Introduction

Integrins are a superfamily of cell-surface receptors that mediate cell–cell and cell–extracellular matrix (ECM) interactions, and control metastasis by regulating the adhesive ability of tumor cells (Hynes and Lander, 1992; Miranti and Brugge, 2002). They are heterotypic adhesion receptors composed of noncovalently associated and subunits (Tamkun et al., 1986) and may crosstalk with each other, thereby regulating their functionalities. Integrin 4 is composed of the cell-surface heterodimers a41 and a47, is expressed mainly on lymphoma, melanoma and fibroblasts, and is involved in cell migration and attachment (Chan et al., 1992; Hamann et al., 1994; Michetti et al., 2000). Several studies have suggested that altered integrin 4 expression is correlated with transformation or metastasis (Rincon et al., 1992; Holzmann et al., 1998). For example, a high expression of integrin 4 was found to be associated with the loss of tumorigenicity and a reduction in matrigel invasion by lymphoma and melanoma cells (Qian et al., 1994; Gosslar et al., 1996). However, the mechanism of the loss of integrin 4 from cancer cells has not been explored.

Epigenetic mechanisms that involve DNA methylation and alterations of chromatin structure represent an important way of transcriptionally silencing many genes, especially tumor suppressor genes, and act as alternatives to genetic defects in human cancers. Recently, several studies have found that the methylation of the CpG island is associated with gene silencing in tumors; these include pRb, p15^Ink4b, p16^Ink4a, hMLH1 and TGF- type I receptor (Herman et al., 1996a, 1996b, 1998; Stirzaker et al., 1997; Kang et al., 1999; Song et al., 2000). Moreover, genes silenced by DNA methylation can be reactivated by treatment with 5-aza-2'-deoxycytidine (5-Aza-dC), a well-established inhibitor of DNA methyltransferase (Jones and Taylor, 1980).

In the present study, we investigated whether methylation of the integrin 4 gene inactivates integrin 4 expression in gastric cancer cells. Our findings demonstrate that integrin 4 mRNA expression is lost in the majority of gastric cancer cell lines and in primary tumors by DNA methylation, suggesting that the epigenetic inactivation of the integrin 4 gene may play a role in gastric carcinogenesis.

Results

Loss of integrin 4 expression in gastric carcinoma cells

Integrin 4 mRNA expression was analysed in nine gastric carcinoma cell lines by RT–PCR using specific primers for integrin 4 (exons 4–8). Of these gastric carcinoma cell lines, only one cell line was found to express integrin 4 mRNA (Figure 1a). We used the demethylating agent 5-Aza-dC to investigate whether the loss of integrin 4 expression was caused by DNA methylation in integrin 4-nonexpressing cells. Treatment with 5-Aza-dC was found to restore integrin 4 expression in eight cell lines, without basal integrin 4 expression. AGS cells were exposed to increasing concentrations of 5-Aza-dC for 4 days, which resulted in the dose-dependent induction of integrin 4 mRNA (Figure 1b), and similar results were obtained in other cell lines, that is, SNU-1, SNU-5, SNU-719 and MKN-28 (data not shown). These results suggest that the lack of integrin 4 expression is attributable to a transcription block caused by the methylation of integrin 4 DNA in human gastric carcinoma cells.

Figure 1.

Restoration of integrin 4 mRNA expression after 5-Aza-dC treatment. (a) Cells were treated for 4 days with DMSO (vehicle) or 5 M of 5-Aza-dC, as described in 'Materials and methods'. RT–PCR was performed using specific primers for integrin 4. -Actin (actin) was used as a loading control. (b) AGS was treated with 1, 5, or 10 M of 5-Aza-dC for 4 days. (c) AGS cell line was treated with DMSO, 5 M 5-Aza-dC (Aza), 50 nM TSA (TSA 50), 100 nM (TSA 100), or 5 M 5-Aza-dC followed by 50 nM (Aza+TSA 50) or 100 nM TSA (Aza+TSA 100), as described in 'Materials and methods'. Water (DW) was used as negative control for the PCR reactions. (d) Relative integrin 4 gene expression by quantitative real-time PCR. Expression levels were normalized to the human GAPDH control gene, an endogenous control. All expression levels are shown relative to the 'Aza' sample (set to a value of 1.0). Data represents one of two separate experiments with similar results

Full figure and legend (159K)

To ascertain whether the activities of both DNA methyltransferase and HDAC play a role in the loss of integrin 4 expression, a dual treatment strategy was adopted in the AGS cell line. Treatment of AGS with 5-Aza-dC alone weakly induced integrin 4 mRNA expression, but combined treatment with trichostatin A (TSA) and 5-Aza-dC synergistically increased this gene expression (Figure 1c). However, treatment with TSA alone had no effect on integrin 4 gene expression. This synergistic effect of 5-aza and TSA was further validated by quantitative real-time PCR (Figure 1d).

