Voltage-Gated Sodium Channel (Vgsc) Mutation-Based Pyrethroid Resistance in Aedes aegypti Populations of Three Endemic Dengue Risk Areas of Sri Lanka

Background Pyrethroid insecticides are widely used in many countries for chemical-based control of Ae. aegypti. Regardless of their efficacy, the constant use of insecticides has induced insecticide resistance mechanisms, such as knockdown resistance (kdr) in mosquitoes. Sri Lankan Vector Controlling Entities (VCE) have been using a variety of pyrethroid insecticides as the primary approach for dengue control. However, development of any resistance among the Aedes mosquitoes has been limitedly studied in the country. Therefore, the current study was conducted to evaluate the prevalence of F1534C, V1016G, and S989P mutations among Ae. aegypti mosquito populations in three dengue endemic high-risk regions of Sri Lanka. Methodology. Immature (both pupae and larvae) stages of Ae. aegypti mosquitoes were collected from Colombo, Gampaha, and Kandy districts of Sri Lanka from February 2018 to December 2019. Polymerase Chain Reaction- (PCR-) based assay for molecular genotyping of mutations was performed to identify the prevalence of kdr mutations in collected Ae. aegypti populations, separately. The frequencies of the resistant and susceptible kdr alleles were determined by using the Hardy–Weinberg equilibrium. Results The Ae. aegypti populations from Colombo, Gampaha, and Kandy districts showed 46%, 42%, and 22% of F1534C mutation allele frequencies, along with 15%, 12%, and 6% of V1016G mutation allele frequencies, respectively. The mutation allele frequencies of S989 in Colombo, Gampaha, and Kandy districts were 9.5%, 8.5%, and 4.5%, respectively. The wild-type (PP) genotype remained predominant within all the three districts, whereas the homogenous (QQ) mutation genotype occurred only in minority. The abundance of Q allele frequency in Ae. aegypti mosquitoes was relatively higher for all the three mutations in Colombo. Conclusions The findings clearly indicate that long-term insecticide applications and multiple use of pyrethroids have led to the acquisition of kdr mutations, leading to the development of insecticide resistance among local Ae. aegypti populations, especially in the Colombo and Gampaha districts. Therefore, evaluation of the prevalence levels of these kdr mutations highlights the necessity for shifting towards novel vector control strategies.


