Background/Purpose: Anti-peptidylarginine deiminase (PAD)-4 antibodies (Abs) are present in a portion of RA patients and associate with more severe joint disease, suggesting that they play a role in pathogenesis. A subset of anti-PAD4 Abs cross-react with PAD3, and these Abs are found to enhance PAD4 enzyme activity at physiologic calcium concentrations which can lead to increased citrullination. Anti-PAD3/4 cross-reactive Abs have also been associated with more severe lung disease in RA, suggesting they may have a direct effect in the lung. Because sputum anti-CCP Abs have been identified in RA, we sought to explore the presence and activity of anti-PAD4 Abs in the sputum of RA patients.

Methods: We studied 48 serum anti-CCP+ RA patients and 24 healthy controls. Induced sputum was tested for CCP using commercial ELISAs. Serum and sputum were tested for anti-PAD4 Abs using an established immunoprecipitation method. All anti-PAD4+ samples underwent testing for the presence of PAD3 cross-reactivity. To determine the effect of anti-PAD Abs on PAD4 enzymatic activity, Igs were purified from each anti-PAD+ sample and their effect on citrullination of the histone H3 substrate by PAD4 at increasing calcium concentrations (i.e. 0.2 and 2 mM) was measured by an anti-citrullinated histone H3 immunoblot.

Results: Serum anti-PAD4 Abs were detected in 6/48 (13%) of RA patients and 0/24 (0%) controls. Of the positive patients, 2/6 (33%) were also serum anti-PAD3+. Sputum anti-PAD4 Abs were detected in 3/48 (6%) of RA patients, and of those positive, 1/3 (33%) was also sputum anti-PAD3+. Serum anti-PAD4 Abs were more prevalent in RA patients with sputum anti-CCP Abs, with all anti-PAD4+ RA patients demonstrating sputum anti-CCP positivity (Table). In serum, anti-PAD4 Abs were predominately IgG, whereas in sputum, anti-PAD4 Abs were predominately IgA (Figure). Interestingly, all three samples (serum and sputum) with measurable anti-PAD3/4 Abs were able to increase PAD4 activity at 0.2 mM calcium (i.e. physiologic levels).

Conclusion: We identified serum anti-PAD4 Abs in a portion of RA patients, of which a subset was also anti-PAD3+. We demonstrate for the first time that anti-PAD4 Abs are also present in the sputum in a portion of RA patients, and predominately IgA. Importantly, we found that anti-PAD3/4 IgA present in the sputum was able to lower the calcium threshold required for PAD4 enzymatic activity. These findings suggest that anti-PAD4 Abs may have pathogenic activity directly in the lung, although larger studies are needed to understand the relationship between anti-PAD3/4 and underlying lung disease.