/ (L1014S) mutation in DDT and Pyrethriods resistant An. gambiae s.s. for first time Nigeria. L1014S mutation had been previously characterized as the East African type of kdr 14 but later study discovered that both knockdown mutation co-occur in Gabon and Uganda 18,19 . Recently however, L1014S mutation had been detected in malaria vectors from some parts of West Africa [44][45][46][47] . L1014F and L1014S mutations have been associated with DDT and Pyrethriods cross resistance in An. gambiae 13,14 though it has been argued that this mutation alone may not solely be responsible for this phenotypic response 48 . Highlighting the importance of multiple resistance mechanisms in the high level of resistance reported in An. gambiae s.s. to DDT and Pyrethriods in this study, the allele frequencies of L1014F recorded in this study is higher than what was previously reported in Senegal 49 but lower than that of Burkina Faso 42 . The observed genotypic frequencies of L1014F mutations from all the locations in this study did not depart significantly from the Hardy-Weinberg proportions. Previous studies from Cameroon and Burkina Faso reported varies departure from Hardy-Weinberg proportions with locations 42,50 .
The Ace-1 R (G119S) mutation was detected in An. gambiae s.s. from Alimosho and Kosofe LGAs of Lagos State, though the genotype frequency significantly departs from the Hardy-Weinberg equilibrium which maybe result from the excess heterozygotes resistant genotype in the An. gambiae s.s. population. Previous studies on malaria vectors in southern Nigeria have reported resistance to carbamates 10,31 but did not detect the Ace-1 R mutation that has been linked to carbamates and organophosphate cross resistance in malaria vectors. The detection of Ace-1 R mutation in An. gambiae s.s. in Lagos State could be detrimental to the utilization and efficacy of IRS, a major strategy in the control of malaria vector in Sub Saharan Africa. Previous studies have associated the extensive use of agricultural pesticides 42,51,52 and spread of resistance gene from neighbouring countries 53 with development of carbamates and organophosphates cross resistance, but this may not be applicable in this study. The two LGAs where Ace-1 R mutation was detected are densely populated with little or no agricultural activities. The development of Ace-1 R mutation in locations can be attributed to widespread use of dichlorvos (DDVP) an organophosphate pesticide for the control of mosquitoes and other household pests 54 .
Findings from this study shows that An. gambiae s.s. from Lagos State exhibits multiple resistance mechanism to the different classes of insecticides available for control. It is important that regular insecticides monitoring be carried out if the WHO and SDGs goals of 2030 is to be achieved. The use PBO and other synergists incorporated into LLINs and other malaria vector control strategies should be encouraged in this areas as metabolic resistance mechanism are also proven to contribute to high level of insecticides resistance reported in the study areas.

Materials and Methods
Study area and sample collection. The study was carried out in four Local Government Areas (LGAs) of Lagos State, two densely populated LGAs: Alimosho and Kosofe and two less densely populated LGAs: Ibeju-Lekki and Badagry. According to the National Population Commission (NPC) in 2006, Badagry has an estimated population of 237,731 spanning a 443 km² area. It is the second largest town in Lagos State and surrounded by lakes, creeks and island. The major occupations known include fishing farming and salt making which is due to the abundance of trees and ocean water. Alimosho has an estimated 1,319,571 inhabitants occupying a 138 km² area of and it is the largest LGA in Lagos State. Ibeju-Lekki located in the north eastern part in the Epe Division of Lagos State, Nigeria has a total area of 455 km 2 and a population of 117481 and Kosofe LGA with population of 66393 and an area of 81 km 2 (Fig. 1). collection of anopheles mosquito. Immature stages was collected from selected sites using "dipping" technique 55 , Anopheles mosquito eggs, larvae and pupa were retrieved from automobile tyre tracks, small pools www.nature.com/scientificreports www.nature.com/scientificreports/ and puddles. Immediately after scooping, the larvae were kept in well-labeled containers and subsequently transferred to the insectary at the Nigerian Institute of Medical Research where they were allowed to emerge into adults under standard insectary conditions. Susceptibility and synergistic assay. The tests were performed using WHO test filter papers impregnated with the selected insecticides and piperonyl butoxide (PBO) from the Vector Control Research Unit (VCRU), University Sains Malaysia (http:/www. inreskit.usm.my). Non-blood fed, two to three days old female Anopheles were exposed to DDT (4%), Permethrin (0.75%) and Bendiocarb (0.1%) using the WHO standard procedure. A total of 25 female mosquitoes were pre-exposed to 4% piperonyl butoxide (PBO) and this was replicated in four places. PBO treated mosquitoes were exposed to either DDT (4%) or Permethrin (0.75%) for another one hour, each experiment consisted of four replicates. Knockdown rates of mosquitoes were recorded at intervals for one hour. Mosquitoes were later transferred into holding tubes with untreated papers; allowed a 24-hour recovery period and supplied with a 10% sugar meal during this period after which mortality was recorded. All the bioassays were accompanied by negative control.
