Introduction

Aedes albopictus Skuse, 1895 (Diptera: Culicidae)1, also called the Asian tiger mosquito, is a widely distributed and acknowledged dangerous vector of many arboviruses, including chikungunya virus (CHIKV), dengue virus (DENV), yellow fever virus, and Zika virus (ZIKV)2. Ae. albopictus is highly adapted to different ecosystems and therefore tends to develop higher densities, which will exacerbate the spread of the virus. This species, especially in Guizhou province has been identified as the major vector of dengue. High densities of Ae. albopictus were monitored from 2016 to 2020, with MOIs ranging from 5.82 to 7.95 and sting indexes ranging from 11.69 to 17.01 mosquitoes/person · h3.

Controlling the density of vector insects significantly reduces the prevalence of diseases such as dengue fever because of the inadequacy of treatments and vaccines4. Chemical insecticides, including pyrethroids, carbamates and organophosphorus insecticides, are widely used to reduce the density of Ae. albopictus larvae and adults, preventing the transmission of arboviral diseases, especially during disease outbreaks in China5,6,7. Pyrethroids have been the most widely used insecticides in government campaigns and by citizens since the 1980s due to their low toxicity and high efficiency6,8,9,10. However, years of preference have led to resistance to pyrethroids in most provinces and cities in China11,12,13. Insecticides resistance poses a significant threat to mosquito control efforts, and there is a risk of its increasing. The Global Plan for Insecticide Resistance Management aims to prevent such incidents from occurring14,15. Understanding the insecticide-resistance and its molecular mechanism is crucial for the effective implementation of insecticides. However, the resistance level along with the mechanisms involved in Ae. albopictus population of Guizhou Province are poorly known.

Ae. albopictus populations have demonstrated multiple resistance mechanisms under sustained selection pressure from different insecticide species16, including target insensitivity, behavioral resistance, and metabolic detoxification. Pyrethroids affect voltage-gated sodium channels (VGSC), disrupting the transmission of electrical signals and causing immediate knockdown and eventual death of the mosquito17. Target insensitivity is caused by mutations in the insecticide target site, specifically in the VGSC gene18. These mutations alter the configuration of the site, weakening the binding of pyrethroid insecticides to the sodium channel, resulting in target resistance.

Amino acid substitutions in VGSC lead to resistance to pyrethroid insecticides, called knockdown resistance (kdr)19. Kdr mutations have been associated with pyrethroid resistance in Anopheles sinensis Wiedemann, Anopheles gambiae, Culex pipiens pallens, and Aedes aegypti for several decades20,21,22,23. Since 2011, the F1534C mutant allele was first reported by Kasai et al. in Ae. albopictus from Singapore24. Additionally, F1534S/L/R/W mutations have also been detected in Ae. albopictus24,25,26. Furthermore, I1532T and V1016G/I mutations have also been identified26,27,28,29. In addition, it has been confirmed that the F1534S allele is associated with pyrethroid resistance25,30. The mutant alleles V1016G, F1534S and F1534C have been shown to be associated with strong pyrethroid resistance phenotypes9,29. Moreover, several mutations acting synergistically may lead to an increased extent of resistance16. However, it is worth noting that many studies on Ae. albopictus kdr mutations have been conducted mainly in southern and coastal cities of China. Guizhou Province is an inland city located in southwest China. Although resistance to pyrethroids has been found in recent years, there have been few studies on kdr mutations.

Ae. albopictus is an invasive species with a distribution that covers large parts of China. Its range is gradually expanding from the southeast to the west and north regions31. The susceptible status of Ae. albopictus has been reported in many provinces throughout China25,32,33. The objectives of this study were to determine the insecticide resistance status of Ae. albopictus populations from dengue surveillance areas in Guizhou Province to pyrethroid insecticides (deltamethrin, beta-cypermethrin, permethrin) and the corresponding kdr mutation. Additionally, we analyzed the correlation between pyrethroid resistance and kdr mutations.

