Epigenetic identification of ubiquitin carboxyl-terminal hydrolase L1 as a functional tumor suppressor and biomarker for hepatocellular carcinoma and other digestive tumors

Authors

  • Jun Yu,

    1. Institute of Digestive Disease and Department of Medicine, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Qian Tao,

    Corresponding author
    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    • Room 315, Cancer Center, Department of Clinical Oncology, PWH, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
    • fax: (852)-2648-8842.

  • Kin F. Cheung,

    1. Institute of Digestive Disease and Department of Medicine, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Hongchuan Jin,

    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Fan Fong Poon,

    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Xian Wang,

    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Hongyu Li,

    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Yuen Y. Cheng,

    1. Institute of Digestive Disease and Department of Medicine, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Christoph Röcken,

    1. Institute of Pathology, Charite University Hospital, Berlin, Germany
    Search for more papers by this author
  • Matthias P. A. Ebert,

    1. Department of Medicine II, Technical University of Munich, Munich, Germany
    Search for more papers by this author
  • Anthony T. C. Chan,

    1. Cancer Epigenetics Laboratory, Department of Clinical Oncology, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    Search for more papers by this author
  • Joseph J. Y. Sung

    Corresponding author
    1. Institute of Digestive Disease and Department of Medicine, Chinese University of Hong Kong, Hong Kong
    2. State Key Laboratory in Oncology in South China, Sir YK Pao Center for Cancer, Chinese University of Hong Kong, Hong Kong
    3. Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, Hong Kong
    • Institute of Digestive Disease, Department of Medicine and Therapeutics, Prince of Wales Hospital, Chinese University of Hong Kong, Shatin, NT, Hong Kong
    Search for more papers by this author
    • fax: (852) 26461852.


  • Potential conflict of interest: Nothing to report.

Abstract

The ubiquitin carboxyl-terminal hydrolase L1 (UCHL1) is a carboxyl-terminal ubiquitin hydrolase regulating cellular ubiquitin levels, recently suggested as a tumor suppressor. However, the role of UCHL1 in hepatocellular carcinoma (HCC) is not clear. We investigated the expression and DNA methylation of the UCHL1 in primary HCC, liver metastases from digestive carcinomas, and primary digestive cancers. UCHL1 is expressed in all normal tissues and immortalized normal epithelial cell lines, but was low or silenced in 77% (10/13) of HCC cell lines, which is well correlated with its promoter methylation status. Methylation was further detected in 44% (12/27) of HCCs, but less in metastatic tumors generated from colorectal and stomach in the liver (19%, 3/16; P < 0.05). Methylation was also detected in primary digestive tumors, including 71% (22/31) of colon, 77% (53/69) of gastric, and 40% (18/45) of esophageal carcinomas, but none or occasionally in paired adjacent nontumor tissues. Detailed methylation analysis of 49 CpG sites at a 540-bp promoter region by bisulfite genomic sequencing confirmed the methylation. UCHL1 silencing could be reversed by chemical or genetic demethylation of the promoter, indicating direct epigenetic silencing. Restoring UCHL1 expression in silenced cell lines significantly inhibited their growth and colony formation ability by inhibiting cell proliferation, causing cell cycle arrest in G2/M phase and inducing apoptosis through the intrinsic caspase-dependent pathway. Moreover, UCHL1 directly interacts with p53 and stabilizes p53 through the ubiquitination pathway. Conclusion: Epigenetic inactivation of UCHL1 is common in primary HCCs and other digestive tumors. UCHL1 appears to be a functional tumor suppressor involved in the tumorigenesis of HCCs and other digestive cancers. (HEPATOLOGY 2008;48:508–518.)

Hepatocellular carcinoma (HCC) remains the third leading cause of cancer death worldwide and the second most common malignancy in China. The molecular pathogenesis of HCC remains largely unknown. It has been hypothesized that development and progression of HCC are the consequence of cumulative genetic and epigenetic events. CpG hypermethylation acts as an alternative mechanism to gene inactivation, and it is now recognized as an important mechanism during tumor initiation and progression, including liver cancer.1, 2 It is important and intriguing to identify new genes silenced by promoter hypermethylation in liver cancer because aberrant tumor suppressor gene (TSG) hypermethylation is both a mechanism and a biomarker for tumorigenesis.3, 4

A few techniques have been developed to screen genome-wide for methylation-silenced genes,5 including restriction landmark genomic scanning,6 methylation-sensitive arbitrarily primed polymerase chain reaction,7 methylated CpG island amplification combined with representational difference analysis,8 and chemical genomic screening through whole-genome expression profiling combined with global demethylation by 5-aza-2′-deoxycytidine (Aza).9–12 We took advantage of the colon cancer cell line HCT116 and its gene knockout subline HCT116-DKO, which was generated by biallelic knockout of both DNMT1 and DNMT3B.13). DNMT1 and DNMT3B are the two major DNA methyltransferases responsible for the maintenance and de novo CpG methylation. The disruption of these two genes results in more than 95% loss of overall genomic methylation and CpG island demethylation, thus reactivating the expression of epigenetically silenced TSGs.13 We used microarray expression profiling to analyze gene expression changes between HCT116 and DKO cells and identified the ubiquitin carboxyl-terminal hydrolase L1 (UCHL1) gene (also known as PGP9.5 or PARK5) as a potential TSG in human cancer.

