Abstract
Objective:
Mitochondria are found in the extracellular space in rheumatoid arthritis (RA). However, whether mitochondria are a source of autoantigens in RA has not been carefully addressed. Thus, we undertook this study to investigate the presence and significance of antimitochondrial antibodies (AMAs) in patients with RA.
Methods:
AMAs were measured in serum samples from 3 cross-sectional cohorts of RA patients (n = 95, n = 192, and n = 117) and healthy individuals (n = 38, n = 72, and n = 50) using a flow cytometry-based assay. Further, AMAs were detected using an anti-mitofusin-1 (anti-MFN-1) IgG enzyme-linked immunosorbent assay and Western blot analysis. A longitudinal inception cohort, followed up for a median of 8 years, was used to study disease progression.
Results:
AMA levels were elevated in RA patients from all 3 cohorts as compared to healthy individuals (P < 0.001, P < 0.05, and P < 0.01), with a range of 14-26% positivity. Levels of anti-MFN-1 antibodies correlated with AMA levels (r = 0.31, P = 0.006) and were elevated in RA patients as compared to healthy individuals (P < 0.001). The presence of AMAs was associated with erosive disease (P < 0.05) and interstitial lung disease (P < 0.01). Further, AMA levels were found to predict erosive disease (odds ratio [OR] 4.59, P = 0.006) and joint space narrowing (OR 3.08, P = 0.02) independent of anti-citrullinated protein antibodies. Finally, anti-MFN-1 antibodies identified seronegative patients developing erosive disease (OR 9.33; P = 0.02).
Conclusion:
Our findings demonstrate the presence of novel autoantibodies targeting mitochondria in the setting of RA. AMAs were used to stratify patients based on disease phenotype and to predict development of erosive disease, including in patients with seronegative disease. Our results highlight the essential role of mitochondria in the pathogenesis of RA and suggest a possible benefit of therapies targeting mitochondrial-mediated inflammation and clearance in these patients.