Cis non-coding genetic variation drives gene expression changes in the E. coli and P. aeruginosa pangenomes

Bacteria use gene regulation to dynamically adapt to changes in their environment, including resistance to stress and the occupation of new niches. Gene expression is known to vary within a species pangenome, but the extent to which these changes could be explained by genetic variants in cis non-coding regions has so far been poorly investigated. Statistical genetics offers a hypothesis-free approach to this problem, as opposed to mechanistic models, which can be used only for reference isolates that are not representative of the whole species. In this study, we assembled two genomic and transcriptomic datasets for Escherichia coli (N=117) and Pseudomonas aeruginosa (N=413) and identified associations between genetic variants in cis non-coding regions and recorded gene expression variation. We identified at least one associated variant in up to 39% of the tested genes in both species. We partly validated the associations in-silico and in-vitro for E. coli, reinforcing the difficulty of identifying a single mechanism generating gene expression diversity. We then investigated the relevance of non-coding variants in explaining the variability in antimicrobial resistance in both species using two additional publicly available datasets, identifying a large number of these variants across antimicrobial compounds. This work confirms the role of genetic variation in often overlooked regions of bacterial genomes in influencing molecular and clinically relevant phenotypes.