Voltage-sensitive sodium channel mutations S989P + V1016G in Aedes aegypti confer variable resistance to pyrethroids, DDT and oxadiazines.

BACKGROUND: Aedes aegypti is a vector of several important human pathogens. Control efforts rely primarily on pyrethroid insecticides for adult mosquito control, especially during disease outbreaks. A. aegypti has developed resistance nearly everywhere it occurs and insecticides are used. An important mechanism of resistance is due to mutations in the voltage-sensitive sodium channel (Vssc) gene. Two mutations, in particular, S989P + V1016G, commonly occur together in parts of Asia.
RESULTS: We have created a strain (KDR:ROCK) that contains the Vssc mutations S989P + V1016G as the only mechanism of pyrethroid resistance within the genetic background of Rockefeller (ROCK), a susceptible lab strain. We created KDR:ROCK by crossing the pyrethroid-resistant strain Singapore with ROCK followed by four backcrosses with ROCK and Vssc S989P + V1016G genotype selections. We determined the levels of resistance conferred to 17 structurally diverse pyrethroids, the organochloride DDT, and oxadiazines (VSSC blockers) indoxacarb (proinsecticide) and DCJW (the active metabolite of indoxacarb). Levels of resistance to the pyrethroids were variable, ranging from 21- to 107-fold, but no clear pattern between resistance and chemical structure was observed. Resistance is inherited as an incompletely recessive trait. KDR:ROCK had a > 2000-fold resistance to DDT, 37.5-fold cross-resistance to indoxacarb and 13.4-fold cross-resistance to DCJW.
CONCLUSION: Etofenprox (and DDT) should be avoided in areas where Vssc mutations S989P + V1016G exist at high frequencies. We found that pyrethroid structure cannot be used to predict the level of resistance conferred by kdr. These results provide useful information for resistance management and for better understanding pyrethroid interactions with VSSC.

INTRODUCTION

Aedes aegypti is an anthropophilic mosquito and vector of four important human disease viruses: dengue, yellow fever, chikungunya, and Zika. It is estimated that dengue is a risk to >40% of the world's population. There are an estimated 50 to 100 million dengue virus infections a year worldwide.1, 2, 3 Yellow fever, like dengue, is a viral hemorrhagic fever. This can be a lethal disease if not properly treated and despite the existence of a preventative vaccine, there are still ∼ 200 000 cases of yellow fever each year, resulting in 30 000 deaths worldwide.4 Chikungunya and Zika viruses are new to the Americas as of 20135 and 2015,6 respectively. Chikungunya is no longer a disease of developing and low‐income countries, and can be endemic even in the wealthy and developed countries of temperate regions.7 Zika is an emerging disease that has existed in Africa and Asia with little apparent impact on human health prior to 2007.8 This disease has generated a great deal of human health concern since it reached the Americas in 2014 because of its rapid spread and associations with microcephaly and Guillain‐Barré syndrome.9 Given that A. aegypti has a wide global distribution and thrives in urban environments, it poses a serious risk to human health.

Insecticide resistance is a problem affecting the control of a wide range of pests and almost every aspect of human life including health, agriculture and recreation. As of 2012, over 450 insecticide‐resistant arthropods species have been reported, with at least 198 of these being of medical importance.10, 11 In regions where A. aegypti‐transmitted diseases are most prevalent, mosquitoes have high levels of resistance to insecticides due to intensive selection pressure.12 This compromises our ability to control these mosquitoes and the diseases they spread.

Pyrethroid insecticides, such as cypermethrin, deltamethrin and permethrin, are widely used for control of mosquitoes, including A. aegypti.13 Pyrethroids are one of the most commonly used insecticide classes because of their high toxicity to insects and low risk to mammals. Insecticides are still the primary means by which to control A. aegypti in endemic areas, and this has led to extensive use of pyrethroids for the last three decades.

A. aegypti has developed two major resistance mechanisms against pyrethroids, and resistance is found worldwide.12 The two major mechanisms are mutations in the voltage‐sensitive sodium channel (Vssc) gene (generally known as knockdown resistance or kdr) and detoxification by cytochrome P450 monooxygenases (CYPs).14 Pyrethroids act directly on the insect VSSC15 and certain mutations in Vssc (such as S989P + V1016G) make the channel less sensitive to pyrethroids. The S989P + V1016G Vssc mutations (herein referred to as kdr) are common in Asia,12 yet the level of resistance these mutations confer to pyrethroids is not known.

