Aberrant Phase Separation of Endothelial MAML1 Causes Congenital Heart Disease by Suppressing Notch Activity

BACKGROUND: Congenital heart disease (CHD), the most common birth defect and a leading cause of infant mortality, is frequently linked to dysregulated Notch signaling. However, the role of the Notch transcriptional coactivator Mastermind-like 1 (MAML1) in CHD pathogenesis and the underlying molecular mechanism remain unclear. METHODS: We investigated the role of MAML1 in CHD by focusing on a patient-derived Q401K mutation with a knock-in mouse model and an endocardium-specific Maml1 knockout mouse model, complemented by CRISPR-edited human heart organoids. Cardiac phenotypes were assessed by echocardiography and histological analysis. The underlying molecular mechanisms were dissected through biochemical assays, microscopy to analyze liquid-liquid phase separation (LLPS), and mass spectrometry to identify posttranslational modifications and the upstream kinase of MAML1. RESULTS: In a clinical cohort of patients with CHD, we identified rare missense variants of MAML1 associated with ventricular septal defects. Modeling a patient-derived variant (Q401K) was sufficient to recapitulate key ventricular septal defect–related phenotypes in both knock-in mice and human heart organoids. To confirm the tissue-specific pathogenicity, we showed that endocardium-specific knockout of MAML1 in mice and MAML1 deletion in human heart organoids caused similar septal and valvular defects by disrupting Notch-driven endocardial-to-mesenchymal transition. Mechanistically, we discovered that MAML1 activity depends on LLPS, which forms nuclear condensates, required for efficient interaction with the NOTCH1 intracellular domain and activation of downstream transcriptional targets. Crucially, patient-derived pathogenic variants, including Q401K, function as charge-altering mutations within the intrinsically disordered region 2, a core region for MAML1 LLPS, pathologically abrogating LLPS to downregulate Notch signaling. Furthermore, we identified a regulatory axis in which PKN2 phosphorylates MAML1 at Ser314, which destabilizes MAML1 condensates and consequently attenuates Notch transcriptional output. CONCLUSIONS: These findings support MAML1 as a candidate gene for CHD and identify MAML1 LLPS as a critical biophysical determinant of Notch transcriptional output in endocardial cells. The electrostatic integrity of MAML1 condensates is essential for proper regulation of Notch signaling during cardiac morphogenesis. Dysregulation of this state, whether through CHD-associated charge-altering variants or aberrant PKN2-mediated phosphorylation, impairs Notch signaling and disrupts endocardial-to-mesenchymal transition, thereby establishing a converged molecular mechanism underlying congenital cardiac malformations.