Background/Purpose: Systemic sclerosis (SSc) is a rare autoimmune disease characterized by skin and internal organ fibrosis, vascular abnormalities, and autoantibody formation. Single cell genomics studies have identified multiple fibroblasts populations in normal and fibrotic skin in vivo. Here we report a 3D skin-like tissue that reproduces multiple fibroblasts populations similar to those observed in human skin in vivo.

Methods: We have developed in vitro 3D skin-like tissues that reproduce phenotypic and molecular aspects of SSc. Two different types of 3D tissues were analyzed including a self-assembly (SA) tissue with only fibroblasts and a skin equivalent (saSE) tissue constructed with fibroblasts, keratinocytes, macrophages, and plasma isolated from donors. SA or saSE 3D skin-like tissues were seeded with either SSc-derived or control cells; the same cells were seeded into 2D culture in parallel. 3D skin-like tissues were dissociated into single cell suspensions and processed using the 10X genomics platform. Single cell RNA-sequencing (scRNA-seq) of the saSE tissues yielded expression profiles for 15,921 SSc and 31,632 control cells, the SA tissue model included expression profiles for 27,217 SSc and 26,415 control cells, 2D cells in cultures included 36,419 SSc and 26,991 control cells.

Results: scRNA-seq was performed to compare cells grown in 2D cultures to 3D SA, and saSE tissues. The most complex model, the saSE model, exhibited four distinct fibroblasts populations that parallel those observed in normal-human skin (Figure 1A). Analysis of fibroblasts grown in 2D culture showed the least diversity, consisting of only two fibroblast subsets. Fibroblast subsets in the more complex saSE tissue model were characterized by greater expression of mediators implicated in SSc pathogenesis. Fibroblast subpopulations were named by their top two differentially expressed genes and are referred to as “Fibroblast VEGFA, STC1”, “Fibroblast APOE, CDF”, “Fibroblast LUM, TPM1”, and “Fibroblast MMP2, PTGDS”. The “Fibroblast LUM, TPM1” cluster was marked by expression of collagen genes, notably COL1A1 and COL3A1, that are over-expressed in SSc (Figure 1B). Gene set enrichment analysis (GSEA) of each fibroblast subpopulation demonstrated upregulation of TGFβ signaling in clusters “Fibroblast LUM, TPM1” and “Fibroblast VEGFA, STC1”. IL2-STAT5 signaling, Hypoxia, and PI3K signaling were also reported as differentially-expressed pathways among these fibroblast subpopulations. Comparison to scRNA-seq data from human skin biopsies shows parallel populations of fibroblasts highly expressing APOE, PTGDS, and LUM.

Conclusion: We have used scRNA-seq to capture fibroblasts diversity in a fabricated 3D human skin model that recapitulates SSc dermal fibrosis. We identified four fibroblast subsets with potential roles in SSc. GSEA demonstrated that each of these populations is characterized by distinct signaling pathway activation. These fibroblast subsets likely make unique contributions to underlying SSc pathology and may represent targetable cell types for therapeutic intervention in 3D skin models that more closely approximate the human condition.

Figure 1: A. The saSE model system revealed four distinct fibroblast populations expressing biologically and pathologically relevant genes. We also identified a small populations of macrophages and keratinocytes. B. Differential gene expression between SSc and healthy control samples revealed upregulation of collagen genes COL1A1, COL3A1, and COL10A1 in SSc samples. This differential activation of fibroblasts is likely contributing to SSc pathogenesis.

« Back to ACR Convergence 2021

ACR Meeting Abstracts - https://acrabstracts.org/abstract/identification-of-distinct-fibroblast-populations-in-systemic-sclerosis-3d-skin-tissues-with-single-cell-omics/