Genetic Analysis of Adult-Onset Autoimmune Diabetes
Article Figures & Tables
Figures
- FIG. 1.
Frequency histogram of age at diagnosis in the 1,384 German autoimmune diabetic case subjects. The curve represents the normal density.
- FIG. 2.
Comparison of effects of the minor allele in German autoimmune diabetic case subjects diagnosed between 17 and 89 years of age (X) and British pediatric-onset type 1 diabetic case subjects diagnosed before 17 years of age (♦). Effects in type 1 diabetic subjects are taken from Smyth et al. (38), Fung et al. (39), and Barrett et al. (1) and are also available from www.t1dbase.org/page/regions. Effects for all SNPs tested in Table 2 are given for British pediatric-onset type 1 diabetic case subjects diagnosed before 17 years of age in Supplementary Table 2.
Tables
- TABLE 1
HLA associations in a maximum of 1,177 adult-onset autoimmune diabetic case subjects and 2,210 control subjects
SNP (allele or genotype) N (frequency) OR [95% CI] P Case subjects Control subjects rs2187668 (DR3)* 585 (0.25) 395 (0.09) 3.37 [2.92–3.90] 2.31 × 10−65 rs660895 (DR4)* 868 (0.37) 662 (0.16) 3.34 [2.94–3.81] 8.65 × 10−84 DR3† 567 (0.25) 380 (0.09) 5.36 [4.53–6.34] DR4† 852 (0.37) 656 (0.16) 4.98 [4.23–5.77] DRX† 857 (0.38) 3,138 (0.75) 1.00 (ref.) 2.02 × 10−179 3/3‡ 75 (0.07) 15 (0.01) 37.97 [21.26–67.81] 7.22 [4.07–12.79] 3/4‡ 222 (0.20) 64 (0.03) 26.22 [18.94–36.30] 4.98 [3.65–6.80] 4/4‡ 142 (0.12) 47 (0.02) 22.42 [15.48–32.48] 4.26 [2.98–6.10] 4/X‡ 346 (0.30) 498 (0.24) 5.26 [4.23–6.54] 1.00 (ref.) 3/X‡ 195 (0.17) 286 (0.14) 5.15 [4.02–6.60] 0.98 [0.78–1.23] X/X‡ 158 (0.14) 1,177 (0.56) 1.00 (ref.) 0.19 [0.15–0.24] The SNPs rs2187668 and rs660895 were used as proxies for HLA-DRB1*03 (DR3) and HLA-DRB1*04 (DR4) alleles, respectively, and to code for the DR3/4/X genotypes (see research design and methods). The P values reported are for a model that assumes multiplicative allelic effects either for the individual SNPs or for the joint effects of both DR3 and DR4.
*Individual SNP associations of rs2187668 and rs660895 use all samples genotyped at the test SNP.
†A joint effects model including both DR3 (rs2187668) and DR4 (rs660895). Only samples genotyped at both SNPs were included (see research design and methods).
‡The genotype effects model. Only samples genotyped at both SNPs were included (see research design and methods). No P value is reported for this model because a multiplicative allelic effects model was an appropriate approximation (P = 0.55). N, number of chromosomes or genotypes.
