CpG island methylator phenotype association with upregulated telomerase activity in hepatocellular carcinoma

Abstract

CpG island methylator phenotype (CIMP) involves the targeting of multiple genes by promoter hypermethylation. Telomerase plays an important role in the development of cellular immortality and oncogenesis. To gain insight into the role of epigenetic aberration of telomerase-related genes in hepatocarcinogenesis, we determined a hypermethylation profile in HCC. We examined the promoter methylation status of 9 genes associated with telomerase activity in 120 HCC, 120 cirrhosis tissues and 10 normal liver tissues by methylation-specific PCR. Assay of telomerase activity was by TRAP-ELISA. The frequency of promoter methylation of each gene was P21 63.3%, P15 42.5%, P16 62.5%, P53 14.2%, RB 32.5%, P27 48.3%, WTI 54.2%, E2F-1 70.8% and P300 65.8% of 120 HCC. Methylation status of P21, P15, P16, WTI and E2F-1 was significantly associated with HCC and nontumor tissues (p < 0.05). CIMP+ was detected in 61.7% (74/120) HCC and 15% (18/120) cirrhosis tissues, no CIMP+ was present in normal liver tissues (p < 0.001). A significant difference between CIMP status and metastasis was been found in HCC (p < 0.001). Results showed that 94.6% (70/74) HCC and 55.6% (10/18) cirrhosis patients with CIMP+ show expression of high telomerase activity than 45.5% (10/22) HCC and 6.25% (1/16) cirrhosis patients with CIMP− (p < 0.001). CIMP lead to high levels of expression of telomerase activity through the simultaneous inactivation of multiple genes associated with telomerase activity by concordant methylation. © 2008 Wiley-Liss, Inc.

Hepatocellular carcinoma (HCC) is one of the most common malignancies in the world, and the prognosis of patients with HCC is very poor.1 Chronic hepatitis infection and cirrhosis are well-documented risk factors for HCC.2, 3

Telomerase are situated at the ends of linear chromosomes and protect them from degradation as well as end-to-end fusions.4 In contrast to most somatic cells, germ, stem and tumor cells have the ability to maintain telomere length, a prerequisite for their unlimited replication potential. In humans, telomerase are typically 10 kb long and consist of TTAGGG repeats.5, 6 Telomerase plays an important role in the development of cellular immortality and oncogenesis.7 Previous studies have shown that telomerase activity is found in 85–90% of all human tumors, in 22.5%–45.9% of their adjacent normal cells.8, 9 The rate of telomere DNA shortening is regulated by telomerase expression and activity.10 A number of different mechanisms have been shown to regulate telomerase activity. A more general repressor of telomerase activity could be p53,11, 12 which could form complexes with sp1 and may thereby prevent sp1 from binding to the hTERT promoter and downregulate telomerase expression.13 Endogenous P53 represses telomerase in a cell type-specific manner through an indirect mechanism and is mediated by the p21/E2F pathway.14 The studies have shown that inactivation of the tumor suppressor p16 (or pRB) and telomerase activation is both necessary and sufficient for cellular immortalization in a cell culture model.15 Induction of Rb expression is sufficient to downregulate telomerase activity in human carcinoma cell lines.16 P16 and p15 might also be involved in the maintenance of senescence through downregulation of terlomerase.17, 18 Inactivation of p16/p15- and p14ARF-dependent pathways possibly in conjunction with telomerase activation might be critical steps for a meningioma cell towards escape from senescence, that is, immortalization.19 These findings suggest a critical role for inactivation of a cell cycle checkpoint at the G1/S boundary and telomerase activation in cellular immortalization and malignant progression. In addition, human cancer cell lines stably over-expressing E2F-1 exhibited decreased hTERT mRNA expression and telomerase activity.20 A novel tumor suppressor p27 also has been shown to be a negative regulator of hTERT expression and telomerase activity through post-transcriptional upregulation by INF-γ/IRF-1 signaling.21 Meanwhile, the transcription factor p30022 and the Wilms' Tumor 1 suppressor gene (WT1)23 may be, as transcriptional repressor, repress the telomerase enzyme activities at least in some specific cells.

