SEARCH

SEARCH BY CITATION

Keywords:

  • Epstein-Barr virus;
  • gastric cancer;
  • zinc finger E-box binding factor 1;
  • latent-lytic switch;
  • gene modulation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

BACKGROUND:

The role of Epstein-Barr virus (EBV) infection in gastric carcinogenesis remains largely unknown. The authors studied the effects of zinc finger E-box binding factor 1 (ZEB1) on latent-lytic switch of EBV infection in gastric cancer and explored the importance of EBV in gastric carcinogenesis.

METHODS:

Loss or gain of ZEB1 function was obtained by ZEB1 small-interfering RNA (siRNA) knock-down or forced ZEB1 re-expression. Cell growth was evaluated by cell viability and colony formation assays, and the cell cycle was assessed by flow cytometry. EBV was detected using quantitative polymerase chain reaction (PCR) and in situ hybridization analyses.

RESULTS:

ZEB1 knock-down in a latent EBV-infected gastric cancer cell line (YCC10) increased lytic gene BamHI W leftward reading frame 1 (BZLF1) expression and decreased the expression of latent gene EB nuclear antigen 1 (EBNA1) concomitant with the inhibition of cell viability (P < .05) and S-phase DNA synthesis (P < .01). ZEB1 depletion combined with ganciclovir revealed a further reduction in cell viability (P < .001). ZEB1 knock-down induced cell apoptosis and the up-regulation of caspase 3 and poly(adenosine diphosphate-ribose) polymerase cleavage. Conversely, ectopic overexpression of ZEB1 in a lytic EBV-infected gastric cancer cell line (AGS-EBV) inhibited BZLF1 promoter (Zp) activity, BZLF1 expression, and apoptosis and promoted cell growth. EBV infection was detected in 11.3% (80 of 711) of gastric cancers. The presence of EBV was associated with age, men, and intestinal type cancer.

CONCLUSIONS:

ZEB1 was confirmed as a key mediator of the latent-lytic switch of EBV-associated gastric cancer, a distinct subtype with different clinicopathologic features. The current results indicated that inhibition of ZEB1 may be a potential target for EBV-associated gastric cancer therapy. Cancer 2012;. © 2011 American Cancer Society.

Epstein-Barr virus (EBV) has been established as an infective agent that causes gastric cancer.1 EBV infection has 2 distinct forms, latent infection and lytic infection in host cells, which can be distinguished by the different expression of EBV genes.2 However, EBV-infected tumor cells almost always are latently infected. Latent EBV infection is characterized by the minimal expression of viral genes essential for its persistence. Such an immune-escape strategy enables the virus to remain dormant within the host cells.3

The lytic replication of EBV is initiated only after the expression of 1 of the EBV intermediate-early genes, BamHI W leftward reading frame 1 (BZLF1) or BRLF1. Induction of lytic EBV infection results in host cell killing.4 BZLF1 encodes the transcription factor EB1, which functions as a transcriptional activator of viral genes essential for lytic replication.5 During latency, BZLF1 is not expressed. Overexpression of the BZLF1 gene is sufficient to convert cells from a latent form to a lytic form of viral infection.6 Thus, the regulation of BZLF1 expression is crucial to the life cycle of the virus. Recently, zinc finger E-box binding factor (ZEB1) was identified as the transcriptional repressor pivotal for the silencing of BZLF1 promoter (Zp),7, 8 indicating that the aberrant regulation of ZEB1 expression in tumor cells may have an important influence on EBV dormancy and persistence. ZEB1, as a key regulator of latent-lytic switch of EBV infection, may be developed into a novel molecular target for the prevention and treatment of EBV-related gastric cancers.9, 10 However, the mechanism of whether and how ZEB1 regulates the latent-lytic switch of the EBV life cycle in EBV-associated gastric cancer and the role of EBV in gastric carcinogenesis in Chinese remain largely unknown. In the current study, we evaluated the effect of ZEB1 on modulating the latent-lytic switch of EBV infection in gastric cancer cells and explored the potential of ZEB1 as a novel molecular target for EBV-associated gastric cancer. We also addressed the clinical importance of the presence of EBV infection in gastric carcinogenesis in a large-scale cohort of Chinese patients.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

Cell Lines

AGS is a gastric carcinoma cell line, and Akata is an EBV-positive Burkitt lymphoma cell line. The AGS-EBV cell line (a gift from Shannon C. Kenney at the Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC) was obtained from 100-μ/mL hygromycin selection of AGS cells that were infected with a recombinant Akata virus in which a hygromycin resistance cassette had been inserted into the nonessential BDLF3 open reading frame.9 The human gastric cancer cell line YCC10 was naturally infected with EBV (a gift from Dr Qian Tao from the Department of Clinical Oncology, The Chinese University of Hong Kong, Hong Kong).

