PgmNr M293: Genetic control of the epigenetic landscape.

Authors:
Christopher L. Baker 1 ; Guruprasad Ananda 1 ; Michael Walker 1 ; Catrina Spruce 1 ; Karl W. Broman 2 ; Petko Petkov 1 ; Greg Carter 1 ; Kenneth Paigen 1


Institutes
1) The Jackson Laboratory, Bar Harbor, ME, USA; 2) The University of Wisconsin-Madison, Madison, WI, USA.


Abstract:

Cellular differentiation and numerous biological functions are controlled by a wide variety of regulatory elements associated with characteristic epigenetic modifications. Although a variety of proteins are known to act as writers, readers, and erasers of these marks, we know little about the organizing principles behind the systems regulating the establishment and maintenance of this epigenetic landscape. Understanding how these marks are controlled, and ultimately the ability of cells to differentiate and regulate their functions, is a fundamental and open question in biology. As a model to explore these issues, we have leveraged the strength of mouse genetics to understand how natural genetic variation impacts these processes. Studying levels of H3K4me3, an epigenetic mark associated with promoters, enhancers and recombination hotspots, we compared the genomes of C57BL6/J and DBA2/J mice finding considerable differences in activity of individual H3K4me3 sites in spermatocytes. Haplotype-specific analysis in F1 hybrids showed evidence of trans-control at numerous sites where the phenotype at the site did not coincide with the genotype of the parent. To identify the underlying genetic control, we measured H3K4me3 levels at 72,859 sites in a mapping population of 32 BXD recombinant inbred lines, identifying 6,868 QTL at FDR < 0.1. Most QTL map proximally, indicating control of H3K4me3 levels in cis. There were also1,262 trans-regulated sites, with five major QTL controlling greater than 55 sites each, collectively regulating 62% of all trans-regulated sites. The largest trans QTL on Chr 13 controls 452 sites. Importantly, trans-regulated H3K4me3 sites are clustered in domains of 1-10 Mb (R-scan statistic p < 0.001), indicating that trans control involves regional higher order chromatin organization rather than acting on individual H3K4me3 sites. In agreement with this, we find that the same QTL can affect both recombination hotspots where H3K4me3 is deposited by the enzyme PRDM9, and at promoters and enhancers, which are methylated by distinct enzyme systems. Finally, comparing spermatocytes with hepatocytes, we find evidence that these systems are cell and tissue specific, suggesting different systems may play key roles in establishing or maintaining differentiation. Together, these QTL constitute an integrated system controlling the epigenetic landscape in which each QTL controls multiple domains scattered across the genome. Their existence obviously raises questions of the identity of the genes underlying these QTL, their mechanism of action and their roles in regulating cellular differentiation and function.