Background
The evolution and development of human civilizations on Earth have resulted in encroaching of native habitats of many species by humans. As a result, certain insects that breed in human-occupied regions have coevolved with humans, acquiring affinity towards human blood as their dietary requirements [1]. These blood-requiring insects are disastrous as they can cause or spread many dreadful diseases worldwide. Among these, dengue is considered as one of the most important arboviral diseases at present [2]. At the global level, dengue has threatened approximately 3.9 billion people residing in more than 128 countries [3]. Sri Lanka has been engaged in a fight against dengue, since the mid-1960s, and the disease has become the most severe health burden of the country at present [2]. The worst epidemic of dengue in Sri Lanka was witnessed in 2017 resulting in 186,101 dengue cases along with more than 300 deaths. In 2019, a total of 105,049 dengue cases were reported as the second highest dengue epidemic faced by the country. Meanwhile, a total of 31,139 dengue cases were reported in 2020 [4].
With the absence of vaccines, patient management and vector control are crucial to manage dengue outbreaks in many countries, including Sri Lanka [5,6]. Therefore, maintaining the Aedes vector population, that could cause dengue, chikungunya, yellow fever, and Zika, under the threshold level is one of the key preventive approaches against vectorborne diseases [7]. The current vector control strategies in Sri Lanka mainly focus on environmental management [8], biological control [9][10][11], and chemical vector control [12] for the suppression of Aedes populations, below the threshold levels. Among these, application of larvicides remains as the most widely used chemical control method.
Sri Lankan Vector Controlling Entities (VCE) rely upon a variety of pyrethroids, namely, prallethrin, etofenprox, pyrethrins, permethrin, resmethrin, and sumithrin, as the primary larvicides and adulticides for Aedes vector control [13]. However, extensive use of chemical-based vector control methods is known to cause numerous harmful impacts such as the development of insecticide resistance in mosquitoes, unintended side effects on human health or other nontarget populations, and biological magnification [14,15].
Toxic effects of pyrethroids are mediated through preventing the closure of the voltage-gated sodium channels in the axonal membranes [16,17]. Extensive usage of pyrethroids to control Ae. aegypti, the primary vector of dengue, could lead to the selection of insecticide resistance-related mutations in mosquitoes [18]. In insects, the "knockdown resistance" (kdr) is one of the most widespread mechanisms that could cause resistance to pyrethroids and Dichloro-Diphenyl-Trichloroethane (DDT) [19]. The kdr resistance occurs due to mutations in the voltage-sensitive sodium channel (Vssc) or voltage-gated sodium channel (Vgsc), coded by the Vssc gene. The foremost function of Vgsc is the initiation and transmission of action potentials in neuron cells [20]. Due to mutations in the Vssc gene, insecticides are unable to bind and block the function of Vgsc, resulting in resistance [21].
Around eleven mutations linked to pyrethroid resistance in Ae. aegypti mosquitoes have been identified at the global level. Among them, five mutations, namely, S989P, I1011M, V1016G, F1534C, and V410L, have been functionally confirmed to confer pyrethroid resistance [22]. Only three kdr mutations, the V1016G, S989P, and F1534C, have been recognized as commonly occurring amino acid substitutions among Aedes mosquitoes in the Southeast Asian region [23]. The F1534C mutation, which reduces permethrin binding to Vgsc channels, has been reported as the most common kdr mutation in the Southeast Asia [22].
The development of kdr mutation in Ae. aegypti populations could significantly reduce the efficacy of chemicalbased vector control approaches in Sri Lanka. This will cause a direct impact on the success of dengue epidemic management in the country, leading to incidence of more severe dengue epidemics. In a recent study, Fernando et al. [24] have reported the presence of V1016G and S989P mutant alleles in the Vgsc of Sri Lankan Ae. aegypti populations for the first time. A study conducted by Endersby-Harshman et al. [25] to investigate the geographical distributions of three sodium channel mutations found in Ae. aegypti from the Indo-Pacific region has also presented evidence for existence of these mutations in Sri Lanka, which may have been attained by Ae. aegypti via genetic invasion. However, more detailed studies on this aspect are lacking in Sri Lanka. Therefore, the current study was conducted to determine the spatial distribution pattern of V1016G, F1534C, and S989P kdr mutations, among Ae. aegypti mosquito populations in three different high-risk geographical regions for dengue in Sri Lanka.

