Deciphering the transcriptional network of the distinct heart cell types after myocardial infarction

Resumen

Despite the tremendous effort undertaken so far to decipher the cellular response in cardiac repair many questions remain unanswered. The recent advances in omics technology have opened up the possibility to address these questions and to understand the cell type specific functional contribution. However, the massive amount of generated data requires thorough bioinformatics work by implementing a plethora of statistical analysis and mathematical methods that have to be chosen with care to understand underlying molecular mechanisms. Another integral part of the data analysis process is the visualization of the obtained results. Due to the high-dimensionality of the data this step is rather challenging and further improvement is needed to provide advanced methods that facilitate the combination of various data types to enhance graphical representation of context-dependent, relevant information. For this purpose, we developed the R package GOplot to visually combine expression data with functional information (results chapter 1). The R package further supported the exhaustive bioinformatics analysis of ad-hoc generated transcriptomics data of murine cardiac macrophages (Mφ; results chapter 2) and distinct porcine cardiac cells (results chapter 3) in the healthy heart and after myocardial infarction (MI). The analysis revealed that murine cardiac Mφ can be distinguished from other tissue-resident Mφ by cardiac specific transcriptional programs and the expression of cardiac specific markers. The global transcriptional profile of murine post-MI Mφ could be matched to M1/M2 polarization states but systematic gene expression analysis and partial deconvolution revealed that these Mφ are a mix of described in vitro phenotypes. Simulated dynamics and experimental validation of identified core regulatory genes showed that the initial inflammatory response of murine Mφ is limited by hypoxia-mediated up-regulation of Il10. We also found that targeted mRNA degradation contributes to the resolution of inflammation and the Mφ phenotype transition. In swine, gene signatures could be identified for each cell type that were linked to distinct biological functions. In silico analysis of the secretome and receptome indicated a complex network of intercellular communication with numerous potential interactions between Mφ and endothelial cells (EC) and cardiomyocytes (CM) and fibroblasts (FB) post-MI. Collectively, the gathered data provide a useful resource to improve the understanding of the regulatory network of cardiac repair and suggest potential targets for tissue and cell type specific therapeutic interventions to manipulate cardiac repair.