Do CDKN2 p16 540 C>G,CDKN2 p16 580 C>T,and MDM2 SNP309 T>G Gene Variants Act on Colorectal Cancer Development or Progression?

Abstract

CDNK2 p16 plays a pivotal role in G1/S transition by regulating the p53 pathway, which was regulated by a nuclear oncoprotein, mouse double minute 2 (MDM2). Overexpression of the MDM2 gene has been shown in a number of tumor types, its gene amplification is found to associate with accelerated tumor development and failure to treatment in both hereditary and sporadic cancers. Although genetic association studies have revealed the relationship between certain genetic polymorphisms and genes that play important roles in the development and progression of colorectal cancer (CRC), it is still unknown. Therefore, the polymorphisms of p16 540 C>G, 580 C>T, and MDM2 SNP309 T>G designed to investigate the risk of CRC development and progression in a Turkish population. We enrolled 87 patients with CRC and 75 healthy controls into the study. Genotypings were determined using polymerase chain reaction–restriction fragment length polymorphism techniques. Genotype distributions of p16 540 C>G and 580 C>T were found in agreement with the Hardy–Weinberg equilibrium in patients and controls. MDM2 SNP309 T>G was found in agreement with the Hardy–Weinberg equilibrium in controls, but not in patients. The results of our study, the G allele of p16 540 C>G and GG genotype of MDM2 SNP309 T>G were found significantly lower in patients compared with controls (p<0.001, p<0.05, respectively). Haplotype analyses have shown that the C allele of both the CDKN2 p16 540 C>G and 580 C>T variants together indicate a risk haplotype for the patient group; besides, carrying the G allele of p16 540 and G allele of MDM2 also seems a risk haplotype for the patient group. Our study is the first study that investigates the relationship among variants of CDKN2 p16 540 C>G, 580 C>T, and MDM2 SNP309 T>G risk of CRC and the development and progression in the Turkish population.

Introduction

C ancer of the colon or the rectum, together referred to as colorectal cancer (CRC) is one of the most common malignancy with a higher morbidity and mortality prevalence worldwide. CRC can be originated from sporadic (65%–85%), familial (10%–30%), or hereditary (5%–7%) risk factors. However, the development of CRC is influenced by the interactions of numerous genetic and environmental factors (Hardy et al., 2000; Parkin et al., 2002). Traditionally, age, family history, lifestyle and dietary factors, obesity, exposure to chemicals and carcinogens, and some diseases like Crohn's disease and ulcerative colitis are described as the potential epidemiologic risk factors for the development CRC, and all these factors may differ in each race and ethnic group. Furthermore, some familial/hereditary syndromes, including familial adenomatous polyposis and hereditary nonpolyposis colon cancer (Lynch syndrome), which contribute to morphologic and genetic progression to CRC are also well described (Hardy et al., 2000; Houlston, 2001; Grady, 2004).

CRC is increasingly recognized as a heterogeneous disease, therefore, recent studies focused on the identification of the molecular events underlying CRC. Genomic instability, genomic mutations, microRNAs, which are regulators of transcription, and epigenetic changes are some of the molecular mechanisms of primary CRC. The genomic mutations include suppression of tumor suppressor genes such as p16, p53, and APC; activation of tumor oncogenes like k-ras (Hardy et al., 2000; Kanthan et al., 2012); alteration of DNA repair genes (Uchida et al., 1998; Hardy et al., 2000), all of which have important roles in cell cycle proliferation and apoptosis or defects in cell signaling genes like catenins, which alter cell signaling and adhesion (Hardy et al., 2000). Although these genetic association studies have revealed the relationships between the development of CRC and certain genetic polymorphisms, the genes that play important roles in susceptibility to the development and progression of CRC are still unknown. Therefore, the polymorphisms of p16, one of the cell cycle regulatory gene, and mouse double minute 2 (MDM2), an important modulator of tumor suppressor p53, are the interest of this article.

In eukaryotes, the cell cycle progression is regulated by a complex network of protein–protein interactions and protein phosphorylations. Protein phosphorylation is accomplished by the interaction of cyclins with their specific cyclin-dependent kinases (CDK) and negatively regulated by cyclin-dependent kinase inhibitors (CDKIs) (Sherr, 1996). The tumor suppressor p16 (also known as CDKN2/MTS-1/INK4a) is one of the CDKI located on chromosome 9p21 and composed of three exons encoding a 156 amino acid protein (Cobrinik et al., 1992). This inhibitory protein blocks the G1 phase to S phase transition by binding CDK4 and −6, which inhibit the interaction with cyclin D1. This inhibition prevents the phosphorylation of the retinoblastoma (Rb) protein, which is another important tumor suppressor gene product that binds various S phase transcription factors, such as the elongation factor 2, the E2F family of DNA-binding transcription factors (E2F), during the G1 phase. Rb is a very important regulator during G1 to S phase transition in the cell cycle. It can be phosphorylated by CDK4/6- cyclin D1 complex otherwise it made a complex with E2F. After the phosphorylation, the E2Fs bound by Rb are released and the cell was driven to the S phase. Therefore, the inactivation of p16 results in increasing cell proliferation and is thought to contribute to most human cancer malignancies. The inactivation of p16 can be caused by point mutations, small deletions, large hetero- and homozygous deletions, and silencing by methylation of CpG islands in the promoter region (Cobrinik et al., 1992; Shapiro et al., 1995; Sherr, 1996; Sauroja et al., 2000; Yan et al., 2008). Previous reports revealed that polymorphisms in the p16 gene play a critical role in cancer development and prognosis, but their potential impact on CRC has not been well studied. Two adjacent polymorphisms of the CDKN2 p16 gene, 540 C>G (rs11515) and 580 C>T (rs3088440), were identified recently. Both polymorphisms were located in the 3′untranslated (3′UTR) region of exon 3 and shown to affect the activation of p16 and contribute to cancer development and prognosis as well as the tumor aggressiveness (McCloud et al., 2004; Abbaszadegan et al., 2005; Debniak et al., 2005; Fombonne et al., 2005).