Study Area.
Over the past few years, the highest dengue burden in Sri Lanka has occurred in the districts of Colombo, Gampaha, and Kandy. The district of Colombo is the commercial capital of Sri Lanka (6.70°N and 80.22°E). Being located in the lowland of the country, Colombo is the most urbanized metropolitan area of the country. In 2019, 13.7% (n = 4257) of 31,139 dengue cases were observed in Colombo [4] as the most high-risk area for dengue. Meanwhile, Gampaha district (7°05′N and 79°59′E) reported the third highest number of dengue cases as 2655 (8.5%) [4]. Both Colombo and Gampaha districts are located in the low country wet zone of Sri Lanka. Meanwhile, Kandy district (6.93°to 7.50°N and 81.04°E) is located in the central highlands, covering an area of 1940 km 2 with a wide array of natural environmental features in contrast to Colombo. It is of major tourist interest, due to its natural locations and places of historical and religious importance. At present, Kandy district remains as the second high-risk area for dengue transmission in the country, reporting 11.0% (n = 3432) of the dengue cases reported in 2019 [4]. Due to the severe dengue risk, these three districts are characterized by a higher level of insecticide applications to control Aedes vectors. Therefore, the districts of Colombo, Gampaha, and Kandy were selected as the study areas in this study, which is important for evidencebased vector control.     3 BioMed Research International reverse primer (Cys1534; 0.165 μM), common reverse primer (Cpr; 0.5 μM), Taq DNA polymerase enzyme (0.2 U), and template DNA (2 ng). Cycling conditions were hot started at 95°C for 2 min followed by 35 cycles, each of denaturation at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s, followed by the final extension at 72°C for 2 min. Subsequently, 10 μl of amplified products was loaded with 2 μl of 5x gel loading dye in 0.5x TBE buffer, along with a 50 bp DNA marker. Finally, PCR products were visualized by gel electrophoresis (Cell Biosciences, USA) on a 1.5% agarose gel at 150 V for 30 min [24].
2.6. Detection of V1016G and Genotyping. Allele-specific PCR (AS-PCR) was performed to detect the V1016G mutation in Ae. aegypti [24] by amplifying the extracted DNA in a 25 μl of PCR mixture containing 1x PCR buffer, MgCl 2 (1.5 Mm), dNTPs (200 μM), forward primer (V1016G; 0.25 μM), reverse primer (0.125 μM) specific for either glycine (Gly1016R) or valine (Val1016R), Taq DNA polymerase enzyme (0.2 U), and template DNA (2 ng). Cycling conditions were hot started at 95°C for 2 min followed by 35 cycles each of denaturation at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s, followed by the final extension at 72°C for 2 min. Subsequently, 10 μl of amplified products was loaded with 2 μl of 5x gel loading dye in 0.5x TBE buffer along with a 50 bp DNA marker. Finally, PCR products were visualized by gel electrophoresis (Cell Biosciences, USA) on a 1.5% agarose gel at 150 V for 30 min [24].

Detection of S989P and Genotyping.
A similar PCR (AS-PCR) was performed to detect the S989P mutation by amplifying extracted DNA in a 25 μl of PCR mixture containing 1x PCR buffer, MgCl 2 (1.5 Mm), dNTPs (200 μM), forward primer (0.5 μM), reverse primer (0.5 μM), Taq DNA polymerase enzyme (0.2 U), and template DNA (2 ng) [28]. Cycling conditions were hot started at 95°C for 2 min followed by 35 cycles each of denaturation at 95°C for 30 s, annealing at 63°C for 30 s, and extension at 72°C for 30 s, followed by the final extension at 72°C for 2 min. Subsequently, 10 μl of amplified products was loaded with 2 μl of 5x gel loading dye in 0.5x TBE buffer along with a 50 bp DNA marker. Finally, PCR products were visualized by gel electrophoresis (Cell Biosciences, USA) on a 1.5% agarose gel at 150 V for 30 min [24].

Data
Analysis. The frequencies of the resistant (Q) and susceptible (P) kdr alleles for each of the point mutation were determined by the Hardy-Weinberg Equilibrium using an online calculator [29]. The mean P and Q allele frequencies of Ae. aegypti mosquitoes from three study areas in 2018 and 2019 were square root transformed. The Bray-Curtis similarity-based Distance-Based Redundancy Analysis (dbRDA) was performed to highlight and visually represent the underlying segregation patterns of the Ae. aegypti populations from three districts, based on the overall variations in the frequencies of the resistant (Q) and susceptible (P) kdr alleles of F1534C, V1016G, and S989P point mutations. The Plymouth Routines in Multivariate Ecological Research (PRIMER version 6) was used for statistical analysis.

