Objective: To evaluate the contributions of HLA-DQ and -DR polymorphisms to susceptibility to rheumatoid arthritis (RA) in a population in southern Spain, and to compare the value of the shared epitope (SE) and RA protection (RAP) models in accounting for the HLA class II region's contribution to RA predisposition.
Methods: One hundred sixty RA patients and 153 healthy controls were typed for HLA-DRB1 and -DQB1 using high-resolution DNA techniques. Distributions of predisposing DRB1 alleles in patients and control subjects according to the SE model were compared with distributions of predisposing DQ and protective DERAA-positive DRBI alleles according to the RAP model.
Results: DQ3 (DQBI*03 and *04 combined with DQA1*03) and DQ5 (DQB1*0501/DQA1*0101) alleles predisposed individuals to RA independently of SE-positive DRB1 alleles. DQ3/3-homozygous individuals had the strongest risk of developing RA. DQ3 molecules predisposed to RA more than did DQ5 molecules. The weaker predisposition mediated by DQ5 included the DRB1*1001-carrying haplotype; no DRB1*1001-homozygous patients were observed. DRBI*0401 played a unique role in the contribution of DQ3-DR4 haplotypes to RA, in spite of its low frequency in southern Spain.
Conclusion: The low prevalences of RA and of mild disease observed in Spain, and in southern Europe in general, can be explained in great part by the low frequency of DQ3-DR4 haplotypes, especially those carrying DRB1*0401. However, the overall distribution of HLA-DQ and -DR alleles in RA patients compared with control subjects is similar to that in other European and North American populations. A model involving both DQ and DR can best account for the contribution of HLA to RA.