Expression of esophageal cancer related gene 4 (ECRG4), a novel tumor suppressor gene, in esophageal cancer and its inhibitory effect on the tumor growth in vitro and in vivo

Authors

  • Lin-Wei Li,

    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    Current affiliation:
    1. Department of Oncology, Henan Provincial People's Hospital, Zheng-Zhou, People's Republic of China
    Search for more papers by this author
  • Xi-Ying Yu,

    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    Search for more papers by this author
  • Yang Yang,

    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    Search for more papers by this author
  • Chun-Peng Zhang,

    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    Search for more papers by this author
  • Li-Ping Guo,

    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    Search for more papers by this author
  • Shih-Hsin Lu

    Corresponding author
    1. State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, People's Republic of China
    • State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100-021, People's Republic of China
    Search for more papers by this author
    • Fax: +86-10-6771-2368.


Abstract

The ECRG4 gene was initially identified and cloned in our laboratory from human normal esophageal epithelium (GenBank accession no. AF325503). We revealed the expression of ECRG4 protein was downregulated in 68.5% (89/130) ESCC samples using tissue microarray. The low ECRG4 protein expression was significantly associated with regional lymph node metastasis, primary tumor size, and tumor stage in ESCC (p < 0.05). ECRG4 mRNA expression was downregulated in ESCC due to the hypermethylation in the gene promoter. The treatment with 5-aza-2′-deoxycytidine, which is a DNA methyltransferase inhibitor restored ECRG4 mRNA expression in ESCC cells. The result indicated that promoter hypermethylation may be 1 main mechanism leading to the silencing of ECRG4. The high expression of ECRG4 in patients with ESCC was associated with longer survival compared with those with low ECRG4 expression by Kaplan-Meier survival analysis (p < 0.05). ECRG4 protein was an independent prognostic factor for ESCC by multivariable Cox proportional hazards regression analysis (p < 0.05). The restoration of ECRG4 expression in ESCC cells inhibited cell proliferation, colony formation, anchorage-independent growth, cell cycle progression and tumor growth in vivo (p < 0.05). The transfection of ECRG4 gene in ESCC cells inhibited the expression of NF-κB and nuclear translocation, in addition to the expression of COX-2, a NF-κB target gene, was attenuated. Taken together, ECRG4 is a novel candidate tumor suppressor gene in ESCC, and ECRG4 protein is a candidate prognostic marker for ESCC. © 2009 UICC

Esophageal cancer ranks 7th and 6th, respectively, in terms of cancer incidence and mortality rate worldwide.1 Moreover, 50% of esophageal cancer cases in the world occurred in China.2 Esophageal squamous cell carcinoma (ESCC) is the most common histological subtype, which accounts for ∼90% of all esophageal carcinomas diagnosed in China each year. To date, the molecular pathogenesis of ESCC is still unclear. As a result, a major research effort has been directed at identify and cloning novel specific esophageal cancer related genes, which might play important roles in ESCC carcinogenesis.3, 4

The ECRG4 gene (GenBank accession no. AF325503) was initially identified and cloned in our laboratory from human normal esophageal epithelium.5, 6 We compared the differential genes expression between human normal esophageal epithelia and ESCCs from high incidence families in Linxian of Northern China through the effective mRNA differential display technique. Then we obtained a 772-bp full-length cDNA of the ECRG4 gene by SMART™ RACE technique.

The ECRG4 gene is located at chromosome 2q12.2. It contains 4 exons, spans about 12,500 bp and has a 444-bp open reading frame encoding a 148-amino acid polypeptide with molecular weight of 17 kDa. Northern blot assay revealed that ECRG4 was expressed in various tissues including heart, brain, placenta, lung, liver, skeletal muscle, kidney and pancreas.7 Our previous results demonstrated that ECRG4 mRNA level was highly expressed in adult esophageal epithelia, but it was downregulated in ESCC tissues and cell lines.8 These findings suggest that the ECRG4 gene might be involved in ESCC carcinogenesis. However, the mechanism for loss of ECRG4 expression in ESCC remains unclear.

