The genetic architecture of an adaptive phenotype conditions populations evolutionary response to climate change

Populations can be locally adapted to environments that can vary across space and across time. Understanding the genetic architecture of phenotypic variation that enable natural populations to track temporal changes of the environment is crucial in the context of the ongoing climate change. One successful example of ongoing climate adaptation is that of chemotype diversity in common thyme (<i>Thymus vulgaris</i>). Thyme is locally abundant in Mediterranean landscapes and harbors a genetically determined chemotype polymorphism varying at a very local scale. Chemotypes can be grouped by their chemical composition into phenolic and non-phenolic ecotypes, which have been shown to be adapted to different local habitats. In recent years, ecotypes are shifting local frequency as they adaptively respond to ongoing climate change.<br/>To understand the genomic architecture of this adaptive phenotype, we sequenced 1000 thyme individuals from over 70 sites spanning 9 independent gradients of altitude and environmental conditions across Southern France. We find weak population structure across populations, indicating abundant gene flow. We identify 3 genetic loci associated with the ecotype, that together explain more than 90% of phenotypic variance at the ecotype level. All loci fall within genes that encode enzymes involved in monoterpene biosynthesis, and involve structural genomic variation such as copy number variants. We characterize the genetic architecture of the ecotypes and find that it involves epistatic interactions between these three loci.<br/>We use on site temperature and soil measurements to characterize the local environment of some of the sampled transects. We then build genotype-environment association models to detect genetic variants involved in local adaptation. We find the majority of environment associated variants are also involved in determining the chemical ecotypes, confirming chemotype variation as the major phenotype mediating local climate adaptation in common thyme populations.<br/>Finally, we use population genetic forward simulations to explore if and why the inferred genetic architecture of thyme ecotypes provides an advantage over alternative genetic architectures for tracking environmental variation and climate change.