Transcriptional inactivation by cytosine methylation at promoter CpG islands of tumor suppressor genes is thought to be an important mechanism in human carcinogenesis.24 Tumor related genes are silenced by promoter methylation in HCC.25–27 It has previously been reported that a subset of colorectal and gastric tumors display CpG island methylator phenotype (CIMP), affecting multiple genes in a single tumor.28, 29 CIMP tumors are a distinct group of tumors that are defined by a high degree of concordant CpG island hypermethylation of genes. CIMP is now thought to be a new, distinct, yet major pathway of tumorigenesis.24 The concordant CpG island hypermethylation of genes association with telomerase could play important role in HCC with CIMP+, leading to upregulating telomerase activity. But, the role of the CIMP pathway in the tumor evolution of HCC is still due to only one or a few genes were analyzed in the previous epigenetic studies. However, no study to date has examined the relationship between telomerase activation and CIMP in HCC. In this study, we examined inter-relationships between promoter methylation of multiple key genes associated with telomerase activation, telomerase activation and CIMP in HCC by MSP.

Material and methods

Patients and specimens

We studied 120 HCC patients and 120 patients with cirrhosis. Cirrhosis is characterized anatomically by widespread nodules in the liver combined with fibrosis. Cirrhosis patients were selected on the basis of clinical, biochemical and radiological findings and follow-up were performed for at least 6 months to exclude undetected cancers. HCC patients were selected consecutively from those diagnosed according to the criteria of the European association for the study of the liver (EASL).30 The 10 normal liver samples were obtained from 10 adults who suffered accidental liver injury. Tissue specimens were obtained from patients whose frozen tissue samples had been stored in the surgical pathology files of our hospital between September 2002 and December 2005.

The HCC patients consisted of 106 men and 14 women, ranging in age from 31 to 78 (mean ± SD, 52.8 ± 10.2) years. The cirrhosis patients consisted of 102 men and 18 women, ranging in age from 32 to 75 (mean ± SD, 51.3 ± 9.4) years. The diagnosis was confirmed histologically in all cases, based mainly on examinations of sections stained with H&E. All tumors were histologically diagnosed as HCC according to the Edmondson-Steiner classification system31 (20 cases were grade I, 40 cases were grade II, 48 cases were grade III, 12 cases were grade IV). Of the 120 HCC patients, serologic examinations indicated that 81 cases were both hepatitis B surface antigen-positive and hepatitis C virus antibody (anti-HCV)-positive. Written informed consent was obtained from each patient, and the protocol of the study was approved by the local ethics committee.

Sodium bisulfite treatment

Nontumor and tumor samples were digested overnight at 50°C with proteinase K buffered in 1% SDS (pH = 8). Genomic DNA of tumor or normal tissues was isolated by phenol-chloroform extraction and ethanol precipitation. DNA was treated with sodium bisulfites as described previously.32 Briefly, 2 μg of genomic DNA was resuspended in 50 μl of water and then denatured in 2 M NaOH for 10 min at 37°C. The denatured DNA was diluted in 550 μl of freshly prepared solution containing 10 mM hydroquinone (Sigma-Aldrich, St.Louis, USA) and 3 M sodium bisulfite (Sigma-aldrich). The resultant solution was covered with mineral oil and incubated for 16 hr at 50°C. After incubation, the samples were desalted using a Wizard DNA Clean-Up System (Promega, Madison, USA) and treated with 3 M NaOH for 5 min at room temperature. Then 66 μl of NH₄oAc and 2 volumes of 100% ethanol were added, and the DNA was precipitated for at least 1–2 hr at −80°C. After precipitation, the pellets were washed with 70% ethanol, dried, resuspended in 20 μl of water, and stored at −20°C.