In the present study, we investigated the mechanisms of ECRG4 inactivation in ESCC. Furthermore, we evaluated the clinical prognostic significance of ECRG4 expression for ESCC patients and the correlation between downregulation of ECRG4 protein expression and clinicopathologic features using tissue microassay analysis. Then we examined the tumor-suppressing function of ECRG4 in ESCC cells in vitro and in vivo and explored the possible pathway in which ECRG4 might be involved.

Abbrevations

COX-2, cyclooxygenase 2; ECRG4, esophageal cancer related gene 4; ESCC, esophageal squamous cell carcinoma; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; LOH, loss of heterozygosity; NF-(B, NF-kappaB; PBS, phosphate-buffered saline; PCR-SSCP, polymerase chain reaction-single strand conformation polymorphism; RT-PCR, reverse transcriptase-polymerase chain reaction.

Material and methods

Tissue specimens and cell lines

Esophageal tumor tissue and adjacent normal epithelia specimens were collected by the Department of Pathology from 130 patients who underwent surgical resection with histologically confirmed ESCC at Cancer Hospital, Chinese Academy of Medical Science, between 1999 and 2006. None of the patients had received adjuvant chemotherapy, neo-adjuvant chemotherapy or radiation therapy. Primary tumor tissues and the corresponding normal esophageal mucosa from the same patients were separated by experienced pathologists and immediately stored at −70°C until use. This study was approved by the institutional review board of Cancer Hospital, Chinese Academy of Medical Science. All the patients signed informed consent forms for sample collection. After follow-up, the survival information of 73 patients after surgery was obtained. The human esophageal squamous cell line EC9706 was established and studied by Han et al.9 The other 2 esophageal squamous cell lines NEC and EC109 were established and studied previously in our laboratory.8

Construction of eukaryotic expression vector and stable transfection

The coding region of ECRG4 cDNA was subcloned into constitutive mammalian expression vector pcDNA3.1 (Invitrogen). The cDNA was then fully sequenced to ensure that no mutation was introduced during the PCR amplification. The resulting plasmid construct was named pcDNA3.1-ECRG4. EC9706 cells were seeded in 6-cm dishes at 5 × 105 cells/dish and transfected with pcDNA3.1-ECRG4 and pcDNA3.1 using lipofectamine™ 2000 (Invitrogen), according to the manufacturer's protocol. After culturing in medium containing 400 μg/ml of geneticin (Invitrogen) for 3 weeks, individual clones were isolated. Clones that expressed the ECRG4 cDNA coding region were maintained in medium containing 200 μg/ml of geneticin and used for further experiments.

Purification of ECRG4 fusion protein and production of polyclonal antibody

A 688-bp fragment of ECRG4 cDNA was excised from pGEM-T-ECRG4 and subcloned into the pET30a (+) plasmid, producing an inducible expression vector coding for His-tagged ECRG4 soluble protein. Subsequently, the recombinant plasmids were transformed into Escherichia coli BL21 (DE3) cells to produce N-terminal His-tagged soluble ECRG4 protein. Fusion protein expression in Escherichia coli BL21 cells was induced with 0.3 mM isopropyl-D-thiogalactopyranoside (IPTG), and the protein was purified by affinity chromatography with nickel-nitrilotriacetic acid (Ni-NTA) resin (Novagen), according to the manufacturer's protocol. The purified fusion protein was dialyzed in phosphate-buffered saline (PBS; 0.1M sodium phosphate and 0.15M sodium chloride [pH 7.4]) to remove the denaturant. The ECRG4 soluble protein was used to produce polyclonal antibody in rabbit or for functional experiments.10 Rabbit experiments were approved by the institutional review board of animal care of the Animal Center, Peking Union Medical College & Chinese Academy of Medical Sciences.