The p53 tumor suppressor protein plays a central role in the prevention of cancer development by causing growth arrest and/or apoptosis during stress and inhibition of the p53 pathway results in accelerated tumorigenesis (Arva et al., 2005). The nuclear phospho-oncoprotein MDM2, is an E3 ubiquitin ligase and a major negative regulator of p53 that directly binds to the N-terminal transactivation domain of p53 and inactivates the p53 tumor suppressor activity by regulating its location, stability/degradation, and activity as a transcriptional activator (Michael and Oren, 2003; Levav-Cohen et al., 2005). MDM2 has also been shown to promote tumor growth in a p53-independent manner (Oliner et al., 1992). It has negative effects on DNA double-strand break repair (Alt et al., 2005). Overexpression of the MDM2 gene was found in a number of tumor types, and it was reported that the MDM2 gene amplification is associated with accelerated tumor development and failure in the response to treatment in both hereditary and sporadic cancers in humans. (Freedman and Levine, 1999). A single-nucleotide polymorphism in the promoter of MDM2, SNP309 T>G (rs2279744) has been shown to alter protein expression and p53 activity and thus play a role in carcinogenesis (Bond et al., 2004).

This study was designed to investigate the association of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G polymorphisms with the risk of CRC development and progression among the Turkish population.

Materials and Methods

Study participants and clinical investigation

The study groups were composed of 87 cancer patients diagnosed and recruited by the Istanbul University Radiation Oncology and Istanbul Education and Research Hospital (Istanbul, Turkey). They were all newly diagnosed with histopathologically confirmed primary CRC and surgically treated before any radiotherapy and chemotherapy. Pathologic staging information for the CRC patients was confirmed by manual review of the pathology reports and clinical charts. Nodal status was categorized as no regional lymph nodes affected (no lymph node involvement) and the presence of at least one nodal metastasis. Seventy-five healthy volunteers without any malignancy were included as controls. Individuals with a negative family history of cancer and lacking any symptoms of CRC were selected for the control group.

The study protocol was approved by both the Ethics Committee of the Istanbul Faculty of Medicine and the Research Fund of Istanbul University. All participants in the study signed an informed consent form in accordance with ethics guidelines regarding the study.

Polymerase chain reaction-based detection of CDKN2 and MDM2 mutations/genotyping

Blood specimens were collected in tubes containing EDTA, and DNA samples were extracted from whole blood with a salting-out procedure (Miller et al., 1998).

The DNA samples were analyzed for the CDKN2 p16 540 C>G, CDKN2 p16 580C>T, and MDM2 SNP309 T>G polymorphisms by polymerase chain reaction (PCR) with locus-specific primers and subsequent analysis of a restriction fragment length polymorphism (RFLP) created by the mutation as previously reported (Ohmiya et al., 2006; Yan et al., 2008).

The primers for PCR amplification of the CDKN2 p16 exon 3 region spanning the 540 and 580 locus were 5′-GAT GTG CCA CAC ATC TTT GAC CT-3′ and 5′-CTA CGA AAG CGG GGT GGG TTG T-3′, and for the MDM2 promotor region were 5′-CGC GGG AGT TCA GGG TAA AG-3′ and 5′-AGC TGG AGA CAA GTC AGG ACT TAA C-3′. After amplification of the isolated DNAs with PCR, the 540 C>G and 580 C>T substitutions of the CDKN2 p16 gene and the promoter substitution of the MDM2 gene were detected by cutting the PCR products with the proper restriction endonucleases, MspI, HaeIII, and MspA1I (MBI Fermentas, Ontario, Canada), respectively. The digested DNAs were then separated on a 2% agarose gel in the 1XTris borate EDTA buffer followed by staining with the ethidium bromide solution. The genotypes were typed by visualization under ultraviolet light.

Statistical methods

Statistical analysis was performed by using SPSS software package (revision 11.5; SPSS, Inc., Chicago, IL). A univariate analysis was performed to compare the distribution of age and gender and the frequencies of alleles and genotypes. Mean values were compared between patients and controls by the unpaired Student's t-test. Differences in the distribution of genotypes and alleles between patients and controls were tested using the chi-square statistic. The Hardy–Weinberg equilibrium was tested for all polymorphisms. Allele frequencies were estimated by gene counting methods. A multivariate logistic regression model was performed to investigate the associations between the genotypes of the mentioned polymorphisms, patient characteristics, and clinicopathologic parameters. Values of p<0.05 were considered statistically significant. Linkage disequilibrium among CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G polymorphisms was evaluated using the D′ and r² values obtained through the Haploview program (www.broad.mit.edu/mpg/haploview/documentation.php). Values of p<0.05 were considered statistically significant. In our study, we examined the dominant and recessive model each for three SNPs, CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G, because it did not appear which allele was dominant and/or recessive for three SNPs.

Results

The demographic characteristics of our CRC patients and controls are shown in Table 1. We had a 62% power to detect an effect size (W) of 0.20 using two degrees of freedom (α=0.05).

Table 1.

Characteristics of Patients with Colorectal Cancer and Controls

	Patients, n (%)	Controls, n (%)
No. of patients	87	75
Age (year) (mean±standard deviation)	58.3±14.6	54.4±14.6
Male	55 (63.2)	36 (48)
Female	32 (36.8)	39 (52)
Smoking status
Current smokers	6 (6.9)	6 (8)
Never or former smokers	81 (93.1)	69 (92)
Family history of any kind of cancer
Yes	3 (3.4)	0
No	84 (96.6)	75 (100)
Tumor localization
Left colon	11 (12.6)	—
Right colon	14 (16.1)	—
Transverse colon	6 (6.9)	—
Sigmoid	32 (36.8)	—
Caecum	5 (5.7)	—
Rectum	19 (21.8)	—
T stage
T1	5 (5.7)	—
T2	11 (12.6)	—
T3	55 (63.2)	—
T4	16 (18.4)	—
Lymph node status
N0	43 (49.4)	—
N1	24 (27.6)	—
N2	14 (16.1)	—
N3	6 (6.9)	—
The presence of distant metastasis	18 (20.7)	—
The presence of perineural invasion	31 (35.6)	—
The presence of angiolymphatic invasion	39 (44.8)	—
Differentiation status		—
Poorly differentiated	18 (20.7)	—
Moderately differentiated	60 (69.0)	—
Well differentiated	9 (10.3)	—

n, number of individuals.

Distribution of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G genotypes and alleles

The distributions of genotypes and alleles of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G are shown in Table 2.

Table 2.