Results

Larval resistance bioassays

Larval bioassays revealed varying degrees of resistance to the three pyrethroid insecticides (deltamethrin, permethrin, beta-cypermethrin) among nine populations of Ae. albopictus (RR50 ≥ 5) (Table 1). The LC50 values of laboratory sensitive strains of Ae. albopictus against the three above insecticides were 0.00045 mg/l, 0.00129 mg/l, and 0.00073 mg/l, respectively. The LC50 range was 0.007 mg/l (PZ)–0.053 mg/l (ZY), and RR50 was 15.56–117.78 after exposure to deltamethrin in all populations; 0.022 mg/l (PZ)-1.822 mg/l (XY), 17.05–1458.91 to permethrin; 0.005 mg/l (PZ)–0.045 mg/l (ZY), 6.85–61.64 to beta-cypermethrin LC50. All populations tested exhibited high resistance (RR50 ranging from 10.96 to 1458.91), except for the PZ population (RR50 = 6.85) to the three pyrethroid insecticides.

Table 1 Sensitivity of Ae. albopictus larvae to deltamethrin in nine different populations in Guizhou Province, China.
Full size table

Adult resistance bioassays

Overall, all mosquito populations showed resistance to pyrethroid insecticides, with mortality rates ranging from 0.99 to 43.69% (Fig. 1). Among the three pyrethroid insecticides tested, Ae. albopictus exhibited higher resistance to permethrin (24-h mortality range 0.99–23.23%) than to deltamethrin (1.00–26.42%) and beta-cypermethrin (1.00–43.69%). The mortality rates for deltamethrin (1.00%, 2.00%), permethrin (1.01%, 2.00%), and beta-cypermethrin (3.00%, 3.00%) were much lower in the XY and JS populations of Ae. albopictus, indicating higher resistance.

Figure 1
figure 1

Mortality of Ae. albopictus field populations after 24 h exposure to three pyrethroids. Solid line represents mortality at 98%, dashed line as 90%.

Full size image

Frequency of kdr alleles and genotypes in field populations of Ae. albopictus

Partial sequences of structural domains II and III of the VGSC gene were obtained from 2626 mosquitoes exposed to three pyrethroids. Non-synonymous kdr mutations were found at codon 1016 in domain II and codons 1532 and 1534 in domain III of the VGSC gene (Fig. 2). No synonymous mutations were recorded in this study.

Figure 2
figure 2

Sequence chromatograms of Ae. albopictus populations in nine dengue surveillance areas in Guizhou Province. (a–c) at locus 1016, (d–f) at locus 1532, and (g–o) at locus 1534.

Full size image

At codon 1016, two alleles were identified: the wild-type allele GTA/V and the mutant allele GGA/G (Table 2, Fig. 3). The allele V1016G was found in nine populations with a frequency of 13.86%, of which the highest frequency was 23.65% in GY. There existed a total of three genotypes as follows: wild-type genotype V/V (75.67%), wild-type mutant heterozygote V/G (20.94%) and mutant homozygous G/G (3.39%) (Table 3, Fig. 4). At codon 1532, the wild-type allele ATC/I (99.47%) was present in all populations, while the mutant allele ACC/T (0.53%) was only found in GY, CJ, CS, and JS populations (Table 2, Fig. 3). Three genotypes were identified: wild-type genotype I/I (99.20%), wild-type mutant heterozygote I/T (0.53%), and mutant homozygous T/T (0.27%) (Table 3, Fig. 4), of which the mutant genotype T/T was detected only in the CS population (2.37%). The study identified five mutant alleles at codon 1534, including TCC/S (58.02%), TGC/C (11.69%), CCC/P (0.99%), CTC/L (0.04%) and TTG/L (0.02%), in addition to the wild-type allele TTC/F (29.25%) (Table 2, Fig. 3). The nine populations revealed a total of nine genotypes: wild-type genotype F/F (13.10%), wild-type mutant heterozygotes F/S (28.87%), F/L (0.04%), F/C (1.41%), and F/P (1.98%), mutant genotypes S/S (41.55%), L/L (0.04%), and C/C (8.95%), and mutant heterozygotes S/C (4.07%) (Table 3, Fig. 4).

Table 2 Kdr allele mutation frequencies at loci 1016, 1532 and 1534 in the field population of Ae. albopictus in Guizhou Province, China.
Full size table
Figure 3
figure 3

Kdr allele frequencies at loci 1016, 1532 and 1534 in field populations of Ae. albopictus in Guizhou Province, China.

Full size image
Table 3 Kdr genotype frequencies of mutations at loci 1016, 1532 and 1534 in field populations of Ae. albopictus in Guizhou Province, China.
Full size table
Figure 4
figure 4

Kdr genotype frequencies at loci 1016, 1532 and 1534 in field populations of Ae. albopictus in Guizhou Province, China.

