Widespread epistasis regulates glucose homeostasis and gene expression

Abstract

The relative contributions of additive versus non-additive interactions in the regulation of complex traits remains controversial. This may be in part because large-scale epistasis has traditionally been difficult to detect in complex, multi-cellular organisms. We hypothesized that it would be easier to detect interactions using mouse chromosome substitution strains that simultaneously incorporate allelic variation in many genes on a controlled genetic background. Analyzing metabolic traits and gene expression levels in the offspring of a series of crosses between mouse chromosome substitution strains demonstrated that inter-chromosomal epistasis was a dominant feature of these complex traits. Epistasis typically accounted for a larger proportion of the heritable effects than those due solely to additive effects. These epistatic interactions typically resulted in trait values returning to the levels of the parental CSS host strain. Due to the large epistatic effects, analyses that did not account for interactions consistently underestimated the true effect sizes due to allelic variation or failed to detect the loci controlling trait variation. These studies demonstrate that epistatic interactions are a common feature of complex traits and thus identifying these interactions is key to understanding their genetic regulation.

A novel mapping strategy utilizing mouse chromosome substitution strains identifies multiple epistatic interactions that regulate complex traits

“…QTLs accounting for such a high proportion of the variance was not unexpected, as our previous data suggested the widespread importance of epistasis in regulating gene expression (CHEN et al 2017). Of 519 epistatic QTLs (the 510 unique genes and genes regulated by multiple eQTLs), 243 (46.8%) exhibited positive epistasis whereas 276 (53.2%) exhibited negative epistasis.…”

“…The resulting analyses identified widespread epistatic interactions, including specific interacting loci that contributed to the heritable variation in lymphocyte percentage (ly%), red cell distribution width (rdw), and hepatic gene expression of over 500 genes. Intrachromosomal interactions were detected in rdw and hepatic gene expression data that could not be identified in studies of the B6.A4 and B6.A6 CSS strains carrying the entire nonrecombinant A/Jderived substituted chromosomes (CHEN et al 2017). That this number of traits and genes were identified with detectable regulation by epistasis is remarkable considering that only a single time point was examined, gene expression was only measured in the liver, and only one pairwise strain combination of CSSs was examined.…”

“…However, to identify the molecular nature of such QTLs requires identification of the underlying genes, which necessitates higher resolution mapping studies. Therefore, to identify novel epistatic QTLs that regulate complex traits and gene expression levels with greater mapping resolution, we generated and analyzed the N2 generation mice from a modified backcross based on using some of the same CSSs in Chen et al (2017). The resulting analyses demonstrate the ability of a CSS-based backcross to detect and finemap epistatic QTLs for complex traits.…”

The fitness cost of a congenital heart defect shapes its genetic architecture

“…Under an additive model, mice that carry both A/J chromosomes (e.g., chromosomes 5 and 6) should have higher glucose levels, but they actually had normal levels. 28 A/J and C57BL/6J appear to have arrived at different genetic solutions to maintain normoglycemia.…”

Cis-epistasis at the LPA locus and risk of coronary artery disease

“…Less attention has been paid to epistasis in which variants act non-additively or, paradoxically, are dependent on the genetic context 3,4 . Albeit epistasis has broadly been shown to affect multiple traits in fruit flies 5 , mice 6 , and humans 7 , examples demonstrating biological relevance for epistasis via statistical approaches have only been investigated in model organisms [8][9][10] . Earlier attempts were largely unsuccessful given the computational challenges such as the curse of dimensionality, model complexity and bias from linkage disequilibrium (LD).…”