A Polymorphism in the Zinc Transporter Gene SLC30A8 Confers Resistance Against Posttransplantation Diabetes Mellitus in Renal Allograft Recipients

A total of 805 unrelated transplant recipients were recruited from 1989 to 2007. PTDM was diagnosed according to American Diabetes Association criteria (26) at the 3rd month posttransplantation. Patients who were started on antidiabetic medication (oral medication or insulin) after transplantation and continued the medication thereafter were included in the posttransplantation diabetic group. The rest of the patients were assigned to the non- posttransplantation diabetic group. According to our previous study (15), cases of persistent PTDM (patients who developed diabetes within 1 year posttransplantation and remained diabetic) and late PTDM (patients who developed diabetes after 1 year posttransplantation) were assigned to the posttransplantation diabetic group. Transient PTDM cases (patients who developed diabetes within 1 year posttransplantation but eventually recovered to normoglycemia without medication) were classified as non-PTDM.

Patients were eligible to participate in the study if they were the recipients of a kidney allograft with no previous history of organ transplantation. Those with no previous diagnosis of diabetes and a recorded fasting plasma glucose (FPG) level <5.5 mmol/l were included. Patients were excluded if they had a history of diabetes before transplantation, severe metabolic or infectious disease, or a recorded FPG level ≥5.5 mmol/l. A total of 624 unrelated renal allograft recipients were enrolled in this study. Medical histories were obtained during the initial visit, and family history of diabetes refers to first-degree relatives only. Height and weight measurements were taken at the time of transplantation and at 3 and 6 months after transplantation. Blood samples were collected after an overnight fast. Frequency of patient follow-up and data collected at follow-up visits are shown in the online appendix (available at http://dx.doi.org/10.2337/db07-0761). Initial FPG level was measured on the day of transplantation using an enzymatic colorimetric assay. Insulin concentration was measured using a radioimmunoassay kit (Dainabot, Tokyo, Japan). Markers of insulin sensitivity and β-cell insulin secretory function were determined by the homeostasis model assessment (HOMA) of insulin resistance (HOMA-IR) and β-cell function (HOMA-β) (27). HOMA-IR was calculated as (fasting plasma insulin [μU/ml] × FPG [mmol/l])/22.5 and HOMA-β as (20 × fasting plasma insulin [μU/ml])/(FPG [mmol/l] − 3.5). The study protocol was approved by the ethics committee of the Yonsei University College of Medicine. All subjects were provided with adequate information about this study and gave informed consent.

The immunosuppressive regimens consisted of calcineurin inhibitors and glucocorticoids. Immunosuppressive regimens and schedules were as reported previously (15) (online appendix).

Genomic DNA was isolated from peripheral blood lymphocytes. The genotyping of the SLC30A8 polymorphism rs13266634 was done using the TaqMan fluorogenic 5′ nuclease assay (ABI, Foster City, CA). The final reaction volume for the PCR was 5 μl, containing 10 ng genomic DNA, 2.5 μl TaqMan Universal PCR Master Mix, and 0.13 μl 40× Assay Mix (assay ID C2684958_10). Thermal cycling conditions were as follows: 50°C for 2 min to activate the uracil N-glycosylase and prevent carry-over contamination, 95°C for 10 min to activate the DNA polymerase, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min. All PCRs were performed using 384-well plates by a Dual 384-Well GeneAmp PCR System 9700 (ABI), and the endpoint fluorescent readings were performed on an ABI PRISM 7900 HT Sequence Detection System. Forty-eight duplicate samples and negative controls were included to ensure the accuracy of genotyping, and 100% of the duplicates replicated the original genotype. The genotype success rate was 99.48% for rs13266634.

The genotype frequencies were tested for Hardy-Weinberg equilibrium using χ² test. All continuous variables were expressed as means ± SD. Student’s t test was used to compare continuous variables between the PTDM and non-PTDM posttransplantation diabetic groups. Pearson’s χ² test was used to evaluate the difference in the incidence of diabetes between genotypes. ANOVA was used to compare continuous variables among genotypes. Mantel-Haenzel χ² test (P for trend) was used to evaluate the trend of diabetes protection with an increase in number of W-alleles. Cox proportional hazards modeling was used to identify risk factors for PTDM development and calculate the adjusted hazard ratio [HR] and 95% CIs. Diabetes-free survival was calculated from the date of transplantation to the date of PTDM diagnosis or last follow-up. Survival was estimated by Kaplan-Meier method and compared using the log-rank test. Patient baseline characteristics were assessed on transplant day. A P value <0.05 was considered statistically significant. Statistical analyses were performed using SPSS for Windows software (version 12.0; SPSS, Chicago, IL). Power calculations were performed using PASS software version 2005 (NCSS Statistical Software, Kaysville, UT).