Methylation status of the integrin 4 gene in gastric cancer cell lines

Structurally, the 725-bp fragment around the translation start codon of the integrin 4 gene (GenBank Accession no. NT_005265) has a high G+C content (70.1%) and CpG : GpC ratio of 0.84, and thus satisfies the criteria for a CpG island (Figure 2a) (Gardiner-Garden and Frommer, 1987). To explore the potential role of CpG island methylation on the transcriptional silencing of the integrin 4 gene, we checked the methylation status of the gene in gastric cancer cell lines by methylation-specific PCR (MSP), which allows methylated and unmethylated alleles to be specifically amplified after chemically modifying the DNA with sodium bisulfite (Herman et al., 1996a, 1996b). As shown in Figure 2b, SNU-484, expressing integrin 4 mRNA, showed no methylation at the integrin 4 CpG island, whereas the other eight nonexpressing cell lines contained either completely (SNU-1, -5, -601, -638, -719, AGS and MKN-28) or partially methylated CpG sites (SNU-668).

Figure 2.

Methylation status of the integrin 4 CpG island region in gastric cancer cell lines. (a) A map of the CpG island of the integrin 4 gene, which spans from +697 to +1421 with respect to the transcription start site (+1): vertical bars, the locations of the CpG sites. The regions of MSP (MSP) and bisulfite genomic sequencing (BS#1 and BS#2) are shown. (b) Bisulfite-modified DNA derived from gastric cancer cell lines was amplified with primers specific for unmethylated (U) or methylated (M) DNA. No hypermethylation was found in Jurkat, an integrin 4-positive cell line. The 'unmethylated' and 'methylated' products were of 193 and 186 bp, respectively, because of differences in the primer lengths

Full figure and legend (84K)

We examined the precise methylation statuses of the integrin 4 CpG islands by bisulfite genomic sequencing (Figure 3). As shown in Figure 2a, the integrin 4 CpG island was divided into two regions by the PCR primer sets used to amplify the bisulfite-modified genomic DNA. SNU-484 was, as expected, found to be hypomethylated at its integrin 4 CpG island. However, all of the 66 CpG sites examined were completely methylated in SNU-1, -601 and AGS. Consistent with the MSP results (Figure 2b), bisulfite genomic sequencing assays also showed that the integrin 4 CpG island was half-methylated in SNU-668. Taken together, these findings strongly suggest that the methylation status of the integrin 4 CpG island is closely correlated with the expression of the integrin 4 gene in gastric cancer cell lines.

Figure 3.

Bisulfite genomic sequencing data of the integrin 4 CpG island in gastric carcinoma cell lines. The integrin 4 CpG island was divided into two regions (BS#1 and BS#2), as defined by the specific primer sets used for the PCR amplification. The methylation statuses of the 66 CpG sites of integrin 4-positive cells (SNU-484) and of the integrin 4-negative cells (SNU-1, SNU-601, SNU-668 and AGS) were compared. Each circle indicates a CpG site and each line of circles represents the analysis of a single cloned allele. Open and closed circles represent unmethylated and methylated CpG sites

Full figure and legend (324K)

Methylation of the integrin 4 gene in primary gastric tissues

Next, we examined whether the integrin 4 gene is also methylated in primary gastric cancer tissues (Figure 4). MSP analysis showed that 39 of 46 (84.7%) primary gastric cancer tissues were hypermethylated at the integrin 4 CpG island (summarized in Table 1). Compared with cancer tissue, the methylation status of the corresponding noncancerous gastric tissue (12 of 46; 26.1%) was significantly lower. Although the normal tissue counterparts of primary gastric tissues were also sequenced, we were unable to find any aberrant hypermethylation in the CpG island (Figure 4b). These findings demonstrate that the methylation of the integrin 4 CpG island is not an artefact of the in vitro experimental system, but rather a relatively common and cancer-related event in gastric carcinogenesis. Moreover, we correlated integrin 4 methylation status in cancer tissues with the clinicopathological data (Tables 1 and 2). However, aberrant methylation in primary gastric cancer tissue was not found to be significantly correlated with gender, age, histological grade, T stage or N stage.

