Detection of G119S ace-1 ^R mutation in field-collected Anopheles gambiae mosquitoes using allele-specific loop-mediated isothermal amplification (AS-LAMP) method

Allele-specific LAMP (AS-LAMP) primers were designed based on published sequences of the acetylcholinesterase gene carrying the ace-1 ^R mutation [10]. To detect the mutant and wild type of ace-1 ^R , the specific forward inner primer (FIP) and backward inner primer (BIP) were constructed with the mutation on the 5′ end. An additional mismatched nucleotide was added at the penultimate position of each specific BIP to increase the specificity of the primers to each targeted nucleotide site (Fig. 1). For primer mismatch optimization, DNA fragments from An. gambiae mosquitoes carrying either sensitive or resistance-type ace-1 genes were used as the template DNA of the LAMP reaction. The outer primers (B3 and F3) were identical for the two primer sets.

All mosquitoes used in this study were collected during the rainy season of 2008 in the village of Soumousso (N 11°00, W 04°03) located in the Sudan Savannah zone, 380 km from Ouagadougou and 38 km from Bobo-Dioulasso. The village is located within the cotton belt, where insecticide pressure is higher than in other parts of the country, as is the frequency of ace-1 ^R [4]. Mosquitoes were collected as larvae and reared to adults for morphological classification of species and sex. Only the female An. gambiae complex mosquitoes were used for analysis. The mosquitoes were kept in silica gel containing tubes until used.

The genomic DNA of field-collected mosquitoes was extracted by homogenizing individual mosquitoes with a pellet pestle in 100 µl Buffer A [0.1 M Tris HCl (pH 9.0), 0.1 M EDTA, 1 % SDS, and 0.5 % diethyl pyrocarbonate] and incubating the homogenate for 30 min at 70 °C. After incubation, 22.4 µl of 5 M potassium acetate was added, and the mixture was cooled on ice for 30 min. After centrifugation at 15,000 rpm for 15 min at 4 °C, the DNA-containing supernatant was transferred into a new tube and mixed with 45 µl isopropanol. The solution was centrifuged at 15,000 rpm for 20 min at 4 °C, and the supernatant was discarded. The DNA pellet was rinsed with 70 % ethanol and dried. The pellet was then diluted in 30 µl TE buffer.

The LAMP reaction was carried out following manufacturer recommendations (Eiken Chemicals, Tokyo, Japan). A master mix was prepared using 6.25 µl of 2× reaction mix (2.8 mM each of dNTP, 40 mM Tris–HCl (pH 8.8), 20 mM KCl, 16 mM MgSO₄, 20 mM (NH₄)₂SO₄, 0.2 % Tween 20, and 1.6 M betain), 2.75 µl of distilled water and 0.5 µl of each of the primers and the Bst DNA polymerase. The concentration of the primers used was 40 pmol/µl for the inner primers (BIP and FIP) and 5 pmol/µl for the outer primers (B3 and F3). A volume of 11.5 µl of the master mix was placed into test tubes (including a negative control). One µl of DNA solution was added to the sample tubes and 1 µl of distilled water to the negative control tube. All procedures were carried out on ice. The tubes were then incubated in a water bath or LA-200 real-time turbidimeter (Eiken Chemicals) at 63 °C, and turbidity was measured. The reaction was terminated by heating the tube at 95 °C for 5 min. A 2 % agarose gel electrophoresis at 100 V of the LAMP products was performed and the gel was stained with ethidium bromide for examination under UV light to check amplification bands.

The MR4 protocol [25] (based on Weil et al. [10]) was used for ace-1 ^R mutation detection method by PCR-RFLP analysis. The ace-1 ^R SNP region was amplified with two primers (MOUSTDIR1:) (25 pmol/μl) [CCGGGNGCSACYATGTGGAA] MOUSTREV1 (25 pmol/μl) [ACGATMACGTTCTCYTCCGA]. PCR was carried out in a 25-µl tube using KOD FX Neo DNA polymerase (Toyobo, Osaka, Japan). Amplification was performed with 1 µl of genomic DNA as the template using the following PCR programme: 94 °C/2 min × 1 cycle, (98 °C/10 s, 68 °C/30 s) × 35 cycles, 68 °C/2 min × 1 cycle, and 4 °C hold. The PCR product was digested by adding 1 μl AluI (New England Biolabs, Ipswich, MA, USA) restriction enzyme, 2 μl of H₂0, and 2 μl of buffer to 15 μl. An incubation at 37 °C for 8–24 h [25] was observed. Finally, 5 µl of the digestion product was run on 2 % agarose gel and stained in ethidium bromide.