The Distribution of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G Genotypes and Alleles in the Study Groups

Genotype	Controls n (%)	All patients n (%)	p-Value	aOR	95% CI	p-Value
CDKN2 p16 540 C>G
CC	26 (34.7)	58 (66.7)		reference
CG	36 (48.0)	27 (31.0)		0.547	0.375–.797	<0.001
GG	13 (17.3)	2 (2.3)	<0.001	0.100	0.024–0.419	<0.001
C Allele	88 (58.7)	143 (82.2)
G Allele	62 (41.3)	31 (17.8)	<0.001
Recessive model for the G allele
CG+CC	62 (82.7)	85 (97.7)		reference
GG	13 (17.3)	2 (2.3)		0.133	0.031–0.569	<0.001
Dominant model for the G allele
CC	26 (34.7)	58 (66.7)		reference
CG+GG	49 (65.3)	29 (33.3)		0.510	0.363–0.717	<0.001
CDKN2 p16 580 C>T
CC	54 (72.0)	64 (73.6)		reference
CT	18 (24.0)	19 (21.8)		0.916	0.522–1.606	>0.05
TT	3 (4.0)	4 (4.6)	>0.05	1.118	0.261–4.788	>0.05
C Allele	126 (84.0)	147 (84.5)
T Allele	24 (16.0)	27 (15.7)	>0.05
Recessive model for the T allele
CT+CC	72 (96)	83 (95.4)		reference
TT	4 (4)	4 (4.6)		1.149	0.266–4.973	>0.05
Dominant model for the T allele
CC	54 (72)	64 (73.6)		reference
TT+CT	21 (28)	23 (26.4)		0.944	0.570–1.564	>0.05
MDM2 SNP309 T>G
TT	15 (20.0)	17 (19.5)		reference
TG	37 (49.3)	57 (65.5)		1.083	0.875–1.340	>0.05
GG	23 (30.7)	13 (14.9)	0.043	0.716	0.442–1.161	>0.05
T Allele	67 (44.7)	91 (52.3)
G Allele	83 (55.3)	83 (47.7)	>0.05
Recessive model for the G allele
GT+TT	52 (69.3)	74 (85.1)		reference
GG	23 (30.7)	13 (14.9)		0.487	0.266–0.893	0.016
Dominant model for the G allele
TT	15 (20)	17 (19.5)		reference
GG+GT	60 (80)	70 (80.5)		0.977	0.525–1.820	>0.05

Chi-square test was used to compare genotypes in the study group.

CRC, colorectal cancer; n, number of individuals; MDM2, mouse double minute 2; CDK, cyclin-dependent kinase; aOR, adjusted for age and sex.

Genotype distributions of CDKN2 p16 540 C>G and CDKN2 p16 580 C>T were in agreement with the Hardy–Weinberg equilibrium in patients (χ ²=0.31, p=0.577; χ ²=2.43, p=0.12, respectively), and controls (χ ²=0.007, p=0.93; χ ²=0.86, p=0.35, respectively). Also, MDM2 SNP309 T>G was found in agreement in controls, but not in patients (χ ²=0.0002, p=0.98; χ ²=8.52, p=0.0034, respectively).

Significant associations were found in the distributions of CDKN2 p16 540 C>G genotypes and alleles between patients versus controls (p<0.001). As seen in Table 2, the homozygous GG variant of CDKN2 p16 540 C>G was extremely low in the patients. When we analyzed both the recessive and dominant model for CDKN2 p16 540 C>G, it was found statistically significant carrying with GG genotype and G allele were lower in patients than controls (p<0.001). However, there were no significant distributions of genotypes and alleles found for both recessive and dominant models between groups for CDKN2 p16 580 C>T (p>0.05). There were found no significant genotypes and alleles of MDM2 SNP309 T>G between groups (p>0.05). Furthermore, while there was no statistically significant G allele of MDM2 SNP309 T>G in the dominant model, the frequency of the GG genotype was a statistically significant decrease in patients compared to controls in the recessive model (p<0.05). Therefore, the combination of the heterozygous 540CG and homozygous 540GG variant was taken together for the analysis and seen that the frequency of the mutant 540G allele (GG+GC) of CDKN2 p16 540 C>G was found lower in the patient group versus control group (41.3%–7.85%; χ ²=16.52; p<0.001, OR: 0.51, 95% CI: 0.360–0.717). In addition, the frequency of the 540CC genotype in patients was significantly higher than those in controls (66.7%→34.7%; χ ²=16.52; p<0.001, OR: 1.92, 95% CI: 1.363–2.714). Similar to CDKN2 p16 540 C>G, the homozygous variant allele of CDKN2 p16 580 C>T was rare in the study population, however, no significant difference was found in the distribution of CDKN2 p16 580 C>T genotypes and alleles in the study groups (p>0.05).

Although the frequency of controls with the SNP309GG genotype (p=0.016) and patients with the SNP309GT (p=0.037) genotype of MDM2 SNP309 T>G were significantly higher, there was no difference in the SNP309G allele frequency between patients and controls (p>0.05).

Combined genotype analysis of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G

The distribution of CDKN2 p16 540 C>G and CDKN2 p16 580 C>T, MDM2 SNP309 T>G and CDKN2 p16 540 C>G, MDM2 SNP309 T>G and CDKN2 p16 580 C>T combined genotypes are shown in Table 3.

Table 3.

The Distribution of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G Combined Genotypes

Combined genotypes of CDKN2 p16 540 C>G and 580 C>T			Combined genotypes of MDM2 SNP309 T>G and CDKN2 p16 540 C>G			Combined genotypes of MDM2 SNP309 T>G and CDKN2 p16 580 C>T
P16-580/p16-540 combined genotypes	CRC patients n (%)	Controls n (%)	P16-540/MDM2 combined genotypes	CRC patients n (%)	Controls n (%)	P16-580/MDM2 combined genotypes	CRC patients n (%)	Controls n (%)
GGTT	0 (0.0)	2 (2.7)	TTGG	0 (0.0)	3 (4.0)	TTTT	0 (0.0)	0 (0.0)
GGCT	1 (1.1)	3 (4.0)	TTGC	8 (9.2)	7 (9.3)	TTCT	2 (2.3)	4 (5.3)
GGCC	1 (1.1)	8 (10.7)	TTCC	9 (10.3)	5 (6.7)	TTCC	15 (17.2)	11 (14.7)
GCTT	1 (1.1)	0 (0.0)	GTGG	2 (2.3)	4 (5.3)	GTTT	3 (3.4)	1 (1.3)
GCCT	6 (6.9)	7 (9.3)	GTGC	16 (18.4)	19 (25.3)	GTCT	13 (14.9)	9 (12.0)
GCCC	20 (23.0)^a	29 (38.7)	GTCC	39 (44.8)^b	14 (18.7)	GTCC	41 (47.1)	27 (36.0)
CCTT	3 (3.4)	1 (1.3)	GGGG	0 (0.0)	6 (8.0)	GGTT	1 (1.1)	2 (2.7)
CCCT	12 (13.8)	8 (10.7)	GGGC	3 (3.4)	10 (13.3)	GGCT	4 (4.6)	5 (6.7)
CCCC	43 (49.4)^c	17 (22.7)	GGCC	10 (11.5)	7 (9.3)	GGCC	8 (9.2)	16 (21.3)

