Comparative transcriptome analyses of deltamethrin-resistant and -susceptible Anopheles gambiae mosquitoes from Kenya by RNA-Seq

Abstract

Malaria causes more than 300 million clinical cases and 665,000 deaths each year, and the majority of the mortality and morbidity occurs in sub-Saharan Africa. Due to the lack of effective vaccines and wide-spread resistance to antimalarial drugs, mosquito control is the primary method of malaria prevention and control. Currently, malaria vector control relies on the use of insecticides, primarily pyrethroids. The extensive use of insecticides has imposed strong selection pressures for resistance in the mosquito populations. Consequently, resistance to pyrethroids in Anopheles gambiae, the main malaria vector in sub-Saharan Africa, has become a major obstacle for malaria control. A key element of resistance management is the identification of resistance mechanisms and subsequent development of reliable resistance monitoring tools. Field-derived An. gambiae from Western Kenya were phenotyped as deltamethrin-resistant or -susceptible by the standard WHO tube test, and their expression profile compared by RNA-seq. Based on the current annotation of the An. gambiae genome, a total of 1,093 transcripts were detected as significantly differentially accumulated between deltamethrin-resistant and -susceptible mosquitoes. These transcripts are distributed over the entire genome, with a large number mapping in QTLs previously linked to pyrethorid resistance, and correspond to heat-shock proteins, metabolic and transport functions, signal transduction activities, cytoskeleton and others. The detected differences in transcript accumulation levels between resistant and susceptible mosquitoes reflect transcripts directly or indirectly correlated with pyrethroid resistance. RNA-seq data also were used to perform a de-novo Cufflinks assembly of the An. gambiae genome.

Figure 1. Map of Kenya showing the sampling sites.

Anopheles gambiae sensu lato larvae were sampled in different breeding sites from Ahero, Chemelil, Emutete, Chulaimbo, Bungoma and Busia in Western Kenya.

Figure 2. Distribution of differentially accumulated transcripts on chromosome 2R.

Fold-changes in transcript accumulation levels between deltamethrin-resistant and susceptible mosquitoes for the significantly differentially accumulated transcripts mapping on chromosome 2R are presented. Various chromosomal inversions (2Rj, 2Rb, 2Rc and 2Ru) are indicated. The position of the QTL rtp1 is also indicated. For representation purposes, transcripts showing read coverage exclusively in susceptible or resistant mosquitoes were attributed an arbitrary fold-change of −10 or 10, respectively.

Figure 3. Significantly differentially accumulated transcripts associated to metabolic detoxification activities and coding for cuticular proteins.

Absolute quantification level (in FPKM) and fold-changes in transcript accumulation levels (FCA) between deltamethrin-resistant and -susceptible mosquitoes are presented. A: transcripts related to metabolic detoxification activities; and B: transcripts coding for cuticular proteins; C: List of VectorBaseID.

Figure 4. Transcripts with significantly higher accumulation in deltamethrin-resistant mosquitoes than deltamethrin-susceptible mosquitoes or with reads coverage only in deltamethrin-resistant mosquitoes.

A: Transcripts expressed moderately in R mosquitoes (FPMK ≥10) and accumulated ≥5 folds in deltamethrin-resistant mosquitoes in reference to deltamethrin-susceptible mosquitoes; B: Transcripts with read coverage exclusively in the deltamethrin-resistant mosquitoes; and C: list of VectorBase idand best match to PFAM database of the transcrips.

Figure 5. Comparison of RNA-seq and qRT-PCR in transcription determination of eleven selected genes. A: RNA-seq and

qRT-PCR-based fold-changes in transcript accumulation levels. qRT-PCR fold-changes were derived comparing susceptible and resistant mosquitoes from Emutete (in red) and susceptible and resistant mosquitoes from Bungoma (in green); B: a list of VectorBase id and putative functions of the 11 transcripts examined.

Figure 6. NTE validation.

RT-PCR amplification on cDNA from mosquitoes from Emutete (lane 1) and Bungoma (lanes 2–4) of two NTEs identified when Cufflinks is run using the An. gambiae genome annotation as a simple guide of transcript annotation. Lane 5 is the negative control. Results for Cufflinks gene-id 27065 suggest the existence of isoforms.