Background/Purpose: Chronic pain continues to be a significant problem for patients with rheumatoid arthritis (RA), even when the disease is medically controlled or in remission. RA patients often develop a generalized increase in pain sen­sitivity at remote sites from the inflamed joint, suggesting that central sensitization may play an important role. We have previously shown that joint inflammation is associated with an upregulation of inflammatory mediators in the central nervous system (CNS) of humans and mice. Here we sought to investigate if anti-citrullinated protein antibodies (ACPA) can be detected in the cerebrospinal fluid (CSF) of RA patients and if ACPA can be linked to central pain sensitization.

Methods: Clinical study: CSF (via lumbar puncture) and serum was collected from 13 female RA patients with anti-CCP positive (+) (ACPA⁺) antibodies. CSF from age-matched i) individuals with no neuroinflammatory or rheumatic disease (n=10) and ii) multiple sclerosis (MS) patients (n=26) was used as controls. ACPA IgGs were measured using a commercial anti-CCP2 ELISA according to the manufacturer’s protocol (serum diluted 1:100, CSF undiluted).Experimental study: Purification of IgG from plasma and sera from 36 separate ACPA⁺ patients was done by HiTrap Protein G columns and ACPA IgG was further purified using CCP2 affinity columns. IgG not binding to the CPP2-column was used as control and denoted flow through (FT). Balb/c/AnNRJ female mice (>12 weeks old) were used and ACPA or FT (100-300) μg was injected intrathecally into CSF by lumbar puncture. Mechanical hypersensitivity was assessed using von Frey filaments and cold sensitivity by the acetone test. Tissues were collected at day 20 for western blot analysis.

Results: Clinical study: 7/13 serum ACPA⁺ RA patients had ACPA also in CSF (29-154 U/ml). ACPA levels in controls were significantly lower; non-inflammatory controls: 2.19 +/- 0.09U/ml, and MS patients: 2.54 +/- 0.16U/ml. In RA patients, there was a strong correlation between the presence of ACPA in CSF and significantly higher serum ACPA levels (1453 +/- 82U/ml in RA ACPA pos serum and CSF vs 90 +/- 46U/ml; ACPA neg CSF, p<0.01). Experimental study: Intrathecal injection of ACPA, but not FT, led to a reduction in mechanical thresholds from day 10 to day 18.  In addition, 100 μg and 300 μg ACPA, but not FT, induced cold hypersensitivity at day 11. Western blot analysis showed a time dependent elimination of ACPA. Human IgG was not detectable in spinal cords or joints of ACPA injected mice 20 days post injection.

Conclusion: Our data show that ACPA can be detected in CSF of RA patients with high ACPA serum titers. We did not find indications of intrathecal ACPA production. In mice, intrathecal injection of ACPA induced pain-like behavior, indicating changes in spinal/brain facilitation of nociceptive signaling. Interestingly, the delayed onset of pain responses, and the long lasting effect being present despite clearance of antibodies, suggest antibody-induced secondary events at the spinal cord/brain level or redistribution of ACPA to peripheral sites. Altogether, these results give support to the hypothesis that ACPA may enter the CNS and possibly affect central pain mechanisms in RA.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/central-pain-sensitization-in-rheumatoid-arthritis-role-of-acpa/