Epigenetic control of PDX1 and NGN3 by a computationally designed PRC2 inhibitor enforces pancreatic endocrine differentiation from pluripotent stem cells

Directed differentiation of pluripotent stem cells (PSCs) into pancreatic islets is a cornerstone strategy for diabetes cell therapy. This process relies on growth factor–driven activation of core transcriptional regulators, notably PDX1 and NGN3, to restrict the multi-lineage potential of definitive endoderm to pancreatic progenitors and endocrine cell types. Yet differentiation efficiency and lineage fidelity vary markedly across PSC lines. Here, we demonstrate that a dominant constraint is persistent Polycomb Repressive Complex 2 (PRC2)–mediated epigenetic repression at the PDX1 and NGN3 loci, limiting endocrine specification despite inductive signaling. To directly test whether chromatin states at the PDX1 and NGN3 loci gate developmental competence, we deployed a computationally engineered epigenetic effector (EBdCas9) to transiently and sequentially remove H3K27me3 at those loci during defined developmental windows. Targeted epigenetic resolution robustly enhanced endocrine lineage commitment and accelerated β-cell differentiation across genetically diverse PSC lines. In contrast, direct transcriptional activation with VP64dCas9 increased PDX1 and NGN3 expression but did not improve differentiation outcomes. Integrated cell population studies and genome-wide chromatin and transcriptomic analyses reveal that PRC2-targeted remodeling preferentially activates endocrine gene networks while limiting progenitor expansion and lineage-inappropriate programs. These findings establish that gene-targeted manipulation of PRC2-mediated repression at PDX1 and NGN3 can be used to control cell lineage competence. Collectively, our study reframes variability in PSC differentiation as a failure of epigenetic resolution rather than transcriptional insufficiency and introduces locus-specific chromatin remodeling as a generalizable strategy to enforce developmental fidelity.