UCHL1 was first reported as a member of the ubiquitin proteasome pathway controlling intracellular protein degradation, functioning to maintain ubiquitin balance by associating and releasing ubiquitin from tandem conjugated ubiquitin monomers.14UCHL1 is located at chromosome locus 4p14.15 It possesses dimerization-dependent ubiquityl ligase activity15 and increases ubiquitin half-life in vivo,14 suggesting that UCHL1 plays an important role in controlling intracellular ubiquitin levels in cells undergoing ubiquitin-dependent protein degradation. Currently, there is conflicting evidence regarding the role of UCHL1 in tumorigenesis, with antitumor or protumor properties depending on different cancer types.16, 17 Compelling evidence demonstrated that UCHL1 had been identified as a potential TSG, methylated in a cancer-specific manner in human cancers.18–23 However, the functional role of UCHL1 in tumorigenesis of liver cancer remains ambiguous. The expression of UCHL1 and its regulatory mechanisms during the development of liver cancer and other digestive cancers have yet to be determined.

We demonstrate that UCHL1 undergoes promoter CpG hypermethylation-associated silencing in liver cancer and other cancers from the digestive system. The analysis of liver cancer and a panel of other primary digestive tumors showed that UCHL1 hypermethylation is a common event in the tumorigenesis of liver, colon, stomach, and esophageal cancers. We also show that UCHL1 functions as a TSG to suppress tumor cell growth through inhibiting proliferation, inducing apoptosis, and stabilizing p53 from the de-ubiquitination pathway.

Abbreviations

Aza, 5-aza-2′-deoxycytidine; BGS, bisulfite genomic sequencing; ESCC, esophageal squamous cell cancer; DKO, HCT116 DNMT1−/− DNMT3B−/−; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HCC, hepatocellular carcinoma; mRNA, messenger RNA; MSP, methylation-specific polymerase chain reaction; PARP, poly(adenosine diphosphate–ribose) polymerase; PCR, polymerase chain reaction; RT-PCR, reverse transcription polymerase chain reaction; TSG, tumor suppressor gene; TUNEL, terminal deoxynucleotidyl transferase-mediated nick-end labeling; UCHL1, ubiquitin carboxyl-terminal hydrolase L1.

Materials and Methods

Tumor Cell Lines, Primary Tumor and Normal Tissue Samples.

A panel of tumor cell lines from the digestive system were used,10, 24–26 including 13 liver cancer cell lines (Hep3B, HepG2, hUH1, hUH4, hUH6, hUH7, Mahlavu, PLC/PRF/5, SNU387, SNU398, SNU423, SNU449, SNU475), four colon (HCT116, HT-29, LoVo, SW480), 16 stomach (Kato III, YCC1, YCC2, YCC3, YCC6, YCC7, YCC9, YCC10, YCC11, YCC16, SNU719, AGS, MKN28, MKN45, SNU1 and SNU16), and 17 esophageal squamous cell cancer (ESCC; EC1, EC18, EC109, HKESC1, HKESC2, HKESC3, KYSE30, KYSE70, KYSE140, KYSE150, KYSE180, KYSE270, KYSE410, KYSE450, KYSE510, KYSE520 and SLMT-1). Two immortalized normal esophageal epithelial cell lines (NE1, NE3) were used as “normal” controls for ESCC. HCT116 cell line with genetic double knockout of DNA methyltransferases (DNMT1 and DNMT3B): HCT116 DNMT1−/−DNMT3B−/− (DKO) was growing with 0.4 mg/mL geneticin (G418) and 0.05 mg/mL hygromycin.13

Human liver tumors including 27 primary HCC and 16 liver metastases of primary colorectal and stomach tumors were obtained from hepatectomy patients at the time of surgery; matched nontumor liver samples from HCC patients were also obtained at least 2 cm distant from the tumor. Primary ESCC specimens and matched morphologically normal esophageal epithelium tissues (nontumor tissues) at least 6 to 10 cm from the tumors were collected from the patients who underwent resection for ESCC and were not pretreated with chemotherapy or radiation therapy. Nontumor portions were trimmed off from the frozen tumor blocks, and the selected tumor areas had more than 80% tumor cells as confirmed by histology.10, 24 Paired surgical gastric and colon cancers were also obtained. Human normal adult and fetal tissue RNA samples were purchased commercially.10, 24 All patients gave informed consent for obtaining the study specimens.

RNA Extraction and Semiquantitative Reverse Transcription Polymerase Chain Reaction.

