β-Catenin and K-Ras Mutations and RASSF1A Promoter Methylation in Taiwanese Colorectal Cancer Patients

Abstract

Aims: The purpose of this study was to investigate the associations of β-catenin mutations, K-ras mutations, methylations of the RASSF1A promoter, and the survival of Taiwanese colorectal cancer (CRC) subjects who received 5-fluorouracil (5-FU) adjuvant chemotherapy. Results: The complete coding region of the K-ras gene and exon 3 and exon 4 of the β-catenin gene isolated from tumor tissues and adjacent normal colon tissues from 117 CRC subjects were sequenced, respectively. Methylations in the RASSF1A promoter region were also investigated. Various characteristics of the 117 subjects were recorded and used in the Cox proportional-hazard model analyses. Three missense mutations, one nonsense mutation, and one deletion were identified in the β-catenin gene. A 2 bp deletion was identified in the K-ras gene. We found that the frequencies of mutations in the β-catenin and K-ras genes were less pronounced in Taiwanese CRC subjects as compared with other populations. Methylations in the RASSF1A promoter region were detected in 73.5% (n=86/117) of the subjects, which was higher than in other studies. Methylations in the RASSF1A promoter have no significant effect on hazards for all CRC deaths caused in Taiwanese CRC patients. No interaction between 5-FU adjuvant chemotherapy and methylations of the RASSF1A promoter was observed. Conclusions: The mutation frequencies of β-catenin and K-ras genes in Taiwanese CRC patients are very low, which may suggest that they are not the dominant factors for CRC occurrence and prognosis in Taiwanese CRC patients. Methylation of RASSF1A promoter is independent of the prognosis for Taiwanese CRC patients. Taiwanese subjects differ from subjects of other populations with regard to β-catenin, K-ras, and RASSF1A presentations for CRC.

Introduction

Colorectal cancer (CRC) has become particularly prevalent in developed countries and is currently the third most common cause of death due to cancer in Taiwan (Department of Health Executive Yuan, R.O.C., 2012). Approximately 10% of all CRCs are attributed to hereditary factors (Weitz et al., 2005), and the remaining sporadic CRCs may be due to genetic and epigenetic changes such as regional DNA hypermethylation and overall DNA hypomethylation (Feinberg et al., 1988). Genetic alterations in adenomatous polyposis coli (APC) gene are regarded as playing an important role in CRC development (Lamlum et al., 2000). However, Taiwanese CRC subjects presented low mutation frequency of the APC gene in our previous study (Chen et al., 2006), and, thus, mutations in other genes associated with CRC are speculated. Previous studies suggested that the prognosis of chemotherapy treatment was associated with some genetic mutations or methylations, such as APC and K-ras genes (Chen et al., 2009a, 2009b; Kelley et al., 2011). Genetic testing of potential genes may find prognostic genetic predictors for chemotherapy treatment and their interactions with other genetic variations.

Aberrant activation of Wnt signaling has a major contribution in the pathogenesis of CRC (Behrens, 2005; Segditsas and Tomlinson, 2006). β-catenin is an APC down-regulating factor that plays a central role in the Wnt signaling pathway. Some CRCs lack APC mutations but present β-catenin mutations that block phosphorylation and degradation (Morin et al., 1997). Exon 3 of β-catenin harbors the hot spot for β-catenin mutations in various populations and CRC cell lines (Online Mendelian Inheritance in Man, 2012) and is a very important region coding the putative GSK3-mediated phosphorylation site. In the first part of this study, exon 3 and adjacent exon 4 of β-catenin were sequenced to screen the mutations in Taiwanese CRC patients.

The development of CRC was considered as multistage. Mutations in the genes of the Wnt signaling pathway cause the early formation of adenoma, and methylations or mutations in K-ras and other genes enhance CRC development. (Fearon and Vogelstein, 1990; de Vogel et al., 2009). Mutation rate of spread-throughout K-ras gene in colorectal tumors is about 35% to 45% and is associated with promoter hypermethylation of the DNA repair gene O⁶-methylguanine-DNA methyltransferase (MGMT) (Esteller et al., 2000). We have previously detected the promoter methylation status of MGMT in Taiwanese CRC patients (Chen et al., 2009a), and we sequenced the full coding region of K-ras in the second part of this study.