In this study, we determined the levels of resistance conferred to A. aegypti by Vssc mutations S989P + V1016G (kdr) for 20 insecticides that target the VSSC: 17 structurally diverse pyrethroids, the organochloride DDT and two oxadiazines (VSSC blockers indoxacarb [proinsecticide] and DCJW [active metabolite of indoxacarb]).16 Our goal was to discover if there are any relationships between pyrethroid structure and the resistance conferred by these mutations, and to identify insecticides that should be avoided in areas where Vssc mutations S989P + V1016G exist at high frequencies. Knowing the levels of resistance to structurally diverse pyrethroids facilitates a greater understanding of how pyrethroids interact with VSSC and will provide insight into what pyrethroid structures are most affected by the S989P + V1016G mutations. Information from this study has direct application for resistance management practices because it will identify the chemicals that may be the most, or least, effective against a population harboring these resistance mutations.

MATERIALS AND METHODS

Strains

In order to study the effects of Vssc mutations S989P + V1016G (kdr) without the influence of other resistance mechanisms or strain variations, we needed a strain of A. aegypti that was closely related to a susceptible strain but containing kdr. To accomplish this, we used the following parental strains: Rockefeller (ROCK), an insecticide‐susceptible strain that has been reared in laboratories without exposure to insecticides for decades, which originated from the Caribbean,17 and Singapore (SP), a pyrethroid‐resistant strain in which the mechanisms of resistance have been well studied.14 Adult SP are 579‐fold resistant to permethrin (relative to ROCK) due to kdr and CYP‐mediated detoxification, whereas hydrolases and decreased penetration are not involved.14

Mosquitoes were reared at 27 ± 1 °C and 70–80% relative humidity (RH). Females were blood‐fed using membrane‐covered water‐jacketed glass feeders with cow blood obtained locally. Adults were maintained on 10% sugar water in cages ∼ 35 × 25 × 25 cm holding ≤1000 mosquitoes. Larvae were reared in 27.5 × 21.5 × 7.5 cm containers with ∼ 2 L deionized water and fed Cichlid Gold fish food pellets (Hikari, Hayward, CA, USA) (ground pellets for first instar and medium size pellets for second to fourth instars). Food pellets were given daily, as needed depending on larval density (∼ 400–600) and instar.

The strain KDR:ROCK was isolated from crossing ROCK with SP followed by four backcrosses and genotype selections (see Section 2.2). KDR:ROCK is congenic to ROCK, but resistant to pyrethroids due to the Vssc mutations S989P + V1016G while containing no CYP‐mediated resistance. The procedure for isolating KDR:ROCK is illustrated in Fig. 1. In short, unmated SP females were crossed en masse with ROCK males. Unmated F1 females were backcrossed with ROCK and unmated BC1 females were genotyped for the presence of Vssc mutations S989P + V1016G (see Section 2.2). BC1 females that were heterozygous for P989 + G1016 (kdr) were backcrossed to ROCK males. This process was repeated for the BC2 and BC3 generations. At BC4, both males and unmated females were genotyped and individuals that were heterozygous for P989 + G1016 were reared en masse. Males and unmated females from the next generation were genotyped and individuals that were homozygous for P989 + G1016 were reared en masse. The resultant strain was named KDR:ROCK (Fig. 1).

Figure 1

Protocol for isolating the congenic strain containing Vssc mutations V1016G + S989P (KDR:ROCK). Number of ROCK males included in each backcross varied depending on the number of females added to the cage (approximately half the number of females).

Genotype selections

Genomic DNA was extracted from the hind tarsus of unmated adults using an alkali extraction method as follows: legs were placed in individual wells of a 96‐well PCR plate (Bio‐Rad, Hercules, CA, USA) with approximately three 2.3‐mm diameter zirconia/silica beads (BioSpec Products, Bartlesville, OK, USA) and 10 µL of 0.2 m NaOH per tube. The samples were pulverized for 1–2 min on a vortex mixer at maximum speed and then incubated for 10 min at 70 °C. Ten microliters of neutralization buffer (360 mM Tris–HCl, pH 8.0 and 10 mM EDTA) and 80 µL ddH2O were then added to each well. PCR was carried out using 1 µL gDNA, 12.5 µL GoTaq® Green Master Mix 2x (Promega, Madison, WI, USA), 9.5 µL ddH2O, and 2 µL of 10 µM forward and reverse primer mix (see Table 1) using the following thermocycler conditions: 94 °C for 2 min, 35 × (94 °C for 30 s, 60 °C for 30 s, 72 °C for 30 s) and 72 °C for 10 min.

