Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2015 Jul 3;5:11900. doi: 10.1038/srep11900

TOX and CDKN2A/B Gene Polymorphisms Are Associated with Type 2 Diabetes in Han Chinese

Fengjiang Wei 1,*, Chunyou Cai 1,*, Shuzhi Feng 2,*, Jia Lv 1, Shen Li 1, Baocheng Chang 3, Hong Zhang 4, Wentao Shi 1, Hongling Han 2, Chao Ling 1, Ping Yu 1, Yongjun Chen 4, Ning Sun 2, Jianli Tian 2, Hongxiao Jiao 1, Fuhua Yang 1, Mingshan Li 1, Yuhua Wang 1, Lei Zou 1, Long Su 4, Jingbo Li 5, Ran Li 2, Huina Qiu 5, Jingmin Shi 1, Shiying Liu 2, Mingqin Chang 1, Jingna Lin 5,a, Liming Chen 3,b, Wei-Dong Li 1,c
PMCID: PMC4650661  PMID: 26139146

Abstract

To study associations between type 2 diabetes (T2DM) candidate genes and microvascular complications of diabetes (MVCDs), we performed case-control association studies for both T2DM and MVCDs in Han Chinese subjects. We recruited 1,939 unrelated Han Chinese T2DM patients and 918 individuals with normal blood glucose levels as nondiabetic controls. Among T2DM patients, 1116 have MVCDs, 266 have a history of T2DM of >10 years but never developed MVCDs. Eighty-two single-nucleotide polymorphisms (SNPs) in 54 candidate genes were genotyped. Discrete association studies were performed by the PLINK program for T2DM and MVCDs. Significant associations were found among candidate gene SNPs and T2DM, including rs1526167 of the TOX gene (allele A, P = 2.85 × 10−9, OR = 1.44). The SNP rs10811661 of the CDKN2A/B gene was also associated with T2DM (allele T, P = 4.09 × 10−7, OR = 1.36). When we used control patients with >10 years of T2DM history without MVCD, we found that the G allele of SNP rs1526167 of the TOX gene was associated with MVCD (nominal P = 4.33 × 10−4). In our study, significant associations were found between TOX and CDKN2A/B gene SNPs and T2DM. The TOX polymorphism might account for the higher risk of T2DM and the lower risk of MVCDs in the Han Chinese population.

The prevalence of type 2 diabetes (T2DM) has increased dramatically in China in recent years1,2. Many T2DM genes found in European populations have been replicated in Han Chinese; however, many top candidate genes (including TCF7L2) have shown only moderate associations with T2DM in East Asian populations. On the other hand, many Asian-specific T2DM genes have been identified in recent studies3.

With the remarkable rise worldwide in the prevalence of diabetes, an increase in patients suffering from microvascular complications of diabetes (MVCDs) will be inevitable. Diabetic nephropathy (DN) and diabetic retinopathy (DR), two common MVCDs, are leading causes of end-stage renal failure and blindness in diabetes patients4,5. Epidemiology data for MVCDs are relatively limited in China, although a recent study showed a much lower prevalence of DR in the Beijing area than in European populations6.

Although previous studies have shown that the duration of diabetes and plasma glucose levels are MVCD risk factors7,8, some diabetes patients with severe hyperglycemia never develop MVCDs. In contrast, many diabetes patients with well-controlled blood glucose suffer from MVCDs9. Many studies have found family aggregation of MVCDs and significantly increased risk in siblings, suggested that genetic factors play an important role in the etiology of MVCDs10,11,12,13,14. Although many association studies have identified genes related to MVCDs15, the search for MVCD susceptibility genes was less successful than for T2DM susceptibility genes, partially because of un-diagnosed MVCDs and lack of appropriate normal controls.

It is possible that Han Chinese–specific gene variants are associated with T2DM and/or MVCDs. In this study, we aimed to find T2DM and MVCD candidate genes in Han Chinese T2DM patients by studying associations between single-nucleotide polymorphisms (SNPs) of candidate genes and T2DM/MVCDs.

Results

Using 1,939 T2DM patients as cases and 918 nondiabetic individuals (>57 years old, with normal blood glucose) as controls, analyses for T2DM showed association between candidate gene SNPs and T2DM (Table 1), including rs10811661 of the CDKN2A/B gene (allele T, P = 4.09 × 10−7, OR = 1.364, 95% CI = 1.209–1.538), rs1526167 of the TOX gene (allele A, P = 2.85 × 10−9, OR = 1.440, 95% CI = 1.276–1.624), rs4402960 of the IGF2BP2 gene (allele T, P = 7.56 × 10−4, OR = 1.352, 95% CI = 1.134–1.612), rs6856526 of the LPHN3 gene (allele C, P = 1.33 × 10−3, OR = 1.522, 95% CI = 1.176–1.970), rs13266634 of the SLC30A8 gene (allele C, P = 4.56 × 10−3, OR = 1.282, 95% CI = 1.080–1.532), and rs7756992 of the CDKAL1 gene (allele G, P = 8.21 × 10−4, OR = 1.246, 95% CI = 1.095–1.418). The TOX gene SNP rs1526167 association reached genome-wide association level (P < 5 × 10−8), while the CDKN2A/B SNP rs10811661 was significant after Bonferroni correlation (P < 1.22 × 10−4 for 82 SNPs and 5 binary traits, 6.1 × 10−4 for 82 SNPs and 1 binary trait).

Table 1. Association analyses for T2DM (1939 cases, 918 non-T2DM controls, nominal P < 5 × 10−3).

