Malaria control relies heavily on pyrethroid insecticides, to which susceptibility is declining in Anopheles mosquitoes. To combat pyrethroid resistance, application of alternative insecticides is advocated for indoor residual spraying (IRS), and carbamates are increasingly important. Emergence of a very strong carbamate resistance phenotype in Anopheles gambiae from Tiassalé, Côte d'Ivoire, West Africa, is therefore a potentially major operational challenge, particularly because these malaria vectors now exhibit resistance to multiple insecticide classes. We investigated the genetic basis of resistance to the most commonly-applied carbamate, bendiocarb, in An. gambiae from Tiassalé. Geographically-replicated whole genome microarray experiments identified elevated P450 enzyme expression as associated with bendiocarb resistance, most notably genes from the CYP6 subfamily. P450s were further implicated in resistance phenotypes by induction of significantly elevated mortality to bendiocarb by the synergist piperonyl butoxide (PBO), which also enhanced the action of pyrethroids and an organophosphate. CYP6P3 and especially CYP6M2 produced bendiocarb resistance via transgenic expression in Drosophila in addition to pyrethroid resistance for both genes, and DDT resistance for CYP6M2 expression. CYP6M2 can thus cause resistance to three distinct classes of insecticide although the biochemical mechanism for carbamates is unclear because, in contrast to CYP6P3, recombinant CYP6M2 did not metabolise bendiocarb in vitro. Strongly bendiocarb resistant mosquitoes also displayed elevated expression of the acetylcholinesterase ACE-1 gene, arising at least in part from gene duplication, which confers a survival advantage to carriers of additional copies of resistant ACE-1 G119S alleles. Our results are alarming for vector-based malaria control. Extreme carbamate resistance in Tiassalé An. gambiae results from coupling of over-expressed target site allelic variants with heightened CYP6 P450 expression, which also provides resistance across contrasting insecticides. Mosquito populations displaying such a diverse basis of extreme and cross-resistance are likely to be unresponsive to standard insecticide resistance management practices. Malaria control depends heavily on only four classes of insecticide to which Anopheles mosquitoes are increasingly resistant. It is important to manage insecticide application carefully to minimise increases in resistance, for example by using different compounds in combination or rotation. Recently, mosquitoes resistant to all available insecticides have been found in Tiassalé, West Africa, which could be problematic for resistance management, particularly if common genetic mechanisms are responsible (‘cross-resistance’). Tiassalé mosquitoes also exhibit extreme levels of resistance to the two most important classes, pyrethroids and carbamates. We investigated the genetic basis of extreme carbamate resistance and cross-resistance in Tiassalé, and the applicability of results in an additional population from Togo. We find that specific P450 enzymes are involved in both extreme and cross-resistance, including one, CYP6M2, which can cause resistance to three insecticide classes. However, amplification of a mutated version of the gene which codes for acetycholinesterase, the target site of both the carbamate and organophosphate insecticides, also plays an important role. Mechanisms involved in both extreme resistance and cross resistance are likely to be very resilient to insecticide management practices, and represent an alarming scenario for mosquito-targeted malaria control.

The development of resistance to insecticides has become a classic exemplar of evolution occurring within human time scales. In this study we demonstrate how resistance to DDT in the major African malaria vector Anopheles gambiae is a result of both target-site resistance mechanisms that have introgressed between incipient species (the M- and S-molecular forms) and allelic variants in a DDT-detoxifying enzyme. Sequencing of the detoxification enzyme, Gste2, from DDT resistant and susceptible strains of An. gambiae, revealed a non-synonymous polymorphism (I114T), proximal to the DDT binding domain, which segregated with strain phenotype. Recombinant protein expression and DDT metabolism analysis revealed that the proteins from the susceptible strain lost activity at higher DDT concentrations, characteristic of substrate inhibition. The effect of I114T on GSTE2 protein structure was explored through X-ray crystallography. The amino acid exchange in the DDT-resistant strain introduced a hydroxyl group nearby the hydrophobic DDT-binding region. The exchange does not result in structural alterations but is predicted to facilitate local dynamics and enzyme activity. Expression of both wild-type and 114T alleles the allele in Drosophila conferred an increase in DDT tolerance. The 114T mutation was significantly associated with DDT resistance in wild caught M-form populations and acts in concert with target-site mutations in the voltage gated sodium channel (Vgsc-1575Y and Vgsc-1014F) to confer extreme levels of DDT resistance in wild caught An. gambiae.

