Abstract
A candidate gene study was conducted on 10 established type II diabetes genes and 45 genes associated with autoimmune diseases, including type I diabetes (T1D), in a maximum of 1410 affected sib-pair families assembled by the Type I Diabetes Genetics Consortium. Associations at P values < 10−3 were found for three known T1D regions at chromosomes 4q27, 12q13.2 and 12q24.13 (http://www.T1DBase.org). Support was obtained for a newly identified T1D candidate locus on chromosome 12q13.3–12q14.1 (rs1678536/KIF5A: P=8.1 × 10−3; relative risk (RR) for minor allele=0.89, 95% CI=0.82–0.97), which has a separate association from the previously reported T1D candidate locus ERBB3/12q13.2–q13.3. Our new evidence adds to that previously published for the same gene region in a T1D case–control study (rs1678542; P=3.0 × 10−4; odds ratio (OR)=0.92, 95% CI=0.88–0.96). This region, which contains many genes, has also been associated with rheumatoid arthritis.
Keywords: autoimmune disease, type I diabetes, type II diabetes, SNPs, T1DGC
Introduction
The advent of genome-wide association studies has highlighted the commonality and differences between diseases. Genes found to be associated in one disease are now routinely genotyped in related diseases, providing a more focused and often more informative candidate gene approach.1,2 A candidate gene study was performed using a panel of 216 SNPs from 10 known T2D loci (including HHEX and SLC30A8), 126 SNPs from 5 known T1D regions (including CTLA4 and IL2) and 239 SNPs from 40 known other autoimmune disease regions (including IL23R and PADI4). Although T1D and T2D are considered etiologically distinct, characterized by autoimmune destruction of the pancreatic β cells and by impaired β cell function, respectively, T2D loci were included in the study as they may share a common pathophysiological etiology, suggested by the similarities in their clinical manifestation.3 The five known T1D loci were included for further confirmation. All SNPs were genotyped in a maximum of 1410 T1D affected sib-pair families assembled by the Type I Diabetes Genetics Consortium (T1DGC), providing 2198 parent–child trios. A complete list of SNPs, genes and regions is provided in Supplementary Table 1.
Results
A total of 581 SNPs was selected from T2D genes related to β-cell function, T1D and other autoimmune candidate genes and genotyped at the Broad Institute of Harvard/MIT in Cambridge, MA, USA (http://www.broad.mit.edu/) using the Illumina Golden Gate platform. In addition to the standard sample and SNP quality control, we visually inspected the SNP signal intensity plots for each of the family groups (see accompanying paper4), resulting in 334 SNPs that passed quality control (Supplementary Table 1).8
Fifty-one SNPs exhibited some evidence (P < 0.05) of association with T1D (Table 1),8 16 of which were from known T1D regions. Only four SNPs reached the required level of significance for this study (P<0.001) and these SNPs were all from known T1D regions: rs11171747 on chromosome 12q13.2 (P=7.1 × 10−6), located 21 kb telomeric of ERBB3;5 rs4767364 (P=1.6 × 10−4) and rs17696736 (P=6.2 × 10−4) on 12q24.13 in the C12orf30 region, located 597 kb telomeric of SH2B3;5 and rs12510683 (P=7.1 × 10−4) on 4q27 in the KIAA1109 region, located 393 kb centromeric of IL2-IL21.
Table 1.