Table 1

Vssc primers used for genotype selections.

Primer	Sequence	Purpose
AaSCF20	GACAATGTGGATCGCTTCCC	PCR amplification (Forward)*
AaSCR21	GCAATCTGGCTTGTTAACTTG	PCR amplification (Reverse)*
AaSCR22	TTCACGAACTTGAGCGCGTTG	Sequencing
Ser 1F	GCGGCGAGTGGATCG AAT	Allele‐specific susceptible genotype (Forward)†
Val 1R	GCAAGGCTAAGAAAAGGTTAAGTA	Allele‐specific susceptible genotype (Reverse)†
Pro 1F	GCGGCGAGTGGATCGAAC	Allele‐specific resistant genotype (Forward)†
Gly 1R	GCAAGGCTAAGAAAAGGTTAAGTC	Allele‐specific resistant genotype (Reverse)†

PCR product = 614 bp

ASPCR product = 357 bp

The genotypes were determined using two methods: Sanger sequencing (BC1 to BC3) and allele‐specific polymerase chain reaction (ASPCR) (BC4 and BC4F1). For Sanger sequencing genotyping, PCR products were treated with ExoSAP (5 µL PCR product/1 µL ExoSAP mix), and sequenced (6 µL ExoSAP treated PCR product, 1 µL primer, 11 µL ddH2O) at the Cornell University Biotechnology Resource Center. For ASPCR, forward and reverse primers (see Table 1) were made for both the resistant (kdr) and susceptible genotypes based on methods reported by Li et al.18 Susceptible and kdr primers were used on each sample with the following thermocycler conditions: 94 °C for 3 min, 35 × (94 °C for 30 s, 62 °C for 30 s, 72 °C for 1 min) and 72 °C for 7 min. Each ASPCR was evaluated on a 1% agarose gel and was scored as homozygous susceptible (ASPCR band only with susceptible primers), homozygous kdr (ASPCR band only with kdr primers) or heterozygous (ASPCR band with both kdr and susceptible primers). The ASPCR was validated by using samples of known genotype based on sequences acquired from the Sanger sequencing method to ensure ASPCR results matched the sequencing results (Fig. S1). Control (ROCK, SP and F1 [ROCK × SP]) templates were run on every 96‐well plate.

Insecticides

A total of 20 insecticides that target the VSSC and two synergists were used in this study: acrinathrin (88.7%, Chem Service, West Chester, PA, USA), bioallethrin (3.8% cis, 95.7% trans, Chem Service), bioresmethrin (99.5%, Chem Service), cis‐permethrin (99%, Chem Service), cyfluthrin (98%, Bayer CropScience, Leverkusen, Germany), cyhalothrin (90.2%, ICI Americas, Wilmington, DE, USA), cypermethrin (94.6%, ICI Americas), DDT (1,1‐bis(4‐chlorophenyl)‐2,2,2‐trichloroethane) (98%, Sigma‐Aldrich, St. Louis, MO, USA), deltamethrin (99.5%, Chem Service), diethyl maleate (DEM) (97%, Sigma‐Aldrich), etofenprox (96.3%, Mitsui Toatsu, Tokyo, Japan), 1R‐trans fenfluthrin (100%, Bayer CropScience), fenpropathrin (99.5%, Chem Service), flumethrin (100%, Bayer CropScience), indoxacarb (DPX‐MP062, 100%), and its bioactive metabolite DCJW, (98%, DuPont, Wilmington, DE, USA), tau‐fluvalinate (93.8%, Chem Service), permethrin (99.5% pure, 24.1% cis, 75.8% trans, Chem Service), piperonyl butoxide (PBO) (90%, Sigma‐Aldrich), tefluthrin (96%, Sigma‐Aldrich), transfluthrin (99.5%, Chem Service) and trans‐permethrin (99% FMC, Philadelphia, PA, USA). All insecticides were diluted in acetone (VWR, Radnor, PA, USA).