Figure 4.

Analysis of integrin 4 methylation in primary gastric cancer tissues. (a) Representative MSP analysis of the integrin 4 CpG island in primary tissues. Parallel amplification reactions were performed using primers specific for unmethylated (U) or methylated (M) DNA. (b) The methylation status of integrin 4 was determined by bisulfite sequencing in cancer tissue samples and corresponding normal tissue samples. T, cancer tissue DNA; N, noncancerous tissue DNA

Full figure and legend (580K)

Table 1 - Integrin 4 methylation in primary tissues.

Full table

Table 2 - Integrin 4 methylation status in primary cancer tissue.

Full table

Inhibition of invasion by expressing integrin 4

To understand the biological meaning of the methylation-based loss of integrin 4 expression in gastric carcinoma, we transfected integrin 4 cDNA into AGS cells without basal integrin 4 expression by methylation and produced a stable cell line expressing integrin 4 (denoted by AGS-4) (Figures 5a and b). Since the loss of integrin 4 expression has been associated with cellular invasiveness in several cancers (Qian et al., 1994; Holzmann et al., 1998; Zhu et al., 2001), we compared the invasive ability of AGS (parental cells), AGS-c (transfected with pcDNA3.1) and AGS-4 using a matrigel invasive assay. The results presented in Figure 5c show that AGS-4 cells had 80% less invasive ability than the parental AGS and negative-control AGS-c cells. These findings suggest that integrin 4 expression can reduce invasion and control metastasis in gastric cancer cells.

Figure 5.

Effects of integrin 4 expression on the invasiveness of gastric cancer cell lines (a) RT–PCR analysis of integrin 4 in AGS control-transfectants (AGS-c) and in AGS-integrin 4 stable cell lines (AGS-4). RNAs from Jurkat and HeLa cells were used as positive and negative controls, respectively. (b) FACS analysis of integrin 4 protein on AGS and AGS-a4 cells. Integrin 4 was expressed more so on AGS-4 than on AGS. (c) Invasiveness of AGS, AGS-c and AGS-4 cells by the matrigel assay. After incubating for 48 h at 37°C, invasive cells were stained with crystal violet and counted. The invasiveness of the AGS-c cells was considered to be 100 for normalization purposes

Full figure and legend (176K)

Discussion

It is known that the loss of integrin 4 in melanoma cells and the allelic inactivation of the integrin 4 gene in fibrosarcoma cells is associated with the acquisition of a metastatic potential, which implies that the expression of integrin 4 is involved in metastasis (Qian et al., 1994; Lobb and Hemler (1994); Garofalo et al., 1995; Gosslar et al., 1996; Holzmann et al., 1998; Sato et al., 1998; Felding-Habermann (2003); Zhu et al., 2001). However, the molecular mechanism that inactivates or downregulates integrin 4 expression in human cancer is not known. The aberrant methylation of several genes, such as, TIMP-3, E-cadherin and beta-catenin, is known to play an important role in metastasis (Yoshiura et al., 1995; Bachman et al., 1999; Ebert et al., 2003). More recently, several reports have linked epigenetic mechanisms with the inactivation of members of the integrin family. DNA methylation of integrin 4 was identified in a mouse liver tumor by restriction landmark genomic scanning for methylation (RLGS-M). In addition, DNA methylation and the chromatin condensation of integrin L were detected in fibroblasts, and were suggested to contribute to the tissue-specific expression of integrin L (Akama et al., 1997; Lu et al., 2002).

