Identification Of Proangiogenic Fibroblasts In Human Skin
Abra H. Shen, SB, Mimi R. Borrelli, MBBS, MSc, Michael Januszyk, MD, PhD, Nestor M. Diaz Deleon, Sandeep Adem, MS, Geoffrey C. Gurtner, MD, Derrick C. Wan, MD, Michael T. Longaker, MD, MBA.
Department of Surgery, Division of Plastic Surgery, Hagey Laboratory for Pediatric Regenerative Medicine, Stanford University School of Medicine, Stanford, CA, USA.
PURPOSE: Angiogenesis is crucial for healthy wound healing; however, excessive neovascularization may lead to ineffective perfusion and cutaneous scar formation. Identification of a fibroblast subset with proangiogenic qualities may advance our understanding of both fibroblast heterogeneity and its interplay with cutaneous wound healing and may have considerable therapeutic implications. These efforts, however, have been limited by the insufficient granularity inherent to traditional cell capture technologies. Here, for the first time, we apply high-resolution, high-throughput full transcriptome sequencing to identify and characterize novel subpopulations of human fibroblasts with proangiogenic transcriptional profiles.
METHODS: Fresh human foreskin was mechanically and enzymatically digested using collagenase buffer. 10X single cell RNA-sequencing analysis was performed on fibroblasts isolated from foreskin samples using fluorescence-activated cell sorting (FACS), with a combination of a negative (CD45-CD235a-CD31-) and positive (Thy-1 [CD90]) gating strategy. Transcriptional data underwent manifold-based dimensionality reduction and were visualized using UMAP plots to delineate subpopulations of fibroblasts and identify candidates with the strongest expression profiles for proangiogenic genes (i.e. vascular endothelial growth factor [VEGF], angiopoietin [ANGPT], and chemokine ligand [CXCL] family genes). Additionally, cluster-specific genes were identified as those most differentially expressed relative to all other cells. These were then used to perform gene set enrichment analysis (GSEA) against the Reactome pathway database to identify key molecular pathways upregulated in each cluster.
RESULTS: A total of 8 distinct subpopulations of cells were identified (Figure 1A). UMAP plots revealed two clusters of particular interest (clusters #0 and #7). Cells in these clusters had elevated expression of the proangiogenic genes ANGPT1 and HIF1A. Additionally, cluster #0 had a particularly high expression of VEGFA, while cluster #7 had a specific elevation of ANGPT2 gene expression (Figure 1B). The top cluster-specific genes for cluster #0 included SFRP2, RSPO3, and S100A4, which have all been associated with proangiogenic behaviors, while cluster-specific genes for cluster #7 included VEGFB and the angiogenic chemokines CXCL1, CXCL2, and CXCL3 (Figure 1C). GSEA of top scoring cluster-specific genes against the Reactome database identified numerous VEGF-related signalling pathways among top scoring networks (Figure 1D). Both clusters #0 and #7 exhibited high expression of the surface marker MCAM [CD146] (Figure 1E), which has been previously shown in other reports to mark other cell types that have an angiogenic role.
CONCLUSIONS: Transcriptomic analysis of FACS-sorted fibroblasts from fresh human foreskin revealed two subpopulations with proangiogenic transcriptional profiles, distinguished by high expression of the surface marker CD146. Future work will investigate the proangiogenic behavior of prospectively isolated CD146+ fibroblasts in vitro.
Back to 2020 Abstracts