Adult bioassays

Topical adult bioassays were done on 3–7‐day‐old mated females. Mosquitoes were briefly anesthetized with CO2 and held on ice. A 0.22‐µL drop of insecticide in acetone was applied to the thorax of each mosquito using a Hamilton PB‐600 repeating dispenser equipped with a 10‐µL syringe. Controls were treated with acetone only. At least five doses were used per bioassay with at least three giving mortality values between 0 and 100%, and each containing 20 mosquitoes. Mosquitoes were given a cotton ball saturated with distilled water and held at 25 °C. A minimum of four replicates over at least 2 days and two cages were done per strain for each insecticide. Mortality was defined as mosquitoes that were ataxic after 24 h. For indoxacarb, mortality was recorded at 72 h due to the slow action of this insecticide, and mosquitoes were given sugar water‐saturated cotton balls that were rehydrated at 24 and 48 h during the bioassay. Probit analysis,19 as adapted to personal computer use20 using Abbott's21 correction for control mortality, was used to calculate the LD50 and the 95% confidence intervals (CI). All the bioassay data fit a line (chi‐square test). Resistance ratios (RR) were calculated by dividing the LD50 of KDR:ROCK by the LD50 of ROCK. Significant differences were determined by calculating the RRs for the minimum and maximum LD50 values based on the 95% CI. If the minimum and maximum RR values did not overlap, they were deemed significantly different.

Bioassays using a synergist (PBO or DEM) were performed as described above, except that 2.5 µg PBO or 2.5 µg DEM was applied to each mosquito 2 h prior to the insecticide application. For this, the mosquitoes were anesthetized on ice twice, once for PBO or DEM and once for permethrin or DDT application. Controls included a double acetone and an acetone plus synergist application.

Larval bioassays

Bioassays using mosquito larvae were performed with permethrin to allow comparison of the levels of resistance conferred by Vssc mutations S989P + V1016G between life stages. Bioassays were carried out on fourth instar larvae reared as described above. Twenty larvae were placed in a 177‐mL frozen yogurt cup (Solo, Lake Forest, IL, USA) containing 99 mL deionized water and 1.0 mL permethrin in acetone. Controls received 1 mL acetone without insecticide. At least five concentrations were used per bioassay with at least three giving mortality values between 0% and 100%. A total of six replicates from two rearing batches were done for each strain. Larvae that were floating, ataxic or unable to surface were counted as dead.

Degree of dominance

Unmated ROCK females (∼ 400) were crossed with KDR:ROCK males (∼ 200) and the F1s were tested (as described in Section 2.4) to determine the degree of dominance (D) of the Vssc mutations S989P + V1016G. Reciprocal crosses were not needed because resistance is not sex‐linked nor due to maternal factors.14 Permethrin was used to determine D because it was the insecticide used for selecting the parent resistant strain SP. D was calculated using the formula given by Stone 1968.22

RESULTS

To confirm that KDR:ROCK did not have CYP‐mediated resistance, we performed permethrin bioassays with and without PBO. As expected, there was no significant difference between the RR values for permethrin 40 and for PBO + permethrin 42 (Fig. 3). We conclude that KDR:ROCK does not have CYP‐mediated resistance.

We tested 17 pyrethroids against KDR:ROCK and ROCK (Fig. 2). RR values varied by up to fivefold for the different pyrethroids (Fig. 3) ranging from 21 (bioallethrin) to 107 (etofenprox). Permethrin, the insecticide used to select for resistance in the SP strain had a RR of 40. There was a significantly higher RR (1.5‐fold) for trans‐permethrin compared with cis‐permethrin. Interestingly, the RR for permethrin, which is a mix of isomers containing 75.8% of the trans isomer, was not significantly different from cis‐permethrin. Comparing insecticides with a single structural change in the acid moiety revealed some interesting differences and similarities. Cyfluthrin, which differs from flumethrin by a chlorobenzene substitution in the acid moiety (see Fig. 2), had a 2.5‐fold lower resistance relative to flumethrin. Cyhalothrin and cypermethrin differ only due to a trifluoro‐ rather than a chloro‐ substituent in the alkene, yet the RR value for cyhalothrin was approximately twofold higher than cypermethrin. However, we found no significant difference between cypermethrin and deltamethrin, which differ only by chlorines (cypermethrin) versus bromines (deltamethrin) on the alkene substituent. No broad generalization can be made concerning larger structural changes in the acid moiety and resistance levels, despite the variable RRs. Cyhalothrin, for example, which has an identical structure to acrinathrin, except for the substituent on the alkene group, has a nearly twofold greater RR than acrinathrin. However, there was no significant difference between RRs for acrinathrin, deltamethrin and cypermethrin, which also have different alkene substituents. Comparing insecticides with a single structural change in the alcohol moiety revealed no differences in RR values. Transfluthrin and 1‐R trans flumethrin, for example, vary only by a single fluorine (on the benzene), but do not differ significantly in RR values. There was also no trend between the resistance levels of pyrethroids with and without α‐CN groups. Comparing pyrethroids with large structural differences in the alcohol portion of the molecule, but with identical acid moieties, also showed no general trends. In one case, a change in the alcohol caused a threefold increase in resistance (bioresmethrin versus bioallethrin), whereas in others (transfluthrin versus 1‐R trans fenfluthrin), there was no significant difference in resistance. Therefore, despite the variable RR values, we observed no general relationships between the RR values and chemical structures.