Chi-square test was used to compare genotypes in the study group.

p=0.030.

p<0.001 versus controls.

p<0.001.

n, number of individuals

The frequency of the CCCC combined genotype of CDKN2 p16 540 C>G and CDKN2 p16 580C>T was significantly higher in CRC patients than (49.4%) controls (22.7%); in addition, in the patient group, this combined genotype had the highest frequency than the other combined genotypes (p<0.001, OR: 2.181, 95% CI: 1.364–3.485). As for the patient group, the GCCC combined genotype frequency was found to be lower than controls (p=0.030, OR: 0.595, 95% CI: 0.368–0.960).

When the MDM2 SNP309 T>G and CDKN2 p16 540 C>G genotypes were compared, it was seen that the GTCC genotype frequency was significantly higher in CRC patients (44.8%) than controls (18.7%) (p<0.001, OR: 2.401, 95% CI: 1.418–4.067). In addition, a higher GTGC combined genotype frequency was found in the control group, but the significance between controls and patients were not statistically evaluated (p>0.05).

Consequently, for the combined genotypes of MDM2 SNP309 T>G and CDKN2 p16 580C>T in both patient and control groups, it was separately found that GTCC had the highest combined genotype frequency than the other variants. On the other hand, the combined genotype frequency of GTCC was slightly higher in CRC patients than controls (p>0.05).

Results of linkage disequilibrium among CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G

The frequencies of haplotypes of CDKN2 p16 540 C>G and CDKN2 p16 580 C>T are shown in Table 4. We found statistically significant higher in patients than those of controls carrying with both of C allele of p16 540 and p16 580 (p<0.001), but haplotype of G allele of p16 540 and C allele of p16 580 were higher in controls than those of patients (p<0.001).

Table 4.

The Frequencies of Haplotypes of CDKN2 p16 540 C>G and CDKN2 p16 580 C>T in Patients and Controls

Haplotype	Frequencies	Patients, control ratio	χ²	p-Value
CC	0.605	0.697, 0.498	13.251	<0.001
GC	0.238	0.148, 0.342	16.638	<0.001
CT	0.108	0.125, 0.088	1.138	0.2862
GT	0.049	0.030, 0.072	2.989	0.0838

The order of haplotype is CDKN2 p16 540C>G and CDKN2 p16 580 C>T.

The frequencies of haplotypes of CDKN2 p16 540 C>G and MDM2 SNP309 T>G are given in Table 5. They were statistically significant higher in patients than those of controls carrying with both of C allele CDKN2 p16 540 and T allele of MDM2 (p<0.001). We also found haplotypes of G allele of CDKN2 p16 540 and G allele of MDM2 were higher in patients than those of controls (p<0.001).

Table 5.

The Frequencies of Haplotypes of CDKN2 p16 540 C>G and MDM2 SNP309 T>G Genes in Patients and Controls

Haplotype	Frequencies	Patients, control ratio	χ²	p-Value
CG	0.360	0.400, 0.313	2.622	0.1054
CT	0.353	0.422, 0.273	7.785	0.0053
GG	0.153	0.077, 0.240	16.528	<0.001
GT	0.134	0.101, 0.173	3.617	0.0572

The order of haplotype is CDKN2 p16 540 C>G and MDM2 SNP309 T>G.

When we analyzed all SNPs of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G, the frequencies of G allele of CDKN2 p16 540, C allele of CDKN2 p16 580, and G allele of MDM2 were statistically significantly lower in patients than those of controls (p<0.001) (Table 6).

Table 6.

The Frequencies of Haplotypes of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G Genes in Patients and Controls

Haplotype	Frequencies	Patients, control ratio	χ²	p-Value
CCT	0.312	0.378, 0.236	7.57	0.059
CCG	0.292	0.319, 0.262	1.252	0.2634
GCT	0.122	0.092, 0.158	3.293	0.0696
GCG	0.116	0.057, 0.184	12.803	3.10⁻⁴
CTG	0.068	0.081, 0.053	1.005	0.3161
CTT	0.040	0.044, 0.036	0.143	0.7049
GTG	0.036	0.020, 0.054	2.626	0.1051
GTT	0.013	0.009, 0.017	0.378	0.5386

The order of haplotype is CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G.

CDKN2 p16 540 C>G, CDKN2 p16 580C>T, and MDM2 SNP309 T>G haplotypes were evaluated for association between CRC and controls. Haplotype analysis revealed that there was a weak linkage between p16 540 and p16 580 in our study groups (D′:0.037; r2:0.001) (Table 6).

Association of the CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G alleles with clinicopathological parameters

In Table 7, the distributions of clinicopathological parameters according to CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G alleles were presented.

Table 7.

Distribution of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G in Patient's Group