Primary Gastric Cancer and Noncancer Tissue Samples

Gastric cancer tissues were obtained from 711 patients with primary gastric cancer in the First Affiliated Hospital of Sun Yat-sen University, Guangzhou from January 1999 to December 2006. The median patient age was 58 years, and the cohort included 485 men and 226 women. In addition, 97 gastric tissues with precancerous lesions (intestinal metaplasia and/or atrophic gastritis) and 24 normal gastric tissues were collected. Among 711 patients with gastric cancer, 555 were followed regularly, and their median survival was 40.1 months (range, 0.2-97.6 months). In total, 239 patients (43.06%) patients died during follow-up. All patients and controls provided informed consent for participation in this study, and the study protocol was approved by the Clinical Research Ethics Committee of the Sun Yat-sen University of Medical Sciences.

Knockdown of ZEB1 by RNA Interference

YCC10 cell lines were transfected with small interfering RNA (siRNA) against ZEB1 (ZEB1-siRNA) or with control siRNA at 50 nM, 100 nM, or 200 nM for 72 hours using an oligofectamine transfection reagent (Invitrogen, Carlsbad, Calif). In some experiments, YCC10 cells (5 × 103 cells per well) were seeded in 96-well plate and transfected with ZEB1-siRNA or control siRNA 1 day before the cells were treated with or without antivirus the prodrug ganciclovir (GCV) (10 mg/mL). The cells were harvested after incubation with or without GCV for 48 hours.

Construction of ZEB1 Expression Vector

A mammalian expression vector pCI-neo-ZEB1 encoding the full-length open reading frame of the human ZEB1 gene was a gift from Dr. Jennifer Richer (Department of Pathology, University of Colorado Health Sciences Center, Aurora, Colo). The sequence corresponding to the open reading frame clone of ZEB1 was amplified and verified by DNA sequencing.

Ectopic Expression of ZEB1 in the AGS-EBV Gastric Cancer Cell Line

The lytic EBV-infected AGS cell line AGS-EBV (5 × 105 cells each) were transfected with 2 μg of either ZEB1-expressing plasmid or empty plasmid as a control using Lipofectamine 2000 Transfection Reagent (Invitrogen). The transfection efficiency of ZEB1 was evaluated by reverse transcriptase-polymerase chain reaction (RT-PCR) and Western blot analyses after incubation for 24 hours, 48 hours, and 72 hours.

Cell Viability Assay

Cell viability was measured by using the 3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay according to the manufacturer's instruction (Promega, Madison, Wis). Each experiment was performed in triplicate.

Luciferase Reporter Activity Assay

The promoter construct pGL3-Zp containing the BZLF1 promoter sequences of EBV strain B95.8 from nucleotide (nt) −222 to nt +57 was cloned into the luciferase reporter vector pGL3-basic (Promega, Madison, Wis). AGS-EBV cells transfected with pCI-neo-ZEB1 or pCI-neo vector (1 × 105 cells per well) in 24-well plates were cotransfected with luciferase reporter plasmid (0.5 μg per well) and pRL-cytomegalovirus vector (5 ng per well) using lipofectamine 2000 (Invitrogen). Cells were harvested 72 hours post-transfection, and luciferase activity was quantified using the Dual-Luciferase Reporter System (Promega). All experiments were performed in triplicate.

Cell Cycle Analysis

Cell cycle distribution was determined by flow cytometry. After 12 hours of synchronization by serum starvation, the transfected YCC10 cells with siRNA-ZEB1 plasmid or control siRNA were restimulated with 10% fetal bovine serum for 72 hours. The cells were sorted by using a FACSCalibur (BD Biosciences, Franklin Lakes, NJ) after being stained with propidium iodide (PI), and cell cycle profiles were analyzed using the ModFitLT software (Becton Dickinson, San Diego, Calif). All experiments were performed in triplicate.

Cell Apoptosis Assay

Cell apoptosis was determined by dual staining with annexin V/fluorescein isothiocyanate (FITC) and PI (Invitrogen). Annexin V/FITC and PI were added to the cellular suspension, and sample fluorescence of 10,000 cells was analyzed by flow cytometry (Becton Dickinson). All experiments were performed in triplicate.

Colony-Formation Assay

AGS-EBV cells (5 × 105 per well) were seeded onto a 6-well plate and transfected with pCI-neo-ZEB1 expression vector or pCI-neo empty vector (2 μg each) using Lipofectamine 2000 Transfection Reagent (Invitrogen). Forty-eight hours after transfection, the cells were collected, plated at appropriate density onto a 6-well plate, and subjected to G418 (0.5 mg/mL) selection (Merck, Darmstadt, Germany) for 2 weeks. Surviving colonies (>50 cells per colony) were counted after staining with Gentian Violet (ICM Pharma, Singapore). All experiments were performed in triplicate.

