Abstract Background The epicardium, the outer layer of the heart, is an indispensable source of cells and paracrine factors during embryonic heart formation. In the adult heart, the epicardium is quiescent unless there is injury. Cardiac damage results in partial recapitulation of developmental processes including epithelial-to-mesenchymal transition (EMT), expression of Wilms' Tumor-1 (WT1), proliferation, and migration of epicardial-derived cells (EPDCs). Aim Given their vital role during development, EPDCs represent an appealing source for endogenous cardiovascular repair. However, EPDC contribution to cardiac tissue formation in the adult is less efficient than during embryonic development. Our aim is to determine the requirements to optimize the adult epicardial response to injury. Methods Human foetal and adult EPDCs were isolated from cardiac specimens and cultured as epithelial-like cells in the presence of an Alk5-kinase inhibitor (A5ki). EMT was induced by adding 1 ng/mL TGFβ for 5 days. Immunofluorescent staining, qPCR, and cytokine arrays were performed. Cultured adult EPDCs pre- and post-EMT were transplanted into the myocardial wall of NOD-SCID mice after inducing myocardial infarction (MI), and cardiac function was measured by high-frequency ultrasound. Hearts were histologically analysed 3 days and 6 weeks post-MI. Results Both foetal and adult human EPDCs can be expanded in culture and undergo EMT after TGFβ stimulation leading to morphological changes accompanied by downregulation of WT1 and E-cadherin, and upregulation of mesenchymal genes. Importantly, upon removal of Alk5ki, foetal EPDCs display instant spontaneous EMT, suggesting the importance of this process for EPDCs' developmental potential. In vivo, animals receiving intramyocardial transplantation of post-EMT EPDCs displayed a higher ejection fraction 6 weeks after MI compared to pre-EMT EPDC receiving animals (26%±11 n=8 vs. 11%±5 n=9 respectively P<0.05). This corresponded to a smaller infarct size in the post-EMT group (16,4%±4 of the left ventricle versus 26,9%±5 in pre-EMT, p<0.05). This could not be explained by a difference in cell grafting, analysed at 3 days post-MI. After 6 weeks, we observed a small difference in human collagen deposition in the post-EMT group, however very low numbers of human cells were detected suggesting a predominantly short-acting paracrine effect. Analysis of cytokine production of cultured cells revealed a higher production of factors involved in angiogenesis and chemotaxis like VEGF and MCP-3 in post-EMT EPDCs in comparison to pre-EMT EPDCs. Effects on local angiogenesis and inflammation in vivo are being investigated Conclusion EPDCs require EMT to acquire the ability to contribute to cardiac repair, which appears to be predominantly through paracrine processes. Our research now focuses on enhancing EMT of endogenous epicardial cells. Acknowledgement/Funding AMS is funded by a Dekker fellowship from the Dutch Heart Foundation