CRC is characterized by genetic and epigenetic alterations. RASSF1A is a novel Ras-binding protein and is regarded as a tumor suppressor. Hypermethylation of the RASSF1A promoter was considered an early epigenetic aberration for colorectal tumorigenesis (Greenspan et al., 2006), and it occurs predominantly in tumors with wild-type K-ras (van Engeland et al., 2002). Besides β-catenin and K-ras mutations, we also examined RASSF1A promoter methylation in the third part of this study. The correlations between patient characteristics, genetic and epigenetic alterations, the treatment and the survival rate of CRC subjects were analyzed.

Materials and Methods

Subjects

We studied 117 Taiwanese patients who had been diagnosed with CRC and had undergone surgical excision at the Tri-Service General Hospital between January 1, 2001, and December 31, 2004. Survival was calculated as the time from the date of cancer diagnosis to the date of death or end of follow-up (July 31, 2006). The histological diagnosis for CRC was made by two certified anatomical pathology specialists at the Department of Pathology, Tri-Service General Hospital. Colorectal tissues were obtained by surgery performed on the 117 CRC subjects. From each of the subjects, matched tumor and normal colorectal tissues were obtained earlier for DNA analysis. Normal colorectal tissues were collected at least 5 cm distant from the tumor. Some of the subjects received 5-fluorouracil (5-FU)-based adjuvant chemotherapy (5-FU, 600 mg/m² within 22 h for two consecutive days; combined leukovoren, 200 mg/m² within 2 h for two consecutive days). This treatment was performed once every 2 weeks and lasted for 8 months. Depending on the subjects' conditions and preferences, radiotherapy (4500 cGy; 25 fractions) was given. Subjects who received 5-FU-based adjuvant chemotherapy before surgery or within 2 weeks after surgery were recruited to participate in this study. The gender, age of diagnosis, cancer stage (Dukes, Kriklin classification), tumor differentiation, and chemotherapy and radiotherapy history were recorded for all the 117 subjects (Table 1). Written informed consents were obtained from all participants in this study, and the study protocol was approved by The Protection of Human Subjects Institutional Review Board, Tzu-Chi University and Hospital.

Table 1.

Clinical Characteristics of the Subjects Who Participated in this Study

Characteristics	Total number of subjects (n=117)
Age at diagnosis
years (mean±sd)	65.5±14.2
Gender (n/%)
male	65/55.6%
female	52/44.4%
Time of follow-up
months (median/range)	24.37 0.33-64.53
Dukes stage (n/%)
A	9/7.7%
B	49/41.9%
C	34/29.0%
D	25/21.4%
Differentiation^a (n/%)
Well	8/7.4%
Moderate	85/78.7%
Poor	15/13.9%
Tumor location^b (n/%)
Distal	29/29.3%
Proximal	21/21.2%
Rectal	49/49.5%
5-FU-based chemotherapy
(n/%)	61/52.1%
Radiotherapy
(n/%)	25/21.4%

Nine subjects were miss-recorded with regard to the differentiation data.

Eighteen subjects were miss-recorded with regard to the tumor location data.

5-FU, 5-fluorouracil.

DNA extraction

Genomic DNA was extracted from 50 mg of colorectal tumor and normal tissue using the Puregene Genomic DNA Purification Kit (Gentra Systems, Inc.). The extracted DNA was spectrophotometrically quantified (DU 640; Beckman Instruments, Inc.) before being subjected to further analysis.