RNA Isolation, Semiquantitative RT-PCR, and Real-Time PCR

Total RNA was extracted from cell pellets using TRIzol Reagent (Molecular Research Center, Inc., Cincinnati, Ohio). Combinational DNA was synthesized from 2 μg total RNA, and messenger RNA expression of ZEB1, BZLF1, and nuclear antigen 1 (EBNA1) in cell lines was examined by semiquantitative RT-PCR with the specific primers listed in Table 1.

Table 1. List of Primers Used for the Messenger RNA Expression Assay
NamePrimer Sequence
  1. Abbreviations: A, adenine; BZLF1, BamHI W leftward reading frame 1; C, cytosine; EBNA1, Epstein-Barr nuclear antigen 1; G, guanine; T, thymine; ZEB1, zinc finger E-box binding factor 1.

ZEB1 
 Sense5′-GCACCTGAAGAGGACCAGAG-3′
 Antisense5′-GTGTAACTGCACAGGGAGCA-3′
EBNA1 
 Sense5′-CGTTTGGGAGAGCTGATTCT-3′
 Antisense5′-CCCCTCGTCAGACATGATTC-3′
BZLF1 
 Sense5′-GCACATCTGCTTCAACAG-3′
 Antisense5′-CGTGAGGTCAGTATATAC-3′
β-Actin 
 Sense5′-GTCTTCCCCTCCATCGTG-3′
 Antisense5′-AGGGTGAGGATGCCTCTCTT-3′

DNA Isolation and Real-Time PCR Quantitation of EBV-DNA

Genomic DNA was extracted from gastric tissues using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Levels of EBV DNA were measured using 2 real-time quantitative PCR systems with the BamHI-W Taqman probe targeting the BamHI-W fragment region and the EBNA1 Taqman probe targeting the EBNA1 region of the EBV genome (Table 2).11, 12 The gene copy number from the EBV-positive cell line Namalwa was used as a standard.11 Quantification of EBV DNA concentration was determined using an ABI Prism 7700 Sequence Detector (Applied Biosystems, Foster, City, Calif). Results were expressed as the number of EBV genome copies per 105 cells.12

Table 2. List of the Primers Used for Real-Time Polymerase Chain Reaction Analysis to Quantify Epstein-Barr Virus DNA
NamePrimer Sequence
  1. Abbreviations: A, adenine; C, cytosine; EBNA1, Epstein-Barr nuclear antigen 1; FAM, carboxyfluorescein; G, guanosine; T, thymine; TAMRA, carboxytetramethylrhodamine.

BamH1 
 Sense5′-CCCAACACTCCACCACACC-3′
 Antisense5′-TCTTAGGAGCTGTCCGAGGG-3′
 Probes5′-(FAM)CACACACTACACACACCCA CCCGTCTC(TAMRA)-3′
EBNA1 
 Sense5′-TCATCATCATCCGGGTCTCC-3′
 Antisense5′-CCTACAGGGTGGAAAAATGGC-3′
 Probes5′-(FAM)CGCAGGCCCCCTCCAGGTAGAA(TAMRA)-3′
β-Actin 
 Sense5′-GCGCCGTTCCGAAAGTT-3′
 Antisense5′-CGGCGGATCGGCAAA-3′
 Probes5′-(FAM)ACCGCCGAGACCGCG TC(MGB)-3′

Western Blot Analysis

Forty micrograms of protein were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto equilibrated polyvinylidene difluoride membranes (Amersham Biosciences, Buckinghamshire, United Kingdom) by electroblotting. Membranes were incubated with primary antibodies (Santa Cruz Biotechnology, Santa Cruz, Calif) overnight at 4°C. After incubation with the secondary antibody, proteins were detected by enhanced chemiluminescence (Amersham, La Jolla, Calif).

In Situ Hybridization for EBV-Encoded Small RNA

To confirm the existence of EBV in gastric cancer cell lines and in gastric cancer tissues, detection of EBV-encoded small RNA (EBER) was carried out with EBV probe in situ hybridization kit (Novocastra, Newcastle, United Kingdom). The sections were hybridized to the fluorescein-conjugated probes. A hybridization reaction was detected by incubation with antifluorescein antibody tagged with alkaline phosphatase. The color reaction was performed with nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indolyl phosphate solution, and the sections were counterstained with 0.1% methyl green. Positive in situ hybridization results for EBER were indicated by a dark brown color in cell nuclei.

Immunohistochemistry for ZEB1

ZEB1 protein expression was detected in paraffin-embedded sections of 15 EBV-positive gastric cancers and 50 EBV-negative gastric cancers using the specific antibody (1:25 dilution; Santa Cruz Biotechnology) and an avidin-biotin complex immunoperoxidase method. Negative controls were run by replacing the primary antibody with nonimmune serum.