SNP Chr Position Gene Risk allele Risk allele frequencies P OR (95% CI)
rs10811661 9 22134094 CDKN2A/B T 0.56 4.09 × 10−7# 1.364 (1.209 ~ 1.538)
rs1526167 8 59702355 TOX A 0.46 2.85 × 10−9* 1.440 (1.276 ~ 1.624)
rs4402960 3 185511687 IGF2BP2 T 0.25 7.56 × 10−4 1.352 (1.134 ~ 1.612)
rs6856526 4 61057462 LPHN3 C 0.93 1.33 × 10−3 1.522 (1.176 ~ 1.970)
rs13266634 8 118184783 SLC30A8 C 0.64 4.56 × 10−3 1.282 (1.080 ~ 1.532)
rs7756992 6 20679709 CDKAL1 G 0.54 8.21 × 10−4 1.246 (1.095 ~ 1.418)
Open in a new tab

*Significant for genome-wide association (P < 5 × 10−8).

#Significant after multiple tests correction.

In MVCD association studies, DN, DR, proliferative retinopathy (PDR), and MVCD patients were selected as cases, and patients with >10 years T2DM history, without DR or DN, were chosen as controls. Association analyses of dichotomous variables showed that rs1526167 of the TOX gene was associated with DN (P = 0.0011, OR = 1.470, 95% CI = 1.166–1.854), DR (P = 0.0082, OR = 1.412, 95% CI = 1.093–1.825), and MVCD (P = 4.33 × 10−4, OR = 1.498, 95% CI = 1.195–1.878). However, the at-risk allele for MVCDs was G, rather than A for T2DM. The T allele of rs10811661 of the CDKN2A/B gene was associated with DR (P = 0.037, OR = 1.314, 95% CI = 1.016–1.698), PDR (P = 0.026, OR = 1.708, 95% CI = 1.063–2.743), and MVCD (P = 0.025, OR = 1.292, 95% CI = 1.032–1.617). The G allele of rs4402960 of the IGF2BP2 gene was associated with DN (P = 0.0092, OR = 1.499, 95% CI = 1.104–2.034) and MVCD (P = 0.0094, OR = 1.481, 95% CI = 1.100–1.995). The T allele of rs12102171 of the SMAD3 gene was associated with DR and MVCD (P = 0.027, and 0.045, respectively). The A allele of the ESR1 gene was associated with DR and PDR (P = 0.041 and 0.033, respectively) (Table 2). None of these nominal associations for MVCDs remained significant after Bonferroni corrections.

Table 2. Candidate gene association studies for MVCD (266 non-MVCD T2DM patients as controls).

SNP rs4402960 rs1526167 rs12102171 rs722208 rs10811661
chromosome 3 8 15 6 9
position 185511687 59702355 67425033 152322885 22134094
Gene IGF2BP2 TOX SMAD3 ESR1 CDKN2A/B
Risk allele G G T A T
DN (N = 836)
 Risk allele frequencies 0.72 0.48      
 P 0.0092 0.0011      
 OR (95% CI) 1.499 (1.104 ~ 2.034) 1.470 (1.166 ~ 1.854)      
DR (N = 504)
 Risk allele frequencies   0.46 0.31 0.46 0.58
 P   0.0082 0.027 0.041 0.037
 OR (95% CI)   1.412 (1.093 ~ 1.825) 1.471 (1.043 ~ 2.072) 1.305 (1.011 ~ 1.686) 1.314 (1.016 ~ 1.698)
PDR (N = 106)
 Risk allele frequencies       0.44 0.57
 P       0.033 0.026
 OR (95% CI)       1.632 (1.038 ~ 2.566) 1.708 (1.063 ~ 2.743)
MVCD (N = 1116)
 Risk allele frequencies 0.72 0.49 0.31   0.59
 P 0.0094 4.33 × 10−4 0.045   0.025
 OR (95% CI) 1.481 (1.100 ~ 1.995) 1.498 (1.195 ~ 1.878) 1.376 (1.007 ~ 1.879)   1.292 (1.032 ~ 1.617)
Open in a new tab

Only results with nominal P < 0.05 are shown.

Four SNPs of the TOX gene were genotyped (Table 3), rs1526167 was not in linkage disequilibrium with other 3 SNPs (Supplement Table 2, Supplement Figures 1 and 2). The SNP rs17304270 of the TOX gene gave marginal association with diabetic nephropathy: allele “A”, OR = 1.514 (1.003–2.285), P = 0.047.

Table 3. Candidate genes and SNPs genotyped in this study.