MSP (methylation-specific PCR)

DNA methylation was determined by MSP.32 MSP distinguishes unmethylated alleles of a given gene based on DNA sequence alterations after bisulfited treatment of DNA, which converts unmethylated but not methylated cytosines to uracils. Subsequent PCR using primers specific to sequences corresponding to either methylated or unmethylated DNA sequences is then performed. Primers for MSP were reported previously: P2133; P15, P1634; P5335; RB36; P2737; E2F-138; WT127; P300.39 Primer sequences are summarized in Table I. The modified genomic DNA samples were PCR amplified in a total volume of 50 μl. The PCR was performed in a thermal cycler. The PCR products were separated on 2% agarose gels and visualized using ethidium bromide staining.

Table I. Primer Sequences for MSP Analysis

Gene			Primer sequence (5′-3′)	Product Size (bp)	References
1 M, methylated sequence; U, unmethylated sequence. 2 F, forward sequence; R, reverse sequence.
P21	U1	F2	GGATTGGTTGGTTTGTTGGAATTT	161	(33)
		R	ACAACCCTAATATACAACCACCCCA
	M	F	TACGCGAGGTTTCGGGATC	171
		R	CCCTAATATACAACCGCCCCG
P15	U	F	TGTGATGTGTTTGTATTTTGTGGTT	154	(34)
		R	CCATACAATAACCAAACAACCAA
	M	F	GCGTTCGTATTTTGCGGTT	148
		R	CGTACAATAACCGAACGACCGA
P16	U	F	TTATTAGAGGGTGGGGTGGATTGT	151	(34)
		R	CAACCCCAAACCACAACCATAA
	M	F	TTATTAGAGGGTGGGGCGGATCGC	150
		R	GACCCCGAACCGCGACCGTAA
P53	U	F	TTGGTAGGTGGATTATTTGTTT	247	(35)
		R	CCAATCCAAAAAAACATATCAC
	M	F	TTCGGTAGGCGGATTATTTG	193
		R	AAATATCCCCGAAACCCAAC
RB1	U	F	GGGAGTTTTGTGGATGTGAT	172	(36)
		R	ACATCAAAACACACCCCA
	M	F	GGGAGTTTCGCGGACGTGAC	172
		R	ACGTCGAAACACGCCCCG
P27	U	F	ATGGAAGAGGTGAGTTAGT	212	(37)
		R	AAAACCCCAATTAAAAACA
	M	F	AAGAGGCGAGTTAGCGT	195
		R	AAAACGCCGCCGAACGA
E2F-1	U	F	TTAGAGTTTTTTTGTGGTAAAAAGGAG	240	(38)
		R	AATCCTCCTTTTTACCACAAAAAAACT
	M	F	CTAGAGCTCTTTCGCGGCAAAAAGGAG	240
		R	GATCCTCCTTTTTGCCGCGAAAGAGCT
WT1	U	F	TGGGATTTGGGTGGTATTTG	216	(27)
		R	CACCAACACCCACTACACCA
	M	F	GTTAGGCGTCGTCGAGGTTA	206
		R	AAAACGCAAAATCCAACACC
P300	U	F	TGTTGTTTGGTTTGGTTTTTTT	138	(39)
		R	CACAAAAAACTCACCCAAACCA
	M	F	CGTTGTTCGGTTCGGTTTTTTC	138
		R	CGCAAAAAACTCGCCCGAACCG

Assay of telomerase activity by TRAP-ELISA

TRAP-ELISA was performed by the TRAP-ELISA assay as described by Cheng et al.40 using telomerase kit the Telo TAGGG Telomerase PCR ELISA PLUS (Roche Diagnostics, Mannheim, Germany), according to the manufacturer's instructions. As positive control for the assay, extracts from HEK293 cells were used, and cell lysates were heat-inactivated for 10 min at 85°C also used as negative controls. Absorbance values are reported as the A450 nm reading after subtraction of the control value. Absorbance values of less than 0.2 were considered as negative for telomerase activity. Absorbance values in the range of 0.2–0.5 were considered as low telomerase activity. Absorbance values of greater than 0.5 were considered as high telomerase activity.

