Objective. To evaluate the specificity of anti-DEK antibodies for juvenile
rheumatoid arthritis (JRA),
Methods. Anti-DEK autoantibodies were measured by enzyme-linked immunosorbe
nt assay (ELISA) using affinity-purified his6-DEK fusion protein. Sera from
639 subjects (417 patients with systemic autoimmune disease, 13 with sarco
idosis, 44 with pulmonary tuberculosis, 125 with uveitis, and 6 with scleri
tis, and 34 healthy control subjects) were screened. Reactivity was verifie
d by immunoblotting and immunoprecipitation Studies using baculovirus-expre
ssed human DEK.
Results. Anti-DEK activity was found at the following frequencies: JRA 39.4
% (n = 71), systemic lupus erythematosus (SLE) 25.1% (n = 216), sarcoidosis
46.2% (n = 13), rheumatoid arthritis 15.5% (n = 71), systemic sclerosis 36
.0% (n = 22), polymyositis 6.2% (n = 16), and adult Still's disease 0% (n =
21), Autoantibodies also were detected in 9.1% of tuberculosis sera (n = 4
4), but were undetectable in sera from the 34 healthy controls. Western blo
t and immunoprecipitation assay results correlated well with the ELISA find
ings. In general, levels of anti-DEK autoantibodies were higher in SLE than
in other patient subsets, including JRA.
Conclusion. Anti-DEK autoantibodies are less specific for JRA than previous
ly believed. They are produced in association with a variety of inflammator
y conditions, many of which are associated with granuloma formation and/or
predominant Th1 cytokine production. Anti-DEK antibodies may be a marker fo
r a subset of autoimmunity associated with interferon-gamma production rath
er than a particular disease subset.