Results
A total of 974 Ae. aegypti mosquitoes caught from the wild were screened for Vgsc mutations as depicted in Figure 1.
In the F1534C genotyping, the PQ genotype dominated among the mosquito samples of Colombo (43.7%), while PP genotype remained dominant among Ae. aegypti mosquitoes from Gampaha (44.5%) and Kandy (71.7%) in 2018 as shown in Table 1. However, the PQ genotype denoted a notable increment in 2019 at the position 1534 in domain III of the Vgsc gene in Ae. aegypti mosquitoes from all the three districts, when compared to 2018. In fact, the PQ genotype was dominant in both Colombo and Gampaha districts accounting for 56.8% and 53.6% frequencies, respectively, among the screened Ae. aegypti mosquitoes in 2019. When the allele frequencies are considered, it was evident that Q allele frequency (mutation) accounted for around one-third (33.3%) of the screened Ae. aegypti mosquito populations from Colombo and Gampaha in 2018. Further, the Q allele frequency denoted gradual increments in 2019 within the mosquitoes from all study districts ( Figure 2).
The PP genotype remained predominant at position 1016 in domain II, segment 6, of the Vgsc gene in Ae. aegypti mosquito for the V1016G mutations in Ae. aegypti mosquito populations, while the highest frequency of the QQ genotype was observed in Colombo as 5.1% and 5.7% in 2018 and 2019, respectively. In case of allele frequency, P allele was predominant in the mosquito populations from all three districts. However, Q allele was relatively more abundant among the Ae. aegypti mosquitoes of Colombo, followed by Gampaha (Table 2). In general, mosquitoes from all the three districts denoted a gradually increasing trend of the Q allele frequency in 2019 than in 2018, for the V1016G mutation ( Figure 3).
In the positions 1016 and 989 for the S989P mutations, the PP genotype dominated (>80%) in all the Ae. aegypti mosquito populations. However, the homozygous resistance genotype was relatively higher in the Ae. aegypti population from the district of Colombo, denoting 1.9% and 3.6% in 2018. Interestingly, Ae. aegypti mosquitoes from Gampaha accounted for the highest QQ genotypic frequency as 4.5% in 2019 (Table 3). In case of allele frequency, similar to other   Figure 5). Interestingly, Ae. aegypti populations of Colombo and Gampaha districts formed a major cluster, denoting a similarity level of 95.3% in overall P and Q allele frequencies. As depicted in Figure 5, the Ae. aegypti populations of Colombo and Gampaha districts denoted a 25.3% level of dissimilarity from the Ae. aegypti population of Kandy. Based on the loading coefficients, it appears that the relatively higher Q allele frequencies of F1534C, V1016G, and S989P in the Ae. aegypti populations of Colombo and Gampaha districts are the cause for formation of a major cluster at a 95.3% similarity level. Since the Q allele frequencies in the Ae. aegypti populations of Kandy are notably low, it has formed a separate major cluster sharing only 74.6% of similarity ( Figure 5).  (Table 4). Interestingly, the FF VV SS remained prominent in the Ae. aegypti populations from Kandy in 2018 (71.7%) and 2019 (64.1%). On the contrary, the homozygous resistance genotypes CC GG PP were relatively low (<3.5%) in the Ae. aegypti populations from the district of Colombo, Gampaha, and Kandy in 2018 and 2019 (Table 4).