- TABLE 2
Associations at 24 non-HLA loci in a maximum of 1,196 autoimmune diabetic case subjects diagnosed at over 17 years of age and 2,215 control subjects
Candidate gene (region), SNP N Chromosomes (MAF) OR [95% CI] P Power* Case subjects Control subjects α = 0.001 α = 0.05 PTPN22 (1p13.2), rs2476601 (C>T) 376 (0.16) 431 (0.10) 1.77 [1.53–2.06] 8.03 × 10−14 1.00 1.00 IFIH1 (2q24.2), rs1990760 (A>G) 832 (0.36) 1,643 (0.39) 0.87 [0.78–0.96] 0.0077 0.35 0.83 STAT4 (2q32.2), rs7574865 (G>T) 590 (0.25) 938 (0.21) 1.23 [1.09–1.39] 5.93 × 10−4 0.04 0.35 CTLA4 (2q33.2), rs3087243 (G>A) 928(0.39) 2,043 (0.46) 0.75 [0.68–0.83] 3.65 × 10−8 0.74 0.98 IL2 (4q27), rs2069763 (G>T) 786(0.34) 1,484 (0.34) 0.96 [0.86–1.08] 0.51 0.17 0.64 IL2 (4q27), rs2069762 (T>G) 715 (0.30) 1,366 (0.31) 0.96 [0.86–1.08] 0.51 0.12 0.57 BACH2 (6q15), rs11755527 (C>G) 1,144 (0.49) 1,686 (0.45) 1.16 [1.05–1.29] 0.0045 0.20 0.68 GLIS3 (9p24.2), rs7020673 (G>C) 1,052 (0.44) 2,083 (0.48) 0.88 [0.79–0.97] 0.012 0.23 0.72 IL2RA (10p15.1), rs12722495 (A>G) 190 (0.08) 499 (0.11) 0.70 [0.58–0.83] 5.22 × 10−5 0.99 1.00 IL2RA (10p15.1), rs2104286 (A>G) 523 (0.22) 1,083 (0.25) 0.85 [0.76–0.96] 0.031 0.07 0.44 IL2RA (10p15.1), rs11594656 (T>A) 576 (0.25) 1,127 (0.27) 0.92 [0.82–1.03] 0.13 0.20 0.68 RNLS (10q23.31), rs10509540 (T>C) 530 (0.24) 1,133 (0.26) 0.88 [0.78–1.00] 0.041 0.95 1.00 INS (11p15.5), rs689 (A>T) 382 (0.16) 1,280 (0.29) 0.47 [0.42–0.54] 1.97 × 10−32 1.00 1.00 ERBB3 (12q13.2), rs2292239 (C>A) 846 (0.36) 1,376 (0.32) 1.20 [1.08–1.33] 8.40 × 10−4 0.97 100 SH2B3 (12q24.12), rs3184504 (A>G) 1,030 (0.44) 2,114 (0.48) 1.35 [1.22–1.50] 2.67 × 10−9 0.95 100 CLEC16A (16p13.13), rs12708716 (A>G) 763 (0.34) 1,583 (0.38) 0.83 [0.75–0.93] 7.82 × 10−4 0.77 0.98 IL27 (16p11.2), rs4788084 (G>A) 939 (0.39) 1,830 (0.42) 0.90 [0.81–1.00] 0.042 0.37 0.84 CTRB2 (16q23.1), rs7202877 (T>G) 271 (0.12) 468 (0.10) 1.09 [0.92–1.27] 0.32 0.41 0.86 GSDMB (17q12), rs2290400 (G>A) 1,127 (0.49) 2,115 (0.50) 0.97 [0.87–1.07] 0.051 0.30 0.79 PTPN2 (18p11.21), rs478582 (T>C) 1,027 (0.43) 1,947 (0.44) 0.96 [0.87–1.06] 0.55 0.65 0.96 PTPN2 (18p11.21), rs45450798 (G>C) 390 (0.17) 644 (0.15) 1.15 [1.00–1.31] 0.054 0.63 0.95 CD226 (18q22.2), rs763361 (C>T) 1,157 (0.49) 2,178 (0.49) 0.98 [0.89–1.08] 0.68 0.25 0.74 UBASH3A (21q22.3), rs3788013 (C>A) 1,035 (0.44) 1,853 (0.42) 1.06 [0.96–1.17] 0.26 0.19 0.68 Obesity locus FTO (16q12.2), rs9939609 (T>A)† 953 (0.41) 1,798 (0.42) 0.96 [0.87–1.06] 0.45 0.42 0.86 Type 2 diabetes locus TCF7L2 (10q25.2), rs12255372 (G>T) 571 (0.27) 1,147 (0.27) 1.01 [0.89–1.13] 0.93 0.99 0.99 Autoantibody loci GAD2 (10p12.1), rs2839671 (G>A) 412 (0.18) 776 (0.18) 0.96 [0.84–1.09] 0.50 0.13 0.59 FCRL3 (1q23.1), rs7528684 (A>G) 1,045 (0.44) 1,919 (0.44) 1.03 [0.93–1.14] 0.58 0.30 0.79 MAF, minor allele frequency; number of chromosomes is listed for the minor allele.
