Background Hypoxia-inducible factor (HIF)-1α is a critical transcription regulator in cellular response to hypoxic condition. Hypoxic microenvironment exists in rheumatoid arthritis (RA) joints and we have reported that elevated HIF-1α promoted angiogenesis and inflammation in RA. Recent studies showed HIF-1α might promote osteoclast-mediated bone resorption in vitro. However, little was known about its role in RA joint destruction.
Objectives To explore the correlation of synovial HIF-1α with joint destruction and its progression in RA.
Methods Patients with active RA were followed up and X-ray assessment of hand/wrist was repeated at baseline and one year. Erosive disease was defined according to 2013 EULAR definition and radiographic progression was defined as the increase of total Sharp score more than 0.5 from baseline to one year. Synovial tissue was obtained from RA patients as well as OA and orthopedic arthropathies (Orth.A) patients. Serial tissue sections were stained with H&E, immunohistochemically for CD3, CD20, CD38, CD68, CD15, CD34 and HIF-1α. Krenn's synovitis score was semi-quantitatively assessed and the density of positive-staining cells was quantitatively determined.
Results (1) Twenty five RA patients fulfilled 1 year follow-up, 84% were female, age (median and IQR, similarly hereinafter) was 50 (40–57) years, disease duration was 24 (10–48) months and disease activity SDAI was 35.0 (25.3–46.3).
(2) HIF-1α expressed in lining and sublining area with intense nuclear and endochylema staining in RA synovium and the percentage of HIF-1α+ cells [57% (31%–77%)] was significantly higher than that in OA [n=13, 25% (9%–41%), P=0.022]or Orth.A [n=8, 21% (12%–38%), P=0.021, Fig.1]. The percentage of HIF-1α+ cells significantly correlated with subscore of synovial stroma activation, CD34+ microvessel count, CD68+ macrophage count and CD15+ neutrophil count (r=0.420–0.586, all P<0.05).
(3) Eleven (44%) RA patients had erosive disease at baseline. The percentage of HIF-1α+ cells was significantly higher in erosive patients than non-erosive patients [78% (49%–82%) vs 47% (23%–62%), P=0.014]. ROC curve analysis showed that the tradeoff value of synovial HIF-1α for diagnosing erosive disease was 69% with sensitivity 73% and specificity 100% (AUC=0.792, 95%CI: 0.574–1.010, P=0.014). Patients were then divided into high (n=8) and low synovial HIF-1α (n=17) groups. Total Sharp score and erosion subscore were higher in patients with high synovial HIF-1α than that in patients with low synovial HIF-1α [33 (14–69) vs 10 (4–24), 18 (11–25) vs 3 (2–12), respectively, all P<0.05]
(4) Eight (32%) RA patients showed radiographic progression at one year. The percentage of HIF-1α+ cells was significantly higher in progressive patients than non-progressive patients [80% (58%–85%) vs 49% (18%–63%), P=0.018]. ROC curve analysis showed that the tradeoff value of synovial HIF-1α for predicting 1-year radiographic progression was 66% with sensitivity 75% and specificity 82% (AUC=0.798, 95%CI :0.576–1.019, P=0.018). Univariate logistic regression showed that high level of synovial HIF-1α was a significant predictor of 1-year radiographic progression (P=0.011, OR=14.00, 95%CI: 1.841 to 106.465).
Conclusions HIF-1α might play important roles in the pathogenesis of joint destruction in RA.
Acknowledgements This work was supported by National Natural Science Foundation of China (81471597).
Disclosure of Interest None declared