Background/Purpose: Arthritis is a common manifestation of SLE but the inflammatory and immune cells that infiltrate synovium and the signaling pathways activated are not well understood. Previous work delineated an IFN signature in lupus arthritis but not rheumatoid arthritis (RA) or osteoarthritis (OA)¹. A multi-pronged bioinformatic approach was utilized to analyze gene expression from RA, OA and lupus arthritis (LA) samples to gain insight into the precise pathogenic mechanisms involved. 

Methods: Initial analyses examined genes that were similar and different between non-inflammatory OA, LA and RA synovial biopsy samples following LIMMA differential expression (DE) analysis or Weighted Gene Co-expression Network Analysis (WGCNA) followed by interrogation with cell type specific gene signatures using I-Scope^TM and validated by Gene Set Variation Analysis (GSVA). Genes were functionally characterized using BIG-C^TM and pathways elucidated using the upstream regulator (UPR) and canonical pathway functions of IPA^Ò.

Results: From BIG-C^TM and I-Scope^TM analysis, lupus synovitis has a clear gene signature of activated antigen presenting cells including dendritic cells, inflammatory myeloid cells, activated CD4⁺ and CD8⁺ T cells, activated B cells as well as pre- and post- switch plasma cells/plasmablasts as indicated by IgM, IgD and IgG1 heavy chain genes. The presence of both Igk and Igl as well as numerous VL genes indicates the polyclonal nature of the infiltrate. Analysis of gene expression indicated no inflammatory infiltrate in OA and that RA is dominated by activated T and B cell signatures. IPA analysis tools examining canonical pathways and UPRs revealed the role of Sphingosine-1 phosphate receptor and numerous chemokine pathways that may mediate lymphocyte organization and/or recruitment into lupus synovium. In addition, IPA elucidated ongoing signaling induced by inflammatory cytokines. Moreover, signaling pathways including NF-kB, NF-AT and mTOR were evident as well as proliferation and proteasome activity. Important differences were observed in gene expression from lupus compared to RA synovitis, including tendencies of less enrichment of Th17 and T_cyotoxic, less frequent enrichment of T_FH and increased enrichment of myeloid populations. In addition, IPA UPR analysis indicated ongoing signaling by TNF, IFNg, IFNa, CD40L, IL6, IL1b, IL15, IL21 and IL27 in lupus synovitis. 

Conclusion: Bioinformatic analysis of gene expression patterns in lupus synovitis indicate trafficking and potential in situ activation and differentiation of immune cells during pathogenesis of LA. Interestingly, a more prominent myeloid signature is present in LA compared with RA, that might provide a unique target of therapeutic intervention.

¹Nzeusseu-Toukap, A., et al. (2007). Arthritis & Rheumatism, 56(5), 1579–1588.