Background/Purpose:

Synovial fibroblasts (SF) profoundly influence physiological and pathological processes in the joint such as reaction to inflammatory stimuli and production of extracellular matrix. We recently reported that SF from different joints exhibit site specific gene expression patterns; more than 20% of all detected transcripts in RNA sequencing significantly differed between knee, shoulder and hand SF. Here we investigated how these substantial differences in transcriptional programs translate into joint specific function of SF.

Methods:

SF were isolated from synovial tissues of hip, knee, shoulder, elbow and metacarpophalangeal (MCP) joints of RA and OA patients undergoing joint replacement surgery. MetaCore (Thomson Reuters, log2 ratio=2, p=0.05) was used for pathway enrichment analysis of RNA sequencing data (Illumina HiSeq 2000). Basal and TNF-α-induced MMP1 secretion from SF was analyzed with ELISA. Transwell plates (5mm polycarbonate membrane, Corning Costar) were used to measure migration (16h) of leukocytes from healthy donors (n=2) towards supernatants of cultured SF. Adhesion to tissue culture plates 16h after seeding and proliferation of SF were measured in real time with the xCELLigence RTCA DP Instrument (ACEA Biosciences, Inc.). Synovitis was quantified in synovial tissues from hands (n=17), elbows (n=6), shoulders (n=7), knees (n=8) and hips (n=4).

Results: Pathway enrichment analysis of RNA sequencing data between joint localizations (knee, shoulder, MCP) revealed joint specific enrichment of several arthritis-relevant biological processes, such as chemotaxis, proliferation, cell adhesion and matrix degradation. SF from hands, elbows and shoulders (n=4 each) secreted higher levels of MMP1 protein compared to knee SF (n=8) under basal conditions (hand 1808±1008, elbow 1773±1450, shoulder 2073±1083 vs. knee 449±883 pg/ml) and TNF-α stimulation (hand 8431±2339, elbow 8216±2110, shoulder 7176±1428 vs. knee 5369±1515 pg/mL). While adhesion was significantly stronger in shoulder vs. hand SF (cell index 0.9±0.2 vs. 0.6±0.2, p=0.01, n=7 each), shoulder SF (doubling time: 67±15h, p=0.03, n=5) showed a significantly lower proliferative potential than hand (39±3h, n=4) and knee SF (45±9h, n=6). More leukocytes migrated towards the supernatants from cultured knee and hand SF (x fold both: 1.4±0.03 vs. control cell culture medium=1) than shoulder SF (x-fold: 1.2±0.1). The total synovitis score composed of changes in stroma, synovial lining thickness and leukocyte infiltration showed an increased value in hands and elbows (6.2±1.8, 6.2±1.3 vs. shoulder 4.4±2.8, knee 4.1±2.3, hip 4.3±1.5, p=0.06), but the influx of immune cells was significantly enhanced into the synovium of hands versus shoulders and hips (hands 2.1±0.6; shoulders 1±0.4; hips 0.8 ±0.5, p=0.02).

Conclusion:

SF from joints of different anatomic sites are functionally distinct cells and create a unique and specialized microenvironment in each joint localization. Since hand SF secrete higher amounts of MMP1 and exhibit stronger proliferative and chemotactic properties compared with SF from other joints, they may in particular contribute to joint destruction occurring in hand arthritides.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/joint-specific-function-of-synovial-fibroblasts-integrating-positional-transcriptomes-and-anatomic-patterns-of-arthritis/