Background Hypoxia in the inflamed joint is a powerful trigger of synovial fibroblasts and endothelial cell (EC) activation, proliferation and survival. The metabolic turnover of the inflamed synovium outpaces the dysfunctional vascular supply. It may induce a cellular switch in energy production characterized by an increase in glycolytic flux and suppression of mitochondrial functions. Hypoxia-inducible factor 1 (HIF-1) is a master regulator of oxygen homeostasis which mediates a wide range of cellular adaptive physiological responses to reduced oxygen availability. Furthermore, metabolic alterations of inflamed synovium can influence the immune responses and abnormal angiogenesis.
Objectives To identify the relationship between in vivo synovial hypoxia, glucose metabolism, mitochondrial dysfunctions, angiogenesis and inflammation quantified macroscopically and microscopically. Moreover, to determine cellular bioenergetics profile under hypoxia in vitro.
Methods Forty four patients with active inflammatory arthritis were recruited prior to starting biological treatment and underwent arthroscopy, clinical assessment and synovial tissue oxygen (tpO2) measurements. Macroscopic synovitis and vascularity were measured by visual analogue scale. Synovial microscopic levels of glycolysis (GAPDH), inflammation (CD3, CD68) and angiogenesis (Tie2) were quantified by immunohistology. Alterations of the synovial mitochondrial genome were assessed by Random Mutation Capture assay. Secretion levels of VEGF, Ang2, PDGFβ and TNF-α were measured in matched serum and synovial fluid by ELISA. To stabilize HIF-1α expression, primary synovial fibroblasts (RASFC) were stimulated with DMOG. Following stimulation with DMOG, RASFC metabolism was assessed using the XF24 Flux analyzer which simultaneously quantifies real-time measurements of aerobic (Oxygen Consumption Rate) and anaerobic (Extracellular Acidification) bioenergetic profiles.
Results The median tpO2 was 25.4 mmHg (range 3.2-63 mmHg). Synovial oxygen levels in vivo were inversely associated with synovial expression of GAPDH in the lining layer (r=-0.45; p<0.003), sublining layer (r=-0.33; p<0.035) and vascular region (r=-0.42; p<0.007). GAPDH was significantly increased in patients with tpO2<20mmHg compared to patients with tpO2>20mmHg (p<0.05). High synovial GAPDH expression correlated with increased macroscopic vascularity and synovitis (all p<0.05). Elevated microscopic glycolysis was concurrent to higher mitochondrial DNA mutations, indicating altered cellular energy metabolism. Functionally, exposure of RASFC to DMOG increased the rate of glycolysis with concomitant attenuation of mitochondrial respiration. DMOG altered RASFC mitochondrial respiration compared to unstimulated cells, demonstrated by a decrease in basal and maximal respiration and ATP turnover (all p<0.05) with a simultaneous induction in glycolytic cell activity (p<0.001). Finally, elevated synovial glucose assimilation was associated with higher serum levels of TNF-α and VEGF, synovial fluid PDGFβ and Ang2 and with synovial tissue expression of CD3, CD68 and Tie2 (all p<0.05).
Conclusions In the RA joint hypoxia down-regulates mitochondrial respiration and promotes a switch to anaerobic glycolysis. This may enable synovial cells to generate sufficient ATP to support enhanced synovial proliferation, angiogenesis and pannus formation.
Disclosure of Interest M. Biniecka: None declared, C. Ng: None declared, U. Fearon: None declared, D. Veale Grant/research support: AbbVie, Pfizer, MSD, Roche, Consultant for: Pfizer, Roche, Speakers bureau: Abbott, Pfizer, MSD, Roche