	CDKN2 p16 540 C>G			CDKN2 p16 580 C>T			MDM2 SNP309 T>G
	CC n (%)	CG n (%)	GG n (%)	CC n (%)	CT n (%)	TT n (%)	GG n (%)	TG n (%)	TT n (%)
Gender
Female	22 (68.8)	9 (28.1)	1 (3.1)	23 (71.9)	9 (28.1)	0 (0)	3 (9.4)	22 (68.7)	7 (21.9)
Male	36 (65.5)	18 (32.7)	1 (1.8)	41 (74.5)	10 (18.2)	4 (7.3)	10 (18.2)	35 (63.6)	10 (18.2)
Smoking
Presence	5 (71.4)	2 (28.6)	0 (0)	6 (85.7)	1 (14.3)	0 (0)	2 (28.6)	5 (71.4)	0 (0)
Absence	53 (66.3)	25 (31.3)	2 (2.5)	58 (72.5)	18 (22.5)	4 (5)	11 (13.8)	52 (65)	17 (21.2)
Family history of cancer
Presence	3 (100)	0 (0)	0 (0)	2 (66.7)	1 (33.3)	0 (0)	0 (0)	3 (100)	0 (0)
Absence	55 (65.5)	27 (32.1)	2 (2.4)	66 (73.8)	18 (21.4)	4 (4.8)	13 (15.5)	54 (64.3)	17 (20.2)
Tumor grade
T3+T4	49 (68.1)	21 (29.2)	2 (2.8)	54 (75)	15 (20.8)	3 (4.2)	12 (16.7)	46 (63.9)	14 (19.4)
T1+T2	9 (60)	5 (40)	0 (0)	10 (66.7)	4 (24.6)	1 (6.7)	1 (6.7)	11 (73.3)	3 (20)
Lymph node metastasis
Presence	33 (75)	9 (20.5)^a	2 (4.5)	29 (65.9)	11 (25)	4 (9.1)	9 (20.5)	29 (65.9)	6 (13.6)
Absence	25 (58.1)	18 (41.9)	0 (0)	35 (81.4)	8 (18.6)	0 (0)	4 (9.3)	28 (65.1)	11 (25.6)
Distant metastasis
Presence	11 (64.7)	5 (29.4)	1 (5.9)	12 (70.6)	4 (23.5)	1 (5.9)	3 (17.6)	13 (76.5)	1 (5.9)
Absence	47 (67.2)	22 (31.4)	1 (1.4)	52 (74.3)	15 (21.4)	3 (4.3)	10 (14.3)	44 (62.8)	16 (22.9)
Angiolymphatic invasion
Presence	25 (64.1)	13 (33.3)	1 (2.6)	25 (64.1)	12 (30.8)	2 (5.1)	8 (20.5)	26 (66.7)	5 (12.8)
Absence	33 (68.8)	14 (29.2)	1 (2)	39 (81.2)	7 (14.6)	2 (4.2)	5 (10.4)	31 (64.6)	12 (25)
Peritoneural invasion
Presence	21 (67.7)	9 (29.1)	1 (3.2)	25 (80.6)	6 (19.4)	0 (0)	4 (12.9)	23 (74.2)	4 (12.9)
Absence	37 (66.1)	18 (32.1)	1 (1.8)	39 (69.6)	13 (23.3)	4 (7.1)	9 (16.1)	34 (60.7)	13 (23.2)

Chi-square test was used to compare alleles and clinicopathological characteristics in the study group.

n, number.

There was no statistical association between both the 540G and 540C allele of CDKN2A p16 540 C>G with gender, smoking, family history, tumor grade, presence of distant metastasis, angiolymphatic invasion, and perineural invasion in CRC patients. However, the frequency of patients with the homozygous 540GG or 540CC genotype having lymph node metastasis was higher than those without having lymph node metastasis (p=0.031, OR: 1.368; 95% CI: 1.019–1.837).

In the CRC patient group, when the CDKN2A p16 580 C or T allele carriage was compared according to clinicopathological parameters, no statistical association was observed. However, we noticed that while four patients having lymph node metastasis possess 580TT genotype, there was any patients without lymph node metastasis carrying 580TT genotype, but the difference between having 580TT genotype and lymph node metastasis was not statistically significant (p>0.05, OR: 0.909, 95% CI: 0.828–0.998).

There was no statistical association between the allelic distribution of MDM2 SNP309 T>G and clinico-pathological parameters.

Discussion

Like all the other cancer types, CRC was also characterized by the accumulation of the genetic and epigenetic defects mostly seen in growth factors and its receptors, angiogenic factors, cell cycle regulators, or DNA repair genes. It was recently understood that polymorphism studies provide very important information about cancer development and progression as well as the toxicity of anticancer drugs and the efficiency of cancer treatment (Yasui et al., 2006).

The p16 protein, which has an important role in the cell cycle process, was a specific inhibitor of CDK4 and CDK6 and was also referred as a tumor suppressor gene in most malignancies. It was known that the loss of function in p16 results in the inactivation of its growth suppressor effect (Shapiro et al., 1995). Until today the mutations of p16 gene were investigated in many studies (Shapiro et al., 1995; McCloud et al., 2004) and controversial results about cancer development were reported especially for the 3′ UTR region (Kumar et al., 2001; Huber and Ramos, 2006; Larre et al., 2009; Ibarrola-Villava et al., 2010). Accordingly, in the present study, the evaluation of the possible association of CDKN2 p16 540C>G and/or CDKN2 p16 580C>T polymorphisms with the clinical and pathological parameters as well as the development of CRC was aimed.

Kumar et al. reported the prevalence of the heterozygous genotype of CDKN2 p16 540C>G as 29% in a Swedish population and in similar populations, it was reported as 22%. In addition, a 26% prevalence of the CDKN2 p16 580C>T heterozygous genotype was reported in the Swedish population and this was shown in accordance with Australians (Holland et al., 1995; Kumar et al., 1998). In the present report, the genotype distribution of CDKN2 p16 540 C>G was 48% GC, 34.7% CC, and 17.3% GG, and the prevalence of the heterozygous genotype of CDKN2 p16 580 C>T was 24%, which was in accordance with the previous reports of Kumar (Kumar et al., 1998) and Holland (Holland et al., 1995).

Chen et al. (2007) found an association between CDKN2 p16 540 C>G or CDKN2 p16580 C>T genotypes and pancreatic cancer as well as the study by Zheng et al. (2002) performed in head and neck cancer. In the present report, the 540CC genotype and 540C allele frequency of CDKN2 p16 540C>G was found to be significantly higher in patients than those in controls. Furthermore, the finding of lower 540 CG and/or GG frequencies in the patient group suggested that these genotypes may have some protective effects against CRC. On the other hand, we did not observe any association between CRC and CDKN2 p16 580 C>T gene polymorphisms.

Sauroja et al. (2000) declared that the two polymorphisms of CDKN2 p12 540 C>G and CDKN2 p12 580 C>T, were related with tumor aggressiveness. In this report, we observed that the prevalence of homozygous CC and GG genotypes of CDKN2 p16 540 C>G was higher in patients with lymph node metastasis than of those without lymph node metastasis (p=0.03). On the other hand, another interesting finding was the prevalence of TT genotypes of CDKN2 p16 580 C>T in patients without lymph node metastasis. We did not observe TT carriers in patients without having lymph node metastasis. On the other hand, 4/44 of lymph node metastatic patients have TT genotypes.

p16 was one of the transcript of CDKN2A, the other transcript was p14, which was the inhibitor of tumor suppressor Rb by inhibiting the cell cycle activators, CDK4 and −6, MDM2 was a critical molecule, which was responsible from the degradation of tumour suppressor p53 (Sharpless, 2005; Caren et al., 2008). MDM2 was the major regulator of p53 (Harris and Levine, 2005; Toledo and Wahl, 2006) and several mutations of MDM2 were reported. In the promoter region of MDM2, a timidine (T) to guanine (G) transition at position 309 was reported and referred as MDM2 SNP309 T>G polymorphism. Bond et al. (2004) declared that this transition leads to an increase in MDM2 mRNA/protein levels by enhancing the specific binding affinity of a transcriptional activator, Sp1. In addition, Knappskog and Lonning (2011) reported that when compared with the SNP309T allele, SNP309G has much more affinity to the Sp1 transcription factor.

