The euchromatic histone-lysine N-methyltransferases EHMT1/2 suppress post-natal cardiomyocyte proliferation via histone 3 lysine 9 (H3K9) dimethylation

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – EU funding. Main funding source(s): KULeuven C14/17/088 Flemish Research Foundation (FWO) G0C6419N Cardiomyocytes proliferate rapidly during developmental growth in utero, but this proliferative capacity dramatically declines postnatally, rendering the heart unable to repair itself subsequent to injury or during ageing. The mechanisms governing this postnatal loss of cell cycle activity are not fully understood. During maturation, epigenomic remodelling drives wholesale changes in the cardiomyocyte transcriptome from the neonatal to the adult state. This is associated with silencing of foetally-expressed genes and activation of expression of genes required for adulthood. Here, we tested the hypothesis that epigenomic remodelling silences cell cycle-associated genes and hence cell cycle activity postnatally. In this regard, we focused on the role of the repressive H3K9me2 mark. Using immunostaining of Ki-67 and Aurora Kinase B as well as analysis of EdU incorporation to identify cell cycle activity in primary cardiomyocytes culture and in cryosections from rat hearts spanning from embryonic day 18 to adulthood, the abundance of H3K9me2 was found to be inversely correlated with cell proliferation. In line with these data, the expression of a gene panel associated with cell cycle activity decreased postnatally and was inversely related with H3K9me2 abundance in cardiomyocyte nuclei, purified by flow cytometry based on H3K9me2 abundance. Inhibition of the euchromatic histone lysine methyltransferases (Ehmt1/2 aka GLP/G9a), with the highly selective inhibitor A-366, enhanced proliferation of neonatal rat cardiomyocytes in vitro, as well as in vivo, in neonatal mice, via concomitant reduction of the repressive histone mark, H3K9me2. Similar effects were observed in vitro by overexpression of the histone di/tri-methyl demethylase KDM4B. Notably, forced induction of cell cycle activity by overexpression of cyclin/cyclin-dependent kinase significantly reduced H3K9me2 levels, in vitro. Together, these data point to a key role of H3K9me2 and its deposition by Ehmts in post-natal suppression of proliferation. Targeting this repressive mark could provide a strategy to enhance cardiac repair for therapeutic benefit.