Gene amplification and sequencing

The complete coding region of K-ras gene and exon 3 and exon 4 of β-catenin gene was amplified using six primer pairs (Table 2) by a polymerase chain reaction (PCR). All samples were amplified in a 50-μL reaction mixture, containing 50 ng genomic DNA, 10 mM Tris-HCl, 50 mM KCl, 2.5 mM MgCl₂, 0.2 mM of each dNTP, 200 mM of each primer, and 0.15 U AmpliTaq Gold DNA polymerase (Applied Biosystems; Roche Molecular Systems, Inc.). The thermal cycle profile was as follows: initial denaturation at 95°C for 10 min, 35 cycles of 30 s of denaturation at 95°C, 30 s of annealing at 55°C, and 1 min of extension at 72°C. A final 10-min extension at 72°C was added to ensure complete DNA polymerization. The PCR products were subjected to subsequent DNA sequencing. Sequencing reactions were performed using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Each sequencing reaction was amplified in a 10-μL reaction mixture containing 10 to 30 ng amplified DNA. The thermal cycle profile was as follows: initial denaturation at 95°C for 10 s, 25 cycles of 10 s of denaturation at 95°C, 5 s of annealing at 50°C, and 1 min of extension at 60°C. Primers used for direct sequencing reactions were identical to those used in the amplification reactions. Nucleotide sequences were determined using an ABI Prism 3100 Genetic Analyzer (Applied Biosystems) equipped with a long-read sequencing capillary and a POP-4 sequencing polymer.

Table 2.

The Primer Sets Used in this Study

	Forward (5′ to 3′)	Reverse (5′ to 3′)
K-ras 2	GGTACTGGTGGAGTATTTGA	GTATCAAAGAATGGTCCTGC
K-ras 3	GTGCACTGTAATAATCCAG	TGCATGGCATTAGCAAAGAC
K-ras 4	TGACAAAAGTTGTGGACAGG	AAGAAGCAATGCCCTCTCAA
K-ras 5	ACTTCTTGCACATGGCTTTC	GTAGTTCTAAAGTGGTTGCC
K-ras 6	ATTGACAAAACACCTATGCGG	TACCACTTGTACTAGTATGCC
β-catenin	AAGTAACATTTCCAATCTACT	TAGACATTCTGAAACTACTC
RASSF1A-U^a	GGTTTTGTGAGAGTGTGTTTAG	ACACTAACAAACACAAACCAAAC
RASSF1A-M^b	GGGGGTTTTGCGAGAGCGC	CCCGATTAAACCCGTACTTCG

Specific for unmethylated template.

Specific for methylated template.

Methylation-specific PCR

Methylations in the RASSF1A promoter were determined by methylated-specific PCR (MSP). Genomic DNA was modified by the bisulfite-mediated conversion of only unmethylated cytosines using the EpiTect Bisulfite Kit (QIAGEN). Previously reported primer sets were used for the amplification of the RASSF1A promoter (Belinsky et al., 2002). All samples were amplified in 30-μL of a reaction mixture containing 50 ng genomic DNA, 10 mM Tris-HCl, 50 mM KCl, 2.5 mM MgCl₂, 0.2 mM of each dNTP, 200 mM of each primer, and 0.2 U AmpliTaq Gold DNA polymerase (Applied Biosystems; Roche Molecular Systems, Inc.). The thermal cycle profile was as follows: initial denaturation at 95°C for 10 min, 40 cycles of 30 s of denaturation at 95°C, 45 s of annealing at 56°C, and 45 s of extension at 72°C. A final 10-min extension at 72°C was added at the end of the cycle. PCR products were loaded onto 3% NuSieve 3:1 agarose gels (Cambrex Bio Science Rockland, Inc.), stained with ethidium bromide, and visualized by UV transillumination.

Statistical analysis

We performed Spearman correlations to analyze the correlation between factors, including gender, age of diagnosis, cancer stage, tumor differentiation, K-ras mutations, and RASSF1A promoter methylations. We conducted multivariate analysis with the Cox proportional-hazard model to assess the effects of factors, including gender, age of diagnosis, cancer stage, chemotherapy, radiotherapy, Kras mutations, and RASSF1A promoter methylations, on the subjects' survival rate. We transformed the cancer stage into numeral scores to process Spearman correlations and Cox proportional-hazard model analyses using the following rules: for cancer stage, Dukes A=0, Dukes B=1, Dukes C=2, Dukes D=3; and for tumor location, proximal=0, distal=1, rectal=2. All tests were two tailed, and data sets were considered statistically significant when p<0.05. Statistical tests were carried out using the SPSS 10.0 software (SPSS Taiwan Corp.).