Statistical Analysis

The results were expressed as means ± standard deviations. The Mann-Whitney U test was used to compare variables between the 2 sample groups. The correlation between EBV viral load data measured with the BamHI-W region PCR and data measured with the EBNA1 PCR was studied by using the Spearman bivariate correlation test. Selection of a cutoff value was based on the receiver operating characteristics (ROC) curve. The association between patient characteristics and EBV status was analyzed with the chi-square test. Overall survival in relation to EBV status was evaluated by using Kaplan-Meier survival curves and the log-rank test. All analyses were performed using the SAS for Windows software package (version 9; SAS Institute, Inc., Cary, NC). All tests of statistical significance were 2-sided, and P values < .05 were considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

ZEB1 Knock-Down Induced Latent-to-Lytic Switch of EBV Infection

The presence of EBV in AGS-EBV and YCC10 cell lines was confirmed by using an EBER assay (Fig. 1A). The effect of ZEB1 on the expression of BZLF1 in the latent EBV-infected gastric cancer cell line YCC10 (Fig. 1A) was investigated through knock-down ZEB1 with siRNA. ZEB1 was knocked down significantly at 100 nM ZEB1-siRNA (Fig. 1B). Knockdown of ZEB1 markedly enhanced expression of the lytic gene BZLF1 in YCC10 cells compared with the control siRNA-treated cells (Fig. 1C). We also investigated the effect of ZEB1 on the expression of EBNA1, a well known marker for latent EBV infection. EBNA1 expression was inhibited significantly by ZEB1 knockdown (Fig. 1D), as expected. These data provide evidence that ZEB1 plays an essential role in maintaining EBV latency in EBV-positive gastric cancer cells and that ZEB1 depletion can induce the lytic form of EBV infection in gastric cancer cell.

thumbnail image

Figure 1. (A) Epstein-Barr virus (EBV)-encoded small RNA (EBER) in situ hybridization was used to validate the presence of EBV infection in the AGS gastric cancer cell line (original magnification, ×400). Dark brown nuclear staining identifies a positive hybridization signal. (B) The depletion of zinc finger E-box binding factor 1 (ZEB1) in YCC10 cells by ZEB1 small-interfering RNA (siRNA)-mediated knock-down was confirmed by real-time polymerase chain reaction (PCR) analysis. A single asterisk indicates P < .05; double asterisks, P < .001. (C,D) siRNA-mediated knock-down of ZEB1 induced (C) the expression of BamHI W leftward reading frame 1 (BZLF1) and (D) attenuated the expression of Epstein-Barr nuclear antigen 1 (EBNA1) in YCC10 cells by semiquantitative reverse transcriptase-PCR and real-time PCR. Triple asterisks indicate P < .0001. Data are presented as means ± standard deviations.

Download figure to PowerPoint

Knockdown of ZEB1 Induced Significant Suppression of Cell Proliferation and Cell Cycle Arrest in the YCC10 Cell Line

The effect of ZEB1 knockdown on cell viability of YCC10 cells also was examined by MTS assay. We observed that ZEB1 knockdown caused approximately 20% inhibition in cell numbers compared with control siRNA-transfected YCC10 cells (P < .01) (Fig. 2A). Fluorescence-activated cell sorting (FACS) analysis revealed a significant decrease in the number of cells in S-phase in among the YCC10 cells with ZEB1 knockdown compared with the control cells (P < .01) (Fig. 2B), confirming the inhibitory effect of ZEB1 knockdown on cell proliferation. Concomitant with this inhibition of cell proliferation, there was a significant increase in the number of cells accumulating in G2/M-phase after ZEB1 knockdown with YCC10 cells (P < .01) (Fig. 2B).

thumbnail image

Figure 2. Zinc finger E-box binding factor 1 (ZEB1) depletion in YCC10 cells inhibited cell growth. (A) Cell viability of YCC10 cells after transient transfection with ZEB1 small-interfering RNA (siRNA) or control (CTL)-siRNA for 72 hours was determined by 3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay. In some experiments, the cells were treated with the antivirus prodrug ganciclovir (GCV) (10 mg/mL) for the last 48 hours before harvest. (B1) These charts illustrate representative fluorescence-activated cell sorting analysis of YCC10 cells that were transfected with ZEB1 siRNA or control siRNA for 72 hours. (B2) Cell proliferation was calculated as the fraction of cells in S-phase. (B3) The number of cells in G2/M-phase also was determined. A single asterisk indicates P < .01. Data are presented as means ± standard deviations. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

ZEB1 Deletion Enhanced the Effect of GCV on Gastric Cancer Cells

Because the lytic (but not the latent) form of EBV infection converts the antivirus drug GCV into its active form, we examined whether ZEB1 depletion could increase the sensitivity of gastric cancer cells to GCV. After ZEB1 knockdown, as expected, GCV treatment demonstrated a significantly greater additive effect on cell growth with 55% inhibition in cell viability (P < .001) compared with YCC10 cells that were transfected with ZEB1-siRNA alone (Fig. 2A). However, there was no difference in cell viability among YCC10 cells that were treated with GCV alone (Fig. 2A), indicating that depletion of ZEB1 rendered YCC10 cells much more sensitive to the cytotoxic effects of GCV, and the enhanced cytotoxicity was lytic status-dependent on ZEB1 knockdown.

