Integrated multi-omics analysis identifies features that predict human pluripotent stem cell-derived progenitor differentiation to cardiomyocytes
Aaron D Simmons,Claudia Baumann,Xiangyu Zhang,Timothy J Kamp,Rabindranath De La Fuente,Sean P Palecek
DOI: https://doi.org/10.1016/j.yjmcc.2024.08.007
2024-08-31
Abstract:Human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) are advancing cardiovascular development and disease modeling, drug testing, and regenerative therapies. However, hPSC-CM production is hindered by significant variability in the differentiation process. Establishment of early quality markers to monitor lineage progression and predict terminal differentiation outcomes would address this robustness and reproducibility roadblock in hPSC-CM production. An integrated transcriptomic and epigenomic analysis assesses how attributes of the cardiac progenitor cell (CPC) affect CM differentiation outcome. Resulting analysis identifies predictive markers of CPCs that give rise to high purity CM batches, including TTN, TRIM55, DGKI, MEF2C, MAB21L2, MYL7, LDB3, SLC7A11, MAB21L2, and CALD1. Predictive models developed from these genes provide high accuracy in determining terminal CM purities at the CPC stage. Further, insights into mechanisms of batch failure and dominant non-CM cell types generated in failed batches are elucidated. Namely EMT, MAPK, and WNT signaling emerge as significant drivers of batch divergence, giving rise to off-target populations of fibroblasts/mural cells, skeletal myocytes, epicardial cells, and a non-CPC SLC7A11+ subpopulation. This study demonstrates how integrated multi-omic analysis of progenitor cells can identify quality attributes of that progenitor and predict differentiation outcomes, thereby improving differentiation protocols and increasing process robustness.