The epigenomic landscape of single vascular cells reflects developmental origin and identifies disease risk loci

Vascular sites have distinct susceptibility to atherosclerosis and aneurysm, yet the biological underpinning of vascular site-specific disease risk is largely unknown. Vascular tissues have different developmental origins that may influence global chromatin accessibility, and understanding differential chromatin accessibility, gene expression profiles, and gene regulatory networks (GRN) on single cell resolution may give key insight into vascular site-specific disease risk. Here, we performed single cell chromatin accessibility (scATACseq) and gene expression profiling (scRNAseq) of healthy adult mouse vascular tissue from three vascular sites, 1) aortic root and ascending aorta, 2) brachiocephalic and carotid artery, and 3) descending thoracic aorta. Through a comprehensive analysis at single cell resolution, we discovered key regulatory enhancers to not only be cell type, but vascular site specific in vascular smooth muscle (SMC), fibroblasts, and endothelial cells. We revealed an epigenetic 'memory' of embryonic origin with differential chromatin accessibility of key developmental transcription factors such as Tbx20, Hand2, Gata4, and Hoxb family members and discovered transcription factor motif accessibility to be cell type and vascular site specific. Notably, we identify ascending fibroblasts to have distinct epigenomic patterns, highlighting SMAD2/3 function to suggest a differential susceptibility to TGF?, a finding we confirmed through in vitro culture of primary adventitial fibroblasts. Finally, to understand how vascular site-specific enhancers may regulate human genetic risk to disease, we integrated genome wide association study (GWAS) data for ascending and descending aortic dimension, and through using a novel technique, we employed ChromBPNet, a distinct base resolution deep learning model to predict variant effect on chromatin accessibility We trained nine ChromBPNet models to predict variant effect in SMC, Fibroblasts, and Endothelial cells within ascending aorta, carotid, and descending aorta sites of origin. We reveal that although cell type remains a primary influence on variant effect, vascular site modifies effect within cell type and highlights genomic regions that are enriched for specific TF motif footprints — including MEF2A, SMAD3, and HAND2. This work supports a paradigm that the epigenomic and transcriptional landscapes of vascular cells are cell type and vascular site-specific and that these vascular site-specific enhancers govern complex genetic drivers of vascular site-specific disease risk. Overall design: Tissues were collected and dissociated for single cell capture as previously described. Briefly, vascular tissue was washed three times in PBS, tissues were then placed into an enzymatic dissociation cocktail (2 U ml-1 Liberase (5401127001; Sigma–Aldrich) and 2 U ml-1 elastase (LS002279; Worthington) in Hank's Balanced Salt Solution (HBSS)), and minced. After incubation at 37 °C for 1 h, the cell suspension was strained and then pelleted by centrifugation at 500g for 5 min. The enzyme solution was then discarded, and cells were resuspended in fresh HBSS. To increase biological replication, 16 mice were used to obtain single-cell suspensions for each vascular tissue (aortic root/ascending aorta, brachiocephalic/carotid, and descending thoracic aorta) with two captures per vascular site. Cells were FACS sorted and live cells were identified as previously described. Briefly, cells were sorted on a BD Aria II instrument, where cells were gated on forward/side scatter parameters to exclude small debris and then gated on forward scatter height versus forward scatter area to exclude obvious doublet events. Approximately 100-150,000 live cells were sorted for each vascular site, where a portion of cells were taken directly to scRNAseq capture. For single cell ATAC, cells were collected in BSA-coated tubes, and nuclei isolated per 10X recommended protocol, and captured on the 10X scATAC platform. All single-cell capture and library preparation was performed in the Quertermous lab and sequencing was performed at MedGenome (Foster City, CA). Cells were loaded into a 10x Genomics microfluidics chip and encapsulated with barcoded oligo-dT-containing gel beads using the 10x Genomics Chromium controller according to the manufacturer's instructions. Single-cell libraries were then constructed according to the manufacturer's instructions (Illumina). Libraries from individual samples were multiplexed into one lane before sequencing on an Illumina platform with targeted depth of 50,000 reads per cell for RNA and 75,000 reads/cell for ATAC. Post filtering for non-cells, mean number of reads within peaks per cell in scATAC data was 18,000-20,000 as was seen in our prior report.