Full size image

At codon 1534, the mutant allele TCC/S was detected in all populations, with mutation frequencies ranging from 43.46 to 89.97%. Only the BJ population had CTC/L and CCC/P, while TTG/L was exclusively found in the CJ population. The mutant genotype S/S was detected in all populations and was the dominant genotype in LB, ZY, CS, XY, JS and BJ populations with mutation frequencies of 40.94% (122/298), 24.83% (73/294), 36.61% (108/295), 80.97% (234/289), 60.81% (180/296), and 54.65% (147/269), respectively. Mutant homozygotes (C/C) were present in all five populations, with the highest frequency of 38.59% in LB. Mutant heterozygotes (S/C) were also present in all five populations, but not the same, with the highest frequency of 30.61% in ZY.

Frequency and distribution of genotype combinations at loci 1016, 1532 and 1534 in Ae. albopictus

In all populations (2626 individuals in total), we found that some mosquito individuals carried kdr mutations in two codons together, and a total of 23 genotype combinations were identified, with no individuals carrying simultaneous mutations in all three loci (Table 4). Except for the GY and XY populations, one wild type (V/V + I/I + F/F) was present in all populations, with the highest frequency observed in the CJ population (18.27%). Twelve single mutant types were observed, including V/V + I/I + S/S and V/V + I/I + F/S, which were the most widespread (found in 9 populations) with frequencies of occurrence of 41.36% and 15.73%, respectively. Additionally, there were ten double mutant phenotypes, with the highest frequency of V/G + I/I + F/S mutations (12.83%), also found in 9 populations, and the highest frequency of 33.11% in GY.

Table 4 Genotypes and frequencies of three locus combinations of the kdr gene in the Ae. albopictus field population.
Full size table

Correlation between kdr mutant genes and resistance phenotypes

The OR values and 95% CIs were calculated for the six kdr mutant alleles at loci 1016, 1532 and 1534 in Ae. albopictus exposed to deltamethrin, permethrin and beta-cypermethrin. The frequency of mutations (GGA/G) at codon 1016 was 9.29% in susceptible individuals and 14.78% in resistant individuals after exposure to deltamethrin. The OR was 1.69 (95% CI 1.06–2.71) and P < 0.05 (Table 5). The study found that in the Ae. albopictus population, the OR values were 1.33 (95% CI 0.83–2.11) (P > 0.05) and 2.21 (95% CI 1.33–3.70) (P < 0.05) against permethrin and beta-cypermethrin, respectively. Generally, the V1016G mutant allele was positively correlated with deltamethrin and beta-cypermethrin resistance genotypes. No statistical correlation (P > 0.05) was revealed between the I1532T mutant allele and the three pyrethroid resistance phenotypes at locus 1532. At codon 1534, the OR values were 1.95 (P < 0.05), 1.75 (P < 0.05), and 2.90 (P < 0.05) in the Ae. albopictus populations to deltamethrin, permethrin, and beta-cypermethrin, respectively (95% CI 1.44–2.65, 1.28–2.41, and 2.17–3.88, respectively), reflecting that the F1534S mutant allele was positively associated with the three pyrethroid resistance phenotypes. The association between the F1534C mutant allele and beta-cypermethrin insecticide resistance phenotype was positive (OR 5.57, 95% CI 3.01–10.31, P < 0.05). On the other hand, no statistically significant difference was observed between the F1534C mutant allele and deltamethrin and permethrin insecticide resistance phenotypes (P > 0.05). There were no statistically significant differences (P > 0.05) between the F1534P and F1534L mutant alleles and the resistance phenotypes after exposure to deltamethrin, permethrin, and beta-cypermethrin, respectively. However, a positive correlation was found between F1534P and the total pyrethroid resistance phenotype.

Table 5 Relationship between kdr gene mutation and insecticide resistance in the field population of Ae. Albopictus.
Full size table

To explore whether the simultaneous presence of mutations at multiple loci in the one individual was the result of insecticide pressure, we classified the combinatorial genotypes into three categories: 0 (representing no mutations at loci 1016, 1532, and 1534), 1 (mutations at only one of the three loci), and 2 (mutations at only two of the three loci), and used chi-square tests to explore whether the combinatorial genotypes varied for different resistance phenotypes. The results showed significant differences (P < 0.05) in the types of genotypic combinations of different resistance phenotypes under the pressure of the three pyrethroid insecticides. The results of the comparison between the two types are summarized in Table 6.

