Background

Malaria is one of the most serious vector-borne diseases. Despite being preventable and treatable, malaria continues to be a major threat to global public health. In 2017, there were an estimated 219 million cases of malaria and 435,000 deaths from malaria globally, with 5% of the cases occurring in the WHO South-East Asia Region [1]. Guangxi Zhuang Autonomous Region of China was once a malaria-endemic region [2]. After continued efforts for decades, the malaria burden has been substantially reduced in Guangxi [2, 3]. However, the risk of malaria re-emergence remains, due to the importation of malaria parasites in thousands of infected travellers from Africa, and Southeast Asia. In neighbouring Vietnam, the malaria incidence declined significantly between 1991 and 2014, but there were 14,941 confirmed cases in 2014, including 473 confirmed cases in the Northern region [4].

Insecticide-based vector control remains a key preventive strategy in the fight against malaria, credited with the significant reductions in malaria morbidity and mortality since 2000 due to the widespread implementation of insecticidal interventions [5]. However, the continued effectiveness of this strategy has been challenged by the increasing resistance of vectors to available insecticides. To minimize the risk of control failure caused by insecticide resistance, a better understanding of the status of and the genetic mechanisms underpinning resistance to commonly used insecticide is an urgent need.

In the previous study [6], the distribution and frequency of genetic mutations in two targets, acetylcholinesterase (AChE) and the voltage-gated sodium channel (VGSC), were investigated. The G119S mutation within AChE was present at high frequencies (0.61–0.85), but the kdr mutation was rare in the seven Guangxi An. sinensis populations along the China–Vietnam border, suggesting that pyrethroids remain suitable for use against An. sinensis [6]. However, to maintain their effectiveness, the application of pyrethroids should not be taken as the sole measure for vector control, thus insecticides with alternative modes of action should be considered.

The current study was a survey extended to another insecticidal target, gamma-aminobutyric acid (GABA) receptor subunit encoded by the RDL (Resistant to dieldrin) gene, which is the target of multiple types of insecticides, such as dieldrin (cyclodienes), fipronil (phenylpyrazoles), and fluralaner (isoxazolines) [7,8,9]. Efforts were given to detect naturally existing genetic mutations, map their distribution and frequency in seven An. sinensis populations along the China–Vietnam border, and track the origin of the resistance-related mutations. This work focused on amino acid substitutions at three positions (A296G/N/S, V327I and M345M/S) in RDL that are associated with insecticide resistance previously documented in multiple insect species [8,9,10,11,12,13]. The data obtained in this study are of significance for current and future malaria control programs given that the use of GABA targeting insecticides is on the increase.

Methods

Samples

Anopheles sinensis adults were caught in seven different sites from April 2015 to August 2017. The seven sample-collecting sites were located in different villages along the China–Vietnam border. Rice is the main crop planted in these villages. The commonly used insecticides for rice pest control in these areas are diamides (e.g. chlorantraniliprole), neonicotinoids (e.g. imidacloprid) and organophosphorates (e.g. acephate and dimethoate). Methods about sample collection and species identification were described by Yang et al. [6].

AsRDL genotyping

A fragment of the AsRDL gene that includes codons 296 and 327 (named as RDL7) was amplified using primers (Primers ARE-7F and 7I-R, Table 1) and individual genomic DNA as template. The amplification reaction consisted of 25 μl of 2× Taq PCR Master Mix, 0.2 μM each primer, and 50–150 ng of DNA template in a total volume of 50 μl. The amplification was programmed as 95 °C for 5 min; 42 cycles of 95 °C for 30 s, 62 °C for 30 s, 72 °C for 70 s; and 72 °C for 10 min. PCR products were directly sequenced (TSINGKE Biotech, Beijing, China). For clone sequencing, three to ten clones were sequenced. The sequences from clone sequencing were cross-checked with those from direct PCR product sequencing to clarify the two haplotypes in each heterozygote.

