Background and aims A genetic contribution of the human leukocyte antigen (HLA) genes to SLE has been well documented (e.g., OR=1.65, Heritability explained by HLA=2%; Nat Genet 2016). To understand the association in HLA loci within the major histocompatibility complex (MHC) region in large cohorts genotyped for MHC SNPs, we developed an ethnicity-matched HLA reference panel (PLoS One 2014).

Methods Using the HLA imputation and various statistical approaches, we investigated HLA amino acid residues, HLA classical alleles and MHC SNPs simultaneously, and identified that the changes in amino acid positions 11, 13 and 26 of HLA-DRB1 explained the entire HLA association (Nat Commun 2014; Nat Genet 2016). Additionally, all the protein-coding HLA-DR beta genes (HLA-DRB1, HLA-DRB3, HLA-DRB4 and HLA-DRB5) with similar functions were further investigated using imputation-based conditional regression and haplotype analyses. HLA-DRB1 was solely associated with SLE and accounted for the associations of the other HLA-DR beta genes (PLoS One 2016). Finally, we measured allele-specific expression of HLA-DRB1 in blood cells by RNA sequencing followed by an allele-specific read mapping method.

Results Strong allele-specific expression among HLA-DRB1 classical alleles was observed, which caused relatively unequal expression of two heterozygous alleles in individuals. Disease association models, fitted by logistic regression including either the copy number or both the copy number and the relative expression of each allele as predictors, revealed that the SLE association was significantly better explained by adding the variables for the relative expression.

Conclusions These findings indicate that both the qualitative and quantitative effects of HLA-DRB1 variants are driving SLE (Figure).