SNP Gene Chr HWE (P) MAF*
CHB CEU Global
rs7546903 CAMTA1 1 0.603 0.463 0.226 0.368
rs1801133 MTHFR 1 0.236 0.439 0.310 0.325
rs6427665 NOS1AP 1 0.942 0.455 0.233 0.378
rs2661812 NOS1AP 1 0.596 0.475 0.500 0.445
rs16867321 UBE2E3 2 1.000 0.415 0.200 0.271
rs62183937 ABI2 2 1.000 0.475 0.125 0.258
rs11675251 ABI2 2 0.303 0.171 0.482 0.381
rs3731652 ABI2 2 0.666 0.433 0.158 0.375
rs1376877 ABI2 2 0.177 0.171 0.455 0.383
rs11677793 SPAG16 2 0.992 0.200 0.456 0.284
rs7578326 IRS1 2 0.512 0.125 0.350 0.304
rs1678607 VHL 3 0.768 0.111 0.125 0.208
rs13081389 PPARγ 3 0.438 0.022 0.042 0.034
rs35747495 PCAF 3 0.979 0.325 0.300 0.260
rs2929402 PCAF 3 0.227 0.463 0.372 0.419
rs1986917 PCAF 3 0.987 0.433 0.442 0.389
rs4402960 IGF2BP2 3 0.188 0.256 0.280 0.343
rs13129697 SLC2A9 4 0.920 0.439 0.292 0.423
rs1014290 SLC2A9 4 0.892 0.363 0.257 0.308
rs6856526 LPHN3 4 0.484 0.073 0.009 0.129
rs2231142 ABCG2 4 0.989 0.293 0.111 0.139
rs10946398 CDKAL1 6 0.851 0.439 0.336 0.408
rs7756992 CDKAL1 6 0.312 0.488 0.279 0.405
rs1165196 SLC17A1 6 0.205 0.232 0.451 0.260
rs881858 VEGFA 6 0.045 0.189 0.292 0.346
rs9395706 PKHD1 6 0.989 0.476 0.128 0.296
rs722208 ESR1 6 0.583 0.500 0.246 0.412
rs4880 SOD2 6 0.713 0.111 0.300 0.188
rs1581498 SNORD93 7 0.362 0.400 0.467 0.378
rs1799884 GCK 7 0.084 0.171 0.195 0.188
rs705382 PON1 7 0.755 0.415 0.336 0.472
rs1007311 NOS3 7 0.587 0.308 0.500 0.437
rs768403 GBX1 7 0.877 0.463 0.398 0.476
rs7805834 NUB1 7 0.694 0.073 0.102 0.140
rs446886 NUB1 7 0.151 0.073 0.310 0.298
rs386956 NUB1 7 0.972 0.488 0.319 0.478
rs1526167 TOX 8 0.445 0.478 0.467 0.478
rs2726557 TOX 8 0.254 0.427 0.327 0.459
rs11777927 TOX 8 0.757 0.356 0.267 0.356
rs17304270 TOX 8 0.130 0.061 0.288 0.390
rs13266634 SLC30A8 8 0.930 0.476 0.239 0.282
rs10811661 CDKN2A/B 9 0.764 0.415 0.199 0.206
rs3758391 SIRT1 10 0.965 0.195 0.270 0.473
rs7923837 HHEX 10 1.000 0.244 0.367 0.427
rs7903146 TCF7L2 10 0.236 0.024 0.279 0.218
rs2237892 KCNQ1 11 0.287 0.317 0.075 0.170
rs2166706 MTNR1B 11 0.658 0.317 0.389 0.472
rs189037 ATM 11 0.997 0.389 0.485 0.485
rs7312112 IGF1 12 0.304 0.500 0.385 0.462
rs2241220 ACACB 12 0.996 0.341 0.142 0.109
rs11067076 TBX5 12 0.586 0.037 0.257 0.193
rs11067083 TBX5 12 0.545 0.064 0.242 0.199
rs371276 SLITRK5 13 0.615 0.463 0.025 0.244
rs409762 SLITRK5 13 0.832 0.488 0.013 0.244
rs2301113 HIF1A 14 0.925 0.317 0.173 0.428
rs11624704 NRXN3 14 0.492 0.049 0.137 0.115
rs1498506 SMAD3 15 0.840 0.433 0.475 0.455
rs12102171 SMAD3 15 0.353 0.341 0.195 0.276
rs17818920 FTO 16 1.000 0.183 0.250 0.258
rs1876942 FTO 16 0.309 0.341 0.425 0.479
rs708254 FTO 16 0.438 0.350 0.420 0.390
rs2239359 FANCA 16 0.970 0.207 0.416 0.393
rs7190823 FANCA 16 0.488 0.024 0.416 0.343
rs4353 ACE 17 0.953 0.317 0.495 0.498
rs17782313 MC4R 18 0.489 0.232 0.265 0.221
rs8109627 CCDC97 19 0.928 0.390 0.235 0.341
rs4814615 PCSK2 20 1.000 0.488 0.128 0.293
rs3746876 KCNJ15 21 0.906 0.110 0.004 0.046
rs2106294 LIMK2 22 0.824 0.073 0.305 0.178
rs5749286 SFI1 22 0.875 0.378 0.283 0.271
rs5753669 SFI1 22 0.497 0.378 0.283 0.271
rs2295251 SFI1 22 0.652 0.451 0.270 0.404
rs735853 MYH9 22 0.991 0.110 0.477 0.273
rs875726 MYH9 22 0.456 0.305 0.296 0.449
rs2009930 MYH9 22 0.377 0.275 0.283 0.450
rs2239782 MYH9 22 0.067 0.317 0.239 0.349
rs3752462 MYH9 22 0.320 0.305 0.332 0.453
rs2269532 MYH9 22 0.881 0.267 0.358 0.390
rs2071731 MYH9 22 0.999 0.280 0.367 0.423
rs739097 MYH9 22 0.876 0.268 0.456 0.495
rs2285094 PDGFB 22 0.549 0.183 1.000 0.328
rs738409 PNPLA3 22 0.804 0.344 0.233 0.284
Open in a new tab

*MAF: minor allele frequencies, taken from dbSNP. CHB, Han Chinese; CEU, European American.