Previously, we reported that DNA methylation is one of the predominant mechanisms for inactivating various genes, like, TGF- type I receptor, TIMP-3, p16, COX-2 and DLC-1, during gastric carcinogenesis (Kang et al., 1999; Kang et al., 2000; Song et al., 2000, 2001; Kim et al., 2003). These findings led us to investigate whether integrin 4 expression could be regulated by DNA methylation, because integrin 4 expression is very low in our gastric carcinoma cells (Figure 1a). In the present study, we observed a loss of integrin 4 mRNA expression in eight of the nine gastric cancer cell lines examined. Moreover, MSP and bisulfite genomic sequencing showed that eight of the nine gastric cell lines were hypermethylated at the CpG island of integrin 4 (Figure 2b). In addition, treating gastric cancer cell lines, which contained methylated alleles, with 5-Aza-dC restored integrin 4 mRNA expression, demonstrating that DNA methylation is directly involved in the transcriptional silencing of the integrin 4 gene. In addition, we found that the methylation frequency of integrin 4 in cell lines (77.7%) and in primary tissues (87.4%) is very high, indicating that integrin 4 methylation is common feature in gastric cancer.

Interestingly, the methylation of integrin 4 was also observed in the corresponding gastric normal tissues (26.1%). In recent years, an accumulating number of studies have reported that DNA methylation is present in normal tissues (Issa et al., 1994; Toyooka et al., 2002; Kuroki et al., 2003; Li et al., 2003). For example, APC methylation was observed in a majority of normal gastric mucosa (97.5%) tissue samples, and the aberrant methylation of p15, p16 and E-cadherin was commonly detected in adjacent normal gastric tissues (Tsuchiya et al., 2000; Leung et al., 2001). A recent study, involving an investigation of the methylation of 12 genes during multistep gastric carcinogenesis, showed that hypermethylation commenced at an early stage, and that gene methylation appeared to follow a chronological order during the carcinogenesis (Kang et al., 2003). These results imply the existence of precancerous lesions in corresponding normal tissues, and that transcription silencing by DNA methylation may be established during an early stage in multistep process of carcinogenesis.

The detachment of cells from a primary tumor is an important aspect of metastasis (Stetler-Stevenson et al., 1993). Thus, adhesion molecules like integrins, may affect tumor progression and metastasis at many stages (Felding-Habermann, 2003). It was reported that the restored expression of integrin 4 suppressed metastasis in lymphoma and melanoma, and it was suggested that integrin 4 may affect tumor metastasis at many stages (Qian et al., 1994; Gosslar et al., 1996; Zhu et al., 2001). In addition, integrin 4 was found to reduce the secretion of matrix metalloproteinase, which inhibits the invasiveness of tumor cells (Huhtala et al., 1995). In agreement with these results, our finding that the overexpression of integrin 4 reduces the invasive ability of gastric cancer cells supports the hypothesis that the loss of integrin 4 expression may be involved in metastasis.

Recent studies have indicated the existence of a mechanistic linkage between DNA methylation and chromatin alteration, as mediated by, DNMTs, methyl-CpG-binding proteins and HDAC complex (Nan et al., 1998; Burgers et al., 2002). Our results show that combined treatment with 5-Aza-dC and TSA synergistically upregulated integrin 4 mRNA expression more so than 5-Aza-dC alone in AGS cells. This result suggests that methylation plays a dominant role over histone deacetylation in DNA methylation-induced integrin 4 silencing.

In summary, our findings demonstrate that epigenetic alterations of integrin 4 CpG island occur in gastric carcinoma, and that is associated with the transcriptional silencing of the integrin 4 gene. We also show that integrin 4 upregulation may reduce the invasive ability of gastric cancer cells. Taken together, our results suggest that the loss of integrin 4 by DNA methylation may play an important role in gastric carcinogenesis.

Materials and methods

Cell line and tissue sample

Gastric cancer cell lines (SNU-1, -5, -484, -601, -638, -668, and -719) were obtained from the Korean Cell Line Bank (Seoul, Korea). All cell lines were maintained in RPMI1640 supplemented with 10% fetal bovine serum (Hyclone Laboratories, Inc., Logan, UT, USA) and gentamicin (10 /ml) at 37°C in a 5% CO₂ humidified atmosphere. Tumors and the corresponding normal tissues were obtained from 46 patients with primary gastric carcinoma, during surgery at the Seoul National University Hospital, Seoul, Korea. After surgical removal, the tissues were frozen immediately in liquid nitrogen and stored until required.