Table 1 Brief information about PCR
Full size table

Primers 8F-1 and 8R-1 (Table 1) were used to amplify a fragment of 167 bp covering codon 345 of the AsRDL gene (named as RDL8). The reaction mixture (30 μl) included 2× Taq PCR MasterMix, 0.2 μM each primer and 30–100 ng of genomic DNA. The reaction procedure was set as 95 °C for 5 min; 38 cycles of 95 °C for 30 s, 56 °C for 30 s, 72 °C for 10 s; and 72 °C for 10 min. PCR products were digested with the restricted DNA digestion enzyme HpyCH4 III (New England Biolabs) (Fig. 1a) for 2 h in a total volume of 20 μl according to the manufacturer’s instruction. The digestion products were detected by 3% agarose gel electrophoresis. The genotype of each sample was identified based on the resulting gel profiles: one band of 167 bp for homozygous TCG (SS), one band of combined 83 bp and 84 bp for homozygous ACG (TT) (complete digestion of the 167 bp DNA by HpyCH4 III resulted in two fragments with a length of 83 bp and 84 bp respectively, which could not be distinguished on a routine agarose gel); two bands (167 bp + 83 bp) for T/ACG (S/T) heterozygote (Fig. 1b).

Fig. 1
figure 1

Diagnostic PCR-RFLP tests to detect the T345S mutation. The 167 bp amplicon is undigested by HpyCH4 III for homozygous TCG (mutant 345SS), and cut into two fragments of 83 bp and 84 bp for homozygous ACG (wild 345TT) that co-migrate in the agarose gel and cannot be distinguished. A combined pattern (two bands 167 bp + 83/84 bp) is displayed for heterozygotes

Full size image

Data analysis

Each sequencing result of RDL7 was manually checked and the terminal ambiguous regions were trimmed. All the confirmed sequences (360 bp) were aligned using Muscle 3.8 [14], and the polymorphic sites in the sequence were recorded. Hardy–Weinberg equilibrium (HWE) of genotypes in each population was tested by probability test using the online software GENEPOP v.4.2 [15]. The haplotypes of heterozygotes were clarified by clone sequencing. Haplotype network analysis was conducted using Network 5.0 [16].

Results

Nucleotide variations in the RDL7 region of the AsRDL gene

Ten polymorphic sites (PS) were observed in RDL7 in a total length of 360 bp, three PSs in exon 7 and seven in intron 7 (Fig. 2). The 2nd and 3rd PSs represented synonymous mutations, causing amino acid substitutions at sites 296 (A296S) and 327 (V327I), respectively.

Fig. 2
figure 2

The polymorphic sites identified in the AsRDL gene in this study. Non-synonymous sites are indicated by red dots

Full size image

Distribution and frequency of insecticide resistance-related mutations

The distribution and frequency of the three insecticide resistance-related mutations are shown in Table 2 and Fig. 3. Overall, the frequency of A296S was very high (79–100%) in the seven examined populations. Notably, this allele was fixed in four of the seven populations. By contrast, the frequency of the susceptible allele was rare: A296 was only detected in three populations (PXSS, JXTD, JXXW) at frequencies ranging from 1.5 to 21%.

Table 2 Frequency of RDL genotypes in seven An. sinensis populations along the China–Vietnam border
Full size table
Fig. 3
figure 3

The sampling location (a), and distribution frequencies of AsRDL allele (b–d). NPPM Pingmeng, Napo County; JXXW Xinwei, Jingxi County; JXTD Tongde, Jingxi County; DXEC Encheng, Daxin County; LZSK Shuikou, Longzhou County, PXSS Shangshi, Pingxiang County; FCGDX Dongxing town, Fangchenggang City

Full size image

The three possible genotypes for site 327 were detected in all the seven populations, and all genotypes agreed with Hardy–Weinberg equilibrium (Table 2). The frequency of the mutant 327I ranged from 26.9 to 53.2% (Fig. 3). The mutant allele (345S) was detected in five of the seven populations, but absent in DXEC and CZSK. Overall, the frequency of 345S was low (0–28.8%), with the highest frequency being observed in FCGDX (Fig. 3).