RT-PCR detection of ace-1 ^R was carried out based on the methods in the Anopheles Research protocol [25, 26], with the primers (forward: GCCGTCATGCTGTGGATCTT, reverse: GCCCGGTGGTCGTACAC) and probes (wild type allele: VIC-CGGCGGCTTCTAC, mutant type allele: FAM-CGGCAGCTTCTAC) with TaqMan GTXpress master mix (Life Technologies, Carlsbad, CA, USA) and analysed by StepOne RT-PCR machine (Life Technologies). For the positive controls, DNA fragments from An. gambiae mosquitoes carrying either sensitive or resistance-type ace-1 genes cloned into pBSSK vectors were used. PCR alignment scores were taken from respective endpoint scatter plots.

DNA from An. gambiae mosquitoes was used as PCR templates. The target sequence was amplified using the PCR primers BA3 (GCTGGTGGTCAACACGGATA) and BA4 (CCCAGCTGACTCCCAAGAAG), designed to amplify a 603-bp DNA fragment of ace-1 including its SNP region. PCR was performed in a tube with KOD FX Neo DNA polymerase (0.5 µl). The PCR programme was 94 °C/2 min × 1 cycle, (98 °C/10 s, 68 °C/30 s) × 35 cycles, 68 °C/2 min × 1 cycle, and 4 °C hold. The PCR product was purified using a geneclean kit (MP Biomedicals, Santa Ana, CA, USA) and the final product was diluted in 10 µl of distilled water. The nucleotide sequences were determined with a BigDye terminator sequencing kit v. 3.1 (Life Technologies, Carlsbad, CA, USA) using an automated 3100 genetic analyzer (Life Technologies).

Sensitivity was defined as the ratio of true positives to combined true and false positives, and specificity as the ratio of true negatives to combined true negatives and false negatives. Calculation of specificity and sensitivity confidence limits of AS-LAMP and RT-PCR was carried out using the Wilson score method [27].

The BIP and FIP primers, containing the SNP with an additional mismatched nucleotide attached to the penultimate nucleotide at the 3′ (BIP) or 5′ end (FIP), were designed as specific primers to detect the G119S single nucleotide by AS-LAMP. The two specific primers and the mismatched nucleotide increase the specificity of both primers for their target DNA. The different regions targeted by the primers and the primer sequences are shown in Fig. 1.

The primers designed for AS-LAMP were able to distinguish between the wild type allele (ace-1 ^S ) and the mutated type allele (ace-1 ^R ). Detection time for the wild type mutation was ~ 50 min after incubation in the turbidimeter using the wild type primers; detection time for heterozygous samples was 64 min using the same primers. Using these primers, no amplification of the resistant allele (ace-1 ^R ) took place within the 75-min test period (Fig. 2a).

For ace-1 ^R resistant type mutation detection, amplification started at ~51 min after incubation for resistant homozygous samples and at ~55 min for heterozygous samples, with no amplification of the wild type (ace-1 ^S ) taking place within the 75-min test period when using the resistant type primers (Fig. 3a). Gel electrophoresis of the LAMP products confirmed that the amplification was primer DNA-specific, i.e., the primers could only amplify their target-specific DNA (Figs. 2b, 3b).

PCR-RFLP detection based on the protocol of Weill et al. [10] was used to detect the ace-1 ^R mutation in 104 field-collected mosquitoes. Three homozygous resistant samples, nine heterozygous samples and 95 sensitive homozygous samples were identified.

RT-PCR detection of ace-1 ^R , based on the methods in the Anopheles Research protocol [25], found 12 heterozygous samples and 92 homozygous samples for the sensitive alleles, while failing to clearly distinguish between any of the homozygous versus heterozygous samples (Fig. 4; Table 1).

To investigate the comparative sensitivity for detecting each genotype of the ace-1 ^R mutation, ace-1 ^R mutation detection using the AS-LAMP, PCR-RFLP and RT-PCR methods for 104 mosquitoes was performed. The samples were sequenced first using standard PCR (Table 1): these sequences were treated as the ‘gold standard’ for identification. The same samples were then analysed using AS-LAMP, PCR-RFLP and RT-PCR, and the respective results were compared against the corresponding standard PCR-derived sequence. AS-LAMP and PCR-RFLP had the same sensitivity for the homozygous resistant type [100 %, confidence limit (CL) 21–100 %], while RT-PCR could not distinguish clearly any resistant specimen (Fig. 4). For detection sensitivity of the heterozygous type, there was no statistically significant difference between AS-LAMP (89 %, CL 56–99 %), PCR-RFLP (100 %, CL 79–100 %), and RT-PCR (75 %, CL 47–91 %), although the sample number may have been too few to detect significance. The specificity of all three methods was perfect (100 %) for resistant homozygous and heterozygous detection. There were, however, slight (albeit not statistically significant) differences in detection specificity for the sensitive homozygous type (PCR-RFLP and AS-LAMP 90 %, CL 60–98 %; RT-PCR 100 %, CL 72–100 %).

To evaluate whether the AS-LAMP reaction could be done with only a water bath and scored by eye, without a real-time turbidimeter, 24 DNA samples used in Table 1 were randomly chosen and subjected to AS-LAMP with a water bath. As shown in Table 2, the naked eye detection results of all samples were completely identical to that obtained with the real-time turbidimeter.