Western blot analysis

Whole-cell lysates of EC9706 cells were prepared by incubating cells in RIPA buffer (1% NP-40; 0.5% sodium deoxycholate; 0.1% SDS; 50 mM Tris-HCl [pH 7.5]) containing protease inhibitors. Cell lysates were centrifuged at 10,000g for 10 minutes at 4°C. The supernatant was collected, and the protein concentration was measured using the BCA™ Protein Assay Kit (Pierce). Proteins (40 μg) in cell lysates or culture media were separated by 10–15% SDS-polyacrylamide gel electrophoresis and transferred onto PVDF membrane. The membranes were blocked in TBST (0.2 M NaCl; 10 mM Tris pH7.4; 0.2% Tween20)/5% skim milk for 2 hr at room temperature and then incubated with primary antibodies in TBST/5% skim milk. The primary antibodies used for Western blot analysis were polyclonal rabbit anti-ECRG4 (1:2,000), monoclonal mouse anti-p65(1:2,000), polyclonal goat anti-COX-2 (1:2,000), monoclonal mouse anti-IκBα (1:2,000), monoclonal mouse anti-β-tublin (1:4,000) and monoclonal mouse anti-β-actin (1:4,000). The membranes were then washed 3 times with TBST, followed by incubation with horseradish peroxidase-conjugated secondary antibody (1:4,000) in TBST/5% skim milk. Bound antibody was visualized using ECL detection reagent. Nuclear and cytoplasmic protein extraction from EC9706 were done by using commercially available kit (Kit number 78833, Thermo Fisher Scientific Inc., Rockford, IL).

Cell proliferation assay

Stable-transfected cells (pcDNA3.1 and pcDNA3.1-ECRGR4) were seeded into 96-well plates (1.5 × 103 cells/well). After culturing for various durations, cell proliferation was evaluated by thiazolyl blue tetrazolium bromide (MTT) assay, according to the manufacturer's protocol (Sigma-Aldrich Co., St. Louis, MO). In brief, 10 μl MTT solution (5 mg/ml) was added to each well, then the cells were cultured for another 4 hr at 37°C. DMSO (100 μl) was added to each well and mix vigorously to solubilize colored crystals produced within the living cells. The absorbance at 570 nm was measured by using a multi-well scanning spectrophotometer Victor 3.

Colony formation assay

Stable-transfected cells (pcDNA3.1 and pcDNA3.1-ECRG4) were seeded into six-well plates at 200 cells per well in RPMI-1640 with 10% FBS. Fourteen days later, formed colonies were fixed in methanol and stained with 1% crystal violet. Colonies that contained more than 50 cells were counted.

Anchorage-independent growth in soft agar assay

Six-well plates were precoated with 0.6% soft agar in RPMI-1640 with 10% FBS, then stable-transfected cells (pcDNA3.1 and pcDNA3.1-ECRG4) were seeded at 500 cells per well in 0.3% agar/RPMI-1640 with 10% FBS. The plates were incubated for 2 weeks at 37°C. Colonies that contained more than 50 cells were counted.11

Tumor growth in vivo

Six-week-old female BALB/c nude mice were housed at 4 mice per cage and fed ad libitum with autoclaved food. The ECRG4-transfected (pcDNA3.1-ECRG4) or mock-transfected (pcDNA3.1) cells (1×106) were incubated in Trypsin-EDTA (Invitrogen), washed with PBS, centrifuged at 1,500g for 5 minutes, resuspended in PBS, and injected subcutaneously into the armpit region of nude mice. Injected mice were examined every 4 days for tumor appearance, and tumor sizes were estimated from length (a) and width (b) of tumors, as measured by calipers, using the formula V = ab2/2.12 Mouse experiments were approved by the institutional review board of animal care of the Animal Center, Peking Union Medical College & Chinese Academy of Medical Sciences.

