Association of Distinct Fine Specificities of Anti-Citrullinated Peptide Antibodies With Elevated Immune Responses to Prevotella intermedia in a Subgroup of Patients With Rheumatoid Arthritis and Periodontitis

Abstract

Objective: In addition to the long-established link with smoking, periodontitis (PD) is a risk factor for rheumatoid arthritis (RA). This study was undertaken to elucidate the mechanism by which PD could induce antibodies to citrullinated peptides (ACPAs), by examining the antibody response to a novel citrullinated peptide of cytokeratin 13 (CK-13) identified in gingival crevicular fluid (GCF), and comparing the response to 4 other citrullinated peptides in patients with RA who were well-characterized for PD and smoking.

Methods: The citrullinomes of GCF and periodontal tissue from patients with PD were mapped by mass spectrometry. ACPAs of CK13 (cCK13), tenascin-C (cTNC5), vimentin (cVIM), α-enolase (CEP-1), and fibrinogen β (cFIBβ) were examined by enzyme-linked immunosorbent assay in patients with RA (n = 287) and patients with osteoarthritis (n = 330), and cross-reactivity was assessed by inhibition assays.

Results: A novel citrullinated peptide cCK13-1 (₄₄₄ TSNASGR-Cit-TSDV-Cit-RP₄₅₈ ) identified in GCF exhibited elevated antibody responses in RA patients (24%). Anti-cCK13-1 antibody levels correlated with anti-cTNC5 antibody levels, and absorption experiments confirmed this was not due to cross-reactivity. Only anti-cCK13-1 and anti-cTNC5 were associated with antibodies to the periodontal pathogen Prevotella intermedia (P = 0.05 and P = 0.001, respectively), but not with antibodies to Porphyromonas gingivalis arginine gingipains. Levels of antibodies to CEP-1, cFIBβ, and cVIM correlated with each other, and with smoking and shared epitope risk factors in RA.

Conclusion: This study identifies 2 groups of ACPA fine specificities associated with different RA risk factors. One is predominantly linked to smoking and shared epitope, and the other links anti-cTNC5 and cCK13-1 to infection with the periodontal pathogen P intermedia.

Figure 1

A, Spearman's rank correlation coefficients for the correlations between different anti−citrullinated peptide antibodies (ACPAs) in rheumatoid arthritis (RA) patients. P < 0.001 for all correlations. Boldface indicates the correlation between citrullinated cytokeratin 13 (cCK13‐1) and citrullinated tenascin‐C (cTNC5) antibody levels. B, Venn diagrams showing the overlap between cTNC5, cCK13‐1, and cyclic citrullinated peptide 2 (CCP‐2) antibody responses (left) and between cTNC5, cCK13‐1, and citrullinated α‐enolase peptide 1 (CEP‐1) responses (right). Values in the diagrams are the number of patients in each subset, and values in parentheses are the total number of ACPA‐positive patients. cFIBβ = citrullinated fibrinogen β; cVIM = citrullinated vimentin.

Figure 2

Anti–citrullinated cytokeratin 13 (anti–cCK13‐1) antibody cross‐reactivity with citrullinated tenascin‐C (cTNC5), citrullinated fibrinogen β (cFIBβ), citrullinated α‐enolase peptide 1 (CEP‐1), and citrullinated vimentin (cVIM). Serum from patients positive for both anti–cCK13‐1 and the test antibodies were preincubated with increasing concentrations of the indicated test peptide, and the IgG responses to cCK13‐1 were measured by enzyme‐linked immunosorbent assay and presented as the OD. Each symbol represents an individual patient.