*Power is calculated for the susceptible allele, assuming multiplicative allelic effects at all loci except INS. For all type 1 diabetes–associated loci, the effect estimates and minor allele frequencies used to calculate the power were taken from Smyth et al. (38), Fung et al. (39), and Barrett et al. (1) and are also available from www.t1dbase.org/page/regions. Note that if the association of a locus is reduced with increasing age at diagnosis, these power estimates could be inflated. For the TCF7L2 and FTO loci, the effect estimates as reported for type 2 diabetes were used (40,41). For the autoantibody loci, an effect estimate of 1.15, which is not unusual for type 1 diabetes, was used.
†rs9939609 is in LD with rs9931494, the SNP reported as associated with LADA (r2 = 0.91, D′ = 0.98 in Centre d'Etude du Polymorphisme Humaine samples) (14).
- TABLE 3
Age at diagnosis in all 1,384 autoimmune diabetic case subjects, irrespective of age at diagnosis, by HLA genotype (rs9271366 being too rare to analyze for age-at-diagnosis effects by age category)
Mean age at diagnosis (years) (SD) Genotype counts (frequency) by age-group P Pcat 0–21 years 22–31 years 32–43 years ≥44 years rs660895 (DR4) A/A (X/X) 35.8 (16.4) 97 (0.30) 104 (0.33) 130 (0.41) 133 (0.44) 4.67 × 10−6 2.58 × 10−4 G/A (4/X) 32.3 (15.5) 172 (0.53) 174 (0.54) 146 (0.46) 146 (0.48) G/G (4/4) 30.0 (14.7) 54 (0.17) 42 (0.13) 41 (0.13) 27 (0.09) rs2187668 (DR3) G/G (X/X) 34.0 (16.3) 187 (0.56) 176 (0.55) 177 (0.56) 187 (0.61) 0.067 0.51 G/A (3/X) 31.9 (15.3) 128 (0.38) 126 (0.39) 118 (0.37) 104 (0.34) A/A (3/3) 32.9 (13.0) 18 (0.05) 20 (0.06) 21 (0.07) 17 (0.06) rs9271366 (DR15) A/A (X/X) 33.1 (15.9) 320 (0.98) 307 (0.96) 298 (0.95) 297 (0.96) 0.13 NA A/G (15/X) 36.7 (11.1) 4 (0.01) 12 (0.04) 15 (0.05) 13 (0.04) G/G (15/15) 37.3 (18.2) 1 (0.00) 0 (0.00) 1 (0.00) 1 (0.00) 3/3 33.3 (12.9) 16 (0.05) 20 (0.06) 21 (0.07) 17 (0.06) 1.78 × 10−7 0.0029 3/4 30.4 (14.4) 72 (0.23) 73 (0.23) 57 (0.18) 50 (0.17) 4/4 30.0 (14.3) 54 (0.17) 41 (0.13) 40 (0.13) 27 (0.09) 3/X 33.9 (16.3) 51 (0.16) 49 (0.16) 59 (0.19) 50 (0.17) 4/X 33.4 (16.1) 98 (0.31) 99 (0.31) 85 (0.28) 92 (0.31) X/X 39.2 (17.3) 26 (0.08) 33 (0.10) 47 (0.15) 60 (0.20) NA, not analyzed. P, the P value for interaction of age at diagnosis as a continuous trait with genotype. Pcat, P value for interaction of genotype with the four age-groups corresponding to the quartiles of the age-at-diagnosis distribution.
