Utilising high throughput screening techniques to identify small molecule activators of the epicardium

The epicardium, the mesothelial layer covering the surface of the heart, plays an essential role during heart development. Epicardium derived cells (EPDCs) contribute essential cardiovascular cell types including vascular smooth muscle, interstitial fibroblasts and cardiomyocytes via the process of epithelial to mesenchymal transition (EMT) and EPDC paracrine signalling is critical for proper organ formation. Whilst dormant in the adult heart, the epicardium is reactivated in response to injury in both mouse and zebrafish. Activation is characterised by epicardial expansion, EMT, EPDC migration and re-expression of embryonic transcription factors including WT1. Moreover, priming the mouse heart with thymosin β4 (Tβ4) increases the number of Wt1+ EPDCs in vivo following myocardial infarction. Subsequently, small numbers of Tβ4 activated Wt1+ cells migrate into the wound forming functional cardiomyocytes. Unfortunately, the number of functional EPDC-derived cardiomyocytes is suboptimal to restore the lost heart muscle; therefore, the aim of this project was to augment the process using chemical or genetic modulation. To realise this aim, a high throughput phenotypic screen utilising primary human epicardium derived cells (hEPDCs) was developed in collaboration with the Target Discovery Institute. Recombinant transforming growth factor beta (TGFβ) was used to induce EMT in hEPDC cultures; morphological changes indicative of EMT were used as a surrogate read out for hEPDC activation. A rigorous analysis protocol was successfully developed, however, ongoing issues with access to primary human tissue led to investigation of alternate in vitro models of epicardial EMT. Preliminary investigations into an EPDC line derived from rat proved problematic and due to numerous inconsistencies in the model an alternate cell line was sought. Success was found with an immortalised EPDC line derived from mouse (mEPDC), which reproducibly underwent EMT in response to TGFβ treatment. However, mEPDC EMT was not accompanied by morphological changes as pronounced as in hEPDC cultures, thus the mouse line was deemed unsuitable for the previously described phenotypic screen. Moving forward, a high throughput scratch assay was developed which utilised migration as a surrogate read out for mEPDC activation. Upon screening of a small candidate library, two Bromodomain and Extra-Terminal motif (BET) inhibitors were found to reproducibly increase the speed of mEPDC wound closure. Screening a wider panel of BET inhibitors yielded several candidates that increased the rate of closure in a dose-responsive manner. To determine whether increased migration was correlated with a change in mEPDC phenotype the cells were cultured with an 'effective' dose of inhibitor and probed for a panel of differentiation markers. Several BET inhibitors were found to significantly upregulate expression of the EMT associated transcription factor Snai2 after 24 hours of treatment compared with the control. Together, the observed increase in cell migration and expression of the Snai2 transcription factor suggest a role for the BET bromodomains in epicardial activation and EMT.