RT–PCR

Total RNA was isolated from 10⁷ to 10⁸ cultured cells using a TRI reagent kit (Molecular Research Center, Cincinnati, OH, USA). cDNA was synthesized from 1 g of total RNA using MMLV Reverse Transcriptase (Invitrogen) with random hexamers. Integrin 4 cDNA was amplified by PCR using the sense primer 5'-GAA TAG CTC CGT GTT ATC AA-3' and antisense primer 5'-TAT ATG CCT TAC CAA TCT GC-3'. Reactions were performed in 20 lvolumes under the following conditions: 94°C for 5 min; then 35 cycles of 94°C for 30 s, 53°C for 30 s, and 72°C for 30 s; and finally 10 min at 72°C. -Actin was used as an internal control to estimate the efficiency of the cDNA synthesis in each cell line (Jong et al., 1999).

Quantitative real-time PCR was performed using commercially available Assay-on-demand probe-primer sets (Assay ID: Hs00168433_m1, Applied Biosystems), according to the manufacturer's guidelines (comparative C_T method). We used ABI PRISM^®7900HT Sequence Detection System for PCR with 384-well plate.

Methylation-specific PCR and bisulfite genomic sequencing

Bisulfite modification was performed using 'one day MSP Kit' (In2gen, Korea), according to the manufacture's instructions. The following primer sets were used; for methylated DNA, MF_4 (5'-TAG AGT TAT TTC GCG TTT TGC G-3') and MR_4 (5'-CTT CGA ATA CTC GCG CTA CTT-3'), and for unmethylated DNA, UF_4 (5'-GTT TAG AGT TAT TTT GTG TTT TGT G-3') and UR_4 (5'-AAA ACT TCA AAT ACT CAC ACT ACT-3'). For bisulfite genomic sequencing, of modified DNA by PCR, we used the following primer sets: BS1F (5'-TCT TAC TAA ACC CAA AAC CAT C-3') and BS1R (5'-AAG GAG AGA GGG AAG AGG A-3'); BS2F (5'-TCC TCT TCC CTC TCT CCT T-3') and BS2R (5'-GTT GTG GGG GTT TTG GTA AA-3'). PCR products were cloned into TA cloning vectors (Invitrogen) and sequenced.

5-Aza-DC and trichostatin (TSA) treatment

Cells were seeded at a density of 1 10⁶ cells/100 mm dish and treated 24 h later with varying concentrations of 5-Aza-dC (Sigma) for 4 days with media changes at 2-day intervals. For the synergistic study, cells were first incubated with 5 M 5-Aza-dC for 48 h at 37°C and then 50 or 100 nM TSA (Sigma) was added for an additional 16 h.

Plasmid and transfection

pCDNA3.1(-) vector encoding human integrin 4 cDNA was a gift from Ginsburg (The Scripps Research Institute, CA, USA). AGS cells were transfected with the plasmid DNA using Lipofamine™ 2000 (Invitrogen), according to the manufacturer's instructions, and then selected with 0.5 mg/ml G418 (Invitrogen) for 3–4 weeks. cDNA expression was confirmed by RT–PCR and flow cytometry.

Flow cytometry

Cell (5 10⁵ per sample) were washed three times with PBS containing 1% BSA, and incubated for 1 h on ice with primary integrin 4 antibody (HP2/1, Chemicon). The cells were then washed and incubated with FITC-conjugated rabbit anti-mouse IgG (The Jackson Laboratory) for 1 h on ice. Fluorescence was measured using a FACSCalibur flow cytometer.

Invasion assay

The invasion assay was carried out in the 24-well BD BioCoat™ Matrigel™ Invasion Camber (Becton Dickinson). After the matrigel inserts had been preincubated with serum-free medium for 2 h at 37°C, the lower chamber was filled with complete medium. Cells (2.5 10⁴ cells/well) were then seeded into the upper compartment in serum-free medium, incubated for 48 h in a tissue culture incubator, washed with PBS and fixed with 4% paraformaldehyde. The filter was stained with 0.1% crystal violet and the cells in the upper chamber were wiped off. Migrated cells were counted in four separated fields per well. Each data point shown represents the mean of three wells.

