Background Rheumatoid arthritis synovial fibroblasts (RASF) are key mediators of the persistence of inflammation and joint destruction in RA. Their activated phenotype has been likened to that of cancer associated fibroblasts, but its molecular basis remains poorly understood. One key question that remains unanswered is: why does arthritis resolve spontaneously in some patients whilst it persists as RA in others? We hypothesised that (1) the transcriptional profile of RASF would differ from that of SF obtained from patients with resolving arthritis and (2) that these differences would identify key genes responsible for the resolution of inflammation. To test our hypotheses we adopted an unbiased approach using high throughput sequencing SAGE methodology

Objectives To compare the transcriptional profiles of SF in two distinct outcome groups: resolving arthritis and longstanding RA to determine key genes responsible for the resolution of inflammation

Methods Synovial biopsies were obtained from treatment naïve patients with inflammatory arthritis of less than 3 months duration whose arthritis spontaneously resolved in the absence of DMARDs by 18 months of follow up (Resolving, n=8). RA samples were obtained from patients with longstanding disease during joint replacement (n=8). SF were cultured under identical conditions of confluence and passage (P5) using established methods.

mRNA was purified using µMacs columns (Miltenyi Biotech) and SAGE tags prepared (SOLiD SAGE kit). Emulsion PCR was performed and beads were sequenced using a SOLiD 4 analyzer. Tags were mapped to the human genome using the Bioscope analysis pipeline. Gene expression profiling was performed using SOLiD SAGE analysis software. Statistical analysis of SAGE expression data was done using edgeR run under bioconductor

Results 1736 genes were differentially expressed between RA and Resolving groups. Of these 1364 were downregulated and 372 upregulated in RA compared to resolving (p<0.01). Transcriptional profiling confirmed differences in gene expression for molecules already known to be dysregulated in RASF such as factors involved angiogenesis (FGF, TGFβR), ECM destruction (MMP24) and cell cycle regulators and transcription factors (PTEN, fos, sumo-1). It also revealed novel targets of potential interest such as the c-myc target gene MINA involved in cell growth, apoptosis and proliferation and FBLIM-1 involved in cell adhesion and migration

Conclusions most of the work done on RASF has used candidate gene approaches to unravel the molecular mechanisms underpinning their activation. This is the first study to undertake transcriptomic analysis comparing persistent RA to resolving disease. Our initial analysis has identified 1736 differentially expressed genes. Our results not only confirm existing RASF data but have also identified new candidates. Further in depth analysis to elucidate the transcriptional profile of resolving fibroblasts is ongoing

Disclosure of Interest None Declared