In conclusion, this study validates an innovative approach for resistance management of mosquito vectors, based on the development of molecules targeting preferentially enzymes already insensitive to currently used insecticides. PTFs identified here behave as preferential inhibitors of AChE1 mutants insensitive to OP and CX and as preferential killers of OPresistant C. pipiens and A. gambiae larvae in bioassays. This approach should allow both efficient vector and resistance control management: The preferential killing of resistant mosquitoes mimics a situation of negative cross-resistance, in which frequency of resistance alleles decreases much faster than when insecticides acting on other targets are used. Furthermore, since wild type OP-susceptible alleles are, by construction, PTFresistant alleles, odds are against the selection of other resistance events. Wild type alleles are indeed more frequent than any spontaneous mutant and are associated with the highest fitness. Although not easily amenable to high throughput screening, voltage-gated sodium channels represent other interesting candidates for developing a similar strategy; they are targets of pyrethroids, the major insecticide class currently used in malaria control; resistance to pyrethroids is spreading in many species through the selection of a very small number of insensitive alleles (particularly kdr-west, kdr-east), affecting the same amino acid 1014 (house fly numbering) [36,37]. This strategy could also be applied to any pest that acquired resistance through one or a few mutations in a structurally constrained target for which resistance is associated with a fitness cost. Developing new approaches to maintain vector control and maximize the effective lifespan of current and future insecticides is one of the objectives of the Global Plan for Insecticide Resistance Management (WHO 2012). This aim is paramount in a context where more than 500 arthropod species (either medical or agricultural pests) have become resistant to most if not all currently used insecticides [38]. The present study demonstrates that a “hit where it already hurts” strategy could fit the bill.
Materials and Methods
ii) Modeling the impact of PTFs on an OP-resistant population indicates that even moderate selectivity (r = 1.5 to 3) is sufficient to significantly decrease the frequency of OPresistant alleles (i.e. effective resistance control) in just a few generations. PTF-treated populations might therefore rapidly regain OP-susceptibility and be subsequently controlled by OPs at much lower doses.
Chemicals
All chemical compounds were purchased from ChemBridge (San Diego, CA, USA), except eserine, from Sigma-Aldrich, (Saint-Louis, USA), propoxur, from Bayer (Leverkusen, Germany), chlorpyrifos, from CIL Luzeau (France) and tacrine, from ICN Biomedicals, Inc (Eschwege, Germany).
Figure 2. Relative AChE1 activity in OP-resistant SR larvae killed by PTF treatment. C. pipiens larvae from the OP-resistant SR strain were exposed to 300 mM PTF until death and then AChE1 residual activity was measured. Chlorpyrifos (Chlpy) and propoxur (Prpx) were used as positive and negative control inhibitors, respectively. Means and standard errors for three independent experiments are shown. Four to five replicates were performed for each bioassay. LD50 is expressed in mM. 95% confidence intervals are indicated into parentheses. Ratio of LD50 for OP-susceptible to LD50 for OP-resistant strains. Culex pipiens strains: Slab, OP-susceptible ace-1S/S; SR, OP-resistant ace-1R/R; Ducos, OP-resistant ace-1D (duplication). d Anopheles gambiae strains: Kisumu, OP-susceptible ace-1S/S; Acerkis, OP-resistant ace-1R/R. e NA: not analyzed. Figure 3. Modeling the impact of PTF treatment on frequency of the ace-1R allele and on larvae survival. Evolution of an OP-resistant, infinite and panmictic population treated with PTFs was computed as described in Material and Methods. The ace-1R initial frequency was 0.9 and PTF compounds were applied at LD50 for ace-1R homozygotes (i.e. m = 0.5). r represents the mortality ratio of ace-1R vs. ace-1S homozygotes. Panel A represents the evolution of ace-1R frequency when this allele is recessive (d = 0) and panel B, when it is dominant (d = 1). Curves represent the evolution of ace-1R frequency across generations for various PTF mortality r ratios between 1.2 and 100. Insets represent the proportion of individuals killed at each generation.
Strains
Three C. pipiens strains were used: the susceptible reference Slab strain, homozygous for ace-1S [39], the resistant reference SR strain, carrying the genetical background of Slab but homozygous for the G119S mutation, allele ace-1R [21] and the Ducos strain, carrying the genetical background of Slab but homozygous for an ace-1 duplication, with one ace-1S and ace-1R copy in tandem [27]. Culex pipiens heterozygous larvae (Slab6SR) were obtained by crossing Slab males and SR females. Two A. gambiae reference strains (S molecular form) were used: the susceptible strain Kisumu, homozygous for ace-1S, collected in Kenya in 1953 and maintained since then under laboratory conditions [40] and the resistant Acerkis strain, carrying the genetical background of Kisumu but homozygous for an ace-1R allele from a population collected in Bobo-Dioulasso, Burkina Faso [41].AChE Assays and Screening Procedures
Production of C. pipiens WT and mutated AChE1s in Drosophila S2 cells was already described [17]. The chemical library (ChemBridge, 3,000 compounds) screening was performed in duplicate on G119S recombinant AChE1. Chemicals were made soluble in ethanol or dimethylsulphoxide (DMSO), and then diluted at 300 mM in ethanol for storage. Compounds (30 mM final concentration) were incubated for 15 min at room temperature with 100 ml of G119S recombinant AChE1 then 100 ml of substrate (acetylthiocholine, 1.6 mM, Sigma-Aldrich) was added and the residual activity was quantified by measuring the optical density at 412 nm, as described by Ellman et al. [26]. PTF analogs were analyzed in dose-response experiments (10-fold serial dilutions from 3 mM to 30 nM), using recombinant WT and G119S AChE1 or mosquito head extracts (heads cut from frozen mosquitoes, homogenized in 400 ml PB containing 1% Triton X100 and cleared by centrifugation at 9,0006g for 3 min). Depending on compound availability, two to five replicates were performed with distinct batches of enzyme. Concentrations producing 50% enzyme inhibition (IC50) were determined using regression analysis of log-concentrations versus percentage inhibitions. IC50 was estimated by nonlinear least square regression. IC50 was also measured on AChE1 from susceptible and resistant mosquito larvae extracts. To address the residual AChE1 activity after exposure to 300 mM PTF, larvae were collected as soon as mortality was reached to avoid AChE1 degradation. Larvae were rinsed twice with distilled water and homogenized individually in 400 ml PB containing 1% Triton X-100. Homogenates were centrifuged at 9,0006g for 3 min and assayed for AChE1 activity as described above.