References

1. Akama TO, Okazaki Y, Ito M, Okuizumi H, Konno H, Muramatsu M, Plass C, Held WA and Hayashizaki Y. (1997). Cancer Res., 57, 3294–3299. | PubMed | ISI | ChemPort |
2. Bachman KE, Herman JG, Corn PG, Merlo A, Costello JF, Cavenee WK, Baylin SB and Graff JR. (1999). Cancer Res., 59, 798–802. | PubMed | ISI | ChemPort |
3. Burgers WA, Fuks F and Kouzarides T. (2002). Trends Genet., 18, 275–277. | Article | PubMed | ISI | ChemPort |
4. Chan BM, Kassner PD, Schiro JA, Byers HR, Kupper TS and Hemler ME. (1992). Cell, 68, 1051–1060. | Article | PubMed | ISI | ChemPort |
5. Ebert MP, Yu J, Hoffmann J, Rocco A, Rocken C, Kahmann S, Muller O, Korc M, Sung JJ and Malfertheiner P. (2003). J. Clin. Oncol., 21, 1708–1714. | Article | PubMed | ISI | ChemPort |
6. Felding-Habermann B. (2003). Clin. Exp. Metast., 20, 203–213. | Article |
7. Gardiner-Garden M and Frommer M. (1987). J. Mol. Biol., 196, 261–282. | Article | PubMed | ChemPort |
8. Garofalo A, Chirivi RG, Foglieni C, Pigott R, Mortarini R, Martin-Padura I, Anichini A, Gearing AJ, Sanchez-Madrid F and Dejana E. (1995). Cancer Res., 55, 414–419. | PubMed | ISI | ChemPort |
9. Gosslar U, Jonas P, Luz A, Lifka A, Naor D, Hamann A and Holzmann B. (1996). Proc. Natl. Acad. Sci. USA, 93, 4821–4826. | Article | PubMed |
10. Hamann A, Andrew DP, Jablonski-Westrich D, Holzmann B and Butcher EC. (1994). J. Immunol., 152, 3282–3293. | PubMed | ISI | ChemPort |
11. Herman JG, Graff JR, Myohanen S, Nelkin BD and Baylin SB. (1996a). Proc. Natl.Acad. Sci. USA, 93, 9821–9826.
12. Herman JG, Jen J, Merlo A and Baylin SB. (1996b). Cancer Res., 56, 722–727. | PubMed | ISI | ChemPort |
13. Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa JP, Markowitz S, Willson JK, Hamilton SR, Kinzler KW, Kane MF, Kolodner RD, Vogelstein B, Kunkel TA and Baylin SB. (1998). Proc. Natl. Acad. Sci. USA, 95, 6870–6875. | Article | PubMed | ChemPort |
14. Holzmann B, Gosslar U and Bittner M. (1998). Curr. Top Microbiol. Immunol., 231, 125–141. | PubMed | ISI | ChemPort |
15. Huhtala P, Humphries MJ, McCarthy JB, Tremble PM, Werb Z and Damsky CH. (1995). J. Cell Biol., 129, 867–879. | Article | PubMed | ISI | ChemPort |
16. Hynes RO and Lander AD. (1992). Cell, 68, 303–322. | Article | PubMed | ISI | ChemPort |
17. Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE and Baylin SB. (1994). Nat. Genet., 7, 536–540. | Article | PubMed | ISI | ChemPort |
18. Jones PA and Taylor SM. (1980). Cell, 20, 85–93. | Article | PubMed | ISI | ChemPort |
19. Jong HS, Park YI, Kim S, Sohn JH, Kang SH, Song SH, Bang YJ and Kim NK. (1999). Cancer, 86, 559–565. | Article | PubMed | ISI | ChemPort |
20. Kang GH, Lee S, Kim JS and Jung HY. (2003). Lab. Invest., 83, 519–526. | Article | PubMed | ISI | ChemPort |
21. Kang SH, Bang YJ, Im YH, Yang HK, Lee DA, Lee HY, Lee HS, Kim NK and Kim SJ. (1999). Oncogene, 18, 7280–7286. | Article | PubMed | ChemPort |
22. Kang SH, Choi HH, Kim SG, Jong HS, Kim NK, Kim SJ and Bang YJ. (2000). Int. J. Cancer, 86, 632–635. | Article | PubMed | ISI | ChemPort |
23. Kim TY, Jong HS, Song SH, Dimtchev A, Jeong SJ, Lee JW, Kim TY, Kim NK, Jung M and Bang YJ. (2003). Oncogene, 22, 3943–3951. | Article | PubMed | ChemPort |