Discussion

To date, more than 30 genes have been identified as reaching the genome-wide significance threshold (P < 5 × 10−8) for T2DM; 10 of these were replicated in the Han Chinese population16. More than 60% of T2DM genes found in East Asian genome-wide association studies (GWAS) were replicated in Han Chinese17, compared with approximately 30% of European population T2DM genes. Compared with Caucasians, the Han Chinese and East Asian populations are more insulin resistant, are more centrally obese, and develop T2DM more readily even with moderate increases in body mass index18,19,20. Thus, seeking T2DM genes in Han Chinese would help us to better understanding the global genetic background of T2DM.

In this study, as candidate genes we chose 1) genes related to T2DM, obesity, or insulin resistance found by previous GWAS and 2) genes related to glucose and lipid metabolism, insulin secretion, or MVCD, such as the HIF1α/-VEGF pathway.

A significant association (P = 2.85 × 10−9) between a TOX gene SNP and T2DM was identified for the first time by this study. In addition, four previous reported T2DM genes, CDKN2A/B, IGF2BP2, SLC30A8, and CDKAL1, showed moderate to strong associations with T2DM. We did not replicate associations with several well-known T2DM genes, including TCF7L2, FTO, IRS1, and KCNQ1, in our Han Chinese population (In our data set, the FTO gene SNPs yielded some associations with insulin resistance related phenotypes in quantitative analyses. Data not shown). Potential reasons for this failed replication may include genetic heterogeneity, low allele frequencies of tested SNPs in Han Chinese (e.g., the MAF of TCF7L2 gene SNP rs7903146 was 0.046), less covered genes (i.e., too few SNPs were genotyped for certain candidate genes), and the relatively small sample size. We performed a gene-specific power calculations for genes IRS1, TCF7L2, KCNQ1, and FTO using real minor allele frequencies (MAF) in our data set (Supplement Table 1). Except TCF7L2, we have adequate power (>80%) to detect associations at the type I error rate (α) = 0.05 when genotype relative risk (GRR) >1.4. We have limited power when GRR ≤ 1.2, it could be the reason of some failed replications for T2DM candidate genes in our study.

Compared with studies of T2DM, GWAS and association studies of MVCD are relatively limited21,22,23,24,25,26. So far about half a dozen of GWAS were performed for MVCD (including 3 in East Asian populations), although none of the MVCD associations reached the genome-wide significance threshold of P < 5 × 10−8. In order to subtract the influence of T2DM, we selected as MVCD controls subjects with >10 years history of T2DM who never developed DN or DR. We had only 106 PDR patients in the study, but we still achieved moderate nominal P values for CDKN2A/B and ESR1 gene SNPs. Recently, Sheu et al.22 performed a GWAS for DR in a Chinese population, several loci was associated with DR, although no genome wide association with P < 1 × 10−7 was reached. We have not gotten a chance to test their findings in our population, on the other hand, the TOX and CDKN2A/B polymorphisms were not among their top associations. We also tested candidate genes for DN that identified by GWAS in African Americans21,27, including MYH9, SFI1, and LIMK2, but no association was found for MVCD. Given the sample size of our non-MVCD T2DM controls (266), we only have moderate power for DN and DR association studies. For PDR, the detection power was very limited. On the other hand, we have MYH9 and SFI1 gene well covered in our study (8 SNPs for MYH9 and 3 for SFI1, D’ > 0.6), we have >80% power to detect association for MVCD, DN, and DR when GRRs were more than 1.40, 1.42, 1.47, respectively (alpha = 0.05, dominant model). Compare with T2DM, our power for detection of MVCD associations was moderate, mainly because of the limited sample size of the control group. We cannot rule out the MYH9 and SFI1 associations that found in African Americans, although population heterogeneity could be a major reason of the difference between Han Chinese and African Americans.

In this study, SNPs in TOX and CDKN2A/B genes yielded the most significant associations for both T2DM and MVCD. Interestingly, the A allele of the TOX gene SNP rs1526167 was the “at-risk” allele in T2DM and the “protection” allele for MVCD. Recent research carried out in Beijing (very close to Tianjin, where our subjects were collected) showed that the prevalence of NPDR was 18.6% in Han Chinese T2DM patients6, compared with 28.5% in U.S. T2DM patients, based on NHANES 2005–2008 data28. Prevalence of DR was quite different among U.S. ethnic groups and was higher in African Americans than in European Americans28. To determine whether the “protection” allele of the TOX gene accounts for the lower DR prevalence in Han Chinese MVCD patients, studies with larger sample sizes are needed.

The CDKN2A/B gene is located in the chromosome 9p21 region, which has been highlighted as the strongest genetic susceptibility locus for cardiovascular disease (CVD)29,30 and linked to other conditions such as T2DM30,31, Alzheimer’s disease32, glaucoma33, and periodontitis34. Interestingly, the region of the chromosome associated with CVD and diabetes was previously considered a gene desert. Previous studies found that rs10811661 has a strong correlation with T2DM in French35, Japanese36, and Chinese37 populations. The SNP rs10811661 locates ~100 kb upstream of CDKN2A/B, which has been shown to associate with downregulation of antisense noncoding RNA in INK4 locus (ANRIL) expression38. The ANRIL methylates histone H3K27 by interacting with polycomb proteins, therefore suppresses the expression of INK4a (i.e. CDKN2A)39. Recently, a large sample sized candidate gene association study for T2DM was carried out in Chinese population40, 8 T2DM related genes from previous GWASs were replicated. The CDKN2A/B region SNP, rs10811661, yielded the most significant association (P = 1.11 × 10−8). The SNP was also associated with many quantitative glycemic traits.