Mosquito identification. The genomic DNA of Anopheles gambiae that has been identified using morphological keys 56 was extracted according using protocols as earlier described 57 . Further, molecular identification of mosquito samples was carried out 57 . Four primers including; ME, AR, QD, UN, GA (Table 4) were used for Anopheles gambiae. This was done to identify sibling species of the An. gambiae complex. Digestion was achieved by utilizing 0.5 μl HhaI restriction enzyme and 10 μl of the PCR product from the reaction above. It was incubated at 37 °C for 24 hours and the PCR fragments were resolved on a 1.5% agarose gel stained with ethidium bromide and visualized under UV light 58 .
Knockdown resistance (L1014F and L1014S) characterization. L1014F mutations assay (kdr west) was carried out as described by 13 using the following primer pairs: Agd1, Agd2, Agd3 and Agd4 (Table 4). L1014S mutation assay (kdr east) was carried out as described by 14 . The primers Agd1, Agd2, Agd4 and Agd5 (Table 4) were used for the assay. The PCR condition for all assay includes an initial denaturation of 95 °C for 5 minutes, then 40 cycles at 95 °C for one minute, 48 °C for 2 minutes, 72 °C for 2 minutes and final extension at 72 °C for 10 minutes.
The PCR reactions were carried out in a total volume of 20 ul containing 12.5 µl of PCR master mix containing 1 × PCR buffer, 1.5 mM MgCl2, 0.2 mM of each dNTP, 0.4 µM of each primer, one unit of Taq polymerase (Solis BioDyne) and 1 µl of genomic DNA. 10ul of PCR product and 1ul of loading buffer was loaded into each sample well on a 1.5% agarose gel visualised by ethidium bromide stains under Ultra Violet light (UV light).
Metabolic enzyme activity assay. Eight adult female mosquito that were pre-exposed to WHO standard insecticide test paper were homogenized singly in 200 μl of cold distilled water in a 1.5 mL centrifuge tube. The  www.nature.com/scientificreports www.nature.com/scientificreports/ homogenate was centrifuged at 14,000 rpm for 20 seconds and the supernatant stored at −20 °C (Hemingway 1998). Three metabolic enzymes, Cytochrome P450 monooxygenase (P450s), Glutathione S-transferase (GSTs), and non-specific esterase (carboxylesterase) assay (COEs) (hydrolyzing αand βnapthyl acetate) were analyzed on single individuals of field collected Insecticide resistant females mosquitoes, and on the laboratory susceptible strain according to WHO protocols (Hemingway 1998). Mean absorbance values for each tested mosquito and enzyme were converted into enzyme activity and standardized based on the total protein amount. Glutathione S-transferase assay. This test was carried out in two replicates. 10 µl of mosquito homogenate were placed in separate well in microtitre plate and 200 µl of the GSH (reduced glutathione)/CDNB (1-chloro-2,4′-dinitrobenzene) working solution was then added. Three plate blanks containing 10 µl distill water and 200 µl of the GSH/CDNB working solution were used per microtitre plate as negative control. The test was then left at room temperature for 20 minutes and the absorbance value was read at 340 nm at end. Cytochrome P450 monooxygenase assay. Mosquito homogenate (2 µl) of were placed in separate wells of the microplate, 80 µl of 0.625 M potassium phosphate (pH 7.2) was added to each replicate. Then 200 µl of the mixture of 5 ml methanol solution of tetramethyl benzidine with 15 ml of 0.25 M sodium acetate buffer (pH 5.0) was added to each well, 25 µl of 3% hydrogen peroxide was also added to each replicate, the preparation was left for 2 hours at room temperature before reading of absorbance at 650 nm. Control was run at 20 µl of buffer instead of mosquito homogenate and the assay was carried out in duplicate. protein assay. Total protein was measured for each mosquito using Biuret test. All measurements were done in duplicate. Protein concentration in sample was calculated as absorbance of sample/absorbance of standard multiply by concentration of standard (60 g/dl). Enzyme activities were calculated as sample absorbance/g/dl of protein.
Statistical analysis. Susceptibility/resistance to test insecticides was categorized based on the 98-100% mortality criteria of mosquito which implies susceptibility, 80-97% mortality indicates suspected resistance that needs further confirmation through biochemical or molecular assays and <80% mortality implies resistance 60 . Regression probit was used to compute the KDT 50 and KDT 95 . Chi-square was used to compare percentage mortality between insecticide only, and PBO plus insecticide. Analysis of variance (ANOVA) was used to determine the difference in the activities of detoxifying enzymes in wild and laboratory susceptible strain of Anopheles mosquito and Duncan multiple range test was check for statistical similarities/differences between the locations. The frequency of kdr-w. kdr-e and Ace-1 R mutations in An. gambiae s.s. population were compared to Hardy-Weinberg expectations using Pearson's chi-square test. All data analyses were computed using Microsoft Excel version 2016 and IBM SPSS Statistics 23. P-value of <0.05 was considered statistically significant.