Depletion of ZEB1 Induced Apoptosis in YCC10 Cells

To determine whether the decrease in cell viability observed was caused by an induction of apoptosis, the cellular apoptotic rate was determined using annexin-V/FITC/PI double staining. The number of early apoptotic cells at 72 hours after ZEB1-siRNA transfection was increased substantially compared with control siRNA-transfected cells (P < .05) (Fig. 3A). The induction of apoptosis was confirmed further by an analysis of 2 crucial apoptosis-related mediators—caspase 3 and poly(adenosine diphosphate-ribose) (PARP)—by Western blot analysis. Enhanced expression of the active forms of caspase 3 and PARP were demonstrated in YCC10 cells treated with ZEB1-siRNA, as indicated in Figure 3B. These results suggested that apoptosis concomitant with G2/M cell cycle arrest induced by the down-regulation of ZEB1 was a plausible cause leading to the growth inhibition in ZEB1-depleted gastric cancer cells.

thumbnail image

Figure 3. Zinc finger E-box binding factor 1 (ZEB1) depletion in YCC10 cells induced cell apoptosis. (A1) Apoptosis was measured by flow cytometric analysis of annexin V/fluorescein isothiocyanate (FITC) double-labeled cells. (A2) Knock-down of ZEB1 in YCC10 cells after transient transfection with ZEB1 small-interfering RNA (siRNA) induced cell apoptosis compared with control siRNA-transfected cells. Data are presented as means ± standard deviations. A single asterisk indicates P < .05. (B) Protein expression of apoptosis-related genes was determined using Western blot analysis. PARP indicates poly(adenosine diphosphate-ribose) polymerase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Overexpression of ZEB1 Caused Lytic-to-Latent Switch Through Repressing Zp Activity in Lytic AGS-EBV Gastric Cancer Cells

To confirm the effect of ZEB1 in the regulation of the EBV latent-lytic switch, we tested whether the overexpression of ZEB1 could inhibit lytic reactivation. The lytic EBV-infected AGS cells, which had no ZEB1 expression, were transfected with a pCI-neo vector that expressed ZEB1 or with an empty control vector. Ectopic overexpression of ZEB1 in AGS-EBV cells was confirmed by RT-PCR analysis and Western blot analysis, respectively (Fig. 4A). The overexpression of ZEB1 led to a significant inhibition of EBV lytic gene (BZLF1) expression (Fig. 4B) in AGS-EBV cells.

thumbnail image

Figure 4. The overexpression of zinc finger E-box binding factor 1 (ZEB1) induced lytic-to-latent switch in AGS-Epstein-Barr virus (EBV) cells. (A) Ectopic overexpression of ZEB1 in AGS-EBV cells after transient transfection with the mammalian expression vector pCI-neo-ZEB1 was confirmed by (A1) reverse transcriptase-polymerase chain reaction analysis and (A2) Western blot analysis. BZLF1 indicates BamHI W leftward reading frame 1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. (B,C) The overexpression of ZEB1 in AGS-EBV cells (B) suppressed the expression of BZLF1 and (C) inhibited BZLF1 promoter (Zp) activity by the luciferase reporter assay (pGL3).

Download figure to PowerPoint

It was reported previously that transcription of the BZLF1 gene can be initiated from its proximal promoter Zp and that ZEB1 binds to Zp, repressing BZLF1 transcription.13 Next, we examined the activity of Zp before and after ZEB1 overexpression in AGS-EBV cells using a luciferase reporter activity assay. Our results indicated that the activity of Zp was inhibited significantly by ZEB1 re-expression (P < .001) (Fig. 4C), indicating that ZEB1 inhibited BZLF1 transcription by reducing activity of the BZLF1 promoter Zp.