ESCC tissue microarray and immunohistochemical (IHC) staining

Formalin-fixed, paraffin-embedded esophageal tumors and the corresponding normal epithelia were placed on the tissue microarray. For immunohistochemical analysis, the slides were deparaffinized, rehydrated, immersed in 3% hydrogen peroxide solution for 10 minutes and cooled down at room temperature for 60 minutes. The slides were blocked by 10% normal goat serum at 37°C for 30 minutes, and then incubated with anti-ECRG4 rabbit polyclonal antibody at a dilution of 1:500 for 3 hr at 37°C. After washed with PBS, the slides were incubated with biotinylated secondary antibody (diluted 1:100) for 30 minutes at 37°C. Immunolabeling was visualized with a mixture of 3, 3′-diaminobenzidine solution. Counter-staining was carried out with hematoxylin. Expression score was determined by staining intensity and percentage of immunoreactive cells. Tissues with no staining were rated as 0, with a faint staining or moderate to strong staining in <25% of cells as 1, with moderate or strong staining in 25–50% of cells as 2, and strong staining in >50% of cells as 3. Staining (25%) was used as cut-off point to define high ECRG4 expression and low ECRG4 expression.

Flow cytometric analysis of cell cycle

The transfected cells (pcDNA3.1-ECRG4 and pcDNA3.1) were seeded at a density of 106 cssionells/100-mm dish in RPMI-1640 medium with 10% FBS for 48 hr. Then cells were washed with ice-cold PBS, harvested and fixed in 70% ethanol for 30 minutes. Cells were treated with RNase A and stained with 25 μg/ml propidium iodide (PI). Samples were analyzed using a FACScan flow cytometer (Becton Dickinson), according to the manufacturer's protocol.

Demethylation treatment and RT-PCR analysis

EC9706 cells were treated with 10 μM 5-aza-2′-deoxycytidine (Sigma) for up to 6 days. Fresh RPMI-1640 medium containing 5-aza-2′-deoxycytidine was replaced every 24 hr. After treatment, cells were washed with PBS and collected for RT-PCR. The primers designed for ECRG4 were 5′-GGT TCT CCC TCG CAG CAC CT-3′ as forward and 5′-CAG CGT GTG GCA AGT CAT GGT TAG-3′ as reverse.8

Statistical analysis

All statistical analysis was performed with the SPSS statistical program (version 11.0). Survival curve was obtained by the Kaplan-Meier method. Overall survival of patients with low versus high ECRG4 expression was analyzed using the log-rank test. Multivariable Cox proportional hazards regression analysis was performed with overall survival as the response variable. To verify the proportional hazards model assumption, we tested the hazard ratios for covariates changed with time (including age, gender, grade, ECRG4 expression, primary tumor size, regional lymph node and tumor stage). p < 0.05 was considered statistically significant.

Results

ECRG4 protein expression was downregulated in ESCC

Our previous study revealed that ECRG4 mRNA level was downregulated or undetectable in ESCC tissues and cell lines.8 In this study, after the anti-ECRG4 polyclonal antibody was produced, the specificity of the antibody for the purified ECRG4 protein was tested by enzyme-linked immunosorbent assay and western blot analysis (Figs. 1a and 1b). We further evaluated the ECRG4 protein level in ESCC tissues and cell lines by Western blot. The result demonstrated that ECRG4 protein expression level was downregulated in 3 of 4 ESCC tumor samples, compared with adjacent normal tissues (Fig. 1c). All 3 ESCC cell lines (NEC, EC109 and EC9706) showed no detectable ECRG4 protein expression (data not shown). Then ECRG4 protein expression level was examined by ESCC tissue microarray immunohistochemical staining. We found that normal esophageal epithelia showed moderately or strong positive staining, yet cancer tissues demonstrated negative or weak immunoreactions (Fig. 1d). Totally, 89 of 130 (68.5%) ESCC specimens studied exhibited reduced ECRG4 protein expression. The ECRG4 protein was predominantly found in the cytoplasm and cell membrane.

Figure 1.