Frequency and distribution of triple-locus genotype combinations

When considering the three resistance-related sites together, 11 types of triple-locus genotypes (C1 to C11) were observed (Table 3). The number of triple-locus genotype combinations observed in each sampling site ranged from 3 (CZSK) to 9 (JXTD). Three combinations (C1, C4, C8) were distributed in all the seven populations, while C6 and C7 were uniquely found in PXSS and FCGDX, respectively. The triple-locus wild genotype (C11) was distributed in JXTD and JXXW. The alleles carrying 327I were found only in individuals that harbor 296S, and 345S existed in individuals that were homozygous for 296S. Notably, the triple-locus mutant allele was detected (C6 and C7), although at low frequencies.

Table 3 Distribution and frequency of triple-locus genotype combinations in seven An. sinensis populations along the China–Vietnam border
Full size table

Evolutionary origin of 296S and 327I

To track the possible origin of mutant 296S and 327I, the DNA sequences of 360 bp-in-length (Fig. 2) were used for haplotype identification and network analysis. A total of 21 haplotypes (H1 to H21) were identified (Table 4). Among them, 4 (H1 to H4) were wild haplotypes without 296S and 327I mutations, 11 (H5–H15) were haplotypes with only the 296S mutation, and 6 (H16–H21) carried both 296S and 327I mutations. The 327I mutation was found always together with 296S. Network analysis indicated that the 296S allele might have evolved independently from more than one origin (Fig. 4). The single mutation haplotypes H13 and H10 may be evolved from H2 and H3, respectively, while the five double-mutation haplotypes (H17–H21) could be derived from some 296S haplotypes via only one mutational step (Fig. 4).

Table 4 RDL haplotypes in seven An. sinensis populations along the China–Vietnam border
Full size table
Fig. 4
figure 4

The network of RDL haplotypes identified in An. sinensis populations along the China–Vietnam border

Full size image

Discussion

Point mutations (A to S/G/N) at the site equivalent to 301 in Drosophila (site 296 of AsRDL) in the second transmembrane domain of RDL have been documented in diverse insecticide-resistant insects [10,11,12, 17,18,19]. In this study, the resistant mutation 296S was detected and its frequency was very high (79–100%) in the seven examined populations, suggesting a strong risk of resistance to GABA targeting insecticides in these mosquito populations. The prevalence of mutation at 296 was also detected in An. sinensis collected in other locations of Guangxi [13], and in Anopheles funestus in Africa [12]. Given that the use of cyclodienes has been banned, the high level of 296S in An. sinensis in Guangxi may be a consequence of the increasing use of insecticides targeting the GABA receptor (e.g. ethiprole, fipronil) in the agriculture [20], or selection pressures by organochlorine pesticides from unknown origins. Another explanation may be that A296S substitution has no fitness cost under current natural conditions, or the cost mediated by A296S is alleviated by unknown fitness modifiers.

Two mutations (V327I and T345S) previously detected in another nine An. sinensis populations from Guangxi [13] were also found in the current study. Interestingly, either 327I or 345S was present together with 296S. The 327I mutation was also previously identified in dieldrin-resistant An. funestus [12]. Similarly, a conserved mutation at the equivalent site of RDL was also reported in Anopheles gambiae (T345M) [8] and Drosophila (T350M/S) [21, 22]. Although the two conserved mutations have been suggested to play a role in insecticide resistance when they occurs together with an amino acid substitution at the site 296 [12, 22], their impact on the pharmacology of the GABA receptor and contribution to insecticide resistance remains unclear. This would be an interesting question worthy of study.

Network analysis does not support single origin of the resistance-conferring 296S mutation and the related 327I mutation in AsRDL. The two wild A296 haplotypes (H1 and H4) could be considered as two possible ancestors of resistant 296S haplotypes identified in this study (Fig. 4). Multiple origins of 345S were also inferred in the previous study [13].

Conclusion

A PCR-RFLP diagnostic assay was developed to genotype the T345S mutation in the AsRDL gene. Three previously reported mutations (A296S, V327I and T345S) were detected in the seven An. sinensis populations in Guangxi along the China–Vietnam border. The prevalence of the resistance-related A296S mutation within An. sinensis populations along the China–Vietnam border indicates a risk of resistance to insecticides targeting the GABA receptor, such as dieldrin and fipronil. The resistance A296S allele may have multiple evolutionary origins, and the double mutations (A296S + V327I) may have evolved from alleles carrying the A296S mutation by scaffolding the additional mutation V327I.