Ectopic Expression of ZEB1 Promoted the Growth of Gastric Cancer Cells With Lytic EBV Infection

We further evaluated the biologic influence of ectopic ZEB1 in lytic AGS-EBV cells. Ectopic expression of ZEB1 in AGS-EBV cells caused a significant increase in the number of viable cells (P < .01) (Fig. 5A). The promoting effect of ZEB1 on cell growth was confirmed further in a colony-formation assay (Fig. 5B). The colonies that formed in ZEB1-transfected cells were significantly greater in number and larger in size than those formed in empty vector-transfected cells (up to 100% of vector control; P < .001) (Fig. 5B). Moreover, FACS analysis of ZEB1-transfected AGS-EBV cells revealed a significant induction in the number of S-phase cells compared with vector-transfected cells (P < .01) (Fig. 5C). Thus, ZEB1 overexpression induced cell proliferative capacity.

thumbnail image

Figure 5. (A) Transient overexpression of zinc finger E-box binding factor 1 (ZEB1) enhanced the viability of AGS-Epstein-Barr virus (AGS-EBV) cells. A single asterisk indicates P < .01. (B) The overexpression of ZEB1 promoted AGS-EBV cell colony formation after stable transfection with ZEB1-expressing vector. Double asterisks indicate P < .001. (C) ZEB1 enhanced the number of cells in S-phase. Data are presented as mean ± standard deviation. (D) Protein expression levels of apoptosis-related genes were determined using Western blot analysis. PARP indicates poly(adenosine diphosphate-ribose) polymerase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Suppression of Apoptosis by ZEB1 in AGS-EBV Cells

We examined the contribution of apoptosis to the observed growth inhibition in AGS-EBV cells derived by ZEB1. The overexpression of ZEB1 down-regulated protein expression of cleaved caspase 3, cleaved caspase 9, and cleaved PARP compared with vector-transfected AGS-EBV cells, indicating reduced cell apoptosis, as illustrated in Figure 5D.

Quantitative Analysis of EBV DNA Levels in Gastric Cancer, Precancerous Lesions, and Controls

The presence of EBV in 711 primary gastric cancer samples, 97 premalignant lesions, and 24 normal gastric tissue samples was determined with 2 EBV DNA fragments targeting the BamHI-W region and the EBNA1 region. There was a strong, positive correlation between EBV viral load data from the BamHI-W region and from the EBNA1 region (correlation coefficient, 0.683; P < .0001). ROC curve analysis based on EBV viral load levels in gastric cancer samples and normal gastric tissue samples yielded area under the curve values of 0.956 for BamHI-W and 0.930 for EBNA1 (Fig. 6A), respectively. At a cutoff value of 106 copies per 105 cells, both BamHI-W and EBNA1 had sensitivity of 92.3% and specificity of 97.1%.

thumbnail image

Figure 6. (A) Receiver operating characteristics (ROC) curve analysis was based on quantification of (A1) BamHI-W and (A2) Epstein-Barr nuclear antigen 1 (EBNA1) to detect Epstein-Barr virus (EBV) infection in gastric cancer (GC) tissues and normal control tissues. BamHI-W and EBNA1 yielded area under the curve values of 0.956 and 0.930, respectively. (B) Photomicrographs of histologic specimens reveal the features of positive nuclear signals for EBV RNA by EBV-encoded small RNA (EBER) in GC tissues (dark brown color in cell nuclei; original magnification, ×400). (C) Kaplan-Meier survival curves are shown.

Download figure to PowerPoint

EBV DNA was detected in 80 of 711 gastric cancer samples (11.3%), in 4 of 97 precancerous lesions (4.1%), but not in any of 24 healthy controls using both BamHI-W PCR and EBNA1 PCR. The proportion of EBV DNA-positive samples among these groups differed significantly (chi-square statistic, 7.57; P < .05). EBV DNA-positive results were significantly more frequent in gastric cancer samples than in precancerous lesion samples (chi-square statistic, 4.66; P < .05).

EBER assay, which is the gold standard for EBV detection, also was applied to validate the results generated by Taqman PCR assays. Thirteen samples that had measurable EBV DNA in 2 Taqman assays were selected randomly for EBER assay in paraffin-embedded tissues. The results indicated that all 13 samples had detectable EBV by EBER assay (Fig. 6B) consistent with their EBV status detected by PCR.

Clinicopathologic Features of EBV-Associated Gastric Carcinoma

An analysis of the association between clinicopathologic features and EBV infection in human gastric cancers is presented in Table 3. The presence of EBV was associated with age (P < .05), a male predominance (P < .0001), and intestinal histologic type (P = .05) and was marginally associated with well or moderate differentiation of gastric cancer (P = .08). However, we observed no correlation between the presence EBV and tumor location, Helicobacter pylori infection (Table 3), or survival in patients with gastric cancer (Fig. 6C).

Table 3. Clinicopathologic Features of Epstein-Barr Virus in Patients With Gastric Cancer
 EBV Status: No. (%) 
VariablePositive, n = 68Negative, n = 487P
  1. Abbreviations: EBV, Epstein-Barr virus; SD, standard deviation; TNM, tumor, lymph node, metastasis.