ECRG4 expression was downregulated in ESCC. (a) The anti-ECRG4 polyclonal antibody has high potency for purified ECRG4 protein, shown by enzyme-linked immunosorbent assay. (b) The anti-ECRG4 polyclonal antibody has the specificity for binding to his-tagged ECRG4 protein of the whole bacterium lysis product, which was verified by anti-his antibody. (c) ECRG4 protein expression was downregulated in 3 of 4 ESCC tissues compared with adjacent normal epithelia by Western blot. (d) ECRG4 protein expression was downregulated in ESCC tissues compared with adjacent normal epithelia by immunohistochemical staining of tissue microarray (N = normal; T = tumor) (left; ×100). ECRG4 protein was localized in the cytoplasm and cell membrane (right; ×400). Representative photos of tumor tissues (T1, T2 and T3) showed downregulation of ECRG4 at different degrees compared with normal control (N). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

The association between ECRG4 expression and clinicopathologic features

Then we further analyzed the association between ECEG4 protein expression and clinicopathologic features in ESCC. There were significant correlations between low ECRG4 protein expression and primary tumor size, regional lymph node metastasis and clinicopathologic stage in ESCC (Table I).

Table I. The Association Between ECRG4 Expression and Clinicopathologic Features in ESCC
VariablesNumber of casesECRG4 expression (Mean ± SD)p-value
  • 1

    Mann-Whitney test.

  • 2

    Kruskal-Wallis test.

Age  0.9061
 ≤60 years631.0435 ± 0.8982 
 ≥60 years670.8689 ± 0.8846 
Gender  0.2611
 Male990.8700 ± 0.8367 
 Female310.9667 ± 0.9994 
TNM classification   
 pT  0.0361
  pT1/pT2271.2857 ± 0.9023 
  pT3/pT41030.8090 ± 0.8644 
 N  0.0091
  N0681.1159 ± 0.9782 
  N1620.7167 ± 0.7152 
 Stage  0.0181
 I/II741.0615 ± 0.9164 
 III/IV560.6667 ± 0.7977 
 Grade  0.3662
  G1311.0000 ± 0.9813 
  G2520.9167 ± 0.8711 
  G3470.6316 ± 0.8307 

Promoter hypermethylation and ECRG4 transcriptional inactivation

The expression of tumor suppressor gene is often downregulated in cancer tissues due to the hypermethylation in the gene promoter. Our previous study demonstrated that ECRG4 mRNA level was downregulated in ESCC accompanied with gene promoter hypermethylation.8 In this study, we further treated ESCC cells with demethylation drug 5-aza-2′-deoxycytidine and found that demethylation treatment restored ECRG4 mRNA expression (Fig. 2). The result indicated that promoter hypermethylation could be 1 main mechanism leading to the silencing of ECRG4 in ESCC.

Figure 2.

RT-PCR assay showed that 5-aza-2′-deoxycytidine treatment restored ECRG4 mRNA expression in EC9706 cells.

ECRG4 protein expression and survival of ESCC patients

After follow-up of total 73 ESCC patients, association between ECRG4 protein expression and overall survival was determined by using the log-rank test and multivariable Cox proportional hazards regression analysis. ESCC patients with high ECRG4 protein expression had survived longer than those with low ECRG4 protein expression (Fig. 3), as shown by Kaplan-Meier survival analysis and log-rank test (p = 0.013). Univariate analysis indicated that ECRG4 protein expression level and regional lymph node metastasis were statistically significant prognostic factors (Table II). Multivariable Cox proportional hazards regression analysis indicated that ECRG4 protein expression was an independent prognostic factor for ESCC (ESCC patients with high versus low ECRG4 expression, hazard ratio = 0.115, 95% CI = 0.014–0.917, p = 0.041; Table III).

Figure 3.