Table 6 Genotypic combination type comparisons of different resistance phenotypes of Ae. albopictus under the pressure of three pyrethroid insecticides.
Full size table

Discussion

In the study, nine field populations of Ae. albopictus were obtained from dengue surveillance areas in Guizhou Province, including surrounding and central areas, with 2,626 mosquitoes, the largest number in the same study to date. The results showed that Ae. albopictus (larvae and adult mosquitoes) of Guizhou had already accumulated serious resistance to pyrethroid insecticides (deltamethrin, permethrin, and beta-cypermethrin), approaching or exceeding the results of southern and coastal provinces and cities such as Zhejiang and Hainan in China9,25,34. In addition, molecular experiments showed that corresponding kdr mutations were identified in nine Ae. albopictus populations. The mutant alleles GGA/G at loci 1016, ACC/T at loci 1532, TTC/S, CTC/L, TTG/L, TGC/C and CCC/P at loci 1534 were detected in Ae. albopictus with the mutation frequencies 13.86%, 0.53%, 58.02%, 0.04%, 0.02%, 11.69%, and 0.99%, respectively. Correlation analyses of mutations in codons 1016, 1532 and 1534 with resistance phenotypes revealed that the V1016G mutation was significantly associated with Ae. albopictus resistance phenotypes against deltamethrin (OR > 1, P < 0. 05) and beta-cypermethrin (OR > 1, P < 0.05), while the F1534C mutation was positively correlated with beta-cypermethrin (OR > 1, P < 0.05) and the F1534S mutation was positively correlated for resistance to deltamethrin, permethrin and beta-cypermethrin (OR > 1, P < 0.05). However, no correlation was identified between the I1532T mutation and the three pyrethroid resistance phenotypes, which may be due to the low frequency of the mutation. The difference between the appearance of the kdr mutation and the absence of the mutation was significant (P < 0.0167) for all three insecticides, whereas the occurrence of one locus mutation and the occurrence of two showed a significant difference only in the case of beta-cypermethrin, suggesting that the difference between the resistant phenotype and the susceptible phenotype is mainly the production of the kdr mutation, and that simultaneous mutation of multiple loci may be the result of enhanced resistance.

In the absence of reports on the existence of Aedes aegypti in Guizhou currently, Ae. albopictus is the dominant vector for the transmission of dengue virus. The WHO recommended the global promotion of the use of pyrethroid insecticides for mosquito control in the late 1980s35. Since 1992, pyrethroid insecticides have been widely available because of their affordability, rapid knockdown, low toxicity to mammals and relative safety to humans36. We have reported pyrethroid resistance in wild populations of Ae. albopictus in 2018 and 2021, having observed susceptibility to pyrethroids in GY, CS and XY populations. In this study, bioassays showed increased resistance to pyrethroids in Ae. albopictus compared to the GY population in 201837, and little more or slightly increased resistance in the CS and XY populations in 202138. During our interventions, we noticed that householders in residential areas habitually spray commercially purchased insecticides around flowers and plants, which are complex insecticides containing several pyrethroids. The literature reports that 23 types of pyrethroid active ingredients are approved by the government of the People's Republic of China as public health insecticides11. Second, several cities, including ZY, CS and XY, had contracted with PCOs for regular insecticide spraying to reduce mosquito densities and breeding as part of their hygiene city campaigns. In these habitats, mosquitoes may have developed high levels of resistance at the larval and adult stages under the selective pressure of continuous exposure to pyrethroid insecticides. In addition, the XY and JS populations of Ae. albopictus showed lower mortality rates after exposure to the three pyrethroids than other populations, which may explain the difference in geographical location. In recent years, resistance of Ae. albopictus to pyrethroid insecticides has also been reported in other provinces and cities in China2,39,40, but the differences should be carefully analyzed as more factors affect the results of the bioassay, including the rearing sites, the samples tested and the assessment of mortality standards9. The results of this study enriched the information on the resistance of Ae. albopictus to pyrethroids in the dengue surveillance areas of Guizhou Province, providing more direct supporting data to guide the utilization of insecticides.

