Proteomic Characterization of Embryonic Epicardial-Myocardial Signaling

The communication between the epicardium and the underlying myocardium is crucial not only for proper heart development but also for homeostasis and response to injury in the adult heart. Studies of epicardial-myocardial signaling using targeted approaches have yielded many insights into the molecular basis of this communication. Still, the complex molecular interactions and regulatory networks involved in the epicardial-myocardial crosstalk are not well understood. To obtain a comprehensive, unbiased perspective on epicardial/myocardial signaling, we used a discovery-based proteomics approach that complements traditional targeted approaches. This thesis details the generation of an unbiased dataset of proteins secreted into the media by embryonic chicken epicardial-derived cells (EPDC)-heart explant (EHE) co-cultures. A 150-protein secretome dataset was then generated by integrating mass spectrometry with bioinformatics. The large size of the dataset enabled bioinformatics analysis to deduce networks for the regulation of specific biological processes and predicted signal transduction nodes within the networks. Functional analysis was performed on one of the predicted nodes, NF-κB, and demonstrated that NF-κB activation is an essential step in TGFβ2/PDGFBB-induced cardiac epithelial-to-mesenchymal transition. Taken together, we have generated an unbiased EHE secretome dataset for the first time. Bioinformatics analysis of the dataset successfully predicted an essential role for NF-κB in epicardial EMT, indicating that this dataset can be used to guide functional, and targeted investigation. Epicardial-myocardial crosstalk occurs in vivo in the sub-epicardial niche. To complement our ex vivo culture system-based proteomic study, I explored the feasibility of developing a matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI-IMS) protocol for untargeted detection and imaging of endogenous proteins/peptides on mouse heart sections. Intact protein imaging on embryonic heart sections demonstrated the feasibility of achieving images with sufficient spatial resolution targeted to the epicardium/subepicardium area. However, in situ protein identification only detected the most abundant proteins, with no additional secreted proteins identified. These results suggest further technology development is needed to identify low abundance proteins directly from tissue sections.