Abstract
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.
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Rent or buy this article
Prices vary by article type
from$1.95
to$39.95
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Maahs, D.M., West, N.A., Lawrence, J.M. & Mayer-Davis, E.J. Epidemiology of type 1 diabetes. Endocrinol. Metab. Clin. North Am. 39, 481–497 (2010).
Raychaudhuri, S. et al. Five amino acids in three HLA proteins explain most of the association between MHC and seropositive rheumatoid arthritis. Nat. Genet. 44, 291–296 (2012).
Trynka, G., Wijmenga, C. & van Heel, D.A. A genetic perspective on coeliac disease. Trends Mol. Med. 16, 537–550 (2010).
Gourraud, P.A., Harbo, H.F., Hauser, S.L. & Baranzini, S.E. The genetics of multiple sclerosis: an up-to-date review. Immunol. Rev. 248, 87–103 (2012).
Lee, K.H., Wucherpfennig, K.W. & Wiley, D.C. Structure of a human insulin peptide–HLA-DQ8 complex and susceptibility to type 1 diabetes. Nat. Immunol. 2, 501–507 (2001).
Astill, T.P., Ellis, R.J., Arif, S., Tree, T.I. & Peakman, M. Promiscuous binding of proinsulin peptides to Type 1 diabetes–permissive and –protective HLA class II molecules. Diabetologia 46, 496–503 (2003).
Scally, S.W. et al. A molecular basis for the association of the HLA-DRB1 locus, citrullination, and rheumatoid arthritis. J. Exp. Med. 210, 2569–2582 (2013).
van Lummel, M. et al. Posttranslational modification of HLA-DQ binding islet autoantigens in type 1 diabetes. Diabetes 63, 237–247 (2014).
Erlich, H. et al. HLA DR-DQ haplotypes and genotypes and type 1 diabetes risk: analysis of the Type 1 Diabetes Genetics Consortium families. Diabetes 57, 1084–1092 (2008).
Noble, J.A. et al. The role of HLA class II genes in insulin-dependent diabetes mellitus: molecular analysis of 180 Caucasian, multiplex families. Am. J. Hum. Genet. 59, 1134–1148 (1996).
Nejentsev, S. et al. Localization of type 1 diabetes susceptibility to the MHC class I genes HLA-B and HLA-A. Nature 450, 887–892 (2007).
Cucca, F. et al. The HLA-DPB1–associated component of the IDDM1 and its relationship to the major loci HLA-DQB1, -DQA1, and -DRB1. Diabetes 50, 1200–1205 (2001).
Deschamps, I. et al. HLA genotype studies in juvenile insulin-dependent diabetes. Diabetologia 19, 189–193 (1980).
Howson, J.M., Walker, N.M., Clayton, D. & Todd, J.A. & Type 1 Diabetes Genetics Consortium. Confirmation of HLA class II independent type 1 diabetes associations in the major histocompatibility complex including HLA-B and HLA-A. Diabetes Obes. Metab. 11 (suppl. 1), 31–45 (2009).
Todd, J.A., Bell, J.I. & McDevitt, H.O. HLA-DQβ gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 329, 599–604 (1987).
Cucca, F. et al. A correlation between the relative predisposition of MHC class II alleles to type 1 diabetes and the structure of their proteins. Hum. Mol. Genet. 10, 2025–2037 (2001).
Thomson, G. et al. Genetic heterogeneity, modes of inheritance, and risk estimates for a joint study of Caucasians with insulin-dependent diabetes mellitus. Am. J. Hum. Genet. 43, 799–816 (1988).
Svejgaard, A. & Ryder, L.P. HLA genotype distribution and genetic models of insulin-dependent diabetes mellitus. Ann. Hum. Genet. 45, 293–298 (1981).
Koeleman, B.P. et al. Genotype effects and epistasis in type 1 diabetes and HLA-DQ trans dimer associations with disease. Genes Immun. 5, 381–388 (2004).
Onengut-Gumuscu, S. et al. Fine mapping of type 1 diabetes susceptibility loci and evidence for colocalization of causal variants with lymphoid gene enhancers. Nat. Genet. 47, 381–386 (2015).
Jia, X. et al. Imputing amino acid polymorphisms in human leukocyte antigens. PLoS ONE 8, e64683 (2013).
Brown, W.M. et al. Overview of the MHC fine mapping data. Diabetes Obes. Metab. 11 (suppl. 1), 2–7 (2009).
Noble, J.A. & Erlich, H.A. Genetics of type 1 diabetes. Cold Spring Harb. Perspect. Med. 2, a007732 (2012).
Han, B. et al. Fine mapping seronegative and seropositive rheumatoid arthritis to shared and distinct HLA alleles by adjusting for the effects of heterogeneity. Am. J. Hum. Genet. 94, 522–532 (2014).
Foo, J.N. et al. Coding variants at hexa-allelic amino acid 13 of HLA-DRB1 explain independent SNP associations with follicular lymphoma risk. Am. J. Hum. Genet. 93, 167–172 (2013).