The association between ECRG4 protein expression level and overall survival of ESCC patients. Patients with high ECRG4 expression had survived longer (67.50 ± 5.22 months; 95% CI = 57.27–77.73) after surgery than those with low ECRG4 expression (33.86 ± 3.02 months; 95% CI = 27.94–39.78) (p < 0.05, log-rank test). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Table II. Univariate Analysis of the Factors that Influence the Prognosis of ESCC Patients
ParameterNumberRisk ratio95% CIp-value
  1. CI, confidence interval.

Age   0.331
 ≤60361  
 >60371.6140.610–4.272 
Gender   0.568
 Male531  
 Female200.6940.198–2.433 
Primary tumor   0.611
 T1/T2121  
 T3/T4610.7170.199–2.584 
N   0.029
 N0351  
 N1383.4941.137–10.734 
Stage   0.294
 I/II421  
 III/IV310.5710.2–1.627 
G   0.136
 G1211  
 G241   
 G3111.8600.823–4.205 
ECRG4   0.013
 Low561  
 High170.1100.014–0.859 
Table III. Multivariate Analysis of the Factors that Influence the Prognosis of ESCC Patients
ParameterRisk ratio95% CIp-value
  1. CI, confidence interval.

Age  0.508
 ≤601  
 >601.4120.508–3.923 
Gender  0.553
 Male1  
 Female0.6780.188–2.448 
Primary tumor  0.250
 T1/T21  
 T3/T40.5280.178–1.568 
N  0.028
 N01  
 N14.2341.168–15.355 
Stage  0.967
 I/II1  
 III/IV0.9780.337–2.837 
G  0.851
 G11  
 G2   
 G31.0930.431–2.775 
ECRG4  0.041
 Low1  
 High0.1150.014–0.917 

ECRG4 expression inhibited cell proliferation, colony formation, and anchorage-independent growth in soft agar

Then we examined tumor-suppressing function of ECRG4 in vitro. The stable-transfected EC9706/pcDNA3.1-ECRG4 cells exhibited detectable protein expression compared with EC9706/pcDNA3.1 cells (Fig. 4a). We observed that EC9706/pcDNA3.1-ECRG4 cells proliferated more slowly than EC9706/pcDNA3.1 control group, as shown by MTT assay (Fig. 4b). After 2 week's culture, EC9706/pcDNA3.1-ECRG4 group formed colonies smaller in size and fewer in number than EC9706/pcDNA3.1 control group in both colony formation experiment (Fig. 4c) and soft agar assay (Fig. 4d).

Figure 4.

Effects of induced ECRG4 expression on cell proliferation, colony formation, anchorage-independent growth in soft agar in vitro (A-D), and tumor growth in vivo (E-F). (a) Evaluation of ECRG4 protein expression in stable-transfected cells (pcDNA3.1 and pcDNA3.1-ECRG4) by Western blot. ECRG4 protein was detected in pcDNA3.1-ECRG4 cells. (b) Cell growth curves of EC9706/pcDNA3.1 and EC9706/pcDNA3.1-ECRG4 (p < 0.05). Error bars represent standard deviation from mean value. (c) Representative photos and statistic plots of relative colony formation efficiencies of EC9706/pcDNA3.1-ECRG4 and EC9706/pcDNA3.1 for colony formation assay (p < 0.05). Error bars represent standard deviation from mean value. (d) Representative photos and statistic plots of relative colony formation efficiencies of EC9706/pcDNA3.1-ECRG4 and EC9706/pcDNA3.1 for soft agar assay (p < 0.05). Error bars represent standard deviation from mean value. (e) Tumor weights of EC9706/pcDNA3.1-ECRG4 group and EC9706/pcDNA3.1 control group were plotted (p < 0.05). (f) The subcutaneous tumor growth curves of EC9706/pcDNA3.1-ECRG4 group and EC9706/pcDNA3.1 control group in vivo (p < 0.05). Animal experiments were performed twice in duplicate. (g) The percentages of cells in the G1, S and G2/M phase of cell cycle demonstrated that induced expression of ECRG4 in ESCC cells resulted in an accumulation of cells in G0-G1 phase and a decrease in S and G2/M phase compared with control group (p < 0.05). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