The role of inter-specific hybridisation is of particular importance in mosquito disease vectors for predicting the evolution of insecticide resistance. Two molecular forms of Anopheles gambiae s.s., currently recognized as S and M taxa, are considered to be incipient sibling species. Hybrid scarcity in the field was suggested that differentiation of M and S taxa is maintained by limited or absent gene flow. However, recent studies have revealed shared polymorphisms within the M and S forms, and a better understanding of the occurrence of gene flow is needed. One such shared polymorphism is the G119S mutation in the ace-1 gene (which is responsible for insecticide resistance); this mutation has been described in both the M and S forms of A. gambiae s.s. To establish whether the G119S mutation has arisen independently in each form or by genetic introgression, we analysed coding and non-coding sequences of ace-1 alleles in M and S mosquitoes from representative field populations. Our data revealed many polymorphic sites shared by S and M forms, but no diversity was associated with the G119S mutation. These results indicate that the G119S mutation was a unique event and that genetic introgression explains the observed distribution of the G119S mutation within the two forms. However, it was impossible to determine from our data whether the mutation occurred first in the S form or in the M form. Unexpectedly, sequence analysis of some resistant individuals revealed a duplication of the ace-1 gene that was observed in both A. gambiae s.s. M and S forms. Again, the distribution of this duplication in the two forms most likely occurred through introgression. These results highlight the need for more research to understand the forces driving the evolution of insecticide resistance in malaria vectors and to regularly monitor resistance in mosquito populations of Africa.

The widespread insecticide resistance raises concerns for vector control implementation and sustainability particularly for the control of the main vector of human malaria, Anopheles gambiae sensu stricto . However, the extent to which insecticide resistance mechanisms interfere with the development of the malignant malaria parasite in its vector and their impact on overall malaria transmission remains unknown. We explore the impact of insecticide resistance on the outcome of Plasmodium falciparum infection in its natural vector using three An. gambiae strains sharing a common genetic background, one susceptible to insecticides and two resistant, one homozygous for the ace-1 R mutation and one for the kdr mutation. Experimental infections of the three strains were conducted in parallel with field isolates of P. falciparum from Burkina Faso (West Africa) by direct membrane feeding assays. Both insecticide resistant mutations influence the outcome of malaria infection by increasing the prevalence of infection. In contrast, the kdr resistant allele is associated with reduced parasite burden in infected individuals at the oocyst stage, when compared to the susceptible strain, while the ace-1 R resistant allele showing no such association. Thus insecticide resistance, which is particularly problematic for malaria control efforts, impacts vector competence towards P. falciparum and probably parasite transmission through increased sporozoite prevalence in kdr resistant mosquitoes. These results are of great concern for the epidemiology of malaria considering the widespread pyrethroid resistance currently observed in Sub-Saharan Africa and the efforts deployed to control the disease.

Pyrethroid insecticides, carbamate and organophosphate are the classes of insecticides commonly used in agriculture for crop protection in Benin. Pyrethroids remain the only class of insecticides recommended by the WHO for impregnation of bed nets. Unfortunately, the high level of pyrethroid resistance in Anopheles gambiae s.l., threatens to undermine the success of pyrethroid treated nets. This study focuses on the investigation of agricultural practices in cotton growing areas, and their direct impact on larval populations of An. gambiae in surrounding breeding sites.