At present, more studies have been done on the correlation between CDKN2A/B and the pathogenesis of T2DM than on the relationship between CDKN2A/B and MVCD41,42,43. In the present study, we found associations with both T2DM and MVCD for rs10811661, and the allele T contributed to the increased risk for both diseases. Although associations on the CDKN2A/B (ANRIL) locus were well documented, the mechanism by which this locus affects susceptibility for T2DM and MVCD remains to be investigated.

The TOX gene association for T2DM was first discovered in the present study. The TOX gene, a protein-coding gene located in human chromosome 8, is one of the TOX high-mobility-group proteins. In the present study, we tested 4 SNPs of the TOX gene, rs1526167, rs2726557, rs11777927, and rs17304270. The SNP rs1526167 was not in linkage disequilibrium with other 3 SNPs (Supplement Table 2), it could be the reason for the lack of association of T2DM with those 3 SNPs. The SNP rs17304270, however, was nominally associated with diabetic nephropathy.

The history of the SNP rs1526167 was a little complicated. The SNP rs1526167 located 15Kb downstream of the TOX gene and 1Kb upstream of an unknown function mRNA DL491802. The SNP was first identified by Perlegen Sciences. At the time of the International HapMap Project, that SNP was not included. It could be the reason why rs1526167 was not included in most main stream genome-wide SNP genotyping panels. We checked the LD pattern (in r2) of the chr 8: 59,830,000–59,920,000 region in the HapMap for both Chinese (CHB) and Caucasian (CEU) populations (Supplement Figures 1 and 2): the SNP rs1526167 was located between rs10090702 and rs2726588 (indicated by arrows), and it was located in a separate haplotype block, not in LD with SNPs in the TOX gene coding region and introns.

Although there was no rs1526167 association reported for T2DM, there were some associations found on that SNP for obesity and metabolic syndrome related phenotypes. In a published US patent application (Pub No: US 2006/0177847 A1, Pub Date: Aug 10, 2006), Cox et al. found the TOX gene polymorphism and other 27 DNA sequence variations were associated with Olanzapine treatment emergent weight gain and “metabolic syndrome” in a 1.7 million SNPs genome association study. After checking the DNA sequence provided by the patent application, rs1526167 was among the candidate gene SNPs to screen Olanzapine treatment emergent weight gain and other related traits.

The SNP rs1526167 was associated with obesity in European American extremely obese trios by a transmission disequilibrium test (TDT): P = 2.2 × 10−5; all probands had BMI >35 kg/m2, 428 European American trios (Price et al., unpublished data).

Our results showed that rs1526167 was associated with both T2DM and MVCD, although the risk allele was different for those two conditions. The biological connections between TOX and T2DM are poorly understood, although TOX gene polymorphisms are associated with insulin resistance traits in both Han Chinese (present study) and European Americans.

Since we have not genotyped all reported “positive” SNPs for our candidate genes, we failed to replicate many well established associations. However, the main purpose of this study is not to exclude certain genes from the T2DM/MVCD candidates. We have selected most of our SNPs based on minor allele frequencies (MAF) in Han Chinese, although several SNPs that with higher MAFs in Caucasians were genotyped to verify our previous findings.

In our study, we employed a well-phenotyped, ethnically homogenous population of diabetic subjects, although the number of non-MVCD T2DM controls was relatively limited. A U.S. NHANES epidemiological study showed that prevalence of MVCDs reaches its peak 15 years after onset of T2DM, with almost no new MVCDs developing in individuals with >15 years of T2DM history28. In the present study, we considered individuals with a history of T2DM >10 years, without MVCDs, to be unlikely to carry MVCD susceptibility genes. In our study, the duration of T2DM history in non-MVCD controls was 17.0 ± 5.2 years. We therefore selected these individuals as non-MVCD controls. More controls need to be recruited for association studies with larger power, and much more genotyping needs to be done in the subjects to better understand these relationships between SNPs and MVCDs.

Materials and Methods

Study population and design

We recruited 1,939 T2DM diabetic patients and 918 individuals >57 years old with normal blood glucose levels as nondiabetic controls. We collected older subjects for controls since the late onset of T2DM. All the subjects were unrelated Han Chinese collected from the Metabolic Disease Hospital of Tianjin Medical University, General Hospital of Tianjin Medical University, Tianjin People’s Hospital, and Eye Hospital of Tianjin Medical University. We examined the patients for DR and DN: 836 patients had DN, 398 had nonproliferative DR (NPDR), 106 had proliferative DR (PDR), and 504 had DR; 224 patients had both DN and DR (612 and 280 subjects only had DN or DR, respectively). Overall, 1,116 patients had MVCD (DN or DR). As non-MVCD controls, we used 266 patients with a history of T2DM for >10 years who never developed DR or DN; the average duration of T2DM history in non-MVCD controls was 17.0 ± 5.2 years (Table 4).

Table 4. Basic characteristics of T2DM and MVCD cases and controls.

  T2DM Non-diabetic controls DN DR PDR MVCD Non-MVCD T2DM controls
N 1,939 918 836 504 106 1,116 266
Sex (male/female) 1,040/899 740/178 482/354 257/247 44/62 605/511 124/142
Age (yr, mean ± SD) 58.5 ± 12.0 72.4 ± 9.1 58.8 ± 11.1 57.8 ± 11.1 54.6 ± 11.1 58.7 ± 11.1 63.4 ± 9.3
T2DM duration (yr, mean ± SD) 10.3 ± 7.7 N/A 11.3 ± 7.6 12.6 ± 7.6 13.1 ± 8.2 11.6 ± 7.7 17.0 ± 5.2
Open in a new tab

All subjects gave written informed consent prior to this study, and the protocol was approved by the Committee on Studies Involving Human Beings at Tianjin Medical University. The study was carried out in accordance with the approved guidelines.