ECRG4 expression suppressed tumor growth in vivo

We further evaluated the effect of ECRG4 expression on tumor growth in vivo. EC976/pcDNA3.1 and EC9706/pcDNA3.1-ECRG4 cells were subcutaneously injected into athymic nude mice (8 mice per group), respectively. When the mice were sacrificed 3 weeks after injection, the tumor weights of EC9706/pcDNA3.1 control group were heavier than those of EC9706/pcDNA3.1-ECRG4 group (p < 0.01; Fig. 4e). Although both groups developed tumors when sacrificed, the formed tumors from EC9706/pcDNA3.1 control group grew faster than those from EC9706/pcDNA3.1-ECRG4 group (Fig. 4f).

ECRG4 expression blocked cell cycle progression

Cell cycle examination was carried out by flow cytometric analysis in an attempt to explore the mechanism of growth-suppressing function of ECRG4. The result suggested that ECRG4 expression could arrest ESCC cells at the G1-S checkpoint and delay cell cycle into S phase (Fig. 4g). Consequently, induced ECRG4 expression slowed down cell cycle progression and caused cell cycle G1 arrest.

ECRG4 may be involved in NF-κB pathway

In exploring the molecular mechanism of ECRG4 tumor-suppressing function in ESCC, we found that restoration of ECRG4 expression in ESCC cells inhibited NF-κB expression and nuclear translocation and attenuated NF-κB target gene COX-2 expression (Fig. 5). It indicated that ECRG4 may be involved in NF-κB pathway in ESCC. ECRG4 might induce COX-2 downregulation through NF-κB pathway to inhibit tumor growth in ESCC.

Figure 5.

ECRG4 may be involved in NF-κB pathway. (a) Analysis of cell's total proteins by Western blot showed that NF-κB subunit p65 and NF-κB target gene COX-2 expression were decreased in EC9706/pcDNA3.1-ECRG4 cells compared with EC9706/pcDNA3.1 cells. No detectable difference in IκBα expression was seen between 2 groups. (b) Analysis of nuclear proteins by Western blot indicated that nuclear translocation of NF-κB subunit p65 was reduced in EC9706/pcDNA3.1-ECRG4 cells compared with EC9706/pcDNA3.1 cells. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Discussion

ECRG4 is a new gene cloned and identified in our laboratory,5, 6 and its biological function is still unclear. ECRG4 is highly conserved among various species, suggesting an important role for ECRG4 in eukaryotic cells. In our study, ECRG4 expression was downregulated in ESCC with promoter hypermethylation and demethylation treatment restored ECRG4 mRNA expression in ESCC cells. There were significant correlations between ECRG4 expression downregulation and primary tumor size, regional lymph node metastasis and clinicopathologic stage in ESCC. ECRG4 high expression was associated with prolonged overall survival of ESCC patients; restoration of ECRG4 expression in ESCC cells inhibited cell proliferation, colony formation, anchorage-independent growth in soft agar, cell cycle progression and tumor growth in vivo. Collectively, these results supported the idea that ECRG4 is a candidate tumor suppressor gene in ESCC and indicated that loss of ECRG4 function may play an important role in ESCC carcinogenesis.

