Single-cell RNA sequencing analysis of the human fetal epicardium results in novel epicardial-specific markers and reveals a role for CRIP1 in epicardial EMT

Abstract

Abstract Background The outside mesothelial layer of the heart, the epicardium, is essential during cardiac development. Cells from this layer undergo epithelial-to-mesenchymal transition (EMT), migrate into the tissue and partake in cardiogenesis by contributing cells and by secretion of growth factors. In the adult heart, the epicardium is quiescent but upon ischemic injury it is activated and conveys a regenerative response. This activation is important for repair, but suboptimal compared to the cardiogenic potential during development. Understanding the processes and cells that underlie developmental epicardial activation could contribute to enhancing the adult post-injury response, but little is known about the human developing epicardium. The Purpose of this study is to unravel the cellular composition of the human fetal epicardial layer and to identify novel processes that regulate the activation of the epicardium. Methods An epicardial-cell enriched dataset was obtained by specifically dissecting the epicardial layers from 4 fetal human hearts (14-15 weeks gestation, obtained under informed consent and according to local ethical approval). Single live cells were sorted into 384-wells plates and sequenced. Data analysis was performed using R-packages RaceID3 and StemID2. Findings were validated using qPCR and immunohistochemistry on tissue sections and cell culture models. Results Analysis of 2073 cells revealed a clustering of epicardial (epithelial, pre-EMT) and epicardial-derived (mesenchymal, post-EMT) cell populations. Since classical markers for the epicardium such as Wilms’ Tumor 1 and T-box transcription factor 18 were unable to specifically demarcate the epithelial state, we identified and validated several new markers (e.g. C3, NRK) that allow robust classification of epicardium-associated populations. To establish the regulation of epicardial activation and EMT, we applied pseudo-time analysis. We selected genes that were highly regulated (>2 fold change) in this trajectory for further analysis in vitro. qPCR indicated that CRIP1 is expressed in epithelial cells and rapidly downregulated upon EMT (3x reduction, n=3, *p=0.0104). The expression profile of CRIP1 positively correlates with UPK3B (R2:0.98) and negatively with mesenchymal TCF21 (R2: 0.99), together with GO-terms a role for CRIP in maintaining the epithelial-state is likely. Indeed, siRNA experiments in vitro with cultured epicardial cells confirmed an essential role for CRIP1 in EMT regulation. Together with its confirmed protein expression within the epicardial layer in fetal and adult tissue, this protein is of interest for regulating activation of the epicardium. Conclusions We shed light on the composition of the human fetal epicardium using a highly enriched scRNA dataset. We identify specific markers for different cell-states, and reveal novel regulators of activation that in the future may be influenced to optimise the post-injury response.