Background/Purpose: Recent literature has identified different synovial fibroblast (FLS) populations within RA synovium with distinct inflammatory profiles. Despite current advances in classifying heterogeneity of FLS subsets, understanding of the FLS landscape in healthy synovial tissue is limited. We aim to identify homeostatic vs pro-inflammatory FLS signatures in synovial tissue biopsies obtained from healthy controls (HC) and RA patients and identify conversion triggers and activated signalling pathways.

Methods: Single cell (Sc) RNAseq was performed on 21,759 FLS derived from intact synovial biopsies from 5 HC and 4 RA. Subsequently, multiparametric flow cytometric analysis (22 markers) was performed on digested synovial biopsies from HC subjects and RA patients (further stratified between ACPA+/-) to identify FLS subsets and characterize functional phenotypes.

Results: ScRNAseq identified 14 FLS clusters which broadly aligned to 4 main subsets: CD55+THY1(CD90)-FAP+, CD55-THY1+FAP+, CD55+THY1+FAP+, and CD55-THY1-FAP+. Subsequent analysis showed clusters generally fall into lining/sublining layer, immunoregulatory, and regulatory/homeostatic functional FLS subsets. Six clusters showed higher frequency in RA synovium and eight had higher frequencies in HC. Three of the clusters had lining layer markers including THY1-, CD55+, etc. and were enriched for invasive genes (MMP1, MMP3) and chemokines (CXCL1, CXCL8). Five clusters compose the THY1+ sublining layer fibroblasts and included a perivascular subset expressing NOTCH3, TAGLN, and ACTA2, in addition to enrichment of collagen and ECM genes. Two additional clusters sharing lining layer markers are immunoregulatory demonstrating enrichment of HLA-DR genes and genes involved in a chronic inflammatory response (IL7R, IL32, TGFB1). Of the homeostatic/regulatory subsets, two were enriched with transcription factors (TF), specifically those in the AP-1 TF family and mRNA splicing/ lipid homeostasis genes, while the two other clusters showed enrichment of genes involved in metabolic regulation and angiogenic function.

Flow cytometric analysis showed significantly higher frequencies of CD45-CD146-CD31-PDPN+ FLS in RA synovium compared to HC synovium. PDPN+ populations, both lining and sublining (based on the markers of the 4 main populations observed in the ScRNAseq) displayed higher frequency in RA synovium compared to HC. When stratified for ACPA positivity an increased frequency in sublining FLS population was demonstrated in ACPA+ RA compared to ACPA-, in contrast ACPA- RA had higher frequency of lining layer FLS. This suggests that differential FLS subsets are associated with the RA synovium vs with HC, in addition to possibly differentiating between ACPA status.

Conclusion: Identification of differential FLS subsets and their associated function in the RA vs HC synovium will better facilitate our understanding of their contribution to disease pathogenesis in RA. Furthermore, by stratifying RA patients between ACPA+ and ACPA- we can understand differences of these disease states and better tailor treatments.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/identification-of-homeostatic-and-inflammatory-synovial-fibroblast-signatures-in-synovial-tissue-biopsies-of-healthy-controls-and-patients-with-rheumatoid-arthritis/