MDM2 SNP309 Polymorphism as Risk Factor for Susceptibility and Poor Prognosis in Renal Cell Carcinoma

Materials and Methods

Samples. A total of 200 patients (140 male and 60 female) with pathologically confirmed conventional RCC and 200 age- and sex-matched control individuals was enrolled in this study. The mean ages of the patient and control groups were 64 years (range, 29-87) and 63 years (range, 32-89), respectively (P = 0.44; Table 1 ).

To adjust for age, the number of subjects in two age groups was matched between cases and controls (P = 1.0; Table 1). All of the patients tested were diagnosed with RCC based on histopathologic findings. They were classified according to the WHO criteria and staged according to the tumor-node-metastasis classification. Healthy controls consisted of volunteers with no apparent abnormal findings on medical examination at Shimane University Hospital. Peripheral blood samples were obtained from the patients and controls after written informed consent was obtained. All patients and healthy controls were native Japanese and resided in Shimane prefecture or adjacent prefectures. Peripheral blood samples were obtained from the patients and controls after written informed consent was obtained.

The participation rate was 95% and 80% for the patients and controls, respectively. There were no significant differences between patients and control groups with regard to family history of cancer and body mass index.

Genotyping. The polymorphisms were analyzed by PCR-restriction fragment length polymorphism (RFLP) and single-strand conformational polymorphism (SSCP). Genotyping methods, primer sets, and annealing temperatures used for the RFLP and SSCP are shown in Table 2 (27, 28). Each PCR was carried out in a total volume of 20 μL consisting of 0.3 μL of a 10 μmol/L solution of each primer, 1.5 mmol/L MgCl₂, 0.8 mmol/L deoxynucleotide triphosphate, 0.5 unit REDTaq DNA polymerase (Sigma), 1 μL of genomic DNA (80 ng/μL), and 15.7 μL H₂O using a PTC 200 Thermal Cycler (MJ Research). The PCR program had an initial denaturation step of 7 min at 94°C followed by 35 cycles of 30 s at 94°C, 45 s of annealing at 57°C, and 45 s at 72°C. For RFLP analysis, PCR products were digested with BstUI, MspA1I, and BlpI (New England Biolabs) for p53 codon72, MDM2 SNP309, and p21 codon31, respectively, by the manufacturers' protocols. The PCR products were separated by electrophoresis in a 2% agarose gel and subsequently stained with ethidium bromide. SSCP was done using the following nonradioactive method. Briefly, a mixture of 5 μL PCR product and 7 μL SSCP loading dye [composed of 950 μL formamide, 40 μL of 0.5 mol/L EDTA (pH 8.0), and 4 μL of 0.05% bromphenol blue] was used. The mixture was placed on ice immediately after denaturation at 97°C for 5 min. The mixture (5 μL) was loaded onto 12% polyacrylamide gels and electrophoresis was done at 4°C for 10 h at 30 W with Tris-glycine as running buffer. Gels were stained with a 1:10,000 solution of GelStar (Cambrex Bio Science Rockland) in Tris-glycine buffer for 10 min with stirring at room temperature and examined using an UV transilluminator. To confirm the genotype ascribed by RFLP and SSCP, the PCR products were subjected to direct sequencing.

Immunohistochemistry study. Immunostaining of MDM2 was done in formalin-fixed, paraffin-embedded specimens using a mouse monoclonal antibody against human MDM2 (Santa Cruz Biotechnology). The staining procedure was according to a commercial kit (Santa Cruz Biotechnology). The sections were counterstained with Harris' hematoxylin. A pathologist, not involved in the present study, evaluated the immunostaining under blind conditions. Immunohistochemical staining was graded on an arbitrary scale from 0 to 2+; 0 representing negative expression (0-25% positive cells), 1+ representing weakly positive expression (25-50% positive cells), and 2+ representing strongly positive expression (50-100% positive cells). The scale was determined according to the average rate of positive cells in 10 random fields of all slides.