The overall incidence of PTDM in this study population was 27.9%. Baseline clinical characteristics of the recipients are shown in Table 1. The mean age of patients at transplantation was 38.7 ± 9.5 years. Patients in the posttransplantation diabetic group were older than those in the non-posttransplantation diabetic group (42.10 ± 8.99 vs. 37.42 ± 9.43 years old, P = 0.001). Posttransplantation diabetic patients were followed up for a significantly longer period than non-posttransplantation diabetic patients. Baseline mean body weight was 58.83 ± 9.71 kg, and although initial body weight was not significantly different between the two groups, there was a significant difference in the amount of weight gain during the first 6 months after transplantation between the two groups (Table 1). The duration of dialysis, incidence of acute rejection, and serum creatinine levels were not different between the groups (Table 1). Patients who used tacrolimus as an immunosuppressive agent had a higher incidence of PTDM (P = 0.031). There was no significant difference in steroid dose between the groups (online appendix).

The overall R-allele frequency was 0.577 and the overall W-allele frequency was 0.423. Overall genotype distributions were R/R, 33.17%; R/W, 49.04%; and W/W, 17.79%. Genotype distributions were in agreement with Hardy-Weinberg equilibrium (χ² = 0.013, P = 0.910) (Table 2).

PTDM developed in 174 patients (27.2%). The W/W genotype conferred a reduced risk for PTDM compared with the R/R genotype (risk ratio [RR] 0.52 [95% CI 0.32–0.83], P = 0.007; Table 2). The risk of PTDM was marginally lower in carriers of the R/W genotype than in R/R carriers (0.78 [0.57–1.07], P = 0.126, Table 2). The W-allele was associated with a decreased incidence of PTDM in codominant (R/R, 33.8%; R/W 26.8%; W/W, 19.8%; P = 0.025), dominant (R/R, 33.8%; R/W plus W/W, 24.9%; P = 0.024), and recessive (R/R plus R/W, 29.6%; W/W, 19.8%; P = 0.038) models (Table 2).

There was no difference among genotypes in age at transplantation, initial body weight, amount of body weight gain, FPG level (at baseline and at 3, 6, and 12 months after transplantation), duration of dialysis, percentage of tacrolimus use as an immunosuppressant, incidence of acute rejection, or serum creatinine levels (Table 3).

There was a linear increase in the HOMA-β value from the R/R genotype to the W/W genotype. Insulin secretory function, assessed by HOMA-β, was better preserved in patients with the W/W genotype than in those with the R/R genotype (Table 3). In contrast, no significant difference in HOMA-IR value according to genotype was observed (Table 3).

Independent risk factors of PTDM were identified using multiple Cox proportional hazard regression. We included age, sex, amount of body weight gain, type of immunosuppressant, and genotype as covariates in the multiple Cox proportional hazard regression model because they were associated with PTDM in univariate association tests (Table 1). As shown in Table 4, the SLC30A8 genotype was a significant independent risk factor of PTDM together with age at transplantation and tacrolimus use as an immunosuppressant. The effect of the genotype remained significant after adjustments for age, sex, amount of weight gain, and type of immunosuppressant used (R/W, HR 0.77, P = 0.114; W/W, HR 0.58, P = 0.026). Age, amount of weight gain, and tacrolimus use were associated with PTDM development, but sex was not (Table 4).

The Kaplan-Meier curve for the development of PTDM is shown in Fig. 1. Patients with the W/W genotype had a significantly lower risk (P = 0.007) of developing PTDM than those with the R/R genotype.

Zinc is thought to be an important metal for insulin storage and secretion (16). Zinc homeostasis is regulated by the uptake and export of zinc by specialized transporter proteins. Insulin is stored inside secretory granules as solid hexamers bound with two Zn²⁺ ions per hexamer (17,18). Zinc is likely to play a role in the stability of the insulin hexamer and to modulate glucagon secretion in α-cells via paracrine action.

