Autoantibodies against human calpastatin in rheumatoid arthritis: epitope mapping and analysis of patient sera.

Autoantibodies against calpastatin have recently been described to be highly prevalent in sera of patients with rheumatoid arthritis (RA). When the sera of 45 patients with RA were analysed for autoantibodies against calpastatin by a newly developed enzyme-linked immunosorbent assay (ELISA), only four sera (8.9%) tested positive, which is not significantly different from the frequency observed in healthy controls. Since the ELISA is based on a synthetic peptide containing the C-terminal 27 amino acids of calpastatin bound to the solid phase, this negative result might be the consequence of the small antigen used. Therefore, a systematic analysis of the epitopes for autoantibodies in calpastatin was performed using sera from RA patients and healthy individuals. Recombinant fusion proteins containing fragments of calpastatin or the complete protein were produced and sera analysed by Western blots. In the N-terminal portion (amino acids 1-369), at least two major epitopes exist, against which 65% of normal sera as well as 76% of RA sera show reactivity in Western blot assays. These epitopes are not useful for clinical diagnostics. Only five out of 45 (11.1%) RA sera reacted against the C-terminal portion (amino acids 363-708) of calpastatin, while four out of 52 (7.7%) control sera showed reactivity. Three of the five RA sera and two out of four control sera had autoantibodies against the C-terminal 27 amino acids of calpastatin. These three patient sera had already been tested positive in the ELISA. The fourth patient positive in the ELISA was Western blot negative. The differences between the group of RA patients and controls are not statistically significant. When the clinical characteristics of the four patients with autoantibodies against the carboxyl end of calpastatin were analysed, it became apparent that all four had significantly elevated C-reactive protein (>50 mg/l). This observation might indicate that calpastatin autoantibodies are found in RA patients with more active disease. Thus, while the majority of RA patients do not have an increased prevalence of calpastatin autoantibodies, it cannot be ruled out definitively that a small subgroup may be characterized by autoantibodies to the C-terminus of calpastatin.