SRA

Background: The histone demethylase KDM1A is a multi-faceted regulator of critical developmental processes, including mesodermal and cardiac tube formation during mice gastrulation. The fine-tuning of KDM1A splicing has been linked to regulating the transcriptional program of excitable cells such as neurons. However, it is unknown whether modulating the expression of KDM1A isoforms is crucial for the specification and maintenance of cell identity of other cell types sensitive to electrical cues such as the cardiomyocytes. Objective: We investigated the role of ubKDM1A and KDM1A+2a ubiquitous splice variants during cardiomyogenesis and evaluated their impact on the regulation of cardiac differentiation in vitro. Methods and Results: We discovered a temporal modulation of ubKDM1A and KDM1A+2a isoform levels during human and mouse fetal cardiac development. Therefore, we generated human embryonic stem cells (hESCs) exclusively devoid of one or both isoforms and assessed their potential to derive cardiomyocytes. KDM1A depletion severely impaired cardiac differentiation. Conversely, KDM1A+2a-/- hESCs give rise to functional cardiomyocytes, displaying increased beating amplitude and frequency compared to wild-type cells. Transcriptomic profiling revealed that KDM1A-/- cardiomyocytes fail to activate an effective cardiac transcriptional program, while the depletion of KDM1A+2a enhances the expression of key cardiogenic markers. Notably, the impaired cardiac differentiation of KDM1A-/- cells can be rescued by re-expressing ubKDM1A or catalytically deficient ubKDM1A-K661A, but not by KDM1A+2a or KDM1A+2a-K661A. These data demonstrate a divergent role of the two KDM1A isoforms that is independent of their enzymatic activity. Through an exhaustive biochemical and genome-wide binding characterization, we excluded that the opposite ubKDM1A- and KDM1A+2a-mediated regulation of cardiac differentiation resides into differential substrate specificity, H3K4 demethylation efficiency, core-partners binding affinity, or alternative genome binding profiles. Conclusions: Our findings suggest the existence of a divergent scaffolding role of KDM1A splice variants during hESC differentiation into cardiomyocytes. Overall design: Bulk RNA-SEQ of KDM1A Knockout or KDM1A+2a knockout hESC and hESC-derived cardiomyocytes. Bulk RNA-SEQ of KDM1A Knockout hESC rescued with either KDM1A+2a or ubKDM1A splice varinats. For each genotype and differentiation stage two clones have been generated. Six to four independent RNA-Seq libraries have been processed. ChIP-Seq experiments performed at different stages (Day 0=hESC stage; Day 2=mesodermal cells; Day 4 = cardiac precursors; Day 18-20 = cardiomyocytes) of WT hESC differentiation into cardiomyocytes. ChIP-Seq experiments of hESC-derived cardiomyocytes of KDM1A KO cells rescued with either KDM1A+2a or ubKDM1A isoforms. The differentiation into cardiomyocytes has been performed using the The PSC Cardiomyocyte Differentiation Kit (Thermo Fisher Scientific) according to manufacturer' instructions. For each differentiation stage two independent ChiP-Seq experiments have been performed, with the exception of Day 4 (n=1) and the rescue for KDM1A+2a (n=1).