Cell-surface MHC density profiling reveals instability of autoimmunity-associated HLA

HLA-DRB1*15:01-DQA1*01:02-DQB1*06:02 Haplotype Protects Autoantibody-Positive Relatives From Type 1 Diabetes Throughout the Stages of Disease Progression

The HLA-DRB1*15:01-DQA1*01:02-DQB1*06:02 haplotype is linked to protection from the development of type 1 diabetes (T1D). However, it is not known at which stages in the natural history of T1D development this haplotype affords protection. We examined a cohort of 3,358 autoantibody-positive relatives of T1D patients in the Pathway to Prevention (PTP) Study of the Type 1 Diabetes TrialNet. The PTP study examines risk factors for T1D and disease progression in relatives. HLA typing revealed that 155 relatives carried this protective haplotype. A comparison with 60 autoantibody-negative relatives suggested protection from autoantibody development. Moreover, the relatives with DRB1*15:01-DQA1*01:02-DQB1*06:02 less frequently expressed autoantibodies associated with higher T1D risk, were less likely to have multiple autoantibodies at baseline, and rarely converted from single to multiple autoantibody positivity on follow-up. These relatives also had lower frequencies of metabolic abnormalities at baseline and exhibited no overall metabolic worsening on follow-up. Ultimately, they had a very low 5-year cumulative incidence of T1D. In conclusion, the protective influence of DRB1*15:01-DQA1*01:02-DQB1*06:02 spans from autoantibody development through all stages of progression, and relatives with this allele only rarely develop T1D.

HLA class II sequence variants influence tuberculosis risk in populations of European ancestry

Mycobacterium tuberculosis infections cause 9 million new tuberculosis cases and 1.5 million deaths annually. To identify variants conferring risk of tuberculosis, we tested 28.3 million variants identified through whole-genome sequencing of 2,636 Icelanders for association with tuberculosis (8,162 cases and 277,643 controls), pulmonary tuberculosis (PTB) and M. tuberculosis infection. We found association of three variants in the region harboring genes encoding the class II human leukocyte antigens (HLAs): rs557011[T] (minor allele frequency (MAF) = 40.2%), associated with M. tuberculosis infection (odds ratio (OR) = 1.14, P = 3.1 × 10(-13)) and PTB (OR = 1.25, P = 5.8 × 10(-12)), and rs9271378[G] (MAF = 32.5%), associated with PTB (OR = 0.78, P = 2.5 × 10(-12))--both located between HLA-DQA1 and HLA-DRB1--and a missense variant encoding p.Ala210Thr in HLA-DQA1 (MAF = 19.1%, rs9272785), associated with M. tuberculosis infection (P = 9.3 × 10(-9), OR = 1.14). We replicated association of these variants with PTB in samples of European ancestry from Russia and Croatia (P < 5.9 × 10(-4)). These findings show that the HLA class II region contributes to genetic risk of tuberculosis, possibly through reduced presentation of protective M. tuberculosis antigens to T cells.

Questionable expression of unstable DQ heterodimer containing HLA-DQA1*01:07

Human leukocyte antigens (HLA)-DQA1*01:07 was identified as an HLA-DQ blank specificity that segregated with the serological HLA-A2, -B7, -DR14, -DR52 haplotype, which carried DQB1*05:03. The blank specificity of DQA1*01:07-DQB1*05:03 may be because of lack of reactivity of available typing sera, or disruption of proper assembly of DQ heterodimer. The cDNA sequence of DQA1*01:07 is nearly identical to DQA1*01:04 except for a variant at position 304, which results in the replacement of an arginine with a cysteine at 79α. To determine whether the DQA1*01:07 product can be expressed on cell-surface, we co-expressed DQA1*01:07 with various DQB1*05 or *06 alleles in fibroblast cells. Cell-surface expression of DQ was detectable when DQA1*01:07 was co-expressed with DQB1*06:04 but undetectable with other DQB1*05 and DQB1*06 alleles, including DQB1*05:03, to which DQA1*01:07 was encoded in cis. These data suggest that DQA1*01:07 may act as a phenotypically null allele in the DQA1*01:07-DQB1*05:03 haplotype, while it can be expressed at a low level in the presences of certain DQB1*06 alleles, such as DQB1*06:04, in trans. Based on the null or low expression of DQA1*01:07 as shown in the previous and present studies, DQA1*01:07 has recently been renamed to DQA1*01:07Q, indicating its questionable expression.

