Copy number variation (CNV) and insecticide resistance in mosquitoes: evolving knowledge or an evolving problem?

Abstract

Copy number variation (CNV) in insect genomes is a rich source of potentially adaptive polymorphism which may help overcome the constraints of purifying selection on conserved genes and/or permit elevated transcription. Classic studies of amplified esterases and acetylcholinesterase duplication in Culex pipiens quantified evolutionary dynamics of CNV driven by insecticidal selection. A more complex and potentially medically impactful form of CNV is found in Anopheles gambiae, with both heterogeneous duplications and homogeneous amplifications strongly linked with insecticide resistance. Metabolic gene amplification, revealed by shotgun sequencing, appears common in Aedes aegypti, but poorly understood in other mosquito species. Many methodologies have been used to detect CNV in mosquitoes, but relatively few can detect both duplications and amplifications, and contrasting methods should be combined. Genome scans for CNV have been rare to date in mosquitoes, but offer immense potential to determine the overall role of CNV as a component of resistance mechanisms.

Figure 1

Detecting a tandem gene duplication (or amplification) using discordant read alignment (a,b), or an amplification from read coverage irrespective of relative genomic positions. a) Example of concordant read alignment. Paired end sequencing of a DNA fragment produces a pair of reads (blue arrows) which map facing towards each-other when aligned to the reference genome (red arrows). b) Example of discordant read alignment. Paired end sequencing of a DNA fragment that overlaps the duplication break point produces a pair of reads which map facing away from each other at either end of the duplication when aligned to the reference genome. c) Simulated data showing patterns used to detect CNVs using sequencing coverage. Each point represents a genomic window (usually in the order of a hundred or more base pairs) over which coverage was calculated and normalised such that 1 represents the unduplicated state. The red line indicates the true underlying copy number state. The CNV is visible as a marked transient increase in normalised coverage (here from 1 to 2).