Statistical analysis

All data were generated without knowledge of the clinical status of the samples analyzed. Values for the clinical and biological characteristics of patients were expressed as means ± SD. Comparison was done with Student's t-test (unpaired). Univariate analyses of the interaction between methylation and clinical parameters were performed with Pearson's χ² test and Fisher's exact test was used to examine the tissue samples with the low expected values. All p values presented were two-sided, and a p value of less than 0.05 was considered statistically significant. All statistical analyses were carried out using SPSS 13.0 software (SPSS, Inc., Chicago, USA).

Results

Frequency of CpG island hypermethylation in HCC, cirrhosis samples and normal liver tissues

We examined the hypermethylation status of a panel of 9 genes associated with telomerase activity: P21, P15, P16, P53, RB, P27, WTI, E2F-1 and P300. The frequency of promoter methylation of 9 genes was determined using MSP in 120 cases of HCC, 120 cases of cirrhosis tissues, and 10 normal liver tissues. For the methylation status assay results, the hypermethylated contains only methylated PCR product, the partially methylated contains both methylated and unmethylated PCR products, and the unmethylated contains only unmethylated product. Representative examples of methylation-specific PCR assay results are presented in Figure 1 and overall results are summarized in Table II. The frequency of promoter methylation of genes included in the panel was P21 63.3%, P15 42.5%, P16 62.5%, P53 14.2%, RB 32.5%, P27 48.3%, WTI 54.2%, E2F-1 70.8% and P300 65.8% of 120 HCC. The methylation frequency of genes was much lower in both cirrhosis and normal liver tissues. Methylation status of P21, P15, P16, WT1 and E2F-1 was significantly difference between HCC and cirrhosis tissues (p = 0.003, p < 0.001, p < 0.001, p < 0.001 and p < 0.001, respectively). Methylation was also more frequent in HCC than nontumor for P53 (14.2% versus 8.3%), RB (32.5% vs. 25%), P27 (48.3% vs. 36.7%), and P300 (65.8% vs. 54.2%), but these changes were not statistically significant (p = 0.153, p = 0.199, p = 0.068, and p = 0.065, respectively). Only one case was methylated in P21 and one case was methylated in P16 in normal liver tissues. There was no methylation of P15, P53, RB, P27, WTI, E2F-1 and P300 in normal liver tissues.

Figure 1. Representative MSP experiments for methylation analysis of 9 genes. PCR products amplified with unmethylated (U) and methylated (M) sequence-specific primers. Distilled water was used as a negative control, DNA methylated by SssI methylase (Sss DNA) was used as a positive control (pos). “T” indicated HCC samples, “N” indicated cirrhosis tissues, and 105 was a normal liver tissue. Gene names were indicated at the left of each panel.

Table II. Frequency of Promoter Methylation of Each Gene in HCC, Cirrhosis and Normal Liver Tissue

Gene	Number (%) of methylated samples			p value1
	HCC (N = 120)	Cirrhosis (N = 120)	Normal tissue (N = 10)
1 HCC tumor tissues versus cirrhosis tissues.
P21	76 (63.3)	53 (44.2)	1 (10.0)	0.003
P15	51 (42.5)	23 (19.2)	0 (0)	<0.001
P16	75 (62.5)	28 (23.3)	1 (10.0)	<0.001
P53	17 (14.2)	10 (8.3)	0 (0)	0.153
RB	39 (32.5)	30 (25.0)	0 (0)	0.199
P27	58 (48.3)	44 (36.7)	0 (0)	0.068
WT1	65 (54.2)	22 (18.3)	0 (0)	<0.001
E2F-1	85 (70.8)	29 (24.2)	0 (0)	<0.001
P300	79 (65.8)	65 (54.2)	0 (0)	0.065

Frequency of CpG island methylator phenotype (CIMP) in HCC, cirrhosis and normal liver tissue