Figure 2

Chemical structures of the VSSC targeting insecticides used: pyrethroids (a), oxadiazines (b) and DDT (c). Stereochemistry is not shown for all structures.

Figure 3

Resistance ratios (RRs) for the 17 pyrethroids, DDT, DCJW and indoxacarb. RRs were calculated as the LD50 of KDR:ROCK divided by the LD50 of ROCK. Error bars indicate the minimum and maximum 95% confidence interval range for the RRs of each insecticide.

We determined the resistance to DDT because Vssc mutations are known to confer cross‐resistance between pyrethroids and DDT. The RR for DDT was >2000‐fold with only 27% mortality observed at the maximum concentration that was feasible to use (Fig. 3). CYPs and glutathione‐S transferases (GSTs) can both metabolize DDT.23, 24, 25 To determine whether KDR:ROCK had metabolism‐mediated resistance to DDT (in addition to kdr), we used the CYP and GST inhibitors PBO and DEM,26 respectively. Neither PBO nor DEM measurably increased the toxicity of DDT to the KDR:ROCK strain (Table 2), suggesting that neither CYPs nor GSTs played a major role in the observed DDT resistance.

Table 2

Toxicity of insecticides that target the VSSC against susceptible (ROCK) and resistant (KDR:ROCK) strains of Aedes aegypti