Statistical analysis. Hardy-Weinberg equilibrium was evaluated using SNPAlyze version 2.2 (DYNACOM Co. Ltd.). The χ² test was used to compare the genotype frequency between patients and controls. The odds ratio (OR) was obtained by unconditional logistic regression analysis and adjusted for age and gender as a continuous variable. All statistical analyses were done using StatView (version 5; SAS Institute, Inc.). A P value of <0.05 was regarded as statistically significant.

Results

Characteristics of RCC patients and controls.Table 1 depicts the mean age, gender, tumor grade, and pathology of individual RCC patients. Two-tailed Student's t tests were used to compare the distributions of age and gender between patients and control subjects. In the 200 RCC cases, 144 (72%) were localized and 170 (85%) were grade 1 and 2.

Hardy-Weinberg equilibrium. The genotype frequencies of the four polymorphisms in total samples (n = 400), RCC patients (n = 200), and healthy controls (n = 200) were consistent with the Hardy-Weinberg equilibrium distribution (P > 0.05).

p53, MDM2, and p21 polymorphism and RCC. The genotype distributions of the p53 codon72, MDM2 SNP309, and p21 codon31 polymorphisms between the RCC cases and healthy controls are shown in Table 3 . The frequencies of the variant alleles between cases and controls were as follows: p53 codon72 (0.39, 0.39), MDM2 SNP309 (0.53, 0.45), and p21 codon31 (0.56, 0.56). A significant increase in the GG genotype of the MDM2 SNP309 polymorphism was observed in patients compared with controls (OR, 1.80; 95% confidence interval (95% CI), 1.14-2.84; P = 0.012; Table 3). As the median age was 64 years, we divided the case and control groups at that age and compared the genotype frequencies between the two groups. The increased risk was evident among those subjects who were under 65 years old and had the two G alleles (OR, 3.00; 95% CI, 1.42-6.71; Table 4A ). We also examined the combined effect of the p53 codon72 Arg/Pro and the other genes. However, there is no additional effect from the p53 Arg⁷²Pro polymorphism when the risk of combined genotypes was analyzed (Table 4B).

Comparison MDM2 expression according to MDM2 SNP309 genotyping. To investigate the effect of the MDM2 SNP309 polymorphism on MDM2 expression, immunohistochemistry was done in some of the genotyped RCC tissues. Positive staining for MDM2 was detected in 2 of 15 (13%) TT genotype, 4 of 15 (26%) TG genotype, and 5 of 10 (50%) GG genotype carriers. The frequency of MDM2 expression in GG genotype carriers was significantly higher than that in TT genotype carriers (P = 0.04; Fig. 1 ).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Effect of MDM2 genotype on MDM2 expression. Tumors were scored as follows: score 0, no appreciable staining or staining in <25% of cancer cells; score 1+, tumors with weak appreciable incomplete nuclear and cytoplasmic staining in 25% to 50% of cancer cells; score 2+, strong immunoreactivity of the nucleus and cytoplasm in >50% of cancer cells. Tumors classified as 0 were considered “negative,” and those scored as 1 or 2 were classified as “positive.” IHC, immunohistochemistry. **, positive staining: (1+2) /(0+1+2) × 100(%)

Univariate and multivariate Cox proportional hazard analysis for cancer-specific survival in RCC patients. The prognostic value for tumor-related survival of the variables, including gender, age at diagnosis, tumor grade, pathologic tumor-node-metastasis, and SNPs (MDM2 SNP309, p53 codon72, p21 codon31), was analyzed using Cox proportional hazards analysis (Table 6). In the univariate analysis, tumor grade, pT, pN, pM, and MDM2 SNP309 were associated with survival (Table 6; Fig. 2 ). In the multivariate analysis, pT, pN, and pM had the same result. The hazard ratio for MDM2 SNP309 was 1.92 (0.94-3.86), showing a borderline significance as an independent risk factor for survival.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Cancer-specific survival based on Kaplan-Meier curves for 176 RCC patients. A, MDM2 SNP309 GG is an independent predictor of cancer-specific survival (P = 0.0139). Survival of GG carriers was statistically significantly worse than that of carriers with TG + TT genotypes. B to D, pT3/pT4, pN1/pN2, and pM1 are independent predictors of cancer-specific survival (P < 0.0001).

