Background Synovial fibroblasts play a key role in the persistence of inflammation and joint destruction in rheumatoid arthritis (RA). Cell function and proliferation are highly dependent on availability of nutrients and their metabolism. In the joint, fibroblasts have limited access to nutrients within a poorly vascularised hypoxic tissue and yet expansion of fibroblast numbers is a feature of the RA joint. This suggests fibroblasts may adapt their metabolism to this environment and this altered function may be involved in preventing the resolution of chronic inflammation.
Objectives To use NMR spectroscopy to assess differences in metabolite fingerprints in fibroblasts from patients with established RA, early arthritis and healthy controls, and determine how cytokine production by fibroblasts relates to their metabolic profile.
Methods Fibroblasts were cultured from synovial biopsies from 6 newly presenting, disease-modifying drug naive patients with established RA (>12 week symptom duration), 6 healthy controls (HC), and patients with arthritis of ≤12 weeks duration whose disease resolved (n=6) or evolved into RA (n=6) at follow-up. Cell metabolites were extracted for analysis using 1D 1H-NMR spectroscopy and secreted IL6 measured by ELISA. NMR spectra were analysed using partial least squares discriminant analysis (PLSDA) and partial least squares regression (PLS-R) to correlate the metabolite profiles with 1) IL6 production by fibroblasts and 2) the level of CRP, at the time of synovial biopsy, in the serum of patients whose fibroblasts were being assessed.
Results We were able to distinguish the metabolic profiles of fibroblasts from HC and early RA (sensitivity 67%, specificity 67%), HC and established RA (sensitivity 67%, specificity 50%) and resolving arthritis and early RA (sensitivity 67%, specificity 83%). The IL6 production of fibroblasts from patients with inflammatory arthritis was clearly distinct from that of healthy controls. There was a strong correlation between the metabolic profile of synovial fibroblasts and their IL6 production (p<0.001) with several metabolites (in particular citrate, carnosine, pyroglutamate, alanine and lactate) contributing. In patients with inflammatory arthritis the fibroblast metabolic profile correlated strongly (p<0.001) with patient serum CRP at the time of synovial biopsy, with several metabolites (in particular cholesterol, fatty acids, leucine, citrate and pyroglutamate) contributing.
Conclusions There was a significant association between the metabolomic fingerprint of synovial fibroblasts and their IL 6 production, suggesting that IL6 production drives or is driven by significant changes in metabolism. There was also a significant association between CRP levels in the patients’ serum and the metabolic profile of their synovial fibroblasts suggesting that fibroblasts retain their metabolic fingerprint during culture ex vivo and that this is strongly related to systemic measures of inflammation in patients with clinical synovitis.
Disclosure of Interest None Declared