| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Cancer Research Institute [S. H. S., H-S. J., H. H. C., Y-J. B.], and Departments of Tumor Biology [S. H. S., H-S. J., Y-J. B.] and Internal Medicine [N. K. K., Y-J. B.], Seoul National University College of Medicine, Seoul 110-799, Korea, and Department of Pharmacology, National Cardiovascular Center Research Institute, Osaka 565-8565, Japan [H. I., T. T.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
716 knockout mice (9)
. Moreover, in gastric cancer, Cox-2 overexpression is only limited to cancer tissues compared with the accompanying normal mucosa (10
, 11)
, and the intensity of Cox-2 expression is correlated with the metastatic involvement of the lymph nodes (12)
. However, the possibility that Cox-2 has distinct biological functions on cell cycle progression has been suggested recently (13) . Overexpression of wild-type Cox-2 in human vascular endothelial cells suppresses cell cycle progression at the S-phase, with a concomitant increase in G0/G1 population. The same results were obtained by the introduction of two Cox-2 mutant constructs, possessing only peroxidase activity but without cyclooxygenase activity (14) , suggesting Cox-2-mediated cell cycle arrest by a PG-independent mechanism. Therefore, the biological effect of Cox-2 overexpression on cell cycle progression appears to be different in each cell type, and the mechanism by which such differences occurs remains to be elucidated.
Interestingly, it has been demonstrated that Cox-2 overexpression is less frequent in gastric carcinomas with MSI than in those without MSI (15) . This is consistent with a previous finding (16) of a significant reduction in Cox-2 expression levels in colorectal cancers with defective MMR. MMR deficiency is strongly associated with hyper-methylation of the hMLH1 gene, and this type of epigenetic alteration is accompanied by a down-regulation of hMLH1 expression in gastric cancer (17 , 18) and colorectal cancer (19 , 20) . This type of epigenetic gene inactivation is not limited to hMLH1. DNA methylation is one of the predominant mechanisms for inactivating various genes during the tumorigenesis of gastric carcinoma (21, 22, 23, 24, 25, 26) . These findings led us to investigate whether Cox-2 expression could be regulated by DNA methylation. Very recently, Toyota et al. (27) suggested that Cox-2 expression is inhibited by the aberrant methylation of the exon 1-coding region of the Cox-2 gene in a subset of colorectal tumors. However, we and others (21 , 28) have shown that the methylation of a small number of CpG sites in the promoter region, rather than in the coding region, could down-regulate the promoter activity of p16INK4a in gastric and bladder carcinoma. In this regard and for an improved understanding of the transcriptional regulation of the Cox-2 gene, it is also important to determine the critical region that down-regulates Cox-2 expression by DNA methylation.
In this study, we show that the transcriptional silencing of Cox-2 is strongly related with the methylation status of the 5' CpG island of the Cox-2 gene in human gastric carcinoma cell lines. Moreover, functionally active Cox-2 is reinduced by demethylation of the Cox-2 promoter in gastric carcinoma cell lines.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmids.
Cox-2 promoter constructs (-1432/+59, -327/+59, -220/+59, -124/+59, -52/+59, KBM, ILM, CRM, KBM + ILM, and Triple M) have been described previously (4
, 6
, 31)
. Human Cox-2 cDNA (PHS23'UTR; Ref. 32
) was generously provided by Dr. Stephen M. Prescott (University of Utah, Salt Lake City, UT). Human Cox-1 cDNA (33)
was a kind gift from Dr. Timothy Hla (University of Connecticut Health Center, Farmington, CN). The Cox-2-S516M and S516Q mutant constructs (14)
were the kind gifts of Dr. William Smith (Michigan State University, East Lansing, MI). pGL2-Basic control vector and pSV-ß galactosidase control vector were purchased from Promega.
RNA Extraction and Northern Blot Analysis.
Using a TRI REAGENT kit (Molecular Research Center, Cincinnati, OH) and an Oligotex mRNA kit (Qiagen, Valencia, CA), total or poly(A) RNA was isolated from 107 to 108 cultured cells according to the manufacturer’s instructions. Total cellular RNA (20 µg) or 2 µg of poly(A) RNA were electrophoresed on a 1% agarose gel containing formaldehyde and transferred to nylon membranes (Schleicher & Schuell, Keene, NH) by capillary blotting. The blots were hybridized using a Cox-2 and Cox-1 cDNA for RNA expression, as described previously (26)
. ß-actin cDNA signals were used as an internal control to determine the integrity of RNA and the equality of lane loading (30)
.
Western Blot Analysis.