Total RNA was extracted from cell pellets using TriReagent (Molecular Research Center, Inc., Cincinnati, OH). Reverse transcription polymerase chain reaction (RT-PCR) was performed as previously described using the Go-Taq polymerase (Promega, Madison, WI) and the GeneAmp RNA PCR system (Applied Biosystems, Foster City, CA), with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a control.27). The UCHL1-specific primers are: forward 5′-AGCTCAAGCCGATGGAGATC-3′ and reverse 5′-CCCTTCAGCTCTTCAATCTG-3′ (211-bp product). The p53-specific primers are: forward 5′-AAGCAGTCACAGCACATGAC-3′ and reverse 5′-TAGTGGATGGTGGTACAGTC-3′ (212-bp product). RT-PCR was 32 cycles with an annealing temperature of 55°C.

DNA Extraction and Methylation-Specific PCR.

Genomic DNA was extracted from the cell pellets and tissues using TriReagent or the QIAamp DNA Mini kit (Qiagen, Hilden, Germany). DNA was chemically modified with 2.4 mol/L sodium metabisulfite for 4 hours as previously described.27 The bisulfite-modified DNA was amplified using primer pairs that specifically amplify either methylated or unmethylated sequences of the UCHL1 promoter CpG islands. The primers specific for methylated UCHL1 were: 5′-tttatttggtcgcgatcgttc-3′ (forward, position −101 to −81 from the transcriptional start site according to the NCBI database) and 5′-aaactacatcttcgcgaaacg-3′ (reverse, position +54 to +74). The primers specific for unmethylated UCHL1 were: 5′-gtatttatttggttgtgattgttt-3′ (forward, position −104 to −81) and 5′-cttaaactacatcttcacaaaaca-3′ (reverse, position +54 to +77). These primer pairs have been tested before using them for this study for not amplifying without bisulfite-treated DNA. Methylation-specific PCR (MSP) was performed with 40 cycles using the Taq-Gold polymerase (Applied Biosystems).27

Bisulfite Genomic Sequencing.

Briefly, bisulfite-treated DNA was amplified with primers specific for a fragment of the UCHL1 promoter CpG islands containing 49 CpG sites and spanning the region analyzed by MSP with primers: 5′-GCTCCATACACTCAAGGAACACCC-3′ (forward, −300 to −273 from the transcriptional start site) and 5′-CTTGGTCCCTGCCAGCAGC-3′ (reverse, +231 to +240). The PCR products were then cloned into the pCR4-Topo vector (Invitrogen, Carlsbad, CA), with five to eight colonies randomly chosen and sequenced.

Aza and Trichostatin A Treatment.

Cells were seeded at a density of 1 × 106 cells/mL. After overnight culture, cells were treated with 10 μM of the DNA demethylating agent Aza (Sigma-Aldrich, St. Louis, MO) for 72 hours, with or without 300 nmol/L trichostatin A (Sigma-Aldrich) for an additional 24 hours then harvested.

Cloning of UCHL1 and Construction of Expression Vector.

The UCHL1 expression vector was generated by PCR cloning with pcDNA3.1-TOPO TA expression vector (Invitrogen). Complementary DNA corresponding to the full-length UCHL1 was obtained by RT-PCR amplification of normal human thymus RNA (Clontech, Palo Alto, CA), with primers specific to UCHL1. PCR aliquots were subcloned into the pcDNA3.1-TOPO vector. Clones were screened and sequenced using vector-specific primers.

Colony Formation Assay.

Colon cell line HCT116 (2 ×105 cells/well) and ESCC cell line EC109 (2 × 105 cells/well), were transfected with 0.5 μg UCHL1-expressing or control vector (pcDNA3.1) using FuGENE 6 (Roche). Transfected cells were selected with G418 (0.4 mg/mL; Merck, Darmstadt, Germany) for 10 to 14 days. Colonies were then fixed with methanol/acetone (1:1) and stained with gentian violet and counted. All the experiments were performed in triplicate wells three times.

Cell Cycle Analysis.

Cells were seeded (1 × 106 cells/well) in six-well plates and transfected with vector or UCHL1-expressing plasmid. After 24 hours, the cells were trypsinized, washed in phosphate-buffered saline, and fixed in ice-cold 70% ethanol–phosphate-buffered saline. Free nuclei stained with propidium iodide were obtained from hypotonic lysis of cells in a buffer containing sodium citrate (0.1%), Triton X (0.1%), and propidium iodide (50 g/mL). The cells were then sorted by FACScan analysis. Cell cycle profiles were determined using the ModFitLT software (Becton Dickinson, San Diego, CA). Cells undergoing apoptosis were detected as sub-G1 population because of loss of fragmented DNA.

Ki-67 Staining.

Cell proliferation was assayed by immunoperoxidase staining with anti-Ki-67 antibody (ab833; Abcam, Cambridge, UK). Negative controls were run by replacing the primary antibody with nonimmune serum. The proliferation index was determined by counting the numbers of positive staining cells as percentages of the total number of tumor cells. At least 1000 tumor cells were counted each time.

In Situ DNA Nick End Labeling.