Hu et al. (2007) reported that the allele frequency of SNP309G was 10% in Africans and Americans, 40% in Caucasians, 50% in Asians. In another study by Atwal et al. (2007), higher frequencies of the SNP309T allele were reported in Caucasians, African Americans, and Ashkenazi populations than the SNP309G allele. As for this study, the SNP309G frequency was found as 55.3% in healthy subjects, which was in accordance with the Asian population.

Previous studies declared that in diffuse large B cell lymphoma, connective tissue sarcomas, invasive ductal breast carcinomas and in CRC, the MDM2 SNP309 T>G polymorphism resulted in accelerated tumor formation (Bond et al., 2006a, 2006b). In a meta-analysis, it was found that the SNP309GG genotype has a huge effect on cancer development and the authors declared an elevated inhibitory effect of MDM2 on the p53 pathway caused by this polymorphism, in addition, they concluded their study as lung cancer prognosis negatively affected by SNP309GG genotypes (Wilkening et al., 2007). Similar to some previous reports in this study, it was found that the frequencies of SNP309GG genotype alleles were significantly lower in CRC patients than controls. Indeed, there are also some studies that were in accordance with our results. Shinohara et al. reported that the SNP309G allele (TG+GG) has a protective role on CRC (Shinora et al., 2009). Furthermore, Ko et al. (2000) declared the expression of MDM2 mRNA as a prognostic factor in big cell lung carcinoma; however, they also reported null association with clinicopathological features like tumor grade, type, level, and TNM parameters.

Chua et al. (2010) reported a positive association between SNP309TT genotypes and nonsmoking Chinese female lung cancer patients. However, the finding of SNP309TT genotypes increasing the risk of cancer was an unexpected result when it was evaluated with the previous results that declared the association between SNP309GG genotypes and elevated MDM2A mRNA/protein levels so that the latter effects on p53 inhibition (Bond et al., 2004). More recent studies revealed that SNP309TT genotype established in Chinese breast cancer and lung cancer patients increases the risk (Li et al., 2006; Lum et al., 2008), whereas the SNP309GG genotype decreases the risk of leukemia (Phang et al., 2008). Therefore, it was suggested that the relationship between cancer risk and either the SNP309G allele or SNP309T allele was due to environmental and ethnic factors. However, to expand these results, additional extensive studies must be performed. Although it was not understood how the SNP309TT genotype has a role in cancer development, it was shown that the role of MDM2 in tumorigenesis differs according to the gender and smoking habit. Thus, the mechanism as to how MDM2 modulates lung cancer could be variable between populations.

The role of hormones in carcinogenesis was considered as a complex and attractive pathway. It was shown that the SNP309G allele has accelerated the affinity of the transcriptional activator of estrogen receptor, Sp1, as well as the overexpression of MDM2 (Chua et al., 2010). Bond et al. (2006b) reported that the MDM2 polymorphism was gender-specific, thus, it was mostly seen in females who have an active estrogen pathway and accordingly it was possible to suggest the correlation between the SNP309TT genotype and decreased MDM2 levels. On the other hand, there are also some reports concerning the negative effect of MDM2 on the expression of the estrogen receptor (ER). In this respect, it was possible to hypothesize that by triggering ER expression with decreased levels of MDM2 in individuals carrying SNP309TT genotypes, it was possible that the relative risk of cancer increases. Furthermore, if the expressed ER in lung cancer cells, especially in adenocarcinomas, and the differentiation of these cells via response to estrogen are taken into consideration, it was possible to discover the alternative mechanisms which show the effects of the interaction of MDM2 and hormonal pathways on the risk of cancer development (Duong et al., 2007).

In this study, any association between MDM2 SNP309 T>G and clinical parameters such as gender, smoking, alcohol consumption, family history was found.

This study is preliminary to evaluate the association of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G combined genotypes with the clinicopathological parameters to detect both the developmental risk and prognosis of CRC. With this respect, it was found that in CRC patients, GTCC combined genotypes of MDM2 SNP309 T>G and CDKN2 p16 540 C>G were more common than the other combined variants. Furthermore, this combined genotype was significantly higher in CRC patients than controls.

Both of carrying with GG genotype and G allele were seen protective in patient group using with dominant and recessive models in polymorphism of CDKN2 p16 540 C>G. There was no significancy in CDKN2 580 p16 C>T using with genetic models between groups.

However, the GG genotype of MDM2 SNP309 T>G was found to indicate statistical significance more protective in recessive models than dominant models in patients group.

Haplotype analyses of our study have shown that C allele of both CDKN2 p16 540 C>G and CDKN2 p16 580 C>T variants together is indicated a risk haplotype for patients group, besides carrying with G allele of CDKN2 p16 540 and G allele of MDM2 is also seen a risk haplotype for patients group.

The strength of our present study has shown the relationship between CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G polymorphisms with the risk of CRC development and progression in the Turkish population, and is the first study for our population.

On the other hand, one of the limitations of our study is the number of colorectal cancer patients. This may be increased for collaborating with other Universities or Research Centers in the future for Turkish population.

Consequently, the present study was a preliminary study to establish the link between CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G polymorphisms and pathogenesis of CRC among the Turkish population. However, the limitations of this study included two principle domains. First, clinical data of the patients were inadequate. Second, the number of the study group was relatively small. Therefore, the adverse effects of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G polymorphisms may not be significant with respect to control groups, but still may give us clues on the prognosis of the disease. Additional studies with larger sample sizes are needed to define the influence of CDKN2 p16 540 C>G, CDKN2 p16 580 C>T, and MDM2 SNP309 T>G genotyping on clinical outcomes. The authors believe that the results of the present study could be more conclusive with further studies, which determines the interaction of p16 and MDM2 gene.

Footnotes

Acknowledgments

The present work was supported by a grant from the Scientific Research Projects Coordination Unit of Istanbul University (Project No. 4489).

Disclosure Statement

Authors declare that no competing interests exist.

References

Abbaszadegan

M.R.

, Raziee

H.R.

, Ghafarzadegan

, Shakeri

M.T.