Discussion
Despite the continuous application of larvicides and adulticides, severe incidence of dengue epidemics reported from Sri Lanka strongly suggests the possibility of insecticide resistance among Aedes vectors. The present study was undertaken to further investigate the presence of F1534C, V1016G, and S989P kdr mutations, among Ae. aegypti populations in different geographical regions (Colombo, Gampaha, and Kandy) of Sri Lanka and their development during two preceding years (2018 and 2019).
According to the current study, all the three mutations, F1534C, V1016G, and S989P, were detected among the Ae. aegypti populations in Colombo, Gampaha, and Kandy districts from 2018 to 2019. A study conducted by Endersby-Harshman et al. has reported five allelic combinations to be common in Ae. aegypti mosquitoes from the Indo-Pacific region [25]. Among them, F1534C, V1016G, and S989P   BioMed Research International mutations have been recognized as the most commonly found Vgsc mutations in pyrethroid-resistant Ae. aegypti in many Southeast Asian countries such as Cambodia [30], Indonesia [31,32], Laos [33], Malaysia [34], Myanmar [35], Singapore [36], Thailand [37], and Vietnam [38]. In addition to Southeast Asian countries, high prevalence levels of pyrethroid-resistant mutations have been reported in many countries worldwide due to overapplication of pyrethroids [39,40]. In the current study, both Colombo and Gampaha districts remain as urbanized areas with relatively higher risk for dengue incidence. Therefore, the vector control activities are more intensive in these two districts leading to elevated chances for pyrethroid resistance development. Findings of the current study are also supported by Fernando et al., who have reported the presence of F1534C, V1016G, and S989P mutant alleles in the Vgsc genes of Sri Lankan Ae. aegypti [24].
Among the three point mutations, mutant alleles of F1534C were the most predominant, denoting a relatively extensive dispersion across all of the study sites. High prevalence rates of V1016I and F1534C have been reported in Latin America, including Brazil [41], Venezuela [42], Colombia [43], and Jamaica [44]. In Mexico, a study conducted by Aponte et al. [39] has also reported the widespread distribution of F1534C and I1016 mutations and suggested that higher F1534C frequency is more likely to allow other mutations to evolve. Similarly, in the current study, the Colombo district, which had a higher F1534C frequency, showed the presence of V1016G and S989P mutations at relatively higher levels.
A simple AS-PCR technique has been successfully applied in the current study to detect F1534C, V1016G, and S989P in Ae. aegypti populations of Sri Lanka. Genotyping of F1534C exhibited that the heterozygous (PQ) genotype was slightly higher than the homozygous susceptible gene (PP) in both Colombo and Gampaha districts. The frequencies of mutant alleles of all F1534C, V1016G, and S989P point mutations denoted a gradual increment from 2018 to 2019 in Ae. aegypti populations, showing that those mutant alleles are currently been selected in Ae. aegypti populations in Sri Lanka. Several previous studies have shown that the homozygous mutant individuals would be more resistant, while wild-type homozygous would be more susceptible. Further, an intermediate resistance would be displayed by heterozygous mosquitoes to pyrethroids [18]. The intermediate resistance of heterozygous mosquitoes clearly suggests that mutation is not the only mechanism involved and that either other kdr or enzymatic mechanisms may confer cross resistance or enhance the resistance. The cooccurrence of two or three kdr mutations in Ae. aegypti has been reported in many countries and is believed to cause a higher level of resistance to pyrethroids [39][40][41][42]. Further, it is reported that the F1534C mutation could potentially compensate for any reduction in fitness caused by the V1016G and S989P mutations [45].

