Endothelial Cell Heterogeneity and Activation in Atherosclerosis

Abstract

Endothelial-to-mesenchymal transition (EndMT), aberrant angiogenesis, and inflammation are hallmarks of atherosclerosis. These processes involve activation of overlying endothelial cells (ECs) within the tunica intima that leads to plaque formation, progression, erosion, and rupture. There are multiple ways to model these processes in vitro (discussed in Chapter 1), including with the use of primary human aortic ECs (primary HAECs) and immortalized human aortic ECs (teloHAECs). Yet, characterization of these models on human ECs at single cell resolution, and the ability of such model systems to recapitulate multiomic profiles ex vivo, remain poorly understood. To understand EC heterogeneity, we have taken a multiomics approach. We first cultured populations of primary HAECs and teloHAECs with and without perturbations to mimic EndMT (discussed in Chapter 2 and Chapter 4), angiogenesis and inflammation (discussed in Chapter 3). We then performed Single Cell Multiome ATAC + Gene Expression (10X Genomics) to quantify transcript and accessible chromatin levels to understand biological pathways and mechanisms of gene regulation taking place. Finally, to evaluate these models, we test the ability of our in vitro data to recapitulate the transcriptomic profiles observed in atherosclerosis using publicly available scRNA-seq data taken from ex vivo human atherosclerotic plaques. Additionally, we utilize the latest coronary artery disease (CAD) Genome Wide Association Study (GWAS) to test for overlapping chromatin regions between our in vitro data and significant CAD-associated single nucleotide polymorphisms (SNPs). These studies reveal that EC subtype is a major determinant in the ability to recapitulate ‘omic profiles observed in atherosclerosis and underscore the importance of the utilization of single cell multiomics when considering in vitro models of atherosclerosis (Chapters 2-5). Moreover, we elucidate some of the effects of the perturbations discussed on ECs. For example, we find that Matrigel-seeding – used to recapitulate angiogenesis – downregulates biological pathways involved in EndMT and leads to in vitro transcriptomic profiles that are less like those observed within ex vivo atherosclerotic plaques compared to control (Chapter 3). We also discuss the use of teloHAECs within in vitro models of atherosclerosis and compare these cell lines head-to-head with primary HAECs with and without EndMT perturbations (Chapter 4). We find that TGFB2 treatment of ERG knockout teloHAECs may be a useful in vitro model of atherosclerosis. Altogether, this work underscores the vast diversity of ECs with and without perturbations in vitro, and the potential clinical implications of this heterogeneity in the context of the pathobiology of atherosclerosis.