Results

β-catenin and K-ras mutations and methylations of RASSF1A promoter

The exon 3 and exon 4 of β-catenin gene of all 117 subjects enrolled in the study were sequenced, and there were three missense mutations, one nonsense mutation, and one deletion identified (Table 3). The mutations leading to p.T41A and p.S45F had been found in hepatoblastoma and hepatocellular carcinoma in previous studies (Bläker et al., 1999; Koch et al., 1999; Legoix et al., 1999). The mutations leading to p.E77× and p.I82V and the 595 bp deletion leading to frame shift were first found in this study. No reported mutations, such as the mutations leading to p.S33Y (Ilyas et al., 1997), were identified in our samples. For the K-ras gene, we sequenced the full coding region and found only a 2 bp deletion located in exon 4. Detailed characteristics of the patients with these mutations are presented in Table 3.

Table 3.

Characteristics of the Patients with β- Catenin and K-Ras Mutations Discovered Herein

Location	DNA change	Protein change	Sex	Age	Tumor grade ^a	Tumor location
β-catenin exon 3-exon 4	c.53_458del596bp	p.R18fs^b	male	47	A	proximal
β-catenin exon 3	c.121A>G	p.T41A	female	69	D	rectal
β-catenin exon 3	c.134C>T	p.S45F	male	81	A	proximal
β-catenin exon 3	c.229G>T	p.E77X^c	male	48	C	rectal
β-catenin exon 4	c.244A>G	p.I82V	male	81	A	proximal
K-ras exon 4	c.301_302delAA	p.K101fs	male	55	D	rectal

The NCBI Reference Sequence: NM_001904.3 for β-catenin, and M54968.1 for K-ras.

Duke's stage

fs, frame shift.

X, stop codon.

Methylations in the RASSF1A promoter region were detected in 73.5% (n=86/117) of the subjects, which is higher than in other studies (16% to 45%) (van Engeland et al., 2002; Wagner et al., 2002; van Engeland et al., 2003; Lee et al., 2004). With regard to the variables analyzed for Spearman correlation, methylations in the RASSF1A promoter region are not significantly associated with gender, age, cancer stage, or tumor location.

Survival analysis

To understand how gender, age of diagnosis, cancer stage, chemotherapy, radiotherapy, and methylations of RASSF1A promoter affect the survival rate of the CRC subjects, the Cox proportional-hazard model for univariable and multivariable analysis was processed (Table 4). The mutations of β-catenin and K-ras mutations were omitted in the analysis due to their low frequency. It appeared that only the cancer stage significantly enhanced the subjects' hazards (p<0.001; Table 4). The cancer-stage variable and target variables, RASSF1A methylation and chemotherapy, were included in the multivariable analysis done. However, the adjusted hazard ratios for RASSF1A methylation remained nonsignificant.

Table 4.

Cox Proportional-Hazard Model Analysis

	Univariable (s) (n=117)				Multivariable (s) (n=117)
	All cause death		CRC death		All cause death		CRC death
Variables	Hazard ratio (95% CI)	p	Hazard ratio (95% CI)	p	Hazard ratio (95% CI)	p	Hazard ratio (95% CI)	p
Male gender	1.389 (0.674-2.8642)	0.373	1.420 (0.644-3.130)	0.385	NA	NA	NA	NA
Diagnosis age	1.028 (0.999-1.057)	0.061	1.015 (0.986-1.045)	0.318	NA	NA	NA	NA
Tumor location	1.347 (0.818-2.218)	0.242	1.234 (0.731-2.083)	0.431	NA	NA	NA	NA
Duke's stage	2.654 (1.718-4.098)	<0.001^a	3.053 (1.866-4.995)	<0.001^a	3.132 (2.000-4.904)	<0.001^a	3.639 (2.182-6.068)	<0.001^a
RASSF1A methylation	1.137 (0.490-2.638)	0.766	1.810 (0.624-5.254)	0.275	1.762 (0.739-4.203)	0.201	2.905 (0.974-8.664)	0.056
Chemotherapy	0.715 (0.351-1.458)	0.356	0.994 (0.458-2.157)	0.987	0.413 (0.197-0.863)	0.019^a	0.517 (0.231-1.161)	0.110
Radiotherapy	1.027 (0.442-2.388)	0.950	1.036 (0.415-2.583)	0.940	NA	NA	NA	NA

p<0.05.