We examined the promoter methylation status using MSP in 9 genes, and found that 98.3% (118/120) of HCC showed methylation of at least 1 gene, and only 2 tumors showed no methylation of any of the 9 genes. A total of 3.3% (4/120) of HCC had 1 gene, 4.2% (5/120) 2 genes, 8.3% (10/120) 3 genes, 20.8% (25/120) 4 genes, 42.5% (51/120) 5 genes, 12.5% (15/120) 6 genes, and 6.7% (8/120) 7 genes hypermethylated (Fig. 2a). The mean number of genes hypermethylated in each tumor was 4.54. A total of 17.5% (21/120) of cirrhosis had 1 gene, 23.3% (28/120) 2 genes, 20% (24/120) 3 genes, 12.5% (15/120) 4 genes, 10.83% (13/120) 5 genes and 4.2% (5/120) 6 genes hypermethylated (Fig. 2b). Fourteen cirrhosis tissues showed no methylation of any the 9 genes. The mean number of genes hypermethylated in each nontumor was 2.53. However, no normal liver tissue had 2 genes hypermethylated in normal liver tissues (Fig. 2c).

Figure 2. The distributions of the number of promoter methylated genes per sample in 120 HCC (a), 120 cirrhosis tissues (b), and 10 normal liver tissuses (c), and the frequency of the CIMP in HCC, cirrhosis tissue, and normal liver tissue (d). CIMP is defined by the number of CpG islands methylated in each sample as follows: CIMP− (less than 5 methylated genes); CIMP+ (more than 5 methylated genes). CIMP+ was observed in 61.7% (74/120) HCC, 15% (18/120) cirrhosis tissues, and none of normal liver tissues (0/10). CIMP status is strongly associated with HCC, nontumor tissues, and normal liver tissues (p < 0.001).

Originally, CIMP-positive gastric cancer was defined as a tumor with methylation at more than 3 gene methylated in tumors.29 CIMP were analyzed either according to the average number of methylated genes per tumor.41, 42 A new study taking an unbiased approach offered new markers to define a CIMP concept, in which the threshold distinguishing CIMP+ from CIMP− samples was chosen by minimizing the within group sum of squared errors.43 In this study, CIMP status was classified as CIMP+ samples (with 5 or more methylated genes) and CIMP− (samples with 4 or fewer methylated genes) based on above criteria for CIMP status in several types of tumors, because the average number of methylated genes per tumor was 4.54. The five-marker panel was best satisfied all of the criteria described above and retained a high ranking in its ability to explain the percent of variance by the CIMP definition.

We investigated CIMP status in 120 HCC, 120 corresponding nontumor tissues, and 10 normal liver tissues. Of the HCC, 61.7% (74/120) were CIMP+, 38.3% (46/120) were CIMP−. In contrast, CIMP+ was present in 15% (18/120) of cirrhosis tissues; CIMP− was 85% (102/120). Meanwhile, no CIMP+ case was present in normal liver tissues (Fig. 2d). There was a statically significant difference between CIMP status with different samples, including HCC, nontumors, and normal liver tissues (p < 0.001).

CIMP correlates with clinicopathological features

Using statistical analysis, we examined CIMP with regard to HCC patient clinicopathological parameters of gender, tumor size, smoking history, drinking alcohol, HBV, HCV, node number and metastasis (Table III). A significant difference between CIMP status and metastasis was found in HCC (p < 0.001), which suggested that HCC patients with CIMP+ could have a poor prognosis. However, other clinicopathological features did not differ significantly between CIMP+ and CIMP− groups.

