Background/Purpose: The presence of anti-citrullinated protein antibodies (ACPA) is a hallmark of rheumatoid arthritis (RA). ACPAs specifically recognize citrullinated epitopes, a result of a post-translational modification catalyzed by peptidyl arginine deiminases (PAD). Based on the unique feature of the periodontal bacteria Porphyromonas gingivalis (P.g) to express P.PAD it has been suggested that ACPA-positive RA may be precipitated in the gum mucosa. To address this hypothesis, we have investigated the antibody response against a citrullinated peptide derived from P.PAD (CPP3) in patients with RA, patients with chronic periodontitis (PD), and in controls. In addition, we have cloned B cells from gingival tissue of a patient suffering from both PD and RA and generated monoclonal antibodies (mAbs) with an aim to demonstrate that citrulline-specific B cells, previously detected only in RA joints and circulation, may also reside in gingival tissue.

Methods: Gingival tissue-derived single CD19+B cells from an ACPA-positive RA patient with PD were sorted by flow cytometry. Immunoglobulin variable region genes were sequenced and expressed to generate recombinant mAbs. CPP3-reactivity was analysed by ELISA in serum samples from 66 PD patients, 63 periodontally healthy controls (non-PD), 200 RA patients, and 120 systemically healthy controls (non-RA), as well as in 55 mAbs. Differences in antibody levels were examined using Mann-Whitney U test for independent groups.

Results: Anti-CPP3 antibody levels were low in non-PD controls, while 65% of PD patients showed elevated levels (p<0.0001). Significantly increased antibody levels were also detected in 50% of ACPA-positive RA, 20% of ACPA-negative RA, and in 37% of non-RA controls (p<0.0001). Notably, this antibody response was citrulline-specific, as the antibody response against the arginine-containing control peptide RPP3 was significantly lower in all subsets (p<0.0001). Among 55 mAbs from gingival tissue, 14 (25%) unique clones were CPP3-reactive, of which 4 showed cross-reactivity with RPP3. Interestingly, 4 out of 14 (29%) CPP3-reactive clones also bound citrullinated peptides derived from human a-enolase, filaggrin and histone4, demonstrating cross-reactivity between a bacterial epitope and human epitopes on a monoclonal level.

Conclusion: This study shows that a substantial proportion of systemically healthy individuals possess ACPAs directed against P.g, and these ACPAs also bind epitopes on human proteins. Based on our data, we propose that the ACPA response may be triggered by P.g, via an antibody response against CPP3, which cross-reacts with human citrullinated proteins by mechanisms of molecular mimicry.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/molecular-mimicry-and-autoimmunity-anti-p-gingivalis-antibody-response-in-acpa-positive-rheumatoid-arthritis/