ESCC is a heterogeneous disease, thus, ESCC patients with similar clinical-pathologic features often have different prognostic outcomes. Consequently, it is urgently needed to identify those ESCC patients who are at high risk of treatment failure or recurrence years later after surgical resection of tumor. However, the current clinical staging systems for ESCC have limitations in predicting clinical prognosis. So, new prognostic markers will have important clinical implications in identifying ESCC patients at high risk of postsurgical recurrence or predicting the survival of ESCC patients. These specific prognostic molecular biomarkers for ESCC are particularly noteworthy in this regard for the clinicians to evaluate patients and decide individually the best therapy plan. In our study, ESCC patients with high ECRG4 expression had survived longer than those with low ECRG4 expression. Moreover, ECRG4 protein was an independent prognostic factor for ESCC patients in China. We speculate that ESCC with low ECRG4 expression may have growth advantage with a higher proliferative capability, which leads to poor prognosis. Consistent with our findings, Mori et al.13 found that ECRG4 mRNA expression level could also be a candidate prognostic factor for ESCC patients in Japan. Their work strongly supports our findings in Chinese population. In summary, the clinicians may individually select those ESCC patients with high risk of cancer recurrence for adjuvant therapy by utilizing ECRG4 alone or combined with other markers to improve the prognosis of high-risk patients and spare the low-risk patients from unnecessary overtreatment. In future, further studies are required to validate the clinical utility of ECRG4 protein as biomarker for ESCC prognosis.

The downregulation of ECRG4 protein in ESCC could be caused by decreased stability of the protein due to mutation or reduced transcription of the gene and loss of heterozygosity (LOH) or epigenetic mechanisms. In this study, we found no somatic mutation in the coding region of ECRG4 gene through screening 80 matched DNA samples isolated from cancer tissues, adjacent normal epithelia and peripheral blood lymphocytes of ESCC patients from Linxin by PCR-SSCP (data not shown). Moreover, no LOH locus has been reported at 2q1 in ESCC.9, 14, 15 A growing body of evidence has indicated that in addition to mutational inactivation or deletion of tumor suppressor genes, epigenetic gene silencing plays a significant role in carcinogenesis of various human cancers.16–19 Epigenetic events, which are important in normal cellular functions, are also critical factors during initiation and progression of cancer.20, 21 Although genetic abnormalities in several oncogenes and tumor suppressor genes frequently occur in tumorigenesis,22–24 our previous study showed that ECRG4 mRNA was downregulated with gene promoter hypermethylation in ESCC.8 Tumor suppressor genes such as p14, p15, p16, FHIT, CDH1, MGMT, VHL, RASSF1A and E-cadherin are usually downregulated by gene promoter hypermethylation in ESCC.22, 25–29 In this study, we further discovered that the silencing of ECRG4 was reversed by pharmacologic demethylation treatment in ESCC cells. This indicated that promoter hypermethylation could be 1 of the reasons leading to the loss of ECRG4 expression in ESCC.

Cell cycle analysis indicated that ECRG4 might act as a tumor suppressor gene in ESCC by inhibiting tumor cells growth through inducing cell cycle G1 phase arrest. However, the detailed molecular mechanisms still remain unclear. ECRG4 is highly expressed in certain inflammation and immune related diseases,7 and it has high homology with mouse IgG. It appeared plausible that ECRG4 might play a role in tumor inflammation response in ESCC. It is now generally accepted that NF-κB pathway plays a central role between inflammation and carcinogenesis,30–32 and COX-2 is 1 target gene of transcription factor NF-κB.33 We speculated ECRG4 might be involved in NF-κB pathway in ESCC. Consistent with this hypothesis, our study revealed that restoration of ECRG4 expression in ESCC cells attenuated NF-κB expression and nuclear translocation and reduced NF-κB target gene COX-2 expression. Inhibition of NF-κB pathway within malignant cells could induce tumor regression.34 Tumor suppressor ING4 suppressed brain tumor growth through NF-κB and COX-2 pathway.35 ECRG4 might induce COX-2 downregulation through NF-κB pathway to inhibit tumor growth in ESCC. However, further investigation is necessary to determine the exact role played by ECRG4 in NF-κB pathway in ESCC.

Taken together, ECRG4 is a novel candidate tumor suppressor gene in ESCC, and ECRG4 protein is a candidate prognostic biomarker for ESCC.

Acknowledgements

The authors thank Professor Wei Jing of Burnham Institute Cancer Center (La Jolla, CA 92037) for helpful comments on this manuscript. They also thank Dr. Xiao-Chun Wang and Dr. Hong-Yan Chen for the technical assistance.

Ancillary