Terminal deoxynucleotidyl transferase-mediated nick-end labeling (TUNEL) was performed following the Manufacturer's protocol (single-stranded DNA apoptosis TUNEL kit, Roche, Indianapolis, IN). Nuclei with clear brown staining were regarded as apoptotic cells. The apoptosis index was calculated as the percentage of TUNEL-positive nuclei after counting at least 1000 cells.

UCHL1 Protein Immunoprecipitation.

HCT116 cells after 48 hours' transfection with UCHL1-expressing or empty vector (pcDNA3.1) were lysed in lysis buffer supplemented with proteinase inhibitor (Roche). After centrifugation, supernatant of the cell lysate was incubated with 1% fetal bovine serum (Clonetech, Mountain View, CA) and then precleared with protein A Sepharose (GE Healthcare, Piscataway, NJ). The supernatant was incubated with primary antibody anti-UCHL1 (ab10404, Abcam, Cambridge, MA) overnight at 4°C. Protein A sepharose was then applied and incubated 4 hours at 4°C. The bead slurry was washed and the bound proteins were denatured with Laemmli buffer, and the samples were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis.

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis and Western Blot.

Primary antibodies against Histone H3 (#05-499, Upstate, Charlottesville, VA), p53 (M7001, Dako, Glostrup, Denmark), GAPDH, proliferating cell nuclear antigen, PGP9.5 (Abcam), cleaved PARP, cleaved caspase-3, cleaved caspase-7, cleaved caspase-9, phospho-cdc2, phospho-cdc25C, p21 waf1/Cip1, p27 Kip1 (Cell Signaling, Danvers, MA) and second antibodies against horseradish peroxidase–linked anti-mouse immunoglobulin G and horseradish peroxidase–linked anti-rabbit immunoglobulin G (Cell Signaling) were used. Lysates of cells transfected with pcDNA3.1 or pcDNA3.1-UCHL1 were separated on 10% Bis/Tris-polyacrylamide gel through electrophoresis and blotted onto nitrocellulose membranes (GE Healthcare, Piscataway, NJ). Blots were immunostained with primary antibodies overnight at 4°C and secondary antibody for 1 hour at room temperature. Proteins were visualized using ECL Plus Western blotting Detection Reagents (RPN2132, GE Healthcare). The film was scanned and quantitated by Quantity One software (Bio-Rad, Hercules, CA).

Statistical Analysis.

The results were expressed as mean ± SD. Mann-Whitney U test was used to compare the clinicopathologic variables of the two sample groups. Student t test was used to compare the differences of UCHL1 expression on the effect of colony formation, cell apoptosis, and cell proliferation. All statistical calculations were done using SPSS version 11.0 for windows (SPSS, Inc., Chicago, IL). Value of P < 0.05 was taken as statistical significance.

Results

Epigenetic Identification of UCHL1 as Potential TSG Methylated.

We compared the expression profiles of HCT116 and HCT116-DKO cells by Affymetrix microarray expression analysis. UCHL1 was identified as one of the potential tumor suppressor genes. Semiquantitative RT-PCR confirmed that UCHL1 expression is greatly reduced or silenced in most HCC cell lines (10/13, 77%; Fig. 1A). It has been well documented that hypermethylation in CpG-rich promoter region is a critical event leading to the epigenetic inactivation of TSGs in cancers. We examined the role of hypermethylation in the silencing of UCHL1 and demonstrated that hypermethylation was detected in all the HCC cell lines with silenced or low expression (Table 1 and Fig. 1A). In contrast, UCHL1 was readily expressed in all normal adult tissues, including liver and other digestive tissues (Fig. 1B).

Figure 1.

Promoter methylation and expression of UCHL1 in liver cancer cell lines, and mRNA expression in human normal adult tissues. (A) UCHL1 was greatly reduced or silenced in most of the liver cancer cell lines as a result of the hypermethylation of its promoter. U: unmethylated; M: methylated. (B) Broad expression of UCHL1 in human normal adult tissue panel by semiquantitative RT-PCR, with GAPDH as a control.

Table 1. Frequency of UCHL1 Methylation in Tumors and Normal Tissues
 OriginCell LineTissue
 HCC77% (10/13)44% (12/27)
 Carcinomas metastatic to the liver 19% (3/16)
 ESCC69% (11/16)40% (18/45)
TumorColon cancer80% (4/5)71% (22/31)
Gastric carcinoma82% (9/11)77% (53/69) 
 Surgical margin tissues of HCC 7% (2/30)
 Surgical margin tissues of ESCC 9% (4/45)
NontumorSurgical margin tissues of colon cancer 19% (3/16)
 Surgical margin tissues of gastric cancer 8.7% (2/23)

Correlation of UCHL1 Messenger RNA Expression Level with Methylation Status in Digestive Tumor Cell Lines.