We collected the patient’s general information and clinical characteristics, including gender, age, height, weight, biochemistry and lipid profiles, and fasting plasma glucose. Patients were examined for DR and DN. All phenotypes were documented in a Filemaker Pro database. Table 4 presents the basic characteristics of the study population.

Diabetic retinopathy assessment

All patients received a professional fundus examination and fundus photography; the results were checked and graded by two ophthalmologists at the Eye Hospital of Tianjin Medical University. Level of retinopathy was defined according to a new international classification of DR44 developed by the 29th International Congress of Ophthalmology in 2002. This classification comprises five levels: non-DR, mild nonproliferative DR, moderate nonproliferative retinopathy, severe nonproliferative retinopathy and proliferative retinopathy (PDR). If the levels of DR were inconsistent for the two eyes, the worse eye was recorded for the patient. We used as cases both all DR patients and the subset of PDR patients for separate analyses (Table 4).

Diabetic nephropathy measurement

Microalbumin excretion rates were measured for each patient. The diagnostic criteria for DN includes a history of diabetes and microalbumin/creatinine >300 mg/g, or diagnosis by the renal biopsy. All patients with primary glomerular disease and other secondary glomerular diseases were excluded.

Candidate gene selection and genotyping

As candidate genes for case-control association studies, we selected 1) genes associated with T2DM, obesity, MVCD, or insulin resistance in our45 and other previous candidate gene associations and genome wide association studies (GWAS)21,46,47; and 2) genes in biological pathways related to the development of MVCD, glucose and lipid metabolism, or insulin secretion in diabetes. We selected 82 SNPs in 54 candidate genes (Table 3). Minor allele frequencies (MAFs) of Han Chinese, European American, and global populations were taken from dbSNP ( http://www.ncbi.nlm.nih.gov/snp/). For previously reported associations, we selected SNPs with the most significant association rather than genotyping the whole gene. For less studied genes, multiple SNPs were chosen based on the linkage disequilibrium pattern of the gene (D’ > 0.6). In this study, we have not genotyped all tagged SNPs in certain candidate genes due to limited resources. We also performed gene-specific power calculations for IRS1, TCF7L2, KCNQ1, and FTO genes based on real minor allele frequencies (MAF) in our data set (Supplement Table 1). Linkage disequilibrium among candidate gene SNPs (in D’ and r2) was calculated by Haploview48, results were shown as Supplement Table 2. Linkage disequilibrium (LD) patterns (in r2)of the TOX gene region SNPs were shown as Supplement Figure 1 (Chinese, CHB) and Supplement Figure 2 (Caucasian, CEU).

Genomic DNA samples were extracted from peripheral whole blood samples using the high-salt method. Genotyping was performed by primer extension of multiplex products with detection by matrix-assisted laser desorption time-of-flight mass spectrometry.

Association Studies

The Hardy-Weinberg equilibrium (HWE) test was performed before the association analysis (Table 3). Statistical analyses for phenotypes were performed by SPSS, version 17.0. The allelic frequencies between the case group and the control group were compared by chi tests using PLINK49, and odds ratios (ORs) with 95% confidence intervals (CIs) are presented.

Association studies were carried out in two stages: 1) using 1,939 T2DM patients as cases and 918 nondiabetic individuals as controls, to test associations for T2DM; and 2) using 1,116 MVCD patients as cases and 266 patients with a history of T2DM for >10 years, but without MVCDs, as controls, to test associations for MVCDs. Association studies were performed separately for DN, DR, PDR, and MVCD.

Conclusions

In summary, our case-control studies suggest that TOX and CDKN2A/B gene SNPs are associated with T2DM, DN, DR, and MVCD in Han Chinese. A large prospective study is needed to confirm these associations in Han Chinese. A better understanding of genetic factors predisposing individuals to diabetic complications would help identify diabetic patients at risk and also to reveal the pathogenesis of MVCD.

Additional Information

How to cite this article: Wei, F. et al. TOX and CDKN2A/B Gene Polymorphisms Are Associated with Type 2 Diabetes in Han Chinese. Sci. Rep. 5, 11900; doi: 10.1038/srep11900 (2015).

Supplementary Material

Supplementary Information
srep11900-s1.pdf (1.9MB, pdf)

Acknowledgments

This study was supported by Chinese National Natural Science Foundation of China grants 81070576 (W.D.L.), 81072922 (L.C.) and by grant 12JCZDJC24700 from Tianjin Municipal Science and Technology Commission (W.D.L.). We thank all of the patients for their cooperation and also the clinical doctors of Metabolic Diseases Hospital.

Footnotes

Author Contributions W.D.L., L.C. and J.L. conceived and designed the study; W.D.L., S.L. and F.W. wrote the manuscript; F.W., C.C., J.L., W.S., C.L., P.Y., H.J., F.Y., M.L., Y.W., L.Z., J.S. and M.C. performed the experiments; F.W., C.C., J.L., H.J., F.Y., W.S., P.Y., C.L. and W.D.L. analyzed the data; S.F., B.C., H.Z., H.H., Y.C., N.S., J.T., L.S., J.L., R.L., H.Q., S.L., J.L. and L.C. collected subjects and clinical data.

