Pericyte Snai1 ablation prevents pathological angiogenesis in retinopathy via extracellular matrix remodelling-mediated inhibition of endothelial cell activation

Pathological retinal neovascularization (RNV) is the leading cause of blindness. Pericytes are crucial for vascular stability and involved in RNV, but the underlying mechanism remains unclear. This study investigates the role and mechanism of Snai1, a transcription factor involved in endothelial-mesenchymal transition, in regulating pericytes during RNV. By analysing single-cell RNA-seq data from oxygen-induced retinopathy (OIR) mice, we identified a distinct subpopulation of 'dysfunctional' pericytes with enriched pro-angiogenic pathways and upregulated Snai1. Genetic deletion of Snai1 in pericytes markedly reduced pathological neovascular tufts and vascular leakage, while improving retinal function in OIR mice, without affecting physiological angiogenesis. Mechanistically, Snai1 in pericytes directly bound to promoters and upregulated key extracellular matrix (ECM) genes, including collagens and the crosslinking enzyme Lox, which remodelled the perivascular ECM into a pro-angiogenic niche. Indeed, Snai1-overexpressing pericytes enhanced endothelial cell proliferation and tube formation only upon direct cell-cell contacts in vitro, an effect abolished by knocking down Col4a1 or inhibiting Lox activity with β-aminopropionitrile (BAPN). A feedforward loop was identified where hypoxic ECs secreted ANGPT2 and TGFβ1, stabilizing Snai1 protein in pericytes. Importantly, intravitreal injection of BAPN in OIR mice phenocopied the Snai1 ablation in effectively inhibiting RNV and promoting vascular normalization. In summary, Snai1 in pericytes is a master regulator of pathological angiogenesis in RNV, programming pericytes to construct a pro-angiogenic ECM through collagen synthesis and crosslinking, facilitating aberrant vascular growth via contact-dependent pericyte/ECM-endothelial-cell interactions. Targeting the Snai1/ECM axis represents a potential 'vascular normalization' strategy for RNV-related diseases.