Abstract
Prime editing (PE) is an innovative next-generation gene editing tool that has therapeutic potential in post-mitotic organs, such as the human heart. However, its applicability and efficiency in non-proliferating cells, e.g., human cardiomyocytes, is not yet established. Here, we apply PE directly in cardiomyocytes differentiated from human induced pluripotent stem cells (hi-CMs) carrying dilated-cardiomyopathy-causing mutations. A target array (TA) containing the mutations LMNAK117fs (348-349insG), RBM20P633L (c.1898 C>T), and RBM20R634Q (c.1901 G>A) in the safe-harbor locus AAVS1 in HEK293T cells served as a screening platform for prime editing gRNAs (pegRNAs). The pegRNA screen yielded a set of efficient pegRNAs targeting the respective mutations. Using the PE4 system to correct the RBM20P633L-mutation, we achieved 34.8% T-to-C editing efficiency on average in homozygous P633L/P633L-hi-CMs while maintaining low off-target editing. PE restored RBM20's nuclear localization and normalized cardiac splicing of the calcium-/calmodulin-dependent protein kinase II delta (CAMK2D) transcript. We combine a detailed pegRNA screening assay in an easy-to-transfect HEK293T system (TA-HEK) with subsequent functional validation of PE in hi-CMs carrying patient-derived mutations. This strategy yielded the first PE-mediated phenotypic rescue in a human post-mitotic model of DCM and paves the way for an in vivo strategy to treat RBM20P633L-mediated DCM and other inherited cardiac diseases.