Background/Purpose: Arthritis is a common manifestation of SLE and the ability of a new lupus therapy often depends on its ability to suppress joint inflammation. Despite this, an understanding of the underlying pathogenic mechanisms driving lupus synovitis remains incomplete. We, therefore, interrogated gene expression profiles of SLE synovium to gain insight into the nature of joint inflammation in lupus arthritis.

Methods: Biopsied knee synovia from SLE and OA patients were analyzed for differentially expressed genes (DEGs) and also by Weighted Gene Co-expression Network Analysis (WGCNA) to determine similarities and differences between gene profiles and to identify modules of highly co-expressed genes that correlated with clinical features of lupus arthritis. DEGs and correlated modules were interrogated for statistical enrichment of immune and non-immune cell type-specific signatures and validated by Gene Set Variation Analysis (GSVA). Genes were functionally characterized using BIG-C and canonical pathways and upstream regulators operative in lupus synovitis were predicted by IPA®.

Results: DEGs upregulated in lupus arthritis revealed enrichment of numerous immune and inflammatory cell types dominated by a myeloid phentoype, whereas downregulated genes were characteristic of fibroblasts. WGCNA revealed 7 modules of co-expressed genes significantly correlated to lupus arthritis or disease activity (SLEDAI or anti-dsDNA titer). Functional characterization of both DEGs and WGCNA modules by BIG-C revealed consistent co-expression of immune signaling molecules and immune cell surface markers, pattern recognition receptors (PRRs), antigen presentation, and interferon stimulated genes. Although DEGs were predominantly enriched in myeloid cell transcripts, WGCNA also revealed enrichment of activated T cells, B cells, CD8 T and NK cells, and plasma cells/plasmablasts, indicating an adaptive immune response in lupus arthritis. Th1, Th2, and Th17 cells were not identified by transcriptomic analysis although IPA® predicted signaling by the Th1 pathway and numerous innate immune signaling pathways were verified by GSVA. IPA® additionally predicted inflammatory cytokines TNF, CD40L, IFNα, IFNβ, IFNγ, IL27, IL1, IL12, and IL15 as active upstream regulators of the lupus arthritis gene expression profile in addition to the PRRs IRF7, IRF3, TLR7, TICAM1, IRF4, IRF5, TLR9, TLR4, and TLR3. Analysis of chemokine receptor-ligand pairs, adhesion molecules, germinal center (GC) markers and T follicular helper (T_fh) cell markers indicated trafficking of immune cell populations into the synovium by chemokine signaling, but not in situ generation of fully-formed GCs. GSVA confirmed activation of both myeloid and lymphoid cell types and inflammatory signaling pathways in lupus arthritis, whereas OA was characterized by tissue repair/damage.

Conclusion: Bioinformatic analysis of lupus arthritis reveals a pattern of immunopathogenesis in which myeloid cell-mediated inflammation dominates, leading to further recruitment of adaptive immune cells that contribute to the ongoing inflammatory synovitis.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/analysis-of-gene-expression-from-systemic-lupus-erythematosus-synovium-reveals-unique-pathogenic-mechanisms/