Table III. CIMP Status Association with Clinicophthologic Variables in HCC

Variable		Cases	CIMP(+) (N = 74)	CIMP(−) (N = 46)	p value1
1 Statistical significance determined using the Pearson's χ^2 test.
Sex	Male	106	68	38	0.124
	Female	14	6	8
Tobacco	Yes	58	30	28	0.878
	No	62	44	18
Alcohol	Yes	41	26	15	0.777
	No	79	48	31
HBV+HCV	Yes	81	49	32	0.703
	No	39	25	14
Tumor Size (cm)	≥5	67	37	30	0.103
	<5	53	37	16
Node number	Multi	18	10	8	0.563
	Single	102	64	38
Metastasis	Yes	54	43	11	<0.001
	No	66	31	35

CIMP correlates with telomerase activity

CIMP appears to be involved in an important new tumorigenesis pathway that leads to cancer progression by simultaneously inactivating multiple tumor-related genes. Thus, we analyzed telomerase activity associated with CIMP involved in the methylation status of 9 genes associated with telomerase activity. The telomerase activity was detected by TRAP-ELISA assay. The values of telomerase activity in both types of samples could be arbitrarily divided into 3 categories: high, ≥0.5; low, 0.2–0.5 and negative, <0.2. Table IV shows the telomerase activity in 120 HCC samples, 120 cirrhosis tissues and 10 normal liver tissues. 80% (96/120) of HCC patients were telomerase-positive, but only 28.3% (34/120) of cirrhosis patients were telomerase-positive, and all 10 normal liver tissues were telomerase-negative. The telomerase activity had a statistically significant difference between in HCC, cirrhosis and normal liver tissues (p < 0.05).

Table IV. Telomerase Activity in HCC, Cirrhosis, and Normal Liver Samples and CIMP Status in Different Liver Samples

	No. cases	Telomerase activity (A450nm)			p value1
		High (≥ 0.5)	Low (0.2–0.5)	Negative (< 0.2)
1 Statistical significance determined using the Pearson's χ^2 test. 2 HCC tumor tissues versus cirrhosis tissues. 3 HCC tumor tissues versus normal liver tissues. 4 Cirrhosis tissues versus normal liver tissues.
HCC	120	80	16	24	<0.0012
Cirrhosis	120	11	23	86	<0.0013
Normal	10	0	0	10	0.0424
HCC
CIMP+	74	70	4	0	<0.001
CIMP−	46	10	12	24
Cirrhosis
CIMP+	18	10	8	0	0.002
CIMP−	102	1	15	86

The CIMP status and telomerase activity were studied in different samples (Table IV). No CIMP+ case was detected in all samples (included HCC, cirrhosis and normal liver tissues) with telomerase-negative. Thus, we analyzed CIMP status in HCC and cirrhosis with positive expression of telomerase activity. Interestingly, we found that the 94.6% (70/74) HCC (Fig. 3a) and 55.6% (10/18) (Fig. 3b) cirrhosis patients with CIMP+ showed expression of high telomerase activity (Absorbance ≥0.5). Meanwhile, only 45.5% (10/22) of HCC and 6.25% (1/16) of cirrhosis patients with CIMP− showed expression of high telomerase activity. There was a significant difference between CIMP status with telomerase activity (p < 0.05). Tissues with CIMP+ showed more frequency expression of high levels of telomerase activity. Therefore, we tentatively concluded that CIMP could play a potential role for expression of telomerase activity through the occurrence of such promoter hypermethylation genes associated with telomerase activity in HCC.

Figure 3. CIMP status in HCC and cirrhosis with positive expression of telomerase activity. We analyzed CIMP status in HCC and samples with telomerase positive. Telomerase activity was divided into 2 groups according to absorbance (A450 nm): high group (≥0.5) and low group (0.2–0.5). 94.6% (70/74) HCC and 55.6% (10/18) cirrhosis patients with CIMP+ show expression of high telomerase activity (Absorbance ≥0.5) than 45.5% (10/22) HCC and 6.25% (1/16) cirrhosis patients with CIMP−. There was a significant difference between CIMP status with telomerase activity (p < 0.05).

