Author:

Keywords:

TECNEC - 743074;info:eu-repo/grantAgreement/EC/H2020/743074, G0B7920N#55522276

Abstract:

Vascular endothelial cells (ECs) pervasively line the lumen of all blood vessels in the body. ECs are quiescent under physiological homeostasis but adopt one of three states in disease. EC activation includes the induction of cell adhesion molecules involved in immune cell recruitment and is associated with inflammatory diseases. EC dysfunction constitutes a state where ECs are no longer able to perform their homeostatic physiological functions and is a characteristic of prevalent diseases such as diabetes and atherosclerosis. Finally, ECs frequently adopt a blood vessel forming angiogenic state in diseases such as cancer and obesity. Therapeutic targeting of angiogenic ECs using agents that inhibit pro-angiogenic growth factor signaling is clinically approved to treat neovascularization in age-related macular degeneration (AMD) and cancer. Unfortunately, intrinsic and acquired resistance limit efficacy in both indications. Functional genetics in normal ECs showed that growth factor signaling converges on central carbon metabolism, and that EC metabolism can overrule pro-angiogenic growth factor signaling. However, whether metabolic reprogramming also occurs in diseased ECs, and whether this includes pathways beyond central carbon metabolism is almost entirely unknown. The goal of this work was to provide an atlas of metabolic heterogeneity in single diseased ECs in AMD and cancer. Since metabolomics is insufficiently sensitive to measure metabolite levels and fluxes in single ECs, and we previously documented that changes in metabolic gene expression are predictive of changes in metabolism in ECs, we used single cell RNA sequencing (scRNA-seq) to characterize the metabolic transcriptome of ECs at the single cell level. In the first study, we single-cell RNA-sequenced ECs from human and mouse lung tumors and detected 17 known and discovered 16 novel phenotypes. Integrated analysis of scRNA-seq data with orthogonal multi-omics and meta-analysis data across different human tumors, validated by functional analysis, identified collagen metabolism and modification as a putative angiogenic candidate pathway. In a second study, we used scRNA-sequencing to profile choroidal ECs isolated from mouse models of AMD. Trajectory inference suggested that ECs plastically upregulate genes in central carbon metabolism and collagen biosynthesis during angiogenic differentiation. Choroidal EC-tailored genome scale metabolic modeling integrated with scRNA-seq and gene expression meta-analysis identified rate limiting enzymes in cholesterol and proline biosynthesis as top-ranking candidates in proliferation and collagen production, respectively. In line with the stated ambition of this project, we developed user-friendly browseable databases to accompany each study in order to maximize resource value and facilitate the design of follow-up studies on the metabolic targets that can be derived from this work.