Insecticide	Strain	LD50 (95% CI)	Slope (SE)	n
Acrinathrin	ROCK	0.021 (0.019–0.024)	3.9 (0.4)	480
Acrinathrin	KDR:ROCK	0.62 (0.54–0.70)	2.7 (0.2)	580
Bioallethrin	ROCK	3.00 (2.72–3.31)	4.5 (0.4)	480
Bioallethrin	KDR:ROCK	63.2 (56.1–70.9)	4.0 (0.5)	520
Bioresmethrin	ROCK	0.68 (0.61–0.75)	5.5 (0.7)	383
Bioresmethrin	KDR:ROCK	41.5 (39.0–44.3)	4.1 (0.2)	1456
Cis‐permethrin	ROCK	0.55 (0.50–0.62)	4.4 (0.5)	540
Cis‐permethrin	KDR:ROCK	24.4 (21.3–28.0)	2.8 (0.3)	520
Cyfluthrin	ROCK	0.023 (0.021–0.026)	3.5 (0.3)	700
Cyfluthrin	KDR:ROCK	0.54 (0.49–0.60)	3.9 (0.4)	680
Cyhalothrin	ROCK	0.021 (0.02–0.023)	3.5 (0.2)	959
Cyhalothrin	KDR:ROCK	1.24 (1.05–1.49)	1.9 (0.2)	560
Cypermethrin	ROCK	0.062 (0.056–0.068)	3.3 (0.3)	840
Cypermethrin	KDR:ROCK	1.68 (1.51–1.87)	2.8 (0.2)	740
Deltamethrin	ROCK	0.012 (0.010–0.013)	3.2 (0.2)	850
Deltamethrin	KDR:ROCK	0.39 (0.35–0.44)	2.4 (0.1)	1269
Etofenprox	ROCK	2.06 (1.88–2.26)	4.0 (0.4)	620
Etofenprox	KDR:ROCK	221 (198‐ 246)	2.6 (0.2)	1030
1 R‐trans fenfluthrin	ROCK	1.16 (1.00–1.32)	2.7 (0.2)	858
1 R‐trans fenfluthrin	KDR:ROCK	54.8 (48.6–61.6)	3.1 (0.2)	660
Fenpropathrin	ROCK	0.53 (0.46–0.63)	2.7 (0.3)	671
Fenpropathrin	KDR:ROCK	28.7 (26.4–31.2)	2.9 (0.2)	1162
Flumethrin	ROCK	0.50 (0.45–0.55)	3.6 (0.3)	647
Flumethrin	KDR:ROCK	27.7 (24.4–31.3)	2.5 (0.2)	877
Permethrin	ROCK	0.56 (0.52 ‐0.60)	4.7 (0.3)	1040
Permethrin	KDR:ROCK	22.5 (20.8–24.4)	4.4 (0.3)	980
Permethrin	Rock × KDR:ROCK F1	1.80 (1.76‐1.84)	3.8 (0.1)	500
Permethrin + PBO	ROCK	0.22 (0.20–0.24)	3.1 (0.2)	1361
Permethrin + PBO	KDR:ROCK	9.14 (7.00–14.2)	1.7 (0.3)	504
Tau‐fluvalinate	ROCK	0.32 (0.29–0.36)	4.1 (0.4)	680
Tau‐fluvalinate	KDR:ROCK	22.5 (18.8–27.0)	1.7 (0.2)	660
Tefluthrin	ROCK	4.64 (4.14–5.25)	3.3 (0.3)	830
Tefluthrin	KDR:ROCK	134 (120–153)	2.3 (0.2)	1040
Transfluthrin	ROCK	0.66 (0.60–0.72)	3.3 (0.2)	840
Transfluthrin	KDR:ROCK	38.0 (35.1–41.5)	4.0 (0.3)	840
Trans‐permethrin	ROCK	0.46 (0.42–0.50)	4.1 (0.4)	600
Trans‐permethrin	KDR:ROCK	31.3 (28.5–34.3)	3.8 (0.3)	820
DCJW	ROCK	1.30 (1.20–1.41)	5.5 (0.6)	480
DCJW	KDR:ROCK	17.4 (15.1–19.9)	2.4 (0.2)	620
Indoxacarb	ROCK	7.84 (6.84–8.87)	2.7 (0.3)	500
Indocacarb	KDR:ROCK	293 (256–340)	2.0 (0.1)	770
DDT	ROCK	10.1 (9.31–11.0)	5.8 (0.6)	500
DDT	KDR:ROCK	> 22 000*	–	650
DDT + PBO	KDR:ROCK	> 22 000	–	360
DDT + DEM	ROCK	8.61 (7.58–9.68)	3.5 (0.4)	540
DDT + DEM	KDR:ROCK	> 22 000	–	358

Less than 50% kill at 22 µg/mosquito.

LD50, ng/mosquito.

We tested the VSSC blocker (oxadiazine), indoxacarb, to see if kdr conferred any cross‐resistance to this class of insecticide, which targets the VSSC, albeit with a different mode of action from pyrethroids or DDT.27 The RR for indoxacarb was 37.5‐fold (Fig. 3). Because of the high cross‐resistance, we tested DCJW to determine if the resistance was due to the Vssc mutations or the bioactivation of indoxacarb. Based on neurophysiological studies, indoxacarb itself has little toxicity towards VSSC, whereas its metabolite, DCJW, has high insecticidal potency (> 25‐fold increased toxicity).28 The RR to DCJW was 13.4; significantly less (2.8‐fold) than indoxacarb, but still significantly cross‐resistant compared to ROCK mosquitoes. This suggests that indoxacarb is bioactivated less in KDR:ROCK than in ROCK and may be a hint that the gene coding for the bioactivation enzyme is close to Vssc (i.e. at the resistance locus).

The inherent toxicity of the 20 insecticides, as judged by the ROCK LD50 varied over nearly three orders of magnitude (Table 2). There was even a 460‐fold variation in toxicity of the pyrethroids. Deltamethrin (LD50 = 0.01 ng/mosquito), acrinathrin (0.02 ng/mosquito), cyhalothrin (0.02 ng/mosquito), cyfluthrin (0.02 ng/mosquito) and cypermethrin (0.06 ng/mosquito) had highest toxicity (Fig. S2). The pyrethroids with the lowest toxicity were tefluthrin (4.64 ng/mosquito), bioallethrin (3.00 ng/mosquito), etofenprox (2.06 ng/mosquito) and 1‐R trans fenfluthrin (1.16 ng/mosquito). Generally, the pyrethroids with an α‐CN group were the most toxic. DDT was the least toxic insecticide tested, with a ROCK LD50 value of 10.1 ng/mosquito. Indoxacarb was the second least toxic compound (7.84 ng/mosquito), however DCJW had toxicity similar to that of some pyrethroids (1.30 ng/mosquito). We found no relationship between the pyrethroid LD50 to ROCK and the levels or resistance in KDR:ROCK (data not shown).