Associations of human leukocyte antigens with autoimmune diseases: challenges in identifying the mechanism

The mechanism of genetic associations between human leukocyte antigen (HLA) and susceptibility to autoimmune disorders has remained elusive for most of the diseases, including rheumatoid arthritis (RA) and type 1 diabetes (T1D), for which both the genetic associations and pathogenic mechanisms have been extensively analyzed. In this review, we summarize what are currently known about the mechanisms of HLA associations with RA and T1D, and elucidate the potential mechanistic basis of the HLA-autoimmunity associations. In RA, the established association between the shared epitope (SE) and RA risk has been explained, at least in part, by the involvement of SE in the presentation of citrullinated peptides, as confirmed by the structural analysis of DR4-citrullinated peptide complex. Self-peptide(s) that might explain the predispositions of variants at 11β and 13β in DRB1 to RA risk have not currently been identified. Regarding the mechanism of T1D, pancreatic self-peptides that are presented weakly on the susceptible HLA allele products are recognized by self-reactive T cells. Other studies have revealed that DQ proteins encoded by the T1D susceptible DQ haplotypes are intrinsically unstable. These findings indicate that the T1D susceptible DQ haplotypes might confer risk for T1D by facilitating the formation of unstable HLA-self-peptide complex. The studies of RA and T1D reveal the two distinct mechanistic basis that might operate in the HLA-autoimmunity associations. Combination of these mechanisms, together with other functional variations among the DR and DQ alleles, may generate the complex patterns of DR-DQ haplotype associations with autoimmunity.

Widespread non-additive and interaction effects within HLA loci modulate the risk of autoimmune diseases

Human leukocyte antigen (HLA) genes confer strong risk for autoimmune diseases on a log-additive scale. Here we speculated that differences in autoantigen binding repertoires between a heterozygote’s two expressed HLA variants may result in additional non-additive risk effects. We tested non-additive disease contributions of classical HLA alleles in patients and matched controls for five common autoimmune diseases: rheumatoid arthritis (RA, N_cases=5,337), type 1 diabetes (T1D, N_cases=5,567), psoriasis vulgaris (N_cases=3,089), idiopathic achalasia (N_cases=727), and celiac disease (N_cases=11,115). In four out of five diseases, we observed highly significant non-additive dominance effects (RA: P=2.5×10¹²; T1D: P=2.4×10⁻¹⁰; psoriasis: P=5.9×10⁻⁶; celiac disease: P=1.2×10⁻⁸⁷). In three of these diseases, the dominance effects were explained by interactions between specific classical HLA alleles (RA: P=1.8×10⁻³; T1D: P=8.6×10²⁷; celiac disease: P=6.0×10⁻¹⁰⁰). These interactions generally increased disease risk and explained moderate but significant fractions of phenotypic variance (RA: 1.4%, T1D: 4.0%, and celiac disease: 4.1%, beyond a simple additive model).

Additive and interaction effects at three amino acid positions in HLA-DQ and HLA-DR molecules drive type 1 diabetes risk

Variation in the human leukocyte antigen (HLA) genes accounts for one-half of the genetic risk in type 1 diabetes (T1D). Amino acid changes in the HLA-DR and HLA-DQ molecules mediate most of the risk, but extensive linkage disequilibrium complicates the localization of independent effects. Using 18,832 case-control samples, we localized the signal to 3 amino acid positions in HLA-DQ and HLA-DR. HLA-DQβ1 position 57 (previously known; P = 1 × 10(-1,355)) by itself explained 15.2% of the total phenotypic variance. Independent effects at HLA-DRβ1 positions 13 (P = 1 × 10(-721)) and 71 (P = 1 × 10(-95)) increased the proportion of variance explained to 26.9%. The three positions together explained 90% of the phenotypic variance in the HLA-DRB1-HLA-DQA1-HLA-DQB1 locus. Additionally, we observed significant interactions for 11 of 21 pairs of common HLA-DRB1-HLA-DQA1-HLA-DQB1 haplotypes (P = 1.6 × 10(-64)). HLA-DRβ1 positions 13 and 71 implicate the P4 pocket in the antigen-binding groove, thus pointing to another critical protein structure for T1D risk, in addition to the HLA-DQ P9 pocket.

Construction of a population-specific HLA imputation reference panel and its application to Graves' disease risk in Japanese

To fine map association signals of human leukocyte antigen (HLA) variants in the major histocompatibility complex (MHC) region, we constructed a Japanese population-specific reference panel (n = 908). We conducted trans-ancestry comparisons of linkage disequilibrium (LD) and haplotype structure for HLA variants using an entropy-based LD measurement, ɛ, and a visualization tool to capture high-dimensional variables. Our Japanese reference panel exhibited stronger LD between HLA genes than European or other East Asian populations, characterized by one population-specific common long-range HLA haplotype. We applied HLA imputation to genome-wide association study (GWAS) data for Graves' disease in Japanese (n = 9,003) and found that amino acid polymorphisms of multiple class I and class II HLA genes independently contribute to disease risk (HLA-DPB1, HLA-A, HLA-B and HLA-DRB1; P < 2.3 × 10(-6)), with the strongest impact at HLA-DPB1 (P = 1.6 × 10(-42)). Our study illustrates the value of population-specific HLA reference panels.