, Afsharnezhad

, Ghavamnasiry

M.R.

2005. Aberrant p16 methylation, a possible epigenetic risk factor in familial esophageal squamouscell carcinoma. Int J Gastrointest Cancer, 36:47–54.

Alt

J.R.

, Bouska

, Fernandez

M.R.

, Cerny

R.L.

, Xiao

, Eischen

C.M.

2005. MDM2 binds to nbs1 at sites of DNA damage and regulates double strand break repair. J Biol Chem, 280:18771–18781.

Arva

N.C.

, Gopen

T.R.

, Talbott

K.E.

, Campbell

L.E.

, Chicas

, White

D.E.

, Bond

G.L.

, Levine

A.J.

, Bargonetti

2005. A chromatin-associated and transcriptionally inactive p53-Mdm2 complex occurs in mdm2 SNP309 homozygous cells. J Biol Chem, 280:26776–26787.

Atwal

G.S.

, Bond

G.L.

, Metsuyanim

, Papa

, Friedman

, Distelman-Menachem

, Ben

A.E.

, Lancet

, Ross

D.A.

, Sninsky

, White

T.J.

, Levine

A.J.

, Yarden

2007. Haplotype structure and selection of the MDM2 oncogene in humans. Proc Natl Acad Sci U S A, 104:4524–4529.

Bond

G.L.

, Hirshfield

K.M.

, Kirchhoff

, Alexe

, Bond

E.E.

, Robins

, Bartel

, Taubert

, Wuerl

, Hait

, Toppmeyer

, Offit

, Levine

A.J.

2006a. MDM2 SNP309 accelerates tumor formation in a gender-specific and hormonedependent manner. Cancer Res, 66:5104–5110.

Bond

G.L.

, Hu

, Bond

E.E.

, Robins

, Lutzker

S.G.

, Arva

N.C.

, Bargonetti

, Bartel

, Taubert

, Wuerl

, Onel

, Yip

, Hwang

S.J.

, Strong

L.C.

, Lozano

, Lenine

A.J.

2004. A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell, 119:591–602.

Bond

G.L.

, Menin

, Bertorelle

, Alhopuro

, Aaltonen

L.A.

, Levine

A.J.

2006b. MDM2 SNP309 accelerates colorectal tumour formation in women. J Med Genet, 43:950–952.

Caren

, Erichsen

, Olsson

, Enerback

, Sjoberg

R.M.

, Abrahamsson

, Kogner

, Martinsson

2008. High-resolution array copy number analyses for detection of deletion, gain, amplification and copy-neutral LOH in primary neuroblastoma tumors: four cases of homozygous deletions of the CDKN2A gene. BMC Genomics, 9:353.

Chen

, Li

, Wei

, Sen

, Killary

A.M.

, Amos

C.I.

, Evans

D.B.

, Abbruzzese

J.L.

, Frazier

M.L.

2007. Aurora-A and p16 polymorphisms contribute to an earlier age at diagnosis of pancreatic cancer in Caucasians. Clin Cancer Res, 13:3100–3104.

10.

Chua

H.W.

, Ng

, Choo

, Lum

S.S.

, Li

, Soh

L.Y.

, Sabapathy

, Seow

2010. Effet of MDM2 SNP 309 and p53 codon 72 polymorphisms on lung cancer risk and survival among non-smoking Chinese women in Singapure. BMC Cancer, 10:88.

11.

Cobrinik

, Dowdy

S.F.

, Hinds

P.W.

, Mittnacht

, Weinberg

R.A.

1992. The retinoblastoma protein and the regulation of cell cycling. Trends Biochem Sci, 17:312–315.

12.

Debniak

, Scott

R.J.

, Huzarski

, Byrski

, Rozmiarek

, Debniak

, Zaluga

, Maleszka

, Kladny

, Gorski

, Cybulski

, Gronwald

, Kurzawski

, Lubinski

2005. CDKN2A common variants and their association with melanoma risk: a population-based study. Cancer Res, 65:835–839.

13.

Duong

, Boulle

, Daujat

, Chauvet

, Bonnet

, Neel

, Cavailles

2007. Differential regulation of estrogen receptor alpha turnover and transactivation by Mdm2 and stress-inducing agents. Cancer Res, 67:5513–5521.

14.

Fombonne

, Devouassoux

S.M.

, Bouvier

, Droz

J.P.

, Benahmed

, Krantic

2005. Analysis of p16INK4A gene promoter in male germ-cell tumors: identification of a new point mutation. Cancer Detect Prev, 29:1–7.

15.

Freedman

D.A.

, Levine

A.J.

1999. Regulation of the p53 protein by the MDM2 oncoprotein–thirty-eighth G.H.A. Clowes Memorial Award Lecture. Cancer Res, 59:1–7.

16.

Grady

W.M.

2004. Genomic instability and colon cancer. Cancer Metastasis Rev, 23:11–27.

17.

Hardy

R.G.

, Meltzer

S.J.

, Jankowski

J.A.

2000. ABC of colorectal cancer: molecular basis for risk factors. BMJ, 321:886–889.

18.

Harris

S.L.

, Levine

A.J.

2005. The p53 pathway: positive and negative feedback loops. Oncogene, 24:2899–2908.

19.

Holland

E.A.

, Beaton

S.C.

, Becker

T.M.

, Grulet

O.M.

, Peters

B.A.

, Rizos

, Kefford

R.F.

, Mann

G.J.

1995. Analysis of the p16 gene, CDKN2, in 17 Australian melanoma kindreds. Oncogene, 11:2289–2294.

20.

Houlston

R.S.

2001. What we could do now: molecular patology of colorectal cancer. J Clin Pathol Mol Pathol, 54:206–214.

21.

, Jin

, Wang

, Chen

, Wang

, Shen

2007. MDM2 promoter polymorphism SNP309 contributes to tumorsusceptibility: evidence from 21 case-control studies. Cancer Epidemiol Biomarkers Prev, 16:2717–2723.

22.

Huber

, Ramos

E.S.

2006. The P48T germline mutation and polymorphism in the CDKN2A gene of patients with melanoma. Braz J Med Biol Res, 39:237–241.

23.

Ibarrola-Villava

, Fernandez

L.P.

, Pita

, Bravo

, Floristan

, Sendagorta

, Feito

, Aviles

J.A.

, Martin-Gonzalez

, Lazaro

, Benitez

, Ribas

2010. Genetic analysis of three important genes in pigmentation and melanoma susceptibility: CDKN2A, MC1R and HERC2/OCA2. Exp Dermatol, 19:836–844.

