Low levels of antibodies against common viruses associate with anti-citrullinated protein antibody-positive rheumatoid arthritis; implications for disease aetiology
Natalia Sherina,Hulda S Hreggvidsdottir,Camilla Bengtsson,Monika Hansson,Lena Israelsson,Lars Alfredsson,Karin Lundberg
DOI: https://doi.org/10.1186/s13075-017-1423-9
2017-09-30
Abstract:Background: Infection by common viruses has long been discussed in the aetiology of a number of autoimmune diseases, including rheumatoid arthritis (RA). However, studies investigating this hypothesis in RA show conflicting results. These studies often lack well-matched control populations, and many do not include data on autoantibodies, genetic risk factors and other environmental factors, which are known to contribute to disease only in subgroups of patients. In the present study, we have therefore examined the role of Epstein-Barr virus (EBV), cytomegalovirus (CMV) and parvovirus B19 (B19) in RA aetiology, by analysing anti-viral antibodies in relation to anti-citrullinated protein antibodies (ACPA), smoking, HLA-DRB1 shared epitope (SE) alleles, and clinical parameters, in both RA patients and matched controls. Methods: Anti-viral antibodies were measured by ELISA in serum samples from 990 RA patients and 700 controls from the Swedish population-based Epidemiological Investigation of RA (EIRA) cohort. Data on ACPA, smoking, SE, inflammation (C-reactive protein) and disease activity score in 28 joints (DAS28) was obtained from the EIRA database. Fisher's exact test, the chi-squared test, and the Mann-Whitney U test were used to calculate differences in anti-viral antibody frequencies and levels; unconditional logistic regression was used to determine the association of anti-viral antibodies with different RA subsets. Results: Antibodies against all viruses were highly prevalent in EIRA, with no major differences detected between ACPA-positive RA, ACPA-negative RA and controls. However, both anti-B19 and anti-EBV IgG levels were significantly lower in ACPA-positive RA compared to controls, and there were significant interactions between low levels of anti-B19 and anti-EBV antibodies and SE in the development of ACPA-positive RA. Conclusion: We could not detect an association between RA and elevated anti-viral antibody levels, for any of the three common viruses, EBV, CMV or B19. On the contrary, our study demonstrated association between low anti-EBV/anti-B19 antibody levels and ACPA-positive RA, in particular when HLA-DRB1 SE was present. These data could potentially suggest that high anti-viral antibody levels would be protective against ACPA-positive RA. Further investigations are required to address the mechanisms behind these findings.