Discussion

The development and progression of HCC is a multiplestep process,44 the understanding of the molecular pathways of hepatocarcinogenesis is limited, although recent molecular biological studies have led to rapid progress in the understanding of the molecular events involved. Epigenetic gene silencing is increasingly being recognized as a common way in which cancer cells inactivate cancer-related genes.45, 46 The multiple genes inactivated by promoter region hypermethylation provides an opportunity to examine the patterns of inaction of such genes among different tumors.24, 47 CIMP refers to the notion that a subset of tumors has widespread methylation of CpG islands that leads to epigenetic inactivation of tumor suppressor genes by promoter methylation.48

Our results indicate that the methylation of multiple genes associated with telomerase activity is a common phenomenon in HCC. The frequency of promoter methylation of 9 genes varied from 14.2% for P53 and 70.8% for E2F-1, including P21, P15, P16, P53, RB, P27, WTI, E2F-1, and P300. Promoter hypermethylation of a group of 5 genes (including P21, P15, P16, WTI, and E2F-1) in HCC tissues was more common than in cirrhosis tissues, which suggest that these genes play an important role in hepatocarcinogenesis.

Extensive methylation of multiple CpG islands, conforming to the concept of CIMP, was more present in HCC than cirrhosis and normal liver tissues. We demonstrated that 98.3% HCC have concordant methylation of multiple genes associated with telomerase activity (at least 1 gene). In the study we supported CIMP+ in HCC, which leads to the simultaneous inactivation of multiple genes. The present study showed that CIMP, which includes multiple genes associated with telomerase activity, was significantly associated with metastasis. There was no statistical difference between CIMP status and gender, tumor size, and other features. Consistent with our present results, CIMP of multiple tumor-related genes is associated with a poor prognosis in various types of tumors.49–51 Therefore, consecutive inactivation of multiple tumor-related genes by DNA methylation appears to be important in HCC progression. CIMP seems to be a promising new prognostic marker and its evaluation and investigations into the mechanisms underlying in HCC seem warranted.

Telomerase activation is a common feature of tumor tissue and cell lines.52 Telomerase activation has been suggested to be a late event in tumorigenesis53 and high levels of activity result in an unfavorable clinical prognosis.54 It is of great interest to understand how telomerase activity is regulated in normal and pathological conditions in order to evaluate its potential as a therapeutic target.6 A number of tumor suppressors and oncogenic pathways have been shown to negatively regulate telomerase activity.55, 56 Deregulation of methylation seems to act at multiple steps to allow the cell to escape from cell growth control and apoptosis, CIMP is an effective mechanism for gene silencing. In this study, CIMP is defined by concordant methylation of multiple genes associated with telomerase activity such as P21, P15, P16, P53, RB, P27, WTI, E2F-1 and P300. These transcription factors and tumor suppressors play an important role for regulating telomerase activity and maintaining telomeres length. Epigenetic inactivation may affect all of the molecular pathways involved in cell immortalization and transformation.57 In this study CIMP gives us a chance to know that HCC share a set of genes associated with telomerase activity undergoing hypermethylation. It is possible to find simultaneous inactivation of several pathways by aberrant methylation compromising all of the described genes. The combined methylation of these genes may contribute to upregulation expression of telomerase activity by aberrant telomerase-related factors. Our results showed that expression of high levels of telomerase activity is more detected in HCC than cirrhosis and normal liver tissues. Samples with CIMP+ show more frequent expression of high telomerase activity than those with CIMP−. These results suggested that CIMP could play an important role for inactivation of telomerase activity suppressors by promoter methylation of multiple telomerase-associated genes, which provide new avenues for targeting telomerase in HCC patients.

Our profile has demonstrated that aberrant promoter methylation of telomerase-related genes are frequent in hepatocarcinogenesis. CIMP leads to high levels of expression of telomerase activity through the simultaneous inactivation of multiple genes associated with telomerase activity and HCC patients with CIMP+ shows a poor prognosis. Additional studies are necessary to clarify the detailed clinicopathological features of tumors with and without CIMP in the future.

Acknowledgements

We thank Dr. Jingde Zhu for providing invaluable technical assistance and Professor Jia He for the expert help in statistical analysis.