In the current study, the individuals we used for kdr gene testing were mosquitoes after bioassay, facilitating the accurate determination of the association between the mutation and pyrethroid resistance. We identified seven mutant alleles at codons 1016 (13.86%), 1532 (0.53%) and 1534 (70.75%) and further confirmed the correlation of F1534S with pyrethroids, which is consistent with the results of population studies in several geographical regions of China9,25,34. Second, in the face of the emergence of outbreaks in surrounding provinces and the existence of imported Dengue instances41,42,43,44, the higher mutation frequency may represent a serious threat to the control of potential Dengue outbreaks, even if there has been no indigenous Dengue outbreak at present in Guizhou Province. Moreover, the present study identified the V1016G mutation as positively associated with the resistance phenotype of Ae. albopictus to deltamethrin and permethrin, which has been demonstrated to be responsible for the insensitivity against these insecticides in reports of Aedes Aegypti but has hardly been reported in studies of Ae. albopictus. Some studies have documented a positive correlation between the I1532T mutation and the deltamethrin resistance phenotype45, but there are also conflicting views9,34. The I1532T mutation has a low mutation frequency and was not found to be associated with its insecticide resistance in the present results, which needs to be further investigated. Our results showed that F1534C was only insensitive to beta-cypermethrin. Although a resistance relationship between F1534C and pyrethroids has been detected in Aedes aegypti46, few correlations have been reported in Ae. albopictus in China9,13,33. F1534C could be a weaker indicator of resistance. Significantly, a new mutation, F1534P, was identified in the current study, but it was only found in one population, and its association with resistance needs to be further investigated.

A total of 29 different genotypic combinations were observed at codons 1016, 1532 and 1534. Among these, the F1534S single mutation at three loci was the most common combination, distributed across all populations, corresponding to the findings of Zhao et al. on field populations of Ae. albopictus in 11 dengue-endemic provinces12. The interactions between mutations at different loci remain to be further investigated. Differences in mutation rates at different loci in different geographical regions may result from uneven distribution of resistance39 and the presence of multiple resistance mechanisms, resulting in the kdr mutation not playing a dominant role. Further investigation of other resistance mechanisms, such as metabolic resistance1,47, should also be conducted. In addition, Chen and Zhao et al. found that the mutation rates at loci 1016, 1532, and 1534 were correlated with average annual temperature, rainfall, and dengue-endemic areas12,48. Therefore, monitoring the kdr mutation frequency is not only conducive to the monitoring of Ae. albopictus resistance to pyrethroids and controlling dengue vector density, but also to preventing dengue outbreaks in advance.

Conclusion

Current susceptibility status of wild populations of Ae. albopictus to insecticides and a higher frequency of kdr mutations from dengue-monitored areas in Guizhou Province are reported in this paper. The pyrethroid resistance phenotype was clearly associated with the kdr mutation F1534S, which can be considered a molecular marker for resistance. In addition, outcomes of this study can serve as data support for further research and development of effective insecticidal interventions against Ae. albopictus populations in Guizhou Province.

Materials and methods

Collection of wild Ae. albopictus samples

Nine field populations of Ae. albopictus were collected using Pasteur pipettes from breeding sites, such as flowerpots, waste tires, metal containers, waste buckets, water tanks and other water containers from Guiyang (GY), Ziyun (ZY), Jinsha (JS), Libo (LB), Xingyi (XY), Bijiang (BJ), Congjiang (CJ), Panzhou (PZ) and Chishui (CS) in Guizhou Province during July–August 2022. The larvae were identified as Ae. albopictus species by using the taxonomic keys49. All the samples collected in the field in each region were brought back to the laboratory of the Guizhou Provincial Centre for Disease Control and Prevention in buckets and identified again, and finally the selected Ae. albopictus mosquitoes were reared in tubs filled with dechlorinated water. The larvae were reared to adulthood under standard conditions of 26 ± 1 °C, 75 ± 5% relative humidity and 14/10 h light/dark photoperiod. Larvae were fed a special powdered diet (pig liver: steamed bread = 1:1) and adults were fed a 10% sucrose solution. All Ae. albopictus directly collected and identified from the field (F0 generation) were reproduced (F1–F2 generations) before being tested for susceptibility.

The susceptible strains of Ae. albopictus which served as a reference, was kindly provided by National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention. The strain has been maintained in the insectary over a decade, without exposure to any insecticide.

