Abstract
OBJECTIVE: Anti-Sa antibodies have been described to be a highly specific marker for rheumatoid arthritis (RA). We demonstrate the existence of 2 different subsets of anti-Sa antibodies, only one of which is specific for RA. Our objective was to purify the Sa antigen, and to achieve partial characterization of these proteins. METHODS: Saline extract and mitochondrial extract from human placenta were used as antigenic sources. Antigens were purified by immunoaffinity chromatography and studied by ELISA and immunoblotting. RESULTS: Three antigenically active bands of 68, 50, and 46 kDa were purified from the saline extract by immunoaffinity chromatography. Two other bands of 29 and 10 kDa that do not react with anti-Sa antibodies were obtained as well. The 68 kDa band was purified from a mitochondrial extract. These bands are not the same as other known mitochondrial autoantigens such as M2, M4, or M9. The amino terminal sequence of the 68 kDa Sa band is DEPKXEVP. The sequence of the 68 kDa Sa band is not compiled in the databases we searched, as either aminoterminal or internal sequence. Antibodies to 50/46 kDa anti-Sa bands detected by immunoblotting were highly specific for RA, while the 68 kDa antigen reacted in ELISA with sera from patients with RA and systemic lupus erythematosus, the latter showing a marked increase in features of RA. Antibodies directed against the 68 and 50/46 kDa Sa bands fluctuated with time, the 50/46 kDa anti-Sa antibodies present during the active period of the disease, and the 68 kDa anti-Sa antibodies during the remission period. CONCLUSION: At least 2 subsets of autoantibodies are present in anti-Sa sera, one directed against a 68 kDa Sa protein and another to the typical 50/46 bands of the Sa system.
In this issue