Fourteen of the pyrethroids tested in this study were also tested against a housefly strain, ALkdr, which is resistant to pyrethroids due to the classic L1014F mutation in an otherwise susceptible genetic background.29 This provided a unique opportunity to look at how these different Vssc mutations, L1014F and S989P + V1016G, affect pyrethroid resistance. Here, when we compared the RR vales for both housefly and mosquito Vssc mutations, we found no relationship (R 2 = 0.546) between them (Fig. 4). Four pyrethroids (etofenprox, permethrin, cyfluthrin and bioresmethrin) had higher resistance in ALkdr, whereas 10 (fenpropathrin, tau‐fluvalinate, flumethrin, 1‐R trans fenfluthrin, transfluthrin, acrinathrin, cyhalothrin, cypermethrin, tefluthrin and deltamethrin) were higher in KDR:ROCK (Fig. S3). However, there was no pattern between pyrethroid structure and resistance levels for either one.

Figure 4

Comparison of the fold difference in resistance ratios (RRs) for 14 pyrethroids between a strain with the S989P + V1016G mutations (KDR:ROCK; ) and one with the L1014F mutation (ALkdr; housefly).29 Both strains were congenic with the susceptible strains to which they were being compared. Insecticides with higher resistance due to S989P + V1016G are in light gray and those with higher resistance conferred by L1014F are in black.

We performed additional bioassays to compare resistance between the different life stages and to determine the inheritance of Vssc mutations S989P + V1016G. The KDR:ROCK larvae LD50 to permethrin was 48.3 ng mL−1 (95% CI = 42.8–54.5; n = 836), and the ROCK LD50 was 1.66 ng mL−1 (95% CI = 1.48–1.94; n = 573). Thus, the permethrin resistance of the adults (40‐fold) was not significantly different from that in larvae (29‐fold). The degree of dominance (D) for the F1 hybrid was −0.4, indicating an incompletely recessive inheritance of permethrin resistance.

DISCUSSION

There was a fivefold overall difference in RR levels between the structurally diverse pyrethroids. We found no general relationship between pyrethroid structure and RR values, but did observe a few differences and similarities due to single changes in chemical structures. These differences were, however, no more than 2.5‐fold. Similarly, we observed no general resistance patterns for larger alcohol or acid moiety changes in the pyrethroid structures. The major challenge in making a comparison such as this one is that the commercial availability of bioactive compounds with single, specific changes in their structures is limited. The problem is that even small changes in the chemical structure of a pyrethroid may make it lose its toxicity to insects, or fail to increase the toxicity enough to warrant developing it into a new commercial product.

The >2000‐fold DDT resistance appears to be due entirely to the Vssc mutations S989P + V1016G, because PBO and DEM were not able to measurably reduce the level of resistance. This high level of resistance raises the question of whether this pair of mutations evolved in response to DDT use (and subsequently conferred cross‐resistance to pyrethroids), or whether one or both mutations evolved in response to pyrethroid use. A common mechanism of resistance to DDT was initially linked to the gene knockdown resistance (kdr).30 It was demonstrated that kdr could confer cross‐resistance to pyrethrins and pyrethroids in several insects,31, 32, 33 including A. aegypti,34 and that this was due to insensitivity of the nervous system.33, 35, 36 This was subsequently found to be due to a L1014F mutation in the Vssc gene in houseflies and German cockroaches.37, 38 In Culex, the L1014F mutations was found to confer cross‐resistance between permethrin and DDT.39 In A. aegypti, the classic L1014F mutation in Vssc is absent, although other mutations have been found.12 In fact, over two decades the number of Vssc mutations that confer resistance has expanded greatly,40 but the timelines over which these mutations occurred is unclear, particularly because sequencing Vssc only became a reality more than a decade after pyrethroids were introduced. Thus, it is unclear whether DDT or pyrethroids (the two most commonly used pyrethroids in Singapore have been permethrin and cypermethrin41, 42) or both, are responsible for the evolution of the S989P + V1016G mutations in A. aegypti.