Larval resistance bioassays

The susceptibility of larvae was determined using three insecticides (permethrin 90% pure, deltamethrin 98%, beta-cypermethrin 91.16%) from the Chinese Center for Disease Control and Prevention, complying with WHO guidelines50. Firstly, all insecticides were diluted to 5–7 concentrations with acetone to obtain mortality from 10 to 90%. Then, 1 ml diluted insecticide solution and 20–30 late third and fourth instar larvae were added to an experimental beaker that held 199 ml water separately. Furthermore, insecticide solution was replaced by acetone in the control group. Larval of Ae. albopictus mortalities were recorded after 24 h exposure. Larvae that did not move or twitched when strongly stimulated were considered dead. The assay was repeated three times. Larval mortality was calculated by dividing the number of dead larvae by the total number tested. The extent of resistance is defined by the resistance ratio (RR50), which was derived by comparing the median lethal concentration (LC50) value of the field population with the LC50 value of the sensitive laboratory population to the insecticide.

Adult resistance bioassays

The susceptibility of adults was tested using three insecticide-impregnated papers (0.4% permethrin, 0.03% deltamethrin, 0.08% beta-cypermethrin) from the Chinese Center for Disease Control and Prevention complying with WHO guidelines5651. In the assay, 25 female mosquitoes, non-blood-fed, aged 3–5 days were selected to the resting tube without insecticide to adaptation for 30 min. After that they were gently blown into experimental tubes containing insecticide impregnated paper for up for 1 h. Silicone oil-treated papers without insecticides were applied to the control groups. Following insecticide exposure, mosquitoes were returned to resting tubes with 10% sugar water and mortality assessed after 24 h. Approximately 100 female mosquitoes were tested for each insecticide. Mosquitoes that do not move or only tremble in their limbs and wings and cannot survive are considered dead. They are also considered susceptible phenotypes. In contrast, mosquitoes that can survive are considered resistant phenotypes. Dead and survival mosquitoes were collected individually in test tubes filled with 95% alcohol at − 80 °C for subsequent DNA analysis. The bioassay should be repeated if the mortality in the control group was ≥ 20%. The test group mortality rate would be adjusted using Abbott's formula if the control group mortality rate was between 5 and 20%: Corrected mortality (%) = (test group mortality − control group mortality)/(1 − control group mortality) × 10051.

DNA extraction and mutation detection

Ezup Columnar Animal Tissue Genomic DNA Extraction Kit (Sangon Biotech, B518251, shanghai) was used to extract genomic DNA from a single mosquito for identification of the kdr mutations. Partial domain II and III of the voltage-gated sodium channel gene (VGSC) (covering I1532, F1534 and V1016) were amplified using the primer aegSCF20 (forward: 5′-GAC AAT GTG GAT CGC TTC CC-3′) and aegSCR21 (reverse: 5′-GCA ATC TGG CTT GTT AAC TTG-3′), aegSCF7(forward: 5′-GAG AAC TCG CCG ATG AAC TT-3′) and aegSCR7 (reverse: 5′-GAC GAC GAA ATC GAA CAG GT-3′). The PCR cycling parameter includes the following: pre-denaturation at 95 °C for 5 min, then ten cycles of 94 °C for 30 s, 63 °C for 30 s and 72 °C for 30 s, and followed by 30 cycles of 95 °C for 30 s, 58 °C for 30 s and 72 °C for 30 s, ending with an extension at 72 °C for 10 min. After electrophoresis, PCR products were purified using San Prep Column-based DNA Gel Recovery Kit (Sangon, Shanghai) and sequenced reversely with the primer aegSCR22 (5′-TTC ACG AAC TTG AGC GCG TTG-3′) and aegSCR8 (5′-TAG CTT TCA GCG GCT TCT TC-3′) respectively using the ABI 3730XL automatic sequencer (Applied Biosystems, Shanghai, China). Sequencing was accomplished by Sangon Biotech (Shanghai) Co., Ltd. All sequenced data were analyzed with sequence analysis software.

Statistical analysis

The resistance status of adult mosquitoes was also classified according to WHO criteria51: a mortality < 90% was recognized as resistant, a mortality of 90–98% as probable resistance, and a mortality > 98% was susceptible. For larval bioassays, susceptible if RR50 < 5, moderately resistant if 5 < RR50 ≤ 10, and highly resistant if RR50 > 10. A chi-squared test and hazard analysis were used to analyze the correlation between the frequency of mutant alleles and their resistance phenotypes, and the dominance ratio (OR) values were calculated. Differences were determined to be statistically significant at P < 0.05; the relationship between the kdr allele and the resistance phenotype was considered positive when the OR > 1 and negative when the OR < 1, and not yet statistically significant or statistically insignificant if the 95% CI of the OR value ranged between 1 or a P > 0.05. All data were analysed using excel2013 and spss24.0 software.