We further examined UCHL1 expression levels in a series of tumor cell lines from the digestive system mostly not studied before by semiquantitative RT-PCR. UCHL1 expression was reduced or silenced in 80% (4/5) colon, 82% (9/11) gastric, and 69% (11/16) esophageal cell lines (Table 1 and Fig. 2B). Using MSP, UCHL1 hypermethylation was detected in all the cell lines with silenced or low expression, whereas unmethylated promoter was detected in expressing cell lines (Fig. 2B). In contrast, none of the normal control cell lines (NE1 and NE3) showed hypermethylation (Fig. 2B).

Figure 2.

UCHL1 promoter methylation and expression in colon, gastric, esophageal cancer, and normal epithelial cell lines. (A) Locations of the 49 CpG sites analyzed and primers used in the UCHL1 promoter region. Vertical lines indicate individual cytosine residues of the CpG sites. Methylation-specific PCR (MSP) and bisulfite genomic sequencing (BGS) regions are also shown. (B) Representative analyses of UCHL1 expression by semiquantitative RT-PCR and promoter hypermethylation by MSP in multiple tumor cell lines and immortalized normal esophageal cell lines (NE1 and NE3). Ca: Cancer, U: unmethylated; M: methylated. (C) BGS of the methylation status of UCHL1 in tumor and normal cell lines. Each row represents an individual promoter allele analyzed. Open circles denote unmethylated cytosines, and closed circles denote methylated cytosines.

To further verify the MSP results, detailed methylation analysis using bisulfite genomic sequencing was performed. The UCHL1 promoter contains CpG islands. A region around the transcription start site (−300 bp to +240 bp) containing 49 CpG sites was analyzed (Fig. 2A). The results revealed heavy methylation in HCC cell lines (Huh6 and SNU387) and other digestive cancer cell lines, including the ESCC cell line (EC109) and colon cancer cell lines (LoVo, HT29, and HCT116). The normal epithelial cell line NE1 had few methylated CpG sites, confirming the MSP results (Figs. 2C and 3B). Thus, UCHL1 promoter hypermethylation is tightly associated with its transcriptional silencing in HCC cell line and other tumor cell lines.

Figure 3.

UCHL1 expression could be restored in tumor cells with treatment of 5-aza-2′-deoxycytidine (A) and trichostatin A (T), or genetic demethylation. (A) Five heavily methylated and silenced cancer cell lines from liver (SNU387), esophageal (EC109), gastric (SNU719), and colon (SW480, HCT116) cancer were treated. Up-regulated expression was detected. (B) Detailed methylation analysis was performed using BGS on untreated cell lines HTC116, HCT116-DKO (DNMT1 and DNMT3B knockout cell line), SNU387, and A+T-treated cell line SNU387. Untreated HCT116 is fully methylated, whereas the HCT116-DKO cell line is mostly unmethylated. The demethylation of the UCHL1 promoter was observed in A+T-treated SNU387. Open circles: unmethylated cytosines; closed circles: methylated cytosines.

UCHL1 Expression Could Be Restored with Aza and Trichostatin A Treatment in Tumor Cell Lines.

To confirm that CpG hypermethylation is indeed responsible for the silencing of UCHL1, we treated heavily methylated and silenced cell line from liver (SNU387) and other digestive tumor types (EC109 esophageal, SNU719 gastric, HCT116, and SW480 colon) with Aza combined with trichostatin A. UCHL1 expression was dramatically induced after treatment in all of the cell lines (Fig. 3A). Meanwhile the UCHL1 promoter was demethylated as determined by MSP or bisulfite genomic sequencing (BGS; Fig. 3). These results demonstrate that CpG hypermethylation mediates the transcriptional silencing of UCHL1 in liver and other digestive tumor cells.

Frequent UCHL1 Hypermethylation in Primary Carcinomas.

We next investigated UCHL1 hypermethylation in primary tumors. Aberrant hypermethylation was detected in 44% (12/27) of primary HCCs, but was less frequently detected in metastatic tumors generated from colorectal and stomach in the liver (19%, 3/16; P < 0.05; Table 1). Promoter hypermethylation of UCHL1 was observed in 77% (53/69) of primary gastric cancers and 71% (22/31) of primary colon cancers (Fig. 4 and Table 1). There is a significant difference in UCHL1 methylation between the tissues of primary colon and gastric cancers and the livers with metastatic colorectal and gastric cancer (P < 0.001). UCHL1 hypermethylation was also detected in 40% (18/45) of esophageal tumors, but only occasionally in paired adjacent nontumor or surgical margin tissues (Fig. 4, Table 1), indicating that UCHL1 hypermethylation is tumor-specific. However, there was no association between UCHL1 methylation and other clinicopathological characteristics of cancer patients, including age, sex, tumor grade, TNM staging, classification, and differentiation.

Figure 4.

Representative MSP results of primary tumors (T) and paired adjacent normal tissues (N). U: unmethylated; M: methylated; Ca: cancer.

UCHL1 Suppresses Tumor Cell Growth.