24.

Kanthan

, Senger

J.L.

, Kanthan

S.C.

2012. Molecular events in primary and metastatic colorectal carcinoma: a review. Patholog Res Int, 2012:597.

25.

Knappskog

, Lonning

P.E.

2011. MDM2 promoter SNP285 and SNP309; phylogeny and impact on cancer risk. Oncotarget, 2:251–258.

26.

J.L.

, Cheng

Y.W.

, Chang

S.L.

, Su

J.M.

, Chen

C.Y.

, Lee

2000. MDM2 mRNA expression is a favorable prognostic factor in non-small-cell lung cancer. Int J Cancer, 89:265–270.

27.

Kumar

, Lundh

R.B.

, Louhelainen

, Hemminki

1998. Mutations in the CDKN2A (p16INK4a) gene in microdissected sporadic primary melanomas. Int J Cancer, 75:193–198.

28.

Kumar

, Smeds

, Berggren

, Straume

, Rozell

B.L.

, Akslen

L.A.

, Hemminki

2001. A single nucleotide polymorphism in the 3′ untranslated region of the CDKN2A gene is common in sporadic primary melanomas but mutations in the CDKN2B, CDKN2C, CDK4 and p53 genes are rare. Int J Cancer, 95:388–393.

29.

Larre

B.A.

, Cuellar

, Puig-Butille

J.A.

, Scarone

, Delgado

, Badenas

, Mila

, Malvehy

, Barquet

, Nunez

, Laporte

, Fernandez

, Levrero

, Martinez-Asuaga

, Puig

2009. CDKN2A mutations in melanoma families from Uruguay. Br J Dermatol, 161:1423.

30.

Levav-Cohen

, Haupt

2005. Mdm2 in growth signaling and cancer. Growth Factors, 23:183–192.

31.

, Zhai

, Zhang

, Chamberlain

R.M.

, Spitz

M.R.

, Wei

2006. MDM2 gene promoter polymorphisms and risk of lung cancer: a case-control analysis. Carcinogenesis, 27:2028–2033.

32.

Lum

S.S.

, Chua

H.W.

, Li

W.F.

, Rao

, Wei

, Shao

, Sabapathy

2008. MDM2 SNP309 G allele increases risk but the T allele is associated with earlier onset age of sporadic breast cancers in the Chinese population. Carcinogenesis, 29:754–761.

33.

McCloud

J.M.

, Sivakumar

, Greenhough

, Elder

, Jones

P.W.

, Deakin

, Elder

J.B.

, Fryer

A.A.

, Hoban

P.R.

2004. p16INK4a polymorphism: associations with tumour progression in patients with sporadic colorectal cancer. Int J Oncol, 25:1447–1452.

34.

Michael

, Oren

2003. The p53-Mdm2 module and the ubiquitin system. Semin Cancer Biol, 13:49–58.

35.

Miller

S.A.

, Dykes

D.D.

, Polesky

H.F.

1998. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 16:1215.

36.

Ohmiya

, Taquchi

, Mabuchi

, Itoh

, Hirooka

, Niwa

, Goto

2006. MDM2 promoter polymorphism is associated with both an increased susceptibility to gastric carcinoma and poor prognosis. J Clin Oncol, 24:4434–4440.

37.

Oliner

J.D.

, Kinzler

K.W.

, Meltzer

P.S.

, George

D.L.

, Vogelstein

1992. Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature, 358:80–83.

38.

Parkin

D.M.

, Bray

, Ferlay

, Pisani

2002. Global cancer statistics. CA Cancer J Clin, 55:74–108.

39.

Phang

B.H.

, Linn

Y.C.

, Li

, Sabapathy

2008. MDM2 SNP309 G allele decreases risk but does not affect onset age or survival of Chinese leukaemia patients. Eur J Cancer, 44:760–766.

40.

Sauroja

, Smeds

, Vlaykova

, Kumar

, Talve

, Hahka-Kemppinen

, Punnonen

, Jansen

, Hemminki

, Pyrhonen

2000. Analysis of G1/S checkpoint regulators in metastatic melanoma. Genes Chromosomes Cancer, 28:404–414.

41.

Shapiro

G.I.

, Park

J.E.

, Edwards

C.D.

, Mao

, Merlo

, Sidransky

, Ewen

M.E.

, Rollins

B.J.

1995. Multiple mechanisms of p16INK4A inactivation in non-small cell lung cancer cell lines. Cancer Res, 55:6200–6209.

42.

Sharpless

N.E.

2005. INK4a/ARF: a multifunctional tumor suppressor locus. Mutat Res, 576:22–38.

43.

Sherr

C.J.

1996. Cancer cell cycles. Science, 274:1672–1677.

44.

Shinora

, Sakano

, Hinoda

, Nishijima

, Kawai

, Misumi

, Nagao

, Hara

, Matsuyama

2009. Association of tp53 and MDM2 polymorphisms with survival in bladder cancer patients treated with chemoradiotherapy. Cancer Sci, 12:2376–2382.

45.

Toledo

, Wahl

G.M.

2006. Regulating the p53 pathway: in vitro hypotheses, in vivo veritas. Nat Rev Cancer, 6:909–923.

46.

Uchida

, Ando

, Maruyama

, Kobayashi

, Toda

, Ogawa

, Ozawa

, Matsuda

, Sugimura

, Kanno

, Baba

1998. Genetic alterations of mixed hyperplastic adenomatous polyps in the colon and rectum. Jpn J Cancer Res, 89:299–306.

47.

Wilkening

, Bermejo

J.L.

, Hemminki

2007. MDM2 SNP309 and cancer risk: a combined analysis. Carcinogenesis, 28:2262–2267.

48.

Yan

, Na

, Shan

, Xiao-Wei

, Wei

, Shu-Cheng

2008. p16 (CDKN2) gene polymorphism: association with histologic subtypes of epithelial ovarian cancer in China. Int J Gynecol Cancer, 18:30–35.

49.

Yasui

, Sentani

, Motoshita

, Nakayama

2006. Molecular pathobiology of gastric cancer. Scand J Surg, 95:225–231.

50.

Zheng

, Shen

, Sturgis

E.M.

, Wang

L.E.

, Shete

, Spitz

M.R.

, Wei

2002. Haplotypes of two variants in p16 (CDKN2/MTS-1/INK4a) exon 3 and risk of squamous cell carcinoma of the head and neck: a case-control study. Cancer Epidemiol Biomarkers, 11:1511.