CI, confidence interval; CRC, colorectal cancer; NA, not applicable.

To further investigate the interaction between chemotherapy and methylations of the RASSF1A promoter to the survival of the CRC subjects, RASSF1A methylation plus chemotherapy variable (vs. no RASSF1A methylation or Chemotherapy) was added to the multivariable Cox proportional-hazard model. The results revealed no significant interaction between chemotherapy and methylations of the RASSF1A promoter to the survival of the CRC subjects.

Discussion

We found a very low mutation frequency of β-catenin and K-ras genes in Taiwanese CRC patients. In this study and our previous studies (Chen et al., 2006; Chen et al., 2009a, 2009b), we found that Taiwanese CRC patients present a lower frequency of APC, β-catenin, and K-ras mutation than other populations, but methylations in the APC, MGMT, and RASSF1A promoters were higher than in other studies. These data show that the majority of the studied CRCs with RASSF1A promoter methylation lack K-ras mutations and are contrary to a previous report (van Engeland et al., 2002). Therefore, it may suggest that epigenetic aberration may be the dominant factors that impact CRC development in Taiwanese CRC patients, or alternatively, the mutation patterns for CRC in Taiwanese patients are very much different from the patients in other populations.

With regard to the variables analyzed for Spearman correlation, no correlation among the various variables was observed when the relationship between RASSF1A promoter methylation and the cancer stage was considered. The chi-square test also showed that RASSF1A promoter methylation was independent of the cancer stage (p=0.508), suggesting that RASSF1A promoter methylations are similarly distributed over the Dukes A to D stages. It implies that these events may occur at the early stage of tumorigensis, and we propose that RASSF1A promoter methylations may serve as an early stage biomarker for detecting tumorigensis in Taiwanese CRC patients. This result was consistent with APC or MGMT promoter methylations (Chen et al., 2009a). However, the application should be further verified by additional studies with tissues harvested at more various stages of CRC tumor development.

Multivariable Cox proportional-hazard model analyses show that methylations in the RASSF1A promoter have no significant effect on hazards for CRC deaths caused in Taiwanese CRC patients. In addition, we did not observe a significant interaction between chemotherapy and RASSF1A promoter methylations to the survival of the CRC subjects in this study. It seems that methylations in the RASSF1A promoter may be more significantly associated with the occurrence of CRC, instead of having effects on the prognosis of CRC in Taiwanese patients. Taiwanese subjects differ from subjects of other populations with regard to the phenotypic presentation of CRC. The dominant molecular variants associated with CRC in Taiwanese subjects may be other uncommon genes variants. In conclusion, our study showed that Taiwanese CRC subjects have lower incidences of β-catenin and K-ras mutations and higher methylatons in RASSF1A promoters as compared with other ethnic CRC subjects who manifest a similar phenotypic presentation. Further studies with other genetic and/or epigenetic variants are needed to decipher the significance.

Footnotes

Acknowledgments

This study was supported in part by grants from the Buddhist Tzu-Chi General Hospital, Hualien, Taiwan.

Author Disclosure Statement

No competing financial interests exist.

References

Behrens

. 2005. The role of the Wnt signalling pathway in colorectal tumorigenesis. Biochem Soc Trans, 33:672-675.

Belinsky

, Palmisano

, Gilliland

et al. 2002. Aberrant promoter methylation in bronchial epithelium and sputum from current and former smokers. Cancer Res, 62:2370-2377.

Bläker

, Hofmann

, Rieker

et al. 1999. Beta-catenin accumulation and mutation of the CTNNB1 gene in hepatoblastoma. Genes Chromosomes Cancer, 25:399-402.

Chen

, Chiu

, Wu

et al. 2009a. The association of methylation in the promoter of APC and MGMT and the prognosis of Taiwanese CRC patients. Genet Test Mol Biomarkers, 13:67-71.