Age: Mean ± SD, y53.66 ± 13.0857.00 ± 12.60.0418
Sex   
 Men60 (15.87)318 (84.13).0002
 Women8 (4.52)169 (95.48) 
Location   
 Proximal16 (11.12)127 (88.88)>.05
 Distal45 (11.90)333 (88.1) 
Lauren classification   
 Intestinal60 (13.92)371 (86.08).05
 Diffuse8 (6.72)111 (93.28) 
Differentiation   
 Poor51 (14.61)298 (85.39).0845
 Well or moderate10 (8.13)113 (91.87) 
TNM stage   
 I7 (9.59)66 (90.41).6398
 II6 (8.45)65 (91.55) 
 III22 (11.46)170 (88.54) 
 IV26 (13.98)160 (86.02) 
Helicobacter pylori   
 Positive18 (15.79)96 (84.21) 
 Negative16 (8.38)175 (91.62).110

ZEB1 Expression in Primary Gastric Cancer Tissues

We evaluated ZEB1 expression in EBV-positive and EBV-negative primary gastric cancer tissues by using immunohistochemistry (Fig. 7). ZEB1 was detected frequently in EBV-positive gastric cancers (80%; 12 of 15 samples) compared with EBV-negative gastric cancers (10%; 5 of 50 samples; P < .0001).

thumbnail image

Figure 7. These are representative images of zinc finger E-box binding factor 1 (ZEB1) expression in primary gastric cancer (GC) specimens determined by immunohistochemistry. ZEB1 was expressed in the nucleus of Epstein-Barr virus-positive gastric cancer cells (brown color; original magnification, ×400).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

The gastric cancer cell line YCC10 is a natural model of gastric cancer cells with latent EBV infection. We observed that ZEB1, a key repressor in silencing the expression of the latent-lytic switch gene BZLF1 during EBV latency, was highly expressed in this cell line. The down-regulation of ZEB1 in YCC10 cells caused the up-regulation of BZLF1 expression and the down-regulation of latent gene EBNA1 expression (Fig. 1), thus promoting the latent-lytic switch of EBV infection. BZLF1 reportedly regulates the switch from latent infection to virus replication in EBV-infected cells and, thus, acts as a key mediator of reactivation from latency to the viral productive infection of EBV14 through activating the promoter of another EBV gene, BSLF2/BMLF1 (human herpesvirus 4).15 Expression of the BZLF1 gene is necessary and sufficient to disrupt EBV latency.16 Conversely, it also has been reported that EBNA1 is expressed in all types of EBV latent infections that are identified in proliferating cells and tumors.17 EBNA1 is essential for the maintenance of viral latent replication and persistence.18, 19 Thus, loss of ZEB1 may lead to reactivation into lytic replication because of the enhanced expression of BZLF1 and the reduced expression of EBNA1. To better define the effect of ZEB1 on the latent-lytic switch in gastric cancer, we examined its functional consequences by knocking it down in the human gastric cancer cell line YCC10. In the current work, decreased ZEB1 expression in YCC10 cells led to the inhibition of cell growth and S-phase cells (Fig. 2), induced apoptosis, and caused cell cycle arrest in G2/M-phase (Fig. 3). The induction of apoptosis was confirmed further by increased expression of the activated forms of caspase 3 and PARP, which led to the impairment of DNA repair and apoptosis (Fig. 3B). Thus, heightened ZEB1 depletion may diminish EBV-positive gastric cancer cell growth by up-regulating apoptotic cell death pathways. Collectively, knocking down ZEB1 by itself was sufficient to induce EBV lytic replication in latently infected gastric cancer cells.

We observed that GCV alone was barely effective in controlling the YCC10 cells. This is consistent with previously findings that EBV-positive tumor cells infected with latent viral forms do not respond to GCV.20 However, the induction of lytic EBV infection in YCC10 cells induced by the re-expression of an immediate-early gene BZLF1 through knocking down ZEB1 allowed the cells to be killed by GCV. This is because the host cells with lytic (but not latent) EBV infection express virally encoded kinases to phosphorylate the prodrug GCV into its cytotoxic form.20, 21 Phosphorylated GCV is able to inhibit DNA polymerase of the host cells, leading to the apoptosis of EBV-infected tumor cells.20, 21 Because EBV-positive tumor cells are primarily in the latent form of EBV infection, induction of the latent-to-lytic switch of the EBV life cycle by ZEB1 inhibition can improve the clinical efficacy of GCV specifically by killing EBV-positive tumor cells, representing a new option for the treatment of EBV-associated gastric cancer.