Witte, J.S., Visscher, P.M. & Wray, N.R. The contribution of genetic variants to disease depends on the ruler. Nat. Rev. Genet. 15, 765–776 (2014).
Sivertsen, B., Petrie, K.J., Wilhelmsen-Langeland, A. & Hysing, M. Mental health in adolescents with Type 1 diabetes: results from a large population-based study. BMC Endocr. Disord. 14, 83 (2014).
Speed, D., Hemani, G., Johnson, M.R. & Balding, D.J. Improved heritability estimation from genome-wide SNPs. Am. J. Hum. Genet. 91, 1011–1021 (2012).
Reichstetter, S., Kwok, W.W. & Nepom, G.T. Impaired binding of a DQ2 and DQ8-binding HSV VP16 peptide to a DQA1*0501/DQB1*0302 trans class II heterodimer. Tissue Antigens 53, 101–105 (1999).
Ettinger, R.A., Liu, A.W., Nepom, G.T. & Kwok, W.W. Exceptional stability of the HLA-DQA1*0102/DQB1*0602 αβ protein dimer, the class II MHC molecule associated with protection from insulin-dependent diabetes mellitus. J. Immunol. 161, 6439–6445 (1998).
Miyadera, H., Ohashi, J., Lernmark, A., Kitamura, T. & Tokunaga, K. Cell-surface MHC density profiling reveals instability of autoimmunity-associated HLA. J. Clin. Invest. 125, 275–291 (2015).
Knight, R.R. et al. A distinct immunogenic region of glutamic acid decarboxylase 65 is naturally processed and presented by human islet cells to cytotoxic CD8 T cells. Clin. Exp. Immunol. 179, 100–107 (2015).
Kronenberg, D. et al. Circulating preproinsulin signal peptide–specific CD8 T cells restricted by the susceptibility molecule HLA-A24 are expanded at onset of type 1 diabetes and kill β-cells. Diabetes 61, 1752–1759 (2012).
Nakayama, M. et al. Prime role for an insulin epitope in the development of type 1 diabetes in NOD mice. Nature 435, 220–223 (2005).
Baekkeskov, S. et al. Identification of the 64K autoantigen in insulin-dependent diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 347, 151–156 (1990).
Karlsen, A.E. et al. Recombinant glutamic acid decarboxylase (representing the single isoform expressed in human islets) detects IDDM-associated 64,000-Mr autoantibodies. Diabetes 41, 1355–1359 (1992).
Roep, B.O. & Peakman, M. Antigen targets of type 1 diabetes autoimmunity. Cold Spring Harb. Perspect. Med. 2, a007781 (2012).
Acknowledgements
This research makes use of resources provided by the Type 1 Diabetes Genetics Consortium (T1DGC), a collaborative clinical study sponsored by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), the National Institute of Allergy and Infectious Diseases (NIAID), the National Human Genome Research Institute (NHGRI), the National Institute of Child Health and Human Development (NICHHD) and Juvenile Diabetes Research Foundation International (JDRFI) and supported by grant U01DK062418 (NIDDK). This work is supported in part by funding from the US National Institutes of Health (5R01AR062886-02 (P.I.W.d.B.), 1R01AR063759 (S.R.), 5U01GM092691-05 (S.R.), 1UH2AR067677-01 (S.R.) and R01AR065183 (P.I.W.d.B.)), a Doris Duke Clinical Scientist Development Award (S.R.), the Wellcome Trust (J.A.T.) and the UK National Institute for Health Research (NIHR; J.A.T. and J.M.M.H.) and by a Vernieuwingsimpuls VIDI Award (016.126.354) from the Netherlands Organization for Scientific Research (P.I.W.d.B.). T.L.L. was supported by the German Research Foundation (LE 2593/1-1 and LE 2593/2-1).
Author information
Authors and Affiliations
Contributions
X.H. and S.R. conceived the study. X.H., A.J.D., T.L.L., S.R., B.H., P.I.W.d.B. and S.S.R. contributed to the study design and analysis strategy. X.H., A.J.D., T.L.L. and S.R. conducted all analyses. X.H. and A.J.D. wrote the initial manuscript. B.H. contributed critical analytical methods. S.O.-G., W.-M.C. and S.S.R. organized and contributed subject samples and provided SNP genotype data. J.M.M.H., J.A.T., P.I.W.d.B., S.S.R. and S.R. contributed critical writing and review of the manuscript. All authors contributed to the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 2 Association quantile-quantile plots indicate that sex and regional code covariates adequately control for population stratification.
Using the UK data set, we tested the association of 807 reading- and/or writing-related SNPs with T1D status, expecting a null distribution. We assessed the performance of five genomic correction schemes by calculating the λ factor (median χ2 statistic/0.456). “Region” indicates the use of the 13 regional codes provided by T1DGC; “PC” indicates the use of the top ten genotype principal components calculated by EIGENSTRAT. In our final analyses, we included sex and the region codes as covariates.
Supplementary Figure 3 Amino acid position analysis including non-additive terms yielded the same key positions.