SNPs from known T1D, T2D and autoimmune disease regions with evidence of T1D association, P<0.05, ordered by genomic position
| rs number | Chromosome | Gene region | Parent–child trios | MAF | 1-df P | T1D, T2D or autoimmune |
|---|---|---|---|---|---|---|
| rs6745050 | 2q33 | 2764 | 0.392 | 0.023 | T1D | |
| rs12990970 | 2q33 | 2750 | 0.394 | 0.045 | T1D | |
| rs231727 | 2q33 | CTLA4 | 2763 | 0.359 | 0.00122 | T1D |
| rs10197319 | 2q33 | 2644 | 0.469 | 0.0378 | T1D | |
| rs17268364 | 2q33 | 2756 | 0.459 | 0.0268 | T1D | |
| rs4294983 | 2q33 | 2500 | 0.072 | 0.0278 | T1D | |
| rs12640179 | 4q13 | GC | 886 | 0.078 | 0.0167 | Autoimmune |
| rs16847054 | 4q13 | GC | 1657 | 0.062 | 0.0129 | Autoimmune |
| rs12510683 | 4q27 | 2774 | 0.225 | 0.000713 | T1D | |
| rs6827444 | 4q27 | KIAA1109 | 2726 | 0.181 | 0.0199 | T1D |
| rs12511287 | 4q27 | 2767 | 0.307 | 0.0349 | T1D | |
| rs11567751 | 5p13 | IL7R | 2704 | 0.297 | 0.0147 | Autoimmune |
| rs6897932 | 5p13 | IL7R | 2674 | 0.247 | 0.0424 | Autoimmune |
| rs3194051 | 5p13 | IL7R | 2736 | 0.298 | 0.0196 | Autoimmune |
| rs10214237 | 5p13 | 2751 | 0.246 | 0.0338 | Autoimmune | |
| rs700162 | 5p13 | UGT3A1 | 2563 | 0.402 | 0.0436 | Autoimmune |
| rs2447876 | 5p13 | UGT3A1 | 241 | 0.188 | 0.0302 | Autoimmune |
| rs3792876 | 5q31 | SLC22A4 | 2762 | 0.073 | 0.0157 | Autoimmune |
| rs2073838 | 5q31 | SLC22A4 | 2778 | 0.073 | 0.0196 | Autoimmune |
| rs7739974 | 6p22 | CDKAL1 | 2709 | 0.206 | 0.0195 | T2D |
| rs1569699 | 6p22 | CDKAL1 | 2750 | 0.325 | 0.0195 | T2D |
| rs2206736 | 6p22 | CDKAL1 | 2784 | 0.172 | 0.0174 | T2D |
| rs9356747 | 6p22 | CDKAL1 | 2768 | 0.234 | 0.0356 | T2D |
| rs7741604 | 6p22 | CDKAL1 | 2782 | 0.141 | 0.0208 | T2D |
| rs9465873 | 6p22 | CDKAL1 | 2749 | 0.386 | 0.032 | T2D |
| rs12211466 | 6p22 | CDKAL1 | 557 | 0.062 | 0.00157 | T2D |
| rs7738382 | 6p22 | CDKAL1 | 2752 | 0.281 | 0.0433 | T2D |
| rs11970030 | 6p22 | CDKAL1 | 2765 | 0.124 | 0.04 | T2D |
| rs4712569 | 6p22 | CDKAL1 | 2756 | 0.163 | 0.00371 | T2D |
| rs201351 | 6p22 | CDKAL1 | 2739 | 0.119 | 0.00874 | T2D |
| rs201300 | 6p22 | CDKAL1 | 2765 | 0.387 | 0.0422 | T2D |
| rs4389757 | 6p22 | CDKAL1 | 2763 | 0.125 | 0.00234 | T2D |
| rs9465994 | 6p22 | CDKAL1 | 2764 | 0.473 | 0.0402 | T2D |
| rs4710965 | 6p22 | CDKAL1 | 2690 | 0.462 | 0.0422 | T2D |
| rs6942273 | 6p22 | CDKAL1 | 2755 | 0.44 | 0.0213 | T2D |
| rs4876369 | 8q24 | SLC30A8 | 1646 | 0.153 | 0.00162 | T2D |
| rs10811661 | 9p21 | CDKN2B | 2750 | 0.171 | 0.0182 | T2D |
| rs10876864 | 12q13 | 2691 | 0.449 | 0.00109 | T1D | |
| rs2271194 | 12q13 | ERBB3 | 2653 | 0.447 | 0.00892 | T1D |
| rs705708 | 12q13 | ERBB3 | 2763 | 0.488 | 0.000125 | T1D |
| rs11171747 | 12q13 | 2744 | 0.356 | 7.12E-06 | T1D | |
| rs1678536 | 12q13 | KIF5A | 2756 | 0.278 | 0.00811 | Autoimmune |
| rs2640629 | 12q14 | 2705 | 0.331 | 0.0439 | Autoimmune | |
| rs701008 | 12q14 | AGAP2 | 2742 | 0.353 | 0.00252 | Autoimmune |
| rs2301551 | 12q14 | AGAP2 | 2651 | 0.295 | 0.0253 | Autoimmune |
| rs11172349 | 12q14 | 2722 | 0.307 | 0.0172 | Autoimmune | |
| rs12298022 | 12q24 | C12orf30 | 2759 | 0.077 | 0.00102 | T1D |
| rs17696736 | 12q24 | C12orf30 | 2742 | 0.468 | 0.000618 | T1D |
| rs4767364 | 12q24 | C12orf30 | 2745 | 0.273 | 0.000164 | T1D |
| rs6110460 | 20p13 | DEFB129 | 2669 | 0.47 | 0.0302 | Autoimmune |
| rs729749 | 22Q12 | NCF4 | 2658 | 0.248 | 0.049 | Autoimmune |
MAF, minor allele frequency in unaffected parents; T1D, type I diabetes; T2D, type II diabetes.
