A single-cell atlas of coronary artery and bypass grafts: from cellular heterogeneity to therapeutic targets for enhancing graft longevity

Abstract The long-term success of coronary artery bypass grafting (CABG) critically depends on the choice of vascular graft, yet the molecular determinants of their differential clinical performance remain poorly understood. A systematic, high-resolution comparison of these grafts is essential for optimizing graft selection and improving patient outcomes. By integrating public scRNA-seq datasets and performing original immunofluorescence validations, we constructed a comprehensive single-cell transcriptomic atlas comparing the native coronary artery (CA) with the three principal grafts: the internal thoracic artery (ITA), radial artery (RA), and saphenous vein graft (SVG). We mapped their cellular ecosystems and molecular signatures to elucidate the biological basis of their clinical behavior. Our analysis provides a robust molecular rationale for the ITA’s gold-standard status, revealing a superior vasoprotective phenotype characterized by enhanced endothelial vasodilatory function, minimal contractile potential, and a uniquely quiescent cell-cell communication network. Conversely, the RA displays the highest intrinsic vasoreactivity signature, molecularly underpinning its clinical predisposition to vasospasm. The saphenous vein exhibits a multidimensional high-risk profile, with potent pro-proliferative, pro-fibrotic, and pro-inflammatory characteristics that align with its propensity for neointimal hyperplasia and accelerated failure. Crucially, we observed an association between the grafts’ transcriptomic similarity, quantified by a novel Mean Inter-cluster Distance (MID) metric, and their known clinical patency gradient (ITA > RA > SVG). This atlas elucidates the fundamental biological heterogeneity of CABG grafts, proposes key molecular pathways (e.g., RhoA/ROCK, MIF) as candidate targets warranting further investigation for improving graft longevity, and provides a foundational resource for personalizing graft selection. Our work offers a new quantitative framework for evaluating and designing the next generation of vascular grafts.