Discussion

In this study, we showed that the MDM2 SNP309 polymorphism is associated with an increased risk of renal cancer. Moreover, we used immunohistochemistry to investigate MDM2 expression in renal cancer tissues from different genotype carriers because there have been no reports about the relationship between the MDM2 SNP309 polymorphism and expression of MDM2 in RCC. This is a first report documenting that the MDM2 GG genotype carriers have significantly higher MDM2 expression in renal cancer tissues compared with TT or TG genotype carriers (Table 5 ). Haitel et al. (22) investigated MDM2 expression in renal cancer tissues immunohistochemically and showed that overexpression of the MDM2 is an independent predictor for poor survival.

Among other cancers, there is a significant correlation between MDM2 expression and poor prognosis (30). We investigated whether MDM2 SNP309 polymorphism is a predictive variable in RCC. In the univariate analysis, tumor grade, tumor-node-metastasis, and the MDM2 SNP were statistically significant as prognostic factors. In addition, multivariate Cox proportional hazard analysis revealed that this polymorphism could be an independent prognostic factor (Table 6 ; Fig. 2). Among gastric and esophageal cancer patients, MDM2 GG carriers were shown to be significantly associated with poor survival (20, 21). Our present results agree with those findings. Therefore, the MDM2 SNP309 polymorphism may provide important prognostic information about survival in renal cancer.

The p21 is one of major effectors of p53 and negatively controls cell proliferation by inhibiting several cyclin-dependent kinase complexes. In response to DNA damage, p53 up-regulates p21, resulting in G₁ phase cell cycle arrest. The loss of p53 function leads to the attenuation of p21 expression, although the p53-independent pathway also regulates p21 expression (31). Although p21 mutation is rare in RCC cell lines (32), SNPs of p21 are thought to change its function and the p21 codon31 SNP has been reported to be associated with several cancers (27). However, in this study, the p21 codon31 polymorphism was not significantly associated with risk for RCC (Table 3).

The p53 pathway is an important response to oncogenic stress, and p53 regulates its own intracellular levels through an autoregulatory feedback pathway with MDM2 (30). MDM2 binds to p53 and inactivates it through ubiquitination. MDM2 is a proto-oncogene and loss of p53 function is caused by MDM2 overexpression, mutations, and other mechanisms, resulting in malignant transformation or carcinogenesis (33, 34). In the p53 pathway, p53, p21, and MDM2 play a crucial role together. Polymorphisms in p53-MDM2 (19, 20, 25, 29) and p53-p21 (26, 27) have been reported to be associated with other cancers, such as lung, esophageal, colorectal, breast, and gastric cancer. Based on this evidence, we investigated whether these gene polymorphisms and their gene-gene interaction may be important in RCC. In this case-control study for each polymorphism, a significant association with renal cancer was observed only for the MDM2 SNP309, although additional polymorphisms in the other genes have been linked to susceptibility for other cancers (20, 26, 27, 29).

We found that there was no additional effect from the p53 Arg⁷²Pro polymorphism when the risk of combined genotypes was analyzed because the OR was the same as the one from the MDM2 TT/TG compared with the GG genotype. The p53 polymorphism did not have any effect on cancer-specific survival in univariate analysis (Table 6). Bond et al. (18) have shown the SNP of MDM2 309GG to be associated with age at diagnosis. We compared the average age at diagnosis with SNP309 genotype and found no difference between the three genotypes (TT, TG, and GG; data not shown). We also found that there was an increased risk of RCC in those under 65 years old (Table 4A). This significant association may be due to different distribution of genotypes among controls, not from cases.

Therefore, the MDM2 SNP309 GG genotype may be important in early onset of RCC.

In conclusion, this is the first report to show a significant association between a functional polymorphism in MDM2 and increased risk of developing renal cancer. Moreover, the patients with MDM2 309GG genotype showed worse prognosis and low survival.