Background Recent data suggest RA may originate from an autoimmune response in inflamed mucosa. RA is associated with gingival and airway inflammation. Porphyroma gingivalis, a bacterium causing gingivitis, has been implicated in ACPA formation. Furthermore, ACPA of the IgA class, the main secreted mucosal antibody, precede arthritis by years in a proportion of RA patients. When aiming to prevent arthritis it is of central importance to understand the source and regulation of autoantibody production in pre-clinical RA.
Objectives to determine whether anti-citrullinated protein antibodies (ACPA) are produced at mucosal sites in ACPa positive individuals at risk for RA and to compare the ACPA isotype and fine specificity in mucosal fluids to peripheral blood.
Methods Saliva, sputum, faeces and peripheral blood were collected from 10 individuals with anti-cyclic citrullinated protein (anti-CCP2) antibody positive arthralgia who are part of a cohort that is being prospectively followed for the possible development of arthritis. Saliva and sputum were collected during an early morning visit. Sputum was induced by inhalation of sodium chloride aerosols 4.5%. Mucus plugs were selected from sputum for extraction of supernatants. Faeces was collected at the same visit using stool collection kits and extracted using 6% BSA phosphate buffered saline. ACPA of IgA and IgA class were measured in mucosal fluids and blood using two citrullinated fibrinogen peptides, one enolase peptide and one vimentin peptide and their respective arginin peptides as control.
Results All tested individuals tested positive for the anti-CCP2 test and ACPA in their blood at the day of mucosal fluid collection. Two patients had high reactivity against both citrullinated peptides and arginin controls and were considered ACPA negative. Anti-citrullinated vimentin, fibrinogen and enolase antibodies were detected in respectively 1, 8 and 4 individuals. The saliva of 4 patients contained IgA ACPA, two patients being positive for anti-fibrinogen and anti-enolase, two patients being positive for anti-fibrinogen only. All other saliva, sputum and faeces samples tested negative for ACPA. All 4 patients with IgA ACPA in saliva tested positive for the same antibodies in blood. In addition, in 3 of 4 patients ACPA were detected in blood that were not detected in saliva: in two patients anti-citrullinated fibrinogen antibodies and in one patient anti-enolase antibodies.
Conclusions IgA ACPA are detectable in the saliva of a proportion of ACPA positive individuals at risk for RA. In these patients, IgG ACPA in blood are directed against more peptides compared to IgA ACPA in saliva. Future analyses should focus on further defining the precise source and regulation of mucosal ACPA compared to systemic ACPA.
Acknowledgements The Dutch Reumafonds.
Disclosure of Interest None declared