We repeated the forward-search analysis after incorporating non-additive terms into the regression model. In this analysis, each variant is coded as 0, 1 or 2 for allelic dosage. An additional heterozygote factor is added, which equals 1 only if the individual is heterozygotic for this allele or haplotype. HLA-DQβ1 position 57 (P = 10–1,240), HLA-DRβ1 position 13 (P = 10–653) and HLA-DRβ1 position 71 (P = 10–55) remained the top independent amino acid associations with either the additive or non-additive model.
Supplementary Figure 4 Haplotype–amino acid sequence permutation analysis ensures that HLA-DQβ1 position 57, HLA-DRβ1 position 13 and HLA-DRβ1 position 71 are the independent risk drivers.
We performed 10,000 rounds of permutated association analysis; during each permutation, the amino acid sequence corresponding to each HLA-DRB1, HLA-DQA1 and HLA-DQB1 classical allele was reassigned before association analysis. (a) Histogram of 10,000 deviance values (improvement upon the null model) while testing for the best combinations of one, two and three amino acid positions. In 3% of the 10,000 trials, the single best position exceeded the deviance achieved by HLA-DQβ1 position 57. No combination of two or three amino acid positions outperformed the fit of HLA-DQβ1 position 57 + HLA-DRβ1 position 13 and HLA-DQβ1 position 57 + HLA-DRβ1 position 13 + HLA-DRβ1 position 71, respectively. The best model achieved by the combination of any three amino acid positions obtained a Δdeviance of 8,244.29 (P = 10–1,774, df = 41); in comparison, the model without permutation including HLA-DQβ1 position 57, HLA-DRβ1 position 13 and HLA-DRβ1 position 71 obtained a Δdeviance of 10,148.53 (P = 10–2,161, df = 31). Red arrows indicate the deviance achieved by the best combination in actual data. (b) Histogram of 10,000 P values while testing for the best combinations of one, two and three amino acid positions. Similarly, 3% of the permuted amino acid positions achieved better P values than HLA-DQβ1 position 57 in actual data. No combination of two or three amino acid positions outperformed the combinations of HLA-DQβ1 position 57 + HLA-DRβ1 position 13 and HLA-DQβ1 position 57 + HLA-DRβ1 position 13 + HLA-DRβ1 position 71, respectively. Red arrows indicate the P value achieved by the best combination in actual data.
Supplementary Figure 5 Concordance between the UK and European data sets.
We repeated the association analyses in the UK case-control set and the European pseudocase-pseudocontrol set separately. (a,b) We confirmed that the two sets yielded highly correlated effect sizes for all binary variants (Pearson r = 0.952, P < 2.2 × 10–16) (a) and for haplotypes formed by HLA-DQβ1 position 57, HLA-DRβ1 position 13 and HLA-DRβ1 position 71 (Pearson r = 0.989, P = 1.02 × 10–14) (b). Error bars represent the 95% confidence interval.
Supplementary Figure 6 HLA-DRβ1 position 13 and HLA-DRβ1 position 71 show discordant effect sizes in RA and T1D.
HLA-DRβ1 position 13 and HLA-DRβ1 position 71, which line the P4 pocket of the HLA-DR peptide-binding groove, are indicated in both rheumatoid arthritis (RA) and T1D. However, the individual amino acid residues at each position confer differential risk or protection for each disease (P < 10–230). Each cross shows an individual residue’s (adjusted) univariate OR (with 95% confidence interval) in RA and T1D. Darker areas indicate the same direction of effect (risk or protection) across the two diseases, whereas the lighter gray areas indicate opposite effects. The slanted dashed line indicates the identity line on which a residue’s effect sizes in both diseases would be equal. At HLA-DRβ1 position 13, serine, tyrosine and arginine confer relative protection for each disease; however, they are located far away from the identity line. Glycine is protective for RA, although it confers strong risk for T1D.
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–6, Supplementary Tables 1 and 3–10, and Supplementary Note. (PDF 1841 kb)
Supplementary Table 2: Association statistics of 8,961 binary markers.
This table provides the genomic annotation, minor allele frequency, imputation quality and association statistics for all 8,961 binary markers in the study. (XLSX 2310 kb)
Supplementary Table 11: Pairwise amino acid interactions.
This table provides association statistics for all pairwise interactions between all amino acid positions in HLA-DRβ1, HLA-DQα1 and HLA-DQβ1. (XLSX 177 kb)
Rights and permissions
About this article
Cite this article
Hu, X., Deutsch, A., Lenz, T. et al. Additive and interaction effects at three amino acid positions in HLA-DQ and HLA-DR molecules drive type 1 diabetes risk. Nat Genet 47, 898–905 (2015). https://doi.org/10.1038/ng.3353
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/ng.3353
This article is cited by
-
Non-mutational neoantigens in disease
Nature Immunology (2024)
-
Blood donor biobank as a resource in personalised biomedical genetic research
European Journal of Human Genetics (2024)
-
Genomic data in the All of Us Research Program
Nature (2024)
-
CRISPR-based targeted haplotype-resolved assembly of a megabase region
Nature Communications (2023)
-
Machine learning reveals limited contribution of trans-only encoded variants to the HLA-DQ immunopeptidome
Communications Biology (2023)