Gene regions showing some evidence of association (0.05> P> 0.001) with T1D included SNPs in two T2D genes thought to be involved with β-cell function (Table 1): CDKAL1 on chromosome 6p22.3 (P=1.6 × 10−3) and SLC30A8 on 8q24.11 (P=1.6 × 10−3). Recent studies of SLC30A8 have identified an association of T2D with the non-synonymous SNP, rs13266634 (Arg325Trp), an SNP also genotyped in this study. This SNP has also been reported to determine ZnT8 autoantibody specificity in T1D;6 however, no association with T1D was found with this SNP in this study of T1D affected sib-pair families (data not shown) in agreement with an earlier study of 7680 British T1D cases and 7200 controls.7 Similarly, the most strongly associated CDKAL1 SNP with T2D was not associated with T1D in this large British case–control study.7
In addition, some evidence of association with T1D (Table 1) was found for rs231727 in CTLA4 on chromosome 2q33.2 (P=1.2 × 10−3), a known T1D region;7 rs1678536 in KIF5A on 12q13.3 (P=8.11 × 10−3) and rs701008 in AGAP2 on 12q14.1 (P=2.5 × 10−3). The KIF5A and AGAP2 SNPs are not independently associated with T1D: using logistic regression, adding rs1678536/KIF5A to rs701008/AGAP2 gave P=0.37 and rs701008/AGAP2 to rs1678536/KIF5A gave P=0.08. The two SNPs are 138 kb apart and in the same linkage disequilibrium block (D′ =0.78 and r2=0.42 in parents).7 The associated 12q13.3–q14.1 region is approximately 1.6Mb. Nevertheless, the association of the rs1678536/KIF5A and rs701008/AGAP2 SNPs is independent of the established ERBB3/12q13.2–q13.3 region:8 using logistic regression, adding rs11171747/ERBB3 (most associated SNP found in this study) to rs1678536/KIF5A gave P=1.4 × 10−3 and adding rs1678536/KIF5A to rs11171747/ERBB3 gave P=7.3 × 10−3. KIF5A is about 2Mb centromeric of ERBB3 with a single recombination hotspot between these genes.
Discussion
In addition to the previously reported T1D-associated ERBB3/12q13.2–q13.3 region,5 we provide evidence of an independent T1D locus within the 12q13.3–q14.1 region, which contains 47 protein-coding genes, including KIF5A, AGAP2, PIP4K2C and also the vitamin D-associated gene CYP27B1, which has been reported to show some evidence of an association with T1D.9 However, rs1678542/KIF5A (an SNP associated with rheumatoid arthritis)10 and rs10877012/CYP27B1 are not independently associated with T1D in 7455 cases and 9089 controls: using logistic regression, adding rs1678542/KIF5A to rs10877012/CYP27B1 gave P=0.018 and rs10877012/CYP27B1 to rs1678542/KIF5A gave P=0.10. The 12q13.3–q14.1 region has been earlier associated with rheumatoid arthritis (rs1678542/KIF5A)10 and most recently with T1D in 8010 cases and 9733 controls (rs1678542/KIF5A, P=3.0 × 10−4; OR=0.92, 95% CI=0.88–0.96).2 Hence, we have obtained further support for the 12q13.3–q14.1 region in T1D affected sib-pair families. As this region contains 47 protein-coding genes, further sequencing and genotyping will be required to ascertain the contribution of these genes to T1D. We found no convincing evidence of commonality between type I and type II diabetes.
Materials and methods
Subjects
The DNA samples were genotyped at the Broad Institute of Harvard/MIT in Cambridge, MA, USA (http://www.broad.mit.edu/). The samples were assembled by the T1DGC and consist of affected sib-pair families of two parents and two affected offspring. The families were obtained from nine cohorts: Diabetes UK (DUK), Human Biological Data Interchange (HBDI), T1DGC Asia Pacific (AP) Network, T1DGC European (EUR) Network, T1DGC United Kingdom (UK) Network, T1DGC North America (NA) Network, Joslin (JOS) Diabetes Center, Sardinia (SAR) and Denmark (DAN). The AP, EUR, NA and UK collections were newly recruited by the T1DGC, whereas the remainder where part of established collections; 2074 families had at least 1 member who passed sample quality control, 1410 families provided 2798 parent–child trios.