24. Kuroki T, Trapasso F, Yendamuri S, Matsuyama A, Alder H, Williams NN, Kaiser LR and Croce CM. (2003). Cancer Res., 63, 3352–3355. | PubMed | ISI | ChemPort |
25. Leung WK, Yu J, Ng EK, To KF, Ma PK, Lee TL, Go MY, Chung SC and Sung JJ. (2001). Cancer, 91, 2294–2301. | Article | PubMed | ISI | ChemPort |
26. Li H, Myeroff L, Smiraglia D, Romero MF, Pretlow TP, Kasturi L, Lutterbaugh J, Rerko RM, Casey G, Issa JP, Willis J, Willson JK, Plass C and Markowitz SD. (2003). Proc. Natl. Acad. Sci. USA, 100, 8412–8417. | Article | PubMed | ChemPort |
27. Lobb RR and Hemler ME. (1994). J. Clin. Invest., 94, 1722–1728. | PubMed | ISI | ChemPort |
28. Lu Q, Ray D, Gutsch D and Richardson B. (2002). Blood, 99, 4503–4508. | Article | PubMed |
29. Michetti M, Kelly CP, Kraehenbuhl JP, Bouzourene H and Michetti P. (2000). Gastroenterology, 119, 109–118. | PubMed |
30. Miranti CK and Brugge JS. (2002). Nat. Cell Biol., 4, E83–E90. | Article | PubMed | ISI | ChemPort |
31. Nan X, Ng HH, Johnson CA, Laherty CD, Turner BM, Eisenman RN and Bird A. (1998). Nature, 393, 386–389. | Article | PubMed | ISI | ChemPort |
32. Qian F, Vaux DL and Weissman IL. (1994). Cell, 77, 335–347. | Article | PubMed | ISI | ChemPort |
33. Rincon J, Prieto J and Patarroyo M. (1992). Int. J. Cancer, 51, 452–458. | PubMed |
34. Sato S, Tunashima K, Mafune Y, Ohtsuka H, Koide N, Endo Y, Saito Y, Asakura H and Kominami R. (1998). Oncogene, 17, 105–113. | Article | PubMed |
35. Song SH, Jong HS, Choi HH, Inoue H, Tanabe T, Kim NK and Bang YJ. (2001). Cancer Res., 61, 4628–4635. | PubMed | ISI | ChemPort |
36. Song SH, Jong HS, Choi HH, Kang SH, Ryu MH, Kim NK, Kim WH and Bang YJ. (2000). Int. J. Cancer, 87, 236–240. | Article | PubMed | ISI | ChemPort |
37. Stetler-Stevenson WG, Aznavoorian S and Liotta LA. (1993). Annu. Rev. Cell Biol., 9, 541–573. | Article | PubMed | ChemPort |
38. Stirzaker C, Millar DS, Paul CL, Warnecke PM, Harrison J, Vincent PC, Frommer M and Clark SJ. (1997). Cancer Res., 57, 2229–2237. | PubMed | ISI | ChemPort |
39. Tamkun JW, DeSimone DW, Fonda D, Patel RS, Buck C, Horwitz AF and Hynes RO. (1986). Cell, 46, 271–282. | Article | PubMed | ISI | ChemPort |
40. Toyooka S, Toyooka KO, Harada K, Miyajima K, Makarla P, Sathyanarayana UG, Yin J, Sato F, Shivapurkar N, Meltzer SJ and Gazdar AF. (2002). Cancer Res., 62, 3382–3386. | PubMed | ISI | ChemPort |
41. Tsuchiya T, Tamura G, Sato K, Endoh Y, Sakata K, Jin Z, Motoyama T, Usuba O, Kimura W, Nishizuka S, Wilson KT, James SP, Yin J, Fleisher AS, Zou T, Silverberg SG, Kong D and Meltzer SJ. (2000). Oncogene, 19, 3642–3646. | Article | PubMed | ISI | ChemPort |
42. Yoshiura K, Kanai Y, Ochiai A, Shimoyama Y, Sugimura T and Hirohashi S. (1995). Proc. Natl. Acad. Sci. USA, 92, 7416–7419. | Article | PubMed | ChemPort |
43. Zhu Z, Sanchez-Sweatman O, Huang X, Wiltrout R, Khokha R, Zhao Q and Gorelik E. (2001). Cancer Res., 61, 1707–1716. | PubMed | ISI | ChemPort |

Acknowledgements

This work was supported in part by grants from the Ministry of Science & Technology of Korea through the National Research Laboratory Program for Cancer Epigenetics and by 2003 BK21 Project for Medicine, Dentistry and Pharmacy. We thank other lab members for helpful discussions and Tae Young Kim for quantitative real-time PCR.