Gene-Specific Endothelial Programs Drive AVM Pathogenesis in <i>SMAD4</i> and <i>ALK1</i> Loss-of-Function

BACKGROUND: Hereditary hemorrhagic telangiectasia is a genetic disorder caused by loss-of-function mutations in components of the bone morphogenetic protein signaling pathway, leading to arteriovenous malformations. Most prior work has treated BMP (bone morphogenetic protein)-component depletion as mechanistically interchangeable, yet whether distinct genes converge on a shared mechanism remains unclear. We aimed to understand the molecular relationship between BMP signaling and endothelial flow response that leads to arteriovenous malformation formation. METHODS: We expose human endothelial monolayers treated with small interfering RNA against SMAD4 or ALK1 to laminar flow and analyze flow-responsive transcriptomics, flow-responsive BMP signaling activation dynamics, cell polarity, and morphology. We analyze the cell-autonomous and noncell-autonomous migration dynamics of endothelial cells treated with siSMAD4 or siALK1 . Using the postnatal mouse retina model, we study endothelial cell distribution changes over time in mosaic settings, and assess the remodeling capabilities of SMAD4 iECKO or ALK1 iECKO , relative to littermate controls. RESULTS: This study shows that depletion of SMAD4 or ALK1 leads to fundamentally distinct mechanisms of vascular malformation. SMAD4 deficiency enhances endothelial responses to blood flow, including transcriptional activation and migration against flow, causing excessive capillary pruning and the development of single large shunts. In contrast, ALK1 deficiency disrupts flow sensing, impairs cell polarization and migration, and promotes a persistent angiogenic state, resulting in dense, hypervascularized networks. RNA sequencing across static and flow conditions identifies both flow-dependent and flow-independent transcriptional changes, suggesting early defects in endothelial fate specification. Mosaic in vitro models show that mutant cells co-opt neighboring wild-type cells, while in vivo tracking confirms mutation-specific migration behavior. CONCLUSIONS: These findings reveal divergent cellular programs driving arteriovenous malformations and underscore the need for gene-specific diagnostic and therapeutic strategies.