Chen

, Tsai

, Jao

et al. 2006. Single nucleotide polymorphisms of the APC gene and colorectal cancer risk: a case-control study in Taiwan. BMC Cancer, 6:83.

Chen

, Wu

, Lin

et al. 2009b. Prognostic significance of interaction between somatic APC mutations and 5-fluorouracil adjuvant chemotherapy in Taiwanese colorectal cancer subjects. Am J Clin Oncol, 32:122-126.

Department of Health Executive Yuan, R.O.C. 2012. www.doh.gov.tw/EN2006/index_EN.aspx. June 11 2012.

de Vogel

, Weijenberg

, Herman

et al. 2009. M. MGMT and MLH1 promoter methylation versus APC, KRAS and BRAF gene mutations in colorectal cancer: indications for distinct pathways and sequence of events. Ann Oncol, 20:1216-1222.

Esteller

, Toyota

, Sanchez-Cespedes

et al. 2000. Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase by promoter hypermethylation is associated with G to A mutations in K-ras in colorectal tumorigenesis. Cancer Res, 60:2368-2371.

10.

Fearon

, Vogelstein

. 1990. A genetic model for colorectal tumorigenesis. Cell, 61:759-767.

11.

Feinberg

, Gehrke

, Kuo

et al. 1988. Reduced genomic 5-methylcytosine content in human colonic neoplasia. Cancer Res, 48:1159-1161.

12.

Greenspan

, Jablonski

, Rajan

et al. 2006. Epigenetic alterations in RASSF1A in human aberrant crypt foci. Carcinogenesis, 27:1316-1322.

13.

Ilyas

, Tomlinson

, Rowan

et al. 1997. Beta-catenin mutations in cell lines established from human colorectal cancers. Proc Natl Acad Sci U S A, 94:10330-10334.

14.

Kelley

, Wang

, Venook

. 2011. Biomarker use in colorectal cancer therapy. J Natl Compr Canc Netw, 9:1293-1302.

15.

Koch

, Denkhaus

, Albrecht

et al. 1999. Childhood hepatoblastomas frequently carry a mutated degradation targeting box of the beta-catenin gene. Cancer Res, 59:269-273.

16.

Lamlum

, Papadopoulou

, Ilyas

et al. 2000. APC mutations are sufficient for the growth of early colorectal adenomas. Proc Natl Acad Sci U S A, 97:2225-2228.

17.

Lee

, Hwang

, Lee

et al. 2004. Aberrant CpG island hypermethylation of multiple genes in colorectal neoplasia. Lab Invest, 84:884-893.

18.

Legoix

, Bluteau

, Bayer

et al. 1999. Beta-catenin mutations in hepatocellular carcinoma correlate with a low rate of loss of heterozygosity. Oncogene, 18:4044-4046.

19.

Morin

, Sparks

, Korinek

et al. 1997. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science, 275:1787-1790.

20.

Online Mendelian Inheritance in Man. 2012. CATENIN, BETA-1; CTNNB1. omim.org/entry/116806?search=Beta-Catenin&highlight=betacatenin. 2012 June 11.

21.

Segditsas

, Tomlinson

. 2006. Colorectal cancer and genetic alterations in the Wnt pathway. Oncogene, 25:7531-7537.

22.

van Engeland

, Roemen

, Brink

et al. 2002. K-ras mutations and RASSF1A promoter methylation in colorectal cancer. Oncogene, 21:3792-3795.

23.

van Engeland

, Weijenberg

, Roemen

et al. 2003. Effects of dietary folate and alcohol intake on promoter methylation in sporadic colorectal cancer: the Netherlands cohort study on diet and cancer. Cancer Res, 63:3133-3137.

24.

Wagner

, Cooper

, Grundy

et al. 2002. Frequent RASSF1A tumour suppressor gene promoter methylation in Wilms' tumour and colorectal cancer. Oncogene, 21:7277-7282.

25.

Weitz

, Koch

, Debus

et al. 2005. Colorectal cancer. Lancet, 365:153-165.