Cells were washed with ice-cold PBS and suspended in an extraction buffer [20 mM Tris-Cl (pH 7.4), 100 mM NaCl, 1% NP40, 0.5% sodium deoxycholate, 5 mM MgCl 2, 0.1 mM phenylmethylsulfonyl fluoride, 0.1 mM pepstatin A, 0.1 mM antipain, 0.1 mM chymostatin, 0.2 mM leupeptin, 10 µg/ml aprotinin, 0.5 mg/ml soybean trypsin inhibitor, and 1 mM benzamidine] on ice for 15 min. Lysates were cleared by centrifugation at 13,000 rpm for 20 min. Equal amounts of cell extracts (100 µg) were resolved on 10% SDS-polyacrylamide denaturing gels, transferred onto nitrocellulose membranes, and probed with antihuman Cox-1 and Cox-2 antibodies purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-á-tubulin antibody (Sigma Chemical Co., St. Louis, MO) was used as a loading control. Detection was performed using an enhanced chemiluminescence system (Amersham Pharmacia Biotech).
Southern Blot Analysis.
Genomic DNA was prepared using a standard SDS and proteinase K protocol (30)
. Samples of genomic DNAs (10 µg) were digested with excess restriction enzymes (New England Biolabs, Beverly, MA), as described in "Results." Double digestions were performed sequentially so that each restriction enzyme was in optimal incubation buffer, with a precipitation step in between (21)
. The digested DNA fragments were separated by electrophoresis on 1% agarose gels and blotted onto a nylon membrane. The Cox-2 promoter probe, spanning -1432 to +59 with respect to the transcription initiation site (GenBank accession nos. D28235 and AF044206), was prepared by digestion of phPES2(-1432/+59) plasmid (31)
with KpnI and HindIII. The gel-purified insert was labeled with [
-32P]dCTP by random primer extension. After hybridization, the membranes were washed in a series of solutions as described (34)
and exposed to X-ray film at -70°C.
Bisulfite Modification and Sequencing Analysis.
A total of 2 µg of genomic DNA obtained from gastric carcinoma cells was modified by sodium bisulfite (35)
. Primary and secondary PCR reactions were carried out in 50-µl reaction mixtures under the following conditions: an initial denaturation step at 95°C for 5 min, followed by 35 cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 1 min, and a final extension step at 72°C for 10 min. Primers Cox2-A1 (5'-TGT ATA TTG AAG GTA GTT ATT TTA T-3') and Cox2-d2 (5'-ACC AAA TAC TCA CCT ATA TAA CT-3') were used to generate the primary PCR product for regions A to D. To obtain products for sequencing, a secondary round of PCR was performed using this primary PCR product with specific PCR primer sets for each region. Primers Cox2-A1 and Cox2-a2 (5'-AAA CAC TTA ACT TCC TCT CCA A-3') were used for secondary amplification of region A. Primers Cox2-B1 (5'-TTG GAG AGG AAG TTA AGT GTT T-3') and Cox2-b2 (5'-ATC CCC ACT CTC CTA TCT AAT-3') were used for region B. Primers Cox2-C1 (5'-ATT AGA TAG GAG AGT GGG GAT-3') and Cox2-c2 [5'-TCT AAA AAC (A/G)TC TAA CTA TAA AAC T-3'] were used for region C, and primers Cox2-D1 [5'-AAG TGA G(C/T)G TTA GGA GTA (C/T)GT T-3'] and Cox2-d2 were used for region D. Secondary PCR products were gel-purified and cloned into a TA cloning vector (Invitrogen, Carlsbad, CA). Individual plasmid molecules were then sequenced using the ABI PRISM 377 DNA sequencer (Perkin-Elmer, Foster City, CA; Ref. 21
).
5-Aza-CdR Treatments.
Cells were seeded at a density of 1 x 106 cells/100-mm dish and allowed to attach over a 24-h period. 5-aza-CdR (Sigma Chemical Co.) was then added to a final concentration of 1, 5, or 10 µM, and the cells were allowed to grow for the times indicated in "Results." The same concentration of DMSO was also used as a control for nonspecific solvent effects on cells (21)
. At the end of the treatment period, the medium was removed, and the RNA and protein were extracted for Northern and Western analysis.
PGE2 Production and Determination of Cox Activity.
