Gene-by-environment interactions are pervasive among natural genetic variants

Abstract

Gene-by-Environment (GxE) interactions are fundamental to understanding fitness landscapes and evolution, but have been difficult to identify at the single-nucleotide level, precluding understanding of their prevalence and molecular mechanisms. Most examples involving natural genetic variants exist at the level of entire genomes, e.g. measurement of microbial strain growth across environments, or loci encompassing many variants identified by quantitative trait loci mapping. Here, we introduce CRISPEY-BAR, a high-throughput precision-editing strategy, and use it to map base-pair resolution GxE interactions impacting yeast growth under stress conditions. First, we used CRISPEY-BAR to uncover 338 variants with fitness effects within QTLs previously mapped in different environments. We then measured 1432 ergosterol pathway variants from diverse lineages across six environments, identifying 205 natural variants affecting fitness measured in all six conditions, of which 93.7% showed GxE interactions. Finally, we examine pleiotropic cis-regulatory variants suggesting molecular mechanisms of GxE interaction. In sum, our results suggest an extremely complex, context-dependent fitness landscape characterized by pervasive GxE interactions, while also demonstrating high-throughput genome editing as an effective means for investigating this complexity.