gms | German Medical Science

Background: Porphyromonas gingivalis (P g.) is involved in triggering self-reactive immune responses when cirtullinating bacterial or human proteins. However, first evidence to link anti-ribosomal T and B cells responses to rheumatoid arthritis (RA) has been published but the pathomechanism and its influence on therapy is not clear [1]. Infection based autoimmunity induced by citrullination of human proteins with P g. peptidyl arginine deiminase from RA patient (RA-PPAD) and crossreactivity binding induced by P g. was investigated using patient sera, affinity purified RA patient antibodies and monoclonal antibodies to cit-RA-PPAD. Infection induced enzymatic mimicry is the first time linked to target specific bacterial citrullinated human proteins.

Methods: Screening of RA sera was conducted on 37.830 unique human proteins on protein marcoarrays (http://www.engine-gmbh.de) with 30 RA sera. The autoantibody response to 840 different proteins was recorded and bioinformatically evaluated. Protein arrays with 840 RA autoantigens were citrullinated with human peptidyl arginine deiminase PAD 2, 4, rabbit PAD and RA-PPAD from P.g. Sera and affinity purified antibodies were used to detect reactivity to 840 autoantigens and 15aa CCP peptide form RA-PPAD.

Results: A human protein macroarray consisting of 840 indentified autoantigens from RA patients was modified by human PAD2 ,PAD4, rabbit PAD, and RA-PPAD form P g. Out of 840 antigens we identified 157 autoantigens (23 are nativ targets ) and 134 are targeted by RA patient sera specifically after citrullination with RA PPAD from P. gingivalis. Interstingly CCP positive and CCP negative RA patients patient sera tested reacted with the RA-PPAD citrullinated human autoantigens.

Using cit specific monoclonal antibodies we identified the ribosomal proteins (RP), RPL18a, RPS27a, modified by PAD2, RPL18a and MRPS11 modifies by PAD4, and RPL7L1 modified by rabbit PAD specifically targeted. In addition 6 RA patient sera and 3 osteoarthritis (OA) control sera were used to identify the citrullinated RA-PPAD specific modified autoantigens not targeted when modified by human PAD2 or PAD4 or rabbit PAD. We identified the RA-PPAD citrullinated ribosomal Proteins RPL3, RPL21, RPS24, RPL9, RPL15, RPS24.RPS3a, MRPL28 specifically targeted by RA patients. This identifies ribosomal proteins as major specific RA-PPAD citrullination targets. Moreover, affinity purified antibodies bound to native and citrullinated RA-PPAD from 6 RA patient sera and 3 OA patient sera were tested for crossreactvity on citrullinated 840 human proteinarray. Antibodies to citrullinated ribosomal proteins MRPS11, RPL21, RPS3a, RPL18a, RPS27a, MRPL28 were detected in the RA group but not in the OA control group. High antibody titre against the cit-PPAD-peptide of 15aa (CPP) derived from the autocitrullination site (R63) of RA-PPAD correlates with TNFα-inhibitor (TBA) non-response (n=61). DMARD Patients refractory patients to different treatment regimes, receiving anti-TNF (Humira,Enbrel) treatment, do not respond when maintaining high α-CPP IgG level.

Conclusion: Failure of Porphyromonas gingivalis clearance in RA patients leads to infection induced enzymatic mimicry. Autoimmunity to ubiquitous citrullinated self-antigens with bacterial PPAD enzymes may trigger localized tissue damage in RA. TBA non-response leads to the suggestion to clear Porphyromonas gingivalis infection before α-TNF treatment.