References

  1. Xu Y. et al. Prevalence and control of diabetes in Chinese adults. JAMA 310, 948–959, 10.1001/jama.2013.168118 (2013). [DOI] [PubMed] [Google Scholar]
  2. Yang W. et al. Prevalence of diabetes among men and women in China. N Engl J Med 362, 1090–1101, 10.1056/NEJMoa0908292 (2010). [DOI] [PubMed] [Google Scholar]
  3. Cho Y. S. et al. Meta-analysis of genome-wide association studies identifies eight new loci for type 2 diabetes in east Asians. Nat Genet 44, 67–72, 10.1038/ng.1019 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Martini S., Eichinger F., Nair V. & Kretzler M. Defining human diabetic nephropathy on the molecular level: integration of transcriptomic profiles with biological knowledge. Rev Endocr Metab Disord 9, 267–274, 10.1007/s11154-008-9103-3 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Thylefors B., Negrel A. D., Pararajasegaram R. & Dadzie K. Y. Global data on blindness. Bull World Health Organ 73, 115–121 (1995). [PMC free article] [PubMed] [Google Scholar]
  6. Yuan M. X. et al. Low prevalence of diabetic retinopathy in a Chinese population. Diabetes care 35, e61, 10.2337/dc12-0600 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Freeman M. S., Mansfield M. W., Barrett J. H. & Grant P. J. Heritability of features of the insulin resistance syndrome in a community-based study of healthy families. Diabet Med 19, 994–999 (2002). [DOI] [PubMed] [Google Scholar]
  8. Stratton I. M. et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 321, 405–412 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hietala K., Forsblom C., Summanen P. & Groop P. H. Heritability of proliferative diabetic retinopathy. Diabetes 57, 2176–2180, 10.2337/db07-1495 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Quinn M., Angelico M. C., Warram J. H. & Krolewski A. S. Familial factors determine the development of diabetic nephropathy in patients with IDDM. Diabetologia 39, 940–945 (1996). [DOI] [PubMed] [Google Scholar]
  11. Seaquist E. R., Goetz F. C., Rich S. & Barbosa J. Familial clustering of diabetic kidney disease. Evidence for genetic susceptibility to diabetic nephropathy. N Engl J Med 320, 1161–1165 (1989). [DOI] [PubMed] [Google Scholar]
  12. Wong T. Y. et al. Diabetic retinopathy in a multi-ethnic cohort in the United States. Am J Ophthalmol 141, 446–455, 10.1016/j.ajo.2005.08.063 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Pettitt D. J., Saad M. F., Bennett P. H., Nelson R. G. & Knowler W. C. Familial predisposition to renal disease in two generations of Pima Indians with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 33, 438–443 (1990). [DOI] [PubMed] [Google Scholar]
  14. Rema M., Saravanan G., Deepa R. & Mohan V. Familial clustering of diabetic retinopathy in South Indian Type 2 diabetic patients. Diabet Med 19, 910–916, 820 (2002). [DOI] [PubMed] [Google Scholar]
  15. Hu C. et al. CPVL/CHN2 genetic variant is associated with diabetic retinopathy in Chinese type 2 diabetic patients. Diabetes 60, 3085–3089, 10.2337/db11-0028 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hu C. et al. PPARG, KCNJ11, CDKAL1, CDKN2A-CDKN2B, IDE-KIF11-HHEX, IGF2BP2 and SLC30A8 are associated with type 2 diabetes in a Chinese population. PloS one 4, e7643, 10.1371/journal.pone.0007643 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hu C. et al. Variations in KCNQ1 are associated with type 2 diabetes and beta cell function in a Chinese population. Diabetologia 52, 1322–1325, 10.1007/s00125-009-1335-6 (2009). [DOI] [PubMed] [Google Scholar]
  18. Yoon K. H. et al. Epidemic obesity and type 2 diabetes in Asia. Lancet 368, 1681–1688, 10.1016/S0140-6736(06)69703-1 (2006). [DOI] [PubMed] [Google Scholar]
  19. Torrens J. I. et al. Ethnic differences in insulin sensitivity and beta-cell function in premenopausal or early perimenopausal women without diabetes: the Study of Women’s Health Across the Nation (SWAN). Diabetes care 27, 354–361 (2004). [DOI] [PubMed] [Google Scholar]
  20. Chan J. C. et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. JAMA 301, 2129–2140, 10.1001/jama.2009.726 (2009). [DOI] [PubMed] [Google Scholar]
  21. McDonough C. W. et al. A genome-wide association study for diabetic nephropathy genes in African Americans. Kidney Int 79, 563–572, 10.1038/ki.2010.467 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sheu W. H. et al. Genome-wide association study in a Chinese population with diabetic retinopathy. Hum Mol Genet 22, 3165–3173, 10.1093/hmg/ddt161 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hanson R. L. et al. Identification of PVT1 as a candidate gene for end-stage renal disease in type 2 diabetes using a pooling-based genome-wide single nucleotide polymorphism association study. Diabetes 56, 975–983, 10.2337/db06-1072 (2007). [DOI] [PubMed] [Google Scholar]
  24. Shimazaki A. et al. Genetic variations in the gene encoding ELMO1 are associated with susceptibility to diabetic nephropathy. Diabetes 54, 1171–1178, 54/4/1171 (2005). [DOI] [PubMed] [Google Scholar]
  25. Huang Y. C. et al. Genome-wide association study of diabetic retinopathy in a Taiwanese population. Ophthalmology 118, 642–648, 10.1016/j.ophtha.2010.07.020 (2011). [DOI] [PubMed] [Google Scholar]