In addition to a loss-of-function study, we investigated the role of ZEB1 as a transcriptional repressor of BZLF1 and, thus, as a regulator of the latent-to-lytic switch of the EBV life cycle in gastric cancer through a gain-of-function assay. Ectopic overexpression of ZEB1 in AGS-EBV, a gastric cancer cell line that contains the lytic form of EBV infection22 in the absence of ZEB1, led to the down-regulation of BZFL1 (Fig. 4B). We also demonstrated showed that this suppressive effect of ZEB1 on BZFL1 expression was mediated specifically by binding to a specific site (Zp) of the BZLF1 promoter (Fig. 4C). This was supported by recent reports that ZEB1 can directly bind Zp through the ZV element, repressing the transcription of BZLF1 initiated from Zp and, thus, contributing to regulation of the switch between latency and lytic replication of EBV.7, 8, 13, 23 Two recent studies indicated the role of the microRNA molecule miR200 in regulating the EBV life-cycle switch and also observed that ZEB1 repressed Zp activity.24, 25 In addition, ectopic expression of ZEB1 in AGS-EBV cells had a marked promoting effect on cell growth and proliferation (Fig. 5). Moreover, ectopic expression of ZEB1 in AGS-EBV cells reduced the expression of proapoptotic genes, including cleaved caspase 3, caspase 9, and PARP (Fig. 5D). The reduced cell apoptosis in ZEB1-transfected AGS-EBV cells was caused at least in part by the down-regulation of BZLF1, which reportedly induced cell death in EBV-positive gastric cancer cells.6 Taken together, these results indicate that the overexpression of ZEB1 is sufficient to inhibit lytic reactivation by inhibiting the transcription of BZLF1, further confirming that ZEB1 indeed plays a central role in the maintenance of EBV latency in gastric cancer cells.

The association between EBV infection and gastric cancer has not been well documented in a Chinese population. We evaluated whether EBV-positive gastric cancers display distinct clinicopathologic features and indicate a different prognosis relevant to EBV-negative gastric cancers in a large (711 patients), well documented patient population. The presence of EBV in gastric cancer tissues was determined by 2 real-time, quantitative PCR analyses that targeted different parts of the EBV genome (BamHI-W and EBNA1)12 and was validated by an EBER assay. In our cohort, 80 of 711 patients (11.3%) had EBV-positive gastric cancer. This is similar to the prevalence of EBV reported in patients with gastric cancer in other countries.26 EBV-positive tumors were observed more often in men (P < .001) and in younger patients (P < .05) (Table 2). Trends toward the predominance of men and of younger patients have been observed previously in Japanese27 and Dutch28 patients with gastric cancer. Histologically, a greater prevalence of intestinal type tumors was observed in EBV-positive patients. This is consistent with the data published by van Beek et al.28 The development of the intestinal type malignancy is a longer multistep process through gastric atrophy, intestinal metaplasia, dysplasia, and (ultimately) intestinal type carcinoma, which is different from the diffuse type of gastric cancer in terms of age, sex, and the process of carcinogenesis, in which no precursor lesions have been identified to date. The finding that EBV infection was detected in precancerous lesions (atrophy and intestinal metaplasia), although the frequency of EBV infection was distinctly lower in precancerous lesions than in tumors (P < .05), but not in normal gastric tissues, indicated that EBV enters the gastric epithelium at an early stage during the multistep process of gastric carcinogenesis. This is in line with the observations of others who observed EBV in precursor lesions and also in carcinoma cells.29 Thus, it is likely that EBV infects a dysplastic gastric epithelial cell, transforming it into a carcinoma cell as an additional mechanism contributing to gastric malignant progression. Because ZEB1 was detected more frequently in EBV-positive gastric cancers than in EBV-negative gastric cancers (P < .0001), the current results indicate that ZEB1 is essential for the maintenance of latent EBV replication and persistence in gastric cancer.

No difference was observed in the tumor location in our cohort, although it has been reported that EBV-associated gastric tumors demonstrate a preferential location in the proximal part of the stomach.28 In addition, the overall survival of our EBV-positive patients with gastric cancer did not differ from the survival of patients who had EBV-negative tumors (Fig. 6C). However, a better survival had been described in EBV-associated gastric cancers compared with EBV-negative cancers in the Netherlands, where there is a low incidence of gastric cancer.28 The discrepancies in tumor location and prognosis for patients with EBV-positive cancer observed by us and by others most likely are the result of investigating different populations. Collectively, our results suggest that EBV plays a distinct role in gastric carcinogenesis in Chinese patients.

In conclusion, our findings suggested that ZEB1 is a key repressor in the maintenance of EBV latency in gastric cancer cells through silencing BZLF1. EBV-carrying primary gastric cancer is a distinct subtype with different clinicopathologic features. Induction of the latent-to-lytic switch of the EBV life cycle by inhibiting ZEB1 sensitized EBV-positive gastric cancer cells to GCV, thus providing a new paradigm for treating patients with EBV-associated gastric cancer.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

This project was supported by research funds from of the Research Fund for the Control of Infectious Diseases (08070522 and 10090942), the China 973 Program (2010CB529305), the Chinese University of Hong Kong Group Research Scheme (3110043), and Focused Investments (2041423).

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES