A Variant of the Transcription Factor 7-Like 2 (TCF7L2) Gene and the Risk of Posttransplantation Diabetes Mellitus in Renal Allograft Recipients

A total of 681 unrelated transplant recipients were recruited between 1989 and 2006. PTDM was diagnosed according to American Diabetes Association criteria (21) after the third posttransplantation month, and patients who began and continued an antidiabetes medication (oral medication or insulin) after transplantation were included in the PTDM group. The remaining patients belonged to the non-PTDM group. No patients had any previous diagnosis of diabetes or a recorded fasting plasma glucose (FPG) level <100 mg/dl. Patients were eligible to participate in the study if they were the recipients of a kidney allograft with no previous history of organ transplantation.

In accordance with our previous study (11), both persistent PTDM (diagnosed in patients who developed diabetes within 1 year following transplantation and remained diabetic) and late PTDM (diagnosed in patients who developed diabetes later than 1 year posttransplantation) patients were classified in the PTDM group. Transient PTDM patients (who developed diabetes during the first year following transplantation but eventually recovered to normoglycemia without medication) were classified as non-PTDM.

Patients were excluded if they had a history of diabetes before transplantation, had severe metabolic or infectious disease, had received multiple organ transplants, or had repeated kidney transplants. A total of 511 unrelated renal allograft recipients were enrolled in this study. The study protocol was approved by the ethics committee of the Yonsei University College of Medicine. All subjects were provided adequate information about the study and gave informed consent.

The main immunosuppressive regimens consisted of calcineurin inhibitors (cyclosporine A or tacrolimus) and glucocorticoids. Immunosuppressive regimens and schedules were as reported previously (11) (online appendix [available at http://dx.doi.org/10.2337/dc07-1005]).

Anthropometric measurements were taken using standard techniques at the time of transplantation, then 3, 6, and 12 months after transplantation. All measurements were taken using the same equipment and by the same personnel. All samples were taken the morning after overnight fasting. FPG level was determined using an enzymatic colorimetric assay.

Genomic DNA was isolated from peripheral blood lymphocytes. The genotyping of TCF7L2 polymorphisms (rs4506565, rs7901695, rs7903146, rs11196205, rs12243326, and rs12255372) was performed using the TaqMan fluorogenic 5′ nuclease assay (Applied Biosystems, Foster City, CA). These polymorphisms were chosen from a screen of 15 single nucleotide polymorphisms (SNPs) and were selected because their minor allele frequency (MAF) was >2% or because they were reported to be associated with type 2 diabetes in previous studies (12–19,22–27).

Specific methods are shown in the online appendix. Duplicate samples and negative controls were included to ensure the accuracy of genotyping. On average, 99.3% of attempted genotypings were successful (success rates from 98.8 to 99.6% for each SNP).

Statistical analyses were performed using SPSS for Windows software (version 12.0; SPSS, Chicago, IL). All continuous variables were expressed as the mean ± SD. The genotype frequencies were tested for Hardy-Weinberg equilibrium using a χ² test. The Student's t test was used to compare the continuous variables between the PTDM and non-PTDM groups. To control for age and sex effects, multiple regression and logistic regression tests were used. The Pearson's χ² test was used to evaluate differences in the incidence of diabetes between genotypes. A multivariable logistic regression test was used to identify risk factors for PTDM development and calculate the adjusted odds ratio and 95% CIs. Pairwise linkage disequilibrium between TCF7L2 SNPs was assessed and patient baseline characteristics were assessed on the transplant day. A P value <0.05 was considered statistically significant.

Overall PTDM incidence in this study population was 23.3%. Baseline clinical characteristics of the population are shown in Table 1. The mean age of patients at transplantation was 36.9 ± 10.7 years. Patients in the PTDM group were older than those in the non-PTDM group (aged 41.1 ± 9.3 vs. 35.6 ± 10.8 years; P = 0.001). Follow-up duration was longer in the PTDM group compared with the non-PTDM group (P = 0.003). Although initial mean body weight was not different, patients in the PTDM group gained more weight than those in the non-PTDM group after 3 and 6 months following transplantation. These differences remained significant after adjustment for age and sex (Table 1). Initial FPG levels were not significantly different, but FPG levels between the two groups were significantly different at 3, 6, and 12 months after transplantation, despite antidiabetes treatment (Table 1). The duration of dialysis, incidence of acute rejection, percentage of tacrolimus use as an immunosuppressive agent, and serum creatinine levels were not different between the two groups (Table 1).

Genotype distribution was in agreement with Hardy-Weinberg equilibrium (Table 2 and online appendix Table A1). Because rs4506565, rs7901695, and rs7903146 were in complete linkage (D′ = 1, r² = 1) (online appendix Table A2) and the MAFs of rs11196205 (MAF = 0.023), rs12243326 (MAF = 0.002), and rs12255372 (MAF = 0.006) were too small to perform statistical analysis, only rs7903146 was further investigated (online appendix Table A1). The rs7903146 CC genotype was present in 94.3% of the samples and the CT genotype in 5.7% (Table 2).

PTDM developed in 119 patients (23.3%). The distribution of TCF7L2 genotypes was significantly different between patients with and without diabetes. The CT genotype was more common in patients with PTDM (10.1%) than without (4.3%). The incidence of PTDM was significantly higher in patients with the CT genotype than in patients with the CC genotype (41.4 vs. 22.2%) (odds ratio 2.474 [95% CI 1.146–5.341]; P = 0.024) (Table 3). Unlike other studies on TCF7L2, no patients with the TT genotype were identified in this study. There was no difference between genotypes with regard to age at transplantation; initial body weight; FPG at baseline, 3, 6, and 12 months after transplantation; duration of dialysis; percentage of initial tacrolimus use; incidence of acute rejection; and serum creatinine levels (Table 3).

Multivariable logistic regression tests revealed that age at transplantation, amount of body weight gain during the first 6 months, and TCF7L2 genotype are important risk factors of the development of PTDM (Table 4). The effect of the genotype remains significant after multivariable logistic regression for PTDM after adjustment for age and sex (model 1, P = 0.012), after adjustment for age, sex, and amount of weight gain (model 2, P = 0.016), and after adjustment for age, sex, amount of weight gain, and type of immunosuppressant use (model 3, P = 0.020). Although male sex and use of tacrolimus seemed to be risk factors for PTDM, they were not statistically significant (Table 4).

Various risk factors of the development of PTDM have been studied, but there are few reports on its genetic risks. We have previously reported that defects in insulin secretion play a pivotal role in the pathogenesis of PTDM (10,11). Many recent studies have suggested that TCF7L2 may play a role in insulin secretion (17,18). Therefore, we investigated the genetic influence of the TCF7L2 polymorphism on the development of PTDM in a renal transplant cohort.

We initially selected six SNPs in the TCF7L2 gene that are reported to be associated with type 2 diabetes in many populations (12–19,22–27). However, MAFs were significantly lower in this cohort than in populations of European ancestry. Moreover, we did not observe minor allele homozygotes for rs4506565, rs12243326, rs7903146, rs12255372, or rs7901695. This result is consistent with recently reported haplotype structures in East Asians (23). The frequency of the rs7903146 CT genotype was only 5.7% in the study population herein compared with up to 48% in Caucasian populations. This frequency is similar to that seen in Japan, where the heterozygote accounts for 8.47% (27).

A number of previous reports showed that the TCF7L2 rs7903146 variant is associated with type 2 diabetes in the general population. In this study, TCF7L2 rs7903146 was significantly associated with the development of PTDM. The incidence of PTDM was significantly higher in patients with the CT genotype than in patients with the CC genotype (41.4 vs. 22.2%) (odds ratio 2.474 [95% CI 1.146–5.341]; P = 0.024).

In monovariable analysis, age, the amount of body weight gain, and the TCF7L2 genotype were associated with the development of PTDM. The effect of genotype remains significant after multivariable logistic regression for PTDM. Sex and type of immunosuppressant were not shown to be independent risk factors. These results suggest that the TCF7L2 gene is one of the susceptibility genes for PTDM.

Patients with PTDM gained more weight during the follow-up period. This difference may be due to medications (e.g., insulin or thiazolidinediones) rather than reduced kidney function in PTDM patients or older age (online appendix Table A3). It is unlikely that PTDM patients had less improved kidney function, as there was no significant difference in serum creatinine levels between the two groups. It is also unlikely that the difference in weight gain is simply due to the older age of the PTDM patients since, after adjustment for age and sex, the difference in body weight gain remained significant (Table 1).

TCF7L2 is a novel type 2 diabetes susceptibility gene that confers up to a two-fold increase in the risk of developing type 2 diabetes. The TCF7L2 polymorphism is considered to be the most powerfully associated polymorphism with type 2 diabetes to date (18). Our data suggest that TCF7L2 variation plays an important role in the development of PTDM and type 2 diabetes.

A recent report suggests that this genetic variation increases TCF7L2 expression in the β-cell, reducing insulin secretion and predisposing the subject to diabetes (20). There are few studies on the genetic risk factors for PTDM to date. Bamouulid et al. (28) reported that an interleukin-6 promoter polymorphism is associated with a lower risk of PTDM in 349 renal allograft patients. Numakura et al. (29) reported that a vitamin D receptor (VDR) gene polymorphism is associated with PTDM in 70 renal allograft recipients. Interestingly, VDR polymorphism is also reported to have an effect on insulin secretion (30,31). These results are consistent with our previous reports (10,11) showing that defects in insulin secretion play a more crucial role in the pathogenesis of PTDM than increased insulin resistance.

Although many clinical studies (3,32,33) have indicated that tacrolimus is approximately five times more diabetogenic than cyclosporine, there was no difference in the incidence of PTDM according to the kind of immunosuppressive agents used in this study. This is probably because the doses of calcineurin inhibitors and glucocorticoids used in this study were different.

A limitation of this study is that an oral glucose tolerance test was not routinely performed before transplantation. It is possible that preexisting impaired glucose tolerance leads to overdiagnosis of PTDM. While genetic variation in TCF7L2 is associated with PTDM in the current study, the contribution of this genetic variation to the development of PTDM in Asians is relatively modest because of the low MAF. The T allele frequency was 2.84% in the Korean population, which is comparable with such frequencies in Chinese individuals from Beijing (2.2%) or Japanese individuals from Tokyo (2.3%) and in contrast to those of Caucasian populations (25%) and African populations (29.2%) in the HapMap database (34). There might be a more substantial impact on Caucasian populations in which MAF is reported to be ∼25%.

In conclusion, our study results suggest that the TCF7L2 rs7903146 variant is associated with an increased risk of PTDM. To our knowledge, this study is the first and the largest genetic study to investigate the association between the TCF7L2 gene polymorphism and PTDM. The development of PTDM after renal allograft is a critical factor for quality of life and graft survival. Therefore, to minimize the risk of PTDM development in patients carrying this high-risk genotype, it may be critical to consider the use of less diabetogenic immunosuppressants, encourage weight reduction and lifestyle modification, utilize a rapid steroid tapering schedule, or pursue pharmacologic prevention.

This study was supported by the Global 5-5-10 Project, Yonsei University (6-2007-0087).