The 37.5‐fold cross‐resistance to indoxacarb was partially (2.8‐fold) lost when we tested the bioactivated form of this insecticide (DCJW), suggesting that the gene for the indoxacarb bioactivating enzyme was linked to the resistance locus. This led us to examine the genome sequence of A. aegypti 43, 44 for candidate genes. The enzyme that catalyzes the conversion of indoxacarb to DCJW is thought to belong to the esterase/amidase subclass of hydrolases.28 Using NCBI's most current A. aegypti genome assembly Aaeg5.0 (www.ncbi.nlm.nih.gov) we looked for ∼ 2 megabases in both upstream (5′) and downstream (3′) directions from Vssc. We found three genes (AAEL006034, AAEL006024 and AAEL006023) with hydrolase activity (according to gene ontology annotations GO:0016810 and GO:0016811 in VectorBase [www.VectorBase.org]) between ∼ 740 and 860 kb downstream from Vssc (Table S1). These genes are reasonable candidates for the indoxacarb bioactivation enzyme, and worthy of further study. There were no genes in the upstream region that were annotated as having esterase/amidase activity.

The mechanism responsible for the 13.4‐fold cross‐resistance to DCJW is unclear. This could be due to the S989P + V1016G Vssc mutations or to some other factor at the pyrethroid resistance locus. Few other studies have investigated resistance to oxadiazines in pyrethroid resistant strains.24, 45, 46, 47 Of those, only one had a known Vssc mutation (L1014F) in a resistant strain of housefly, and no cross‐resistance was observed.47 However, this is not the same mutation found in A. aegypti, so whether S989P + V1016G confers cross‐resistance is still unclear. One study done in field collected A. albopictus in Pakistan found resistance to multiple classes of insecticides including pyrethroids and indoxacarb;48 however, they did not evaluate the mechanisms of resistance present. Identification of the factor responsible for the DCJW cross‐resistance in KDR:ROCK will require further study.

When comparing the resistance levels in the KDR:ROCK strain to a housefly strain (ALkdr) with the L1014F mutation,29 we found no relationship between the resistance levels for the different pyrethroids. This suggests that L1014F and S989P + V1016G confer pyrethroid protection somewhat differently. In contrast to the difference in pyrethroid protection given by the different Vssc mutations, the inheritance of this trait seems very consistent. Our results indicate the inheritance of resistance conferred by S989P + V1016G is incompletely recessive. This has commonly been observed for the inheritance of resistance for other Vssc mutations such as A1215D + F1538I,49 V1023G + D1794Y50 and L1014F,51 although some variation has been observed between pyrethroids for the L1014F mutation in houseflies.29 The inheritance of a resistance allele could have important implications for resistance management. For example, control problems are likely to be manifest more slowly in a population if the resistance allele is recessive.52

It would be expected that kdr would confer resistance in all life stages, due to the nature of this resistance (mutation in the target site). However, to our knowledge, the levels of resistance conferred by kdr have not been compared between life stages. We found no significant difference in the levels of permethrin resistance in larvae and adults. This was somewhat surprising because different bioassay methods were used, and different methods can sometimes manifest different RRs.26

CONCLUSION

In summary, we report for the first time the levels of pyrethroid, DDT and oxadiazine resistance conferred by Vssc mutations S989P + V1016G in vivo. Etofenprox (and DDT) should be avoided in areas where Vssc mutations S989P + V1016G exist at high frequencies. We also found that pyrethroid structure cannot be used to predict the level of resistance conferred by kdr. Investigations of the type we carried out are hampered because small changes in the structure of a pyrethroid can sometimes lead to dramatic loss of toxicity. These results provide essential background data for future investigations of pyrethroid interactions with VSSC and for resistance management of this important disease vector.

Table S1: Annotated genes in the Vssc containing supercontig (1.186; NW_001810688.1) in Ae. aegypti. Candidate oxadiazines bioactivation genes are in red. Information from www.VectorBase.org.

Figure S1: Validation of ASPCR. Samples with known genotypes (kdr/kdr [, kdr/+ or +/+ based on Sanger sequencing of PCR products) were evaluated using primers specific for kdr (A) or the susceptible Vssc allele (B).

Figure S2: Toxicity, based on the LD50 of the susceptible ROCK strain, of 20 different VSSC targeting insecticides.

Figure S3: Correlation between the RRs of 14 pyrethroids between a strain with the S989P + V1016G mutations (IsoKDR ) and one with the L1014F mutation (ALkdr housefly)43.

Click here for additional data file.