Autoantibodies to RNA polymerases (RNAP) I and III are highly specific for scleroderma (SSc), whereas autoantibodies to RNAP II are associated with systemic lupus erythematosus (SLE) and overlap syndromes, as well as SSc. The specificities of autoantibodies to RNAP I, II, and III in 129 SSc sera were investigated in the present study. Immunoprecipitation and pulse-chase analysis demonstrated several patterns of autoantibody recognition of RNAPs. Some sera immunoprecipitated RNAP II only after its largest subunit was phosphorylated, suggesting that they contained autoantibodies that recognized an epitope carrying a phosphoamino acid. Autoantibody recognition of all three classes of RNAPs was influenced strongly by race. Although in SLE, autoantibodies to the phosphorylated form of RNAP II (RNAP IIO) were identified in all races, in SSc, these autoantibodies were seen in 21% of Japanese and 5% of Black patients, but never in Caucasians. A striking association of anti-RNAP IIO with anti-topoisomerase I (topo I) autoantibodies was found in Japanese and Black SSc, but not SLE, patients. However, anti-topo I Abs were not associated with anti-RNAP IIO in Caucasians. Japanese SSc patients who were positive for both anti-RNAP IIO and anti-topo I Abs had a significantly higher frequency of diffuse disease, pigmentation changes, flexion contractures, and acro-osteolysis than patients having autoantibodies to topo I alone, and were diagnosed at a younger age (p < 0.05). These data suggest that genetic factors (possibly HLA-linked) influence autoantibody specificity, and that different autoantibody fine specificities may either cause, or be predictive of, different clinical outcomes.