- TABLE 4
Association of HLA class II with IA-2A in 904 autoimmune diabetic case subjects, using rs660895 and rs2187668 to model HLA-DR4 and -DR3, respectively
N (frequency) OR [95% CI] P IA-2A positive IA-2A negative rs660895 G (DR4)* 390 (0.42) 291 (0.33) 1.62 [1.31–2.00] 5.45 × 10−6 rs2187668 A (DR3)† 186 (0.20) 260 (0.29) 0.56 [0.44–0.70] 3.22 × 10−7 DR4‡ 382 (0.42) 286 (0.33) 1.34 [1.06–1.70] DR3‡ 184 (0.20) 255 (0.30) 0.64 [0.49–0.82] 1.43 × 10−7 4/4‡ 73 (0.16) 42 (0.10) 1.90 [1.09–3.29] 3/4‡ 81 (0.18) 96 (0.22) 0.84 [0.52–1.36] 3/3‡ 21 (0.05) 34 (0.08) 0.54 [0.28–1.06] 4/X‡ 155 (0.34) 106 (0.25) 1.59 [1.02–2.50] 3/X‡ 61 (0.14) 91 (0.21) 0.68 [0.41–1.12] X/X‡ 61 (0.14) 62 (0.14) 1.00 (ref.) Both DR3 and DR4 were required to model the association with IA-2A. The DR3/4/X genotype coding did not improve on the allele-specific model (P = 0.66).
*Uses all samples genotyped at rs660895.
†Uses all samples genotyped at rs2187668.
‡Uses samples genotyped at both rs660895 and rs2187668. N, number of chromosomes or genotypes.
- TABLE 5
Association of HLA class II with GADA, using rs660895 and rs2187668, to model HLA-DR4 and -DR3 (in 1,327 and 1,342 autoimmune diabetic case subjects, respectively)
SNP N (frequency) OR [95% CI] P GADA positive GADA negative rs2187668 A (DR3)* 586 (0.26) 76 (0.17) 1.90 [1.42–2.55] 6.03 × 10−6 rs660895 G (DR4)* 847 (0.38) 168 (0.36) 1.23 [0.97–1.57] 0.090 DR3† 569 (0.26) 72 (0.16) 1.43 [1.02–1.99] 2.53 [1.82–3.51] DR4† 833 (0.39) 166 (0.38) 1.00 (ref.) 1.78 [1.36–2.33] DRX† 754 (0.35) 202 (0.46) 0.56 [0.43–0.74] 1.00 (ref.) 5.31 × 10−9 4/4 147 (0.14) 22 (0.10) 2.57 [1.44–4.59] 1.37 [0.72–2.59] 3/X 194 (0.18) 23 (0.10) 2.37 [1.38–4.09] 1.26 [0.69–2.32] 3/3 74 (0.07) 6 (0.03) 3.27 [1.25–8.61] 1.74 [0.64–4.76] 3/4 227 (0.21) 37 (0.17) 1.88 [1.18–2.98] 1.00 (ref.) 4/X 312 (0.29) 85 (0.39) 1.00 (ref.) 0.53 [0.34–0.85] X/X 124 (0.12) 47 (0.21) 0.51 [0.32–0.81] 0.27 [0.16–0.47] *Single SNP analyses use all samples genotyped at the test SNP.
†Uses the DR3 and DR4 SNPs to generate a non-DR3, non-DR4 allele, DRX, and to analyze the class II locus HLA-DRB1 as a single locus including alleles using two alleles in the model and using the third as reference. Only samples genotyped at both SNPs (rs2187668 and rs660895) were included in this analysis. The evidence that the DR3/4/X genotype coding was a more appropriate model than the combined DR3 and DR4 allelic effects model was unconvincing (P = 0.024). N, number of chromosomes or genotypes.
- TABLE 6
Association of rs7528684 (A>G) in FCRL3 (1q23.1) with IA-2A positivity in 903 autoimmune diabetic case subjects
Allele or genotype N (frequency) OR [95% CI] P IA-2A positive IA-2A negative G 386 (0.42) 450 (0.50) 0.73 [0.60–0.88] 9.16 × 10−4 A/A 161 (0.35) 116 (0.26) 1.00 (ref.) 1.44 [1.05–1.97] A/G 210 (0.46) 212 (0.48) 0.70 [0.51–0.96] 1.00 (ref.) G/G 88 (0.19) 116 (0.26) 0.53 [0.37–0.78] 0.77 [0.54–1.09] N, number of chromosomes or genotypes.