The frequent silencing of UCHL1 by hypermethylation in liver and other digestive cancer cell lines and primary carcinomas suggests that UCHL1 is likely a tumor suppressor. To test this point, we analyzed the growth characteristics of cells overexpressing UCHL1 by colony formation assay. The cancer cell lines HCT116, EC109, Huh1, and SNU387 with silenced UCHL1 were transfected with UCHL1-expressing constructs, and the numbers of colonies formed were counted after 10 to 14 days' culture with G418 selection. It was found that the colonies formed by UCHL1-transfected cells were significantly less and smaller in size than in vector-transfected cells (P < 0.01; Fig. 5), indicating that UCHL1 indeed has growth inhibitory activity and can function as a tumor suppressor.

Figure 5.

UCHL1 inhibited tumor cell colony formation. (A) The UCHL1-silenced HTC116 and EC109 cells were transfected with UCHL1-expressing or empty vector and maintained in the presence of G418 for 10-14 days; a forced expression of (A1) UCHL1 mRNA level and (A2) protein level were observed in RT-PCR and western blotting, respectively. (B1) Representative of colony formation assays. Experiments were performed in triplicate three times. (B2) Quantitative analyses of colony numbers are shown as values of mean ± standard deviation. P values were calculated using Student t test. The asterisk indicates statistical significant difference (*P < 0.01).

Ectopic Expression of UCHL1 Suppresses Proliferation, Induces G2/M Phase Arrest.

To determine the molecular mechanism by which UCHL1 suppresses colony formation, we investigated the effect of UCHL1 on cell cycle distribution (Fig. 6A,B). After propidium iodide staining, fluorescence-activated cell sorting analysis of UCHL1-transfected HTC116 cells revealed a significant decrease in the number of S-phase cells at 24 hours (P < 0.01; Fig. 6C). Additionally, Ki-67 staining showed a decrease proliferation index in UCHL1-transfected cells compared with vector-transfected cells (9.6 ± 1.8 versus 17.3 ± 1.9, P = 0.001; Fig. 7A), confirming the inhibitory effect of UCHL1 gene on cell proliferation in HTC116 cells. Concomitant with this inhibition, there was a significant increase in the number of cells accumulating in the G2/M phase (P < 0.05; Fig. 6D). To prove these findings, major mediators in cell cycle process were further accessed. Our results showed that reexpression of UCHL1 in the stably transfected HCT116 cells induced phosphorylation of the protein phosphatase cdc25C and subsequently phosphorylated the protein kinase cdc2 (known as Cdk1), which is the key regulator inhibiting the G2/M phase. However, the protein level of G0/G1-phase regulators, including p21 Waf1/cip1 and p27 Kip1, remained unchanged in the stably transfected HCT116 cells (Fig. 8).

Figure 6.

Effect of UCHL1 on the cell cycle of colon cancer cell line HTC116. (A,B) Representative fluorescence-activated cell sorting analysis of HTC116 cells transfected with or without UCHL1. (C) Cell proliferation was calculated as the fraction of cells in S phase. (D) The number of cells in G2/M phases was also determined by flow cytometry. (E) UCHL1 increased the rate of apoptosis, as determined by the number of cells with sub-G1 DNA content. Values are mean ± SD from three replicate experiments. *P < 0.05, **P < 0.01, ***P < 0.001, HTC116/UCHL1-treated versus HTC116/vector.

Figure 7.

Effect of UCHL1 on cell proliferation, apoptosis, and p53 protein levels of colon cell line HTC116. (A) Ki-67–stained slides of HTC116 cells transfected with (A1) vector only or (A2) UCHL1-expressing vector for 24 hours. UCHL1-transfected HTC116 cells exhibited less cell proliferation, as indicated by fewer Ki-67–positive proliferative cells (arrows; ×400). (B) TUNEL staining of HTC116 cells transfected with (B1) vector only or (B2) UCHL1-expressing vector for 24 hours. An increase in the number of TUNEL-positive cells (brown-stained nuclei, arrows) is evident in UCHL1–transfected cells (×400). Summaries of the results in panels A1-A2 and B1-B2 are shown in panels A3 and B3 on the right. (C) Ectopic expression of UCHL1 results in the stabilization and increase of p53 protein in (C1) HTC116 cells, which is not due to any change of the expression levels of (C2) p53 mRNA. The amount of p53 protein in HCT116 and EC109 cells before and after UCHL1 expression was determined by western blot. The relative density is shown as the ratio of p53/Histone H3. (D) Western blotting analysis of p53 after UCHL1 immunoprecipitation (IP-UCHL1) of the HCT116 transient transfected cells. The immunoprecipitation without antibody (IP–no antibody) was served as a negative control for protein immunoprecipitation experiment. *P < 0.01, **P < 0.001.

Figure 8.