BioMed Research International
According to Brengues et al. [19], knockdown resistance against pyrethroids in Ae. aegypti may possibly be conferred by one or more point mutations in the Vgsc locus. For instance, Amelia-Yap et al. [46] have reported the cooccurrence of V1016G/S989P point mutations as a common cooccurrence pattern in tested pyrethroid-resistant Ae. aegypti populations. Hence, the increment of the insensitivity of Vgsc to pyrethroids could be experienced as a result of this phenomenon. Although the S989P mutation was detected, it cannot be considered as the main cause for the insecticide resistance, since the S989P mutation has always been associated with the V1016G mutation, until the S989P mutation occurs alone [34,35]. However, according to a study conducted in Thailand, the V1016G mutation can stand alone despite the absence of the S989P mutation in Ae. aegypti [45]. Hence, it can be extrapolated that the intensifying effect of pyrethroid resistance may have been resulted from the synergized S989P mutation.
Although Du et al. [21] have further proved that the S989P mutation is not having any additive effect to the V1016G mutation, a recent contradictory study has highlighted that the existence of the S989P mutation can highly diminish the sensitivity and the susceptibility of Aedes vectors to pyrethroids [47]. A recent study conducted in Indonesia has also emphasized the cooccurrence of the V1016G and S989P mutations in high frequencies among the Asian Ae. aegypti populations [31,32]. In the current study, only Ae. aegypti mosquito populations from the districts of Colombo and Gampaha, that had relatively higher resistant gene frequencies of V1016G mutation, were also having S989P mutations at relatively notable levels. However, it is still debatable if the S989P mutation was responsible for causing the additive role in pyrethroid resistance development in Ae. aegypti. This signifies the importance of further in-depth studies for the novel discoveries related to this aspect.
The presence of F1534C, V1016G, and S989P kdr mutations in Ae. aegypti mosquito populations from different regions in Sri Lanka indicates the prevalence of insecticide resistance. It is obvious that the findings of the current study are based on only 974 Ae. aegypti mosquito larvae, which is limited. However, the regions selected for the study (Colombo, Gampaha, and Kandy) had been treated with large amounts of insecticide for a long time, compared with 7 BioMed Research International other areas of the country. Therefore, the extensive use of pyrethroids to control dengue should be the cause for this development of pyrethroid resistance in Aedes mosquitoes [48]. Furthermore, the presence of mutated mosquitoes even within a low sample size in the current study predicts that there is a high likelihood of having a notable degree of resistant Ae. aegypti mosquitoes in the wild. If the current rate of pyrethroid use persists in Sri Lanka, prevalence and distribution of kdr mutations could rapidly elevate, challenging the successful management of dengue epidemics in Sri Lanka, through conventional chemical-based control. Although the current study focused on the allele frequency as single mutations, the cooccurrence of the three mutations among the Ae. aegypti mosquitoes was not investigated, due to limitations in funds. Such study would reveal the strength of the resistance to pyrethroid insecticides [21], which is of higher significance. Therefore, in-depth studies are recommended to further investigate the combined effect of the alleles in inducing pyrethroid resistance in Ae. aegypti.
The future holds promising prospects in development of efficient insecticides and new mosquito control strategies. However, their inclusive implementation in tropical regions will require a decade at minimum [49,50]. Thus, characterizing molecular mechanisms that may trigger insecticide resistance in Aedes mosquitoes is vital for tracking down resistance alleles and enhancing resistance management strategies [51]. Therefore, further identification of kdr mutations, that cause insecticide resistance, is crucial in the country. The present study has revealed the presence of insecticide resistance among the wild Ae. aegypti populations through target site resistance mechanisms. Such mutations could exert negative fitness costs influencing the evolution of further insecticide resistance in field populations of mosquitoes. Therefore, it is crucial that such impacts are properly understood and taken into consideration, when designing and implementing future insecticide resistance management strategies. In general, the detection of kdr mutations in Ae. aegypti populations could enable the monitoring of insecticide resistance occurrence and guiding of vector control approaches in Sri Lanka. In addition, the society needs to progressively upgrade their lifestyle to allow better control of Ae. aegypti effectively.

Conclusion
According to the current study, all the three mutations, F1534C, V1016G, and S989P, were detected among the Ae. aegypti populations in Colombo, Gampaha, and Kandy districts. Among the three-point mutations, mutant alleles of F1534C were the most predominant, denoting a relatively extensive dispersion across all the study sites. The heterozygous (PQ) genotype was slightly higher than the homozygous susceptible gene (PP) in both Colombo and Gampaha districts. The frequencies of mutant alleles of all F1534C, V1016G, and S989P point mutations denoted a gradual increment from 2018 to 2019 in Ae. aegypti populations, showing that mutant alleles are being selected within Ae. aegypti populations in Sri Lanka. This could impose a serious threat to the efficacy of current dengue vector control strate-gies, as VCE in Sri Lanka mainly focus on chemical-based dengue control methods. Therefore, evaluation of the prevalence levels of these kdr mutations highlights the necessity for shifting towards novel vector control strategies.
Abbreviations kdr: Knockdown resistance VCE: Vector Controlling Entities WHO: World Health Organization.

Data Availability
Raw data supporting the findings are available from the authors on request.

Ethical Approval
The study did not involve any organism that is characterized with any conservation priority in Sri Lanka.

Consent
All the mosquito specimens were collected with the informed written consent from the household heads permitting the entomological surveillance.

Conflicts of Interest
The authors declare that they have no conflict of interests.

Authors' Contributions
TR contributed in designing the research, supervision of the research work, and writing the manuscript; LU contributed in experimental design, statistical analysis, and writing the manuscript; SS, SK, SN, HR, FFD, and FNS contributed in DNA extraction and kdr mutation genotyping; RD contributed in supervision of the DNA extraction and kdr mutation genotyping; NSG was responsible for overall supervision and reviewing the manuscript. All authors read and approved the final manuscript.