The SLC39 proteins (Zips) control the intracellular uptake of zinc (28), and SLC30 proteins (ZnTs) control the cellular efflux of zinc into the extracellular matrix or intracellular vesicles (29). Among ZnTs, ZnT-8 has recently been identified and cloned as a pancreatic β-cell–specific ZnT and is located in insulin secretory granules (19–21). ZnT-8 is encoded by the gene SLC30A8, which is located at chromosome 8q24.11. The polymorphism rs1326634 in the SLC30A8 gene has been associated with type 2 diabetes (22–25). In addition, recent studies have suggested that SLC30A8 may play a role in insulin secretion in the process of zinc incorporation into insulin vesicles (19,20).

In this study, we demonstrated the genetic influence of the SLC30A8 polymorphism in the development of PTDM in a renal transplant cohort. We calculated the relative risks of the heterozygote R/W genotype and the mutant homozygote W/W genotype compared with the wild homozygote R/R genotype because the risk allele was the major allele (R) in a previously reported French study (22). We found the risk of PTDM to be significantly lower in patients who were W/W homozygotes than in patients who were R/R wild-type homozygotes. There was also a tendency toward a reduction in the risk of PTDM in R/W heterozygote carriers compared with R/R carriers. We observed a linear relationship between the number of R-alleles and an increased incidence of PTDM. This genotype effect remained significant after adjustments for age, sex, amount of body weight gain, and type of immunosuppressant use. These results suggest that the SLC30A8 gene is one susceptibility gene for PTDM. Age at transplantation and tacrolimus use as an immunosuppressant were revealed as important risk factors for PTDM.

Interestingly, there was a higher mean HOMA-β value in patients carrying the W/W genotype than in those carrying the R/R genotype. These results suggest that this polymorphism may be functional and have effects on insulin secretion. There was no significant difference in HOMA-IR value according to genotype. Numakura et al. (13) reported that a vitamin D receptor (VDR) polymorphism was associated with PTDM in 70 renal allograft recipients. The VDR polymorphism is also reported to affect insulin secretion (30–32). Their results and the results of this study are consistent with our previous reports (10,15), in that a defect in insulin secretion plays a more crucial role than increased insulin resistance in the pathogenesis of PTDM.

This SLC30A8 polymorphism site is located at the COOH-terminal region of the protein structure. The rs1326634 SNP results in a nonsynonymous mutation because the nucleotide change from C to T results in an amino acid change from arginine (R) to tryptophan (W) at position 325. This polymorphism might be a gain-of- function mutation that increases protein expression levels, thereby enhancing zinc transport into insulin secretory granules. Facilitated zinc transport could influence insulin stability and secretion. Another possibility is that the polymorphism may affect the posttranslational modification mechanism in the COOH terminus of ZnT-8. Until now, information on ZnT-interacting proteins or phosphorylation events in the intracellular COOH termini of all the ZnT subfamily members has been lacking. Interestingly, the amino acid stretch (TAASR*DSQVV) constituting R325 in the COOH terminus of ZnT-8 has a protein kinase A and protein kinase C recognition motif (R-X-S/T). The R325W polymorphism would disrupt this motif, which may possibly prevent the receptor phosphorylation otherwise regulated by serine protein kinases, leading to changes in the transporter function. There is much that still needs to be learned about the regulation of ZnT, and the exact molecular mechanisms remain to be elucidated.

A limitation of this study is that oral glucose tolerance tests were not routinely performed before transplantation. Preexisting diabetes could lead to an overestimation in the diagnosis of PTDM. There may also be a bias toward the null because non-posttransplantation diabetic patients were followed up for a shorter duration and may have been misclassified. Another limitation is the relatively small sample size. Thus, our study had 51% power to detect an inverted reference odds ratio of 0.65 (1/1.53) at α = 0.05 (22). Further analyses are warranted to replicate the associations in larger populations.

In conclusion, our data suggest that the SLC30A8 rs1326634 variant is associated with a decreased risk of PTDM and may be associated with insulin secretion. The exact molecular mechanisms still need to be clarified.

This work has received a Korea Research Foundation grant funded by the Korean Government (MOEHRD, Basic Research Promotion Fund) (KRF-2007-331-E00081).