Effects of reexpression of UCHL1 on protein expression of regulators involved in G2/M phase cell cycle arrest, induction of apoptosis though intrinsic caspase-dependent pathway. Stably transfected HCT116 cells were prepared after 21 days with G418 selection. Protein expression on p53, a series of intrinsic apoptotic mediators (cleaved-PARP, cleaved-caspase-3, cleaved-caspase-7, cleaved-caspase-9) and cell cycle regulators (phospho-cdc2, phospho-cdc25C, p21 Waf1/Cip1, p27 Kip1) were determined using western blotting analyses. GAPDH was used as internal control.

Ectopic Expression of UCHL1 Induces Apoptosis Through Intrinsic Caspase-Dependent Pathway.

Arrest of cell growth in tumor cells is usually associated with concomitant activation of cell death pathways. We therefore examined the contribution of apoptosis to the observed growth inhibition of UCHL1-transfected cells. We measured the apoptotic rate by flow cytometry and noted that the number of cells with sub-G1 DNA content after 24 hours of UCHL1 transfection was substantially increased, compared with vector-transfected cells (P < 0.001; Fig. 6E). Consistent with this finding, there was an increased number of cells that stained TUNEL-positive after UCHL1 transfection as compared with vector transfection (6.6 ± 0.8 versus 3.5 ± 0.9, P < 0.01; Fig. 7B). These findings suggested that apoptosis in conjunction with cell cycle arrest, as induced by UCHL1, accounts for the growth inhibition in UCHL1-expressing tumor cells.

To elucidate the molecular basis of apoptosis, we accessed the apoptosis mediators, including the active form of nuclear enzyme poly(adenosine diphosphate–ribose) polymerase (PARP), the active form of caspase-3, caspase-7, and caspase-9, in the stably transfected HCT116 cells. Our results indicated that reexpression of UCHL1 enhanced cleaved caspase-9 and induced cleaved caspase-3, caspase-7, and PARP, suggesting that UCHL1-induced apoptosis through the intrinsic caspase-dependent pathway (Fig. 8).

UCHL1 Directly Interacts and Stabilizes p53.

UCHL1 was assumed to work through the ubiquitin-proteasome system to modulate cellular protein stability/degradation, although the associated molecular mechanism is still unclear.14, 15, 28 We examined the effect of ectopic UCHL1 expression on the cellular protein level of p53, which is a protein seriously regulated through its ubiquitination modification.29–31 As shown in Fig. 7C, the p53 protein levels relative to the internal control histone H3 were significantly elevated after transient transfection of UCHL1 plasmid in HCT116 and EC109 cells, which is not due to any change of the expression levels of p53 messenger RNA (mRNA). This was further confirmed in the stably transfected HCT116 cells, which also showed elevated p53 protein (Fig. 8), suggesting that UCHL1 might function as a tumor suppressor through stabilizing p53, most likely through interfering with the ubiquitination-mediated degradation of p53.30, 31 To test this hypothesis, the protein interaction between UCHL1 and p53 was assessed. Strikingly, UCHL1 reexpressing HCT116 cells showed higher levels of p53 protein after UCHL1 immunoprecipitation as compared with the UCHL1 nonexpressing HCT116 cells. P53 protein was not detected in the no-antibody immunoprecipitation (negative control; Fig. 7D). Our findings revealed that UCHL1 direct interacted with p53 and inhibited p53 protein degradation.

Discussion

UCHL1 is frequently silenced or down-regulated with promoter hypermethylation in 77% of HCC cell lines. Hypermethylation was further detected in 44% of the primary HCC. In contrast, UCHL1 hypermethylation was not detected in normal cell lines and occasionally in paired normal tissues, suggesting an important role of UCHL1 in the pathogenesis of liver cancer. We further demonstrated that treatment with the demethylating reagent Aza reactivated UCHL1 expression in silenced tumor cells, and the methylation status was verified by genomic sequencing, indicating that DNA hypermethylation mediates UCHL1 inactivation.

The liver is the most common site of metastatic disease from both gastrointestinal and extraintestinal malignancies. Metastatic cancer cells use a pathway that includes cytoskeleton change, loss of adhesion, enhanced mobility, and degradation of the basement membrane.32UCHL1 methylation has been shown to be an independent prognostic factor for esophageal carcinoma, with higher methylation ratios associated with poorer survival rate and regional lymph node metastases.19 It is interesting to evaluate the hypermethylation of the UCHL1 gene in liver metastases of primary colorectal and gastric tumors. The frequency of UCHL1 hypermethylation is lower in the metastatic liver tumors compared with HCC or primary digestive tumors, suggesting that UCHL1 plays a key role in the progression of HCC and other digestive tumors but may not be a distant metastasis-promoting gene.

The ubiquitin-proteasome system, to which UCHL1 belongs, plays important roles in various cellular processes including cell cycle, apoptosis, and intracellular signaling. The disturbance of the ubiquitin-proteasome system is well recognized in cancer development.33, 34 Many deubiquitination enzymes and critical proteins are known to regulate tumor suppression and are involved in cancer development. UCHL1, as an important member of the ubiquitin-proteasome system, was frequently silenced in cancer cell lines and primary tumors as revealed by the current study and others,18–23 suggesting that UCHL1 silencing could provide a growth advantage to tumor cells. The functional significance of UCHL1 was further evaluated by examining the inhibitory effect of UCHL1 expression on cell growth. Introduction of UCHL1 in silenced tumor cell lines significantly suppressed their growth in colony formation assays.