  26. Fu Y. P. et al. Identification of Diabetic Retinopathy Genes through a Genome-Wide Association Study among Mexican-Americans from Starr County, Texas. J Ophthal 2010, 10.1155/2010/861291 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Palmer N. D. et al. Evaluation of candidate nephropathy susceptibility genes in a genome-wide association study of african american diabetic kidney disease. PloS one 9, e88273, 10.1371/journal.pone.0088273 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Zhang X. et al. Prevalence of diabetic retinopathy in the United States, 2005-2008. JAMA 304, 649–656, 10.1001/jama.2010.1111 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447, 661–678, 10.1038/nature05911 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Broadbent H. M. et al. Susceptibility to coronary artery disease and diabetes is encoded by distinct, tightly linked SNPs in the ANRIL locus on chromosome 9p. Hum Mol Genet 17, 806–814, 10.1093/hmg/ddm352 (2008). [DOI] [PubMed] [Google Scholar]
  31. Cugino D. et al. Type 2 diabetes and polymorphisms on chromosome 9p21: a meta-analysis. Nutr Metab Cardiovasc Dis 22, 619–625, 10.1016/j.numecd.2010.11.010 S0939-4753(10)00294-2 (2012). [DOI] [PubMed] [Google Scholar]
  32. Emanuele E. et al. Chromosome 9p21.3 genotype is associated with vascular dementia and Alzheimer’s disease. Neurobiol Aging 32, 1231–1235, 10.1016/j.neurobiolaging.2009.07.003 S0197-4580(09)00233-4 (2011). [DOI] [PubMed] [Google Scholar]
  33. Burdon K. P. et al. Genome-wide association study identifies susceptibility loci for open angle glaucoma at TMCO1 and CDKN2B-AS1. Nat Genet 43, 574–578, 10.1038/ng.824 (2011). [DOI] [PubMed] [Google Scholar]
  34. Schaefer A. S. et al. Identification of a shared genetic susceptibility locus for coronary heart disease and periodontitis. PLoS Genet 5, e1000378, 10.1371/journal.pgen.1000378 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Duesing K. et al. Strong association of common variants in the CDKN2A/CDKN2B region with type 2 diabetes in French Europids. Diabetologia 51, 821–826, 10.1007/s00125-008-0973-4 (2008). [DOI] [PubMed] [Google Scholar]
  36. Takeuchi F. et al. Confirmation of multiple risk Loci and genetic impacts by a genome-wide association study of type 2 diabetes in the Japanese population. Diabetes 58, 1690–1699, 10.2337/db08-1494 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wu Y. et al. Common variants in CDKAL1, CDKN2A/B, IGF2BP2, SLC30A8, and HHEX/IDE genes are associated with type 2 diabetes and impaired fasting glucose in a Chinese Han population. Diabetes 57, 2834–2842, 10.2337/db08-0047 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Cunnington M. S., Santibanez Koref M., Mayosi B. M., Burn J. & Keavney B. Chromosome 9p21 SNPs Associated with Multiple Disease Phenotypes Correlate with ANRIL Expression. PLoS Genet 6, e1000899, 10.1371/journal.pgen.1000899 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yap K. L. et al. Molecular interplay of the noncoding RNA ANRIL and methylated histone H3 lysine 27 by polycomb CBX7 in transcriptional silencing of INK4a. Mol Cell 38, 662–674, 10.1016/j.molcel.2010.03.021 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zhao Q. et al. Cross-sectional and longitudinal replication analyses of genome-wide association loci of type 2 diabetes in Han Chinese. PloS one 9, e91790, 10.1371/journal.pone.0091790 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lewis J. P. et al. Association analysis in african americans of European-derived type 2 diabetes single nucleotide polymorphisms from whole-genome association studies. Diabetes 57, 2220–2225, 10.2337/db07-1319 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Horikoshi M. et al. Variations in the HHEX gene are associated with increased risk of type 2 diabetes in the Japanese population. Diabetologia 50, 2461–2466, 10.1007/s00125-007-0827-5 (2007). [DOI] [PubMed] [Google Scholar]
  43. Steinthorsdottir V. et al. A variant in CDKAL1 influences insulin response and risk of type 2 diabetes. Nat Genet 39, 770–775, ng2043 10.1038/ng2043 (2007). [DOI] [PubMed] [Google Scholar]
  44. Wilkinson C. P. et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology 110, 1677–1682, S0161-6420(03)00475-5 (2003). [DOI] [PubMed] [Google Scholar]
  45. Wang K. et al. A genome-wide association study on obesity and obesity-related traits. PloS one 6, e18939, 10.1371/journal.pone.0018939 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Grant S. F. et al. Variant of transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet 38, 320–323, 10.1038/ng1732 (2006). [DOI] [PubMed] [Google Scholar]
  47. Voight B. F. et al. Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis. Nat Genet 42, 579–589, 10.1038/ng.609 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Barrett J. C., Fry B., Maller J. & Daly M. J. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21, 263–265, 10.1093/bioinformatics/bth457 (2005). [DOI] [PubMed] [Google Scholar]
  49. Purcell S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81, 559–575, 10.1086/519795 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
srep11900-s1.pdf (1.9MB, pdf)

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES