339T Poster - Quantitative Genetics
Thursday June 09, 9:15 PM - 10:00 PM

Diverse environmental perturbations reveal the evolution and context-dependency of genetic effects on gene expression levels


Authors:
julien Ayroles 1; Amanda Lea 1,2

Affiliations:
1) Princeton University; 2) Vanderbilt University

Keywords:
Complex traits

There is increasing appreciation that complex traits are determined by poorly understood interactions between our genomes and daily environments. These “genotype x environment” (GxE) interactions remain difficult to map at the organismal level but can be uncovered using molecular phenotypes. We address this broad question by asking how prevalent are context-specific versus ubiquitous eQTL? What are the evolutionary forces (e.g., genetic drift, purifying selection, positive selection) that maintain context-specific versus ubiquitous eQTL, and do these forces differ depending on the evolutionary history of the environmental exposure? To address these questions, we embarked on a large-scale study in which we profiled genome-wide gene expression levels across 12 different cellular exposures using 544 lymphoblastoid cell lines (B cell) derived from the 1000 Genomes Project. We used these data to map the genetic basis of gene expression across all 12 conditions and revealed a context-dependent genetic architecture. Our results highlight the extent to which gene expression is fine-tuned by the environment. We found that 23% of eQTLs were context-dependent, that is almost 1 in 4 eQTLs are “response eQTL” or SNPs that do not affect variation in gene expression under the control condition but for which genetic effects are revealed by experimental treatments. We also found that the mean per-gene heritability estimates were significantly higher in almost all treatment conditions relative to their respective controls, emphasizing the important role non-linear genetic effects are likely to play in explaining the missing heritability. These experimental treatments included stimuli familiar to B cells such as immune signaling molecules and hormones, but also man-made chemicals that have not co-evolved with B cells through evolutionary time. Evolutionary analyses revealed that positive selection has shaped GxE loci involved in responding to immune challenges and hormones, but not man-made chemicals, suggesting there is reduced opportunity for selection to act on responses to molecules recently introduced into human environments. Together, this work highlights the importance of considering an exposure’s evolutionary history when studying and interpreting GxE interactions and provides new insight into the evolutionary mechanisms that maintain GxE loci in natural populations.