Fluorescence-activated cell sorting analysis of the effects of UCHL1 expression on the cell cycle in transfected cells revealed a decrease in cell proliferation, which was further confirmed by Ki-67 immunostaining, a concomitant and proportionate increase of cells in G2/M. On the basis of the immunoblot analysis of cell-cycle regulators, G2/M phase arrest by UCHL1 was most likely associated with the phosphorylation of Ser216 and the cdc25C phosphatase regulation. When phosphorylated at Ser216, cdc25C binds to members of the 14-3-3 family of proteins, sequestering cdc25C in the cytoplasm, preventing pre-mature mitosis.35 In addition, the cdc25C phosphatase is responsible for tyrosine 15 dephosphorylation and activation of cdc2. Thus, the phosphorylation at cdc25C would allow the accumulation of tyrosine 15 phosphorylation and inactivation of cdc2 kinase and thus results in the blocking the entry of cells from G2 phase into mitosis (Fig. 9A).

Figure 9.

Proposed scheme for regulation of G2/M phase arrest, induction of cell apoptosis and ubiquitination of p53 gene by UCHL1. (A) The G2/M phase arrest was controlled by the reexpression of UCHL1 through the phosphorylation of Ser216 and inactivation of phosphatase cdc25C, which would facilitate the accumulation of phosphorylated tyrosine 15 and inactivation of cdc2. (B) In the caspase-dependent pathway, reexpression of UCHL1 activated the caspase-9, which further processes other effector caspase members, including caspase-3 and caspase-7, to initiate a caspase cascade. These effectors further initiated the cleavage of PARP, which caused loss of DNA repair, cellular disassembly, and finally undergoing apoptosis. (C) Under polyubiquitination of p53, ubiquitin conjugates to the target proteins of p53 and marks them for the degradation by the proteasome. In contrast, on de-ubiquitination, the UCHL1 can hydrolyze the bond between ubiquitin and the target protein. Finally, the target protein undergoes de-ubiquitination and p53 stabilization.

Our study also revealed an increase of cells in the sub-G1 fraction by UCHL1. This finding indicates increased apoptosis and is further confirmed by positive TUNEL staining. It was of interest to note that the apoptosis induction by UCHL1 is mediated through an intrinsic caspase-dependent pathway. Our results indicated that reexpression of UCHL1 activates caspase-9, which further processes other effector caspase members, including caspase-3 and caspase-7, to initiate a caspase cascade. These effectors further initiate the proteolytic cleavage of the PARP, which caused loss of DNA repair, cellular disassembly, and finally apoptosis (Fig. 9B).

In addition to causing cell cycle arrest at the G2/M phase and the induction of cell apoptosis, importantly in this study, the growth inhibitory effect of UCHL1 as a tumor suppressor may also be mediated through stabilizing p53, probably by interfering with its ubiquitination-mediated degradation, similar to the reported stabilization effect of USP7/HAUSP.29, 30 The relevance of physical interaction of UCHL1 and p53 with respect to the p53 protein stabilization has not been shown before. We first demonstrated that the p53 is up-regulated in the UCHL1 reexpressing cells in both transient and stable transfection assays; we further identified that UCHL1 directly interacts with p53 based on the UCHL1 immunoprecipitation. It is well known that UCHL1 is a de-ubiquitinating enzyme that hydrolyzes the ubiquitin from the substrate and controls the balance of ubiquitination.29, 30 Collectively, these observations indicated the molecular function of UCHL1, which directly interacts and stabilizes p53 through the de-ubiquitination pathway (Fig. 9C).

In summary, we found that UCHL1 is frequently silenced or reduced by promoter hypermethylation in most liver cancer cell lines and primary HCC. Our results suggest that epigenetic inactivation of UCHL1 is an important factor in the tumorigenesis of primary HCC and other primary digestive cancers. We also demonstrated that promoter hypermethylation-mediated silencing of UCHL1 could be reversed by pharmacologic demethylation, and restoration of UCHL1 suppressed tumor cell growth through inducing G2/M arrest and apoptosis as well as through stabilizing p53, providing direct evidence that UCHL1 functions as a tumor suppressor. It therefore will be valuable to explore the possible application of UCHL1 as a molecular marker for the detection and treatment of these malignancies.

Acknowledgements

The authors thank Bert Vogelstein for the HCT116 cell lines with knockout of different DNMTs and George Tsao for the NE1 and NE3 cell lines. We also thank Prof Gopesh Srivastava for providing DNA/RNA of KYSE-cell lines (Shimada et al. Cancer 1992;69:277-284) which were obtained from DSMZ (German Collection of Microorganisms & Cell Cultures and Dr. Sun Young Rha for some gastric cell lines.

Ancillary