Geographic patterns of functional categories of HLA-DRB1 alleles: a new approach to analyse associations between HLA-DRB1 and disease

Because specific amino acids found within the peptide-binding cleft of human leukocyte antigen (HLA) molecules have been implicated in HLA/disease associations, an approach which consists in grouping the alleles according to their functional properties at the protein level may enable us to better understand HLA associations than the conventional allelic classification. In this study, we applied this methodology to investigate the associations between HLA-DRB1 and rheumatoid arthritis. The alleles were first classified into seven functional categories [restrictive supertype patterns (RSPs)], among which three were known to be significantly associated with susceptibility (one category) or resistance (two categories) to rheumatoid arthritis. The frequencies of these categories were then estimated in 104 population samples previously tested for HLA-DRB1, and their variability was analysed spatially on a worldwide scale by applying an original methodology for detecting discontinuities in geographically patterned data. RSP frequencies were also compared to known values of rheumatoid arthritis prevalence in some populations. The results indicated that the three RSP frequency distributions were geographically structured, and that these patterns could generally be explained by the history of human migrations. However, the peculiar pattern observed for RSP 'A' (conferring susceptibility to rheumatoid arthritis) indicated a possible association with some latitude-dependent disease. Furthermore, the very high correlation coefficient found between RSP 'A' frequencies and rheumatoid arthritis prevalence confirmed the significant disease association of this functional category In contrast, the putative protective effect of the other RSPs ('De' and 'Q') was not detectable at the worldwide level, but may be significant in specific geographic areas. This study shows that population genetic diversity analyses based on a functional grouping of HLA alleles provide an efficient way to explore the mutual influence of HLA genetic variation and disease.