Objective. To find evidence of a potential genetic predisposition to the anti-Sm or anti-RNP precipitin autoantibody responses.Methods. HLA-DR and DQ alleles determined by restriction fragment length polymorphism andor oligotyping in 49 subjects with either anti-Sm alone or anti-RNP alone were compared with those in 139 racematched normal control subjects and 59 race-matched lupus patients without anti-Sm and anti-RNP autoantibodies.Results. Black patients with anti-Sm precipitin had increased frequencies of HLA-DR2 and the DQw6-associated DQAl*0102 (P = 0.007, odds ratio [OR] = 6.7) and DQB1*0602 (P = 0.001, OR = 9.1) chain alleles compared with normal black control subjects. Black patients with anti-RNP precipitin showed significant increases in the DQwS-associated DQA1*0101 (P = 0.03, OR = 5.5) and DQB1*0501 (P = 0.002, OR = 23.3) chain alleles compared with lupus patients without anti8m or RNP. White patients with anti-RNP precipitin showed an increased frequency of the DQw8-associated allele DQB1*0302 (P = 0.02, OR = 3.7) compared with normal controls, as well as an increased frequency of the DQwS-associated alleles DQA1*0101 and DQB1*0501 (P = 0.05, OR = 4.2) compared with lupus patients without anti-Sm or RNP. There were no specific HLA-DR2 or DR4 subtype associations found with either anti-Sm or RNP precipitin autoantibodies.Conclusion. There are distinct patterns of major histocompatibility complex class I1 allele associations with the anti-Sm versus the anti-RNP precipitin autoantibody responses, and HLA-DQ associations may be more primary than HLA-DR associations.Autoantibodies to the uridine-rich small nuclear ribonucleoproteins (U snRNP), transcriptional enzymes for splicing messenger RNA, arise spontaneously in patients with the autoimmune disease systemic lupus erythematosus (SLE). In fact, anti-Sm (Smith) autoantibodies, the autoepitopes of which reside on U l , U2, U4, U5, and U6 snRNP, are highly specific for SLE. Anti-RNP, which recognizes different antigens on U1 RNP, is often co-expressed with anti-Sm in lupus patients or may occur alone in SLE or in other systemic diseases, including mixed connective tissue disease, scleroderma, and polymyositis (1). Both anti-Sm and anti-RNP appear to occur more commonly in blacks than whites with SLE (2).SLE is a clinically and serologically heterogeneous disease in which genetic factors appear to play an important role. Earlier studies have demonstrated weak associations with the HLA-DR2 and DR3 alleles in Caucasians (3,4). Stronger HLA class I1 correlations have been found for certain SLE autoantibody subsets than for the disease itself (5-7). Such findings suggest that the immune responses to these autoanti-