SNP selection
A total of 581 SNPs were selected from T2D genes related to β-cell function, recent T1D and other autoimmune candidate genes (Table 1; www.T1DBase.org). SNPs were genotyped using the Illumina Golden Gate platform at the Broad Institute of Harvard/MIT in Cambridge, MA, USA (http://www.broad.mit.edu/). In addition to standard sample and SNP quality control, SNP signal intensity plots for each of the family groups were visually inspected (see accompanying paper4). This process provided 334 SNPs with well-separated signal clouds—148 of 239 SNPs from autoimmune disease regions, 114 of 216 SNPs from T2D regions and 72 of 126 SNPs from T1D regions. Genotype signal intensity cluster plots are available in T1DBase.8
Statistics
All analyses were carried out in the R statistical environment using the snpMatrix package from the bioConductor project.11 Family groups were analyzed using the transmission/disequilibrium test configured as a score test. The scores and their variances were summed over family groups and genotyping centers to pool information.
Supplementary Material
Acknowledgements
The Type I Diabetes Genetics Consortium is 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 (NICHD) and supported by U01 DK062418. JDC, NMW, DJS, KD, BCH and JAT are funded by the Juvenile Diabetes Research Foundation International (JDRF), the Wellcome Trust and the National Institute for Health Research Cambridge Biomedical Centre. The Cambridge Institute for Medical Research is in receipt of a Wellcome Trust Strategic Award (079895). Genotyping was performed at the Broad Institute Center for Genotyping and Analysis is supported by grant U54 RR020278 from the National Center for Research Resources.
Footnotes
Conflict of interest
The authors declare no conflict of interest.
References
- 1.Smyth DJ, Plagnol V, Walker NM, Cooper JD, Downes K, Yang JH, et al. Shared and distinct genetic variants in type 1 diabetes and celiac disease. N Engl J Med. 2008;359:2767–2777. doi: 10.1056/NEJMoa0807917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fung EYMG, Smyth DJ, Howson JMM, Cooper JD, Walker NM, Stevens H, et al. Analysis of 17 autoimmune disease-associated Analysis of 17 autoimmune disease-associated 6q23/TNFAIP3 as a susceptibility locus. Genes Immun. 2009;10:188–191. doi: 10.1038/gene.2008.99. [DOI] [PubMed] [Google Scholar]
- 3.Donath MY, Storling J, Maedler K, Mandrup-Poulsen T. Inflammatory mediators and islet beta-cell failure: a link between type 1 and type 2 diabetes. J MolMed. 2003;81:455–470. doi: 10.1007/s00109-003-0450-y. [DOI] [PubMed] [Google Scholar]
- 4.Cooper JD, Walker NM, Smyth DJ, Downes K, Healy BC, Todd JA The Type I Diabetes Genetics Consortium. Follow-up of 1715 SNPs from the Wellcome Trust Case Control Consortium genome-wide association study in type I diabetes families. Genes Immun. 2009;10 Suppl 1:S85–S94. doi: 10.1038/gene.2009.97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Todd JA, Walker NM, Cooper JD, Smyth DJ, Downes K, Plagnol V, et al. Robust associations of four new chromosome regions from genome-wide analyses of type 1 diabetes. Nat Genet. 2007;39:857–864. doi: 10.1038/ng2068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wenzlau JM, Liu Y, Yu L, Moua O, Fowler KT, Rangasamy S, et al. A common nonsynonymous single nucleotide polymorphism in the SLC30A8 gene determines ZnT8 autoantibody specificity in type 1 diabetes. Diabetes. 2008;57:2693–2697. doi: 10.2337/db08-0522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Raj SM, Howson JMM, Walker NM, Cooper JD, Smyth DJ, Field SF, et al. Type 1 diabetes is genetically distinct from type 2 diabetes. Diabetologia. 2009;52:2109–2116. doi: 10.1007/s00125-009-1391-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. http://www.T1DBase.org.
- 9.Bailey R, Cooper JD, Zeitels L, Smyth DJ, Yang JH, Walker NM, et al. Association of the vitamin D metabolism gene CYP27B1 with type 1 diabetes. Diabetes. 2007;56:2616–2621. doi: 10.2337/db07-0652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Raychaudhuri S, Remmers EF, Lee AT, Hackett R, Guiducci C, Burtt NP, et al. Common variants at CD40 and other loci confer risk of rheumatoid arthritis. Nat Genet. 2008;40:1216–1223. doi: 10.1038/ng.233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Clayton D, Leung HT. An R package for analysis of whole-genome association studies. Hum Hered. 2007;64:45–51. doi: 10.1159/000101422. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.