Cells were treated for 3 days with DMSO (0.1%) or 5-aza-CdR (10 µM) in complete growth medium containing 10% FBS. At the end of the treatment period, the culture media were collected to determine the amounts of PGE2 secreted by the cells. For the Cox-2 catalytic activity measurement, cells were washed twice, and fresh mediums (without FBS) containing 30 µM of arachidonic acid (Sigma Chemical Co.) were added for 30 min at 37°C. The media were then collected for PGE2 analysis. To determine whether 5-aza-CdR-induced alterations in PGE2 concentration are mediated by a change in Cox-2 activity, NS-398 (50 µM; Cayman Chemical, Ann Arbor, MI) was added to cultures 30 min before the Cox-2 activity was determined. The amounts of PGE2 released by these cells were measured using a PGE2 enzyme immunoassay kit (Cayman Chemical) according to the manufacturer’s instructions. The production of PGE2 was normalized according to the number of adherent cells present in the culture at the time of sampling. Results are expressed as pg of PGE2/106 cells ± SD.
Transient Transfection and Luciferase Assays.
SNU-601 and HCT-116 were seeded at a density of 3 x 105/well in a 6-well dish and grown to 60–70% confluence in complete growth media containing 10% FBS. For each well, 2 µg of plasmid DNA and 0.5 µg of pSV-ß galactosidase control vector were cotransfected into cells with Lipofectin (Life Technologies, Inc.) according to the manufacturer’s instructions. After 6 h, complete medium (with 10% FBS) was added, and the cells were further incubated for 18 h. Cells were then serum starved by replacing the growth mediums with serum-free medium for an additional 24 h. Luciferase activities were measured using a TR717 Microplate Luminometer (Tropix, Inc.) with a Bioluminescent Reporter Gene Assay System (Tropix, Inc.) according to the manufacturer’s instructions. A pGL2-Basic control vector without insert was used as a negative control in the transfection experiments. Luciferase activities were normalized using ß-galactosidase.
In Vitro Methylation Assay.
In vitro methylation assays were carried out according to the methods described by Robertson and Ambinder (36)
and Kudo (37)
. Briefly, Cox-2 promoter constructs (-1432/+59, -327/+59, and -220/+59) were incubated overnight with three units of SssI (CpG) methylase (New England Biolabs)/µg of plasmid DNA in the presence (methylated) or absence (mock-methylated) of 1 mM S-adenosylmethionine, as recommended by the manufacturer. After phenol extraction and ethanol precipitation, equal amounts (2 µg) of methylated and mock-methylated reporter constructs were transiently transfected into SNU-601, and luciferase activities were examined as described above. Individual reactions were monitored by digestion with HpaII or HhaI restriction enzymes.
Data Analysis.
Results are representative of at least three independent experiments performed in triplicate and are presented as the means ± SD. Comparisons between groups were made using Student’s paired t test.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Upon the addition of IL-1ß (3
, 5)
, Cox-2 mRNA was also induced after 2 h in SNU-484 and -638 (Fig. 1B)
. But treatment with IL-1ß had no effect on Cox-2 expression in SNU-601. Compared with the variable expression of Cox-2, all of the cell lines expressed relatively equal amounts of Cox-1 mRNA and protein.
|
A restriction map of the Cox-2 CpG island is shown in Fig. 2A
. Genomic DNAs from gastric carcinoma cell lines and HELs were treated with EcoRI. This resulted in a 4.3-kb fragment containing the promoter region of Cox-2, which was then digested with a methylation-sensitive restriction enzyme, namely HpaII, HaeII, or HhaI. The genomic DNA of HEL was digested into a smaller fragment of approximately 1.5 kb, which suggested that the recognition sites for the three methylation-sensitive restriction enzymes are free of methylation (Fig. 2B)
. The digestion patterns obtained from SNU-484 and SNU-668 were consistent with that of HEL. However, three cell lines (SNU-601, -620, and -719) without Cox-2 expression showed resistance to digestion with the three methylation-sensitive restriction enzymes, indicating that heavily methylated CpG-rich regions are present in the Cox-2 CpG island. In the case of the SNU-638 with low Cox-2 expression, a methylation-protected 4.3-kb fragment coexisted with a 1.5-kb band, showing that the methylation status of the Cox-2 CpG island is somewhat heterogeneous in SNU-638. The existences of 2-kb fragments may indicate that some CpG sites located near the downstream of exon 1 are not completely methylated in SNU-601, -620, and -719. These findings strongly suggest that the methylation status of the Cox-2 CpG island is related with the transcriptional silencing of Cox-2 in gastric carcinoma cell lines.
|
, Cox-2 CpG island was divided into four regions (A to D) as defined by the PCR primers sets used to amplify bisulfite-modified genomic DNA. After the amplification of each region, PCR products were cloned, and individual plasmids were sequenced, as described in "Materials and Methods." SNU-484 was found to be hypo-methylated at the proximal 5'-end of the CpG island (64% of 15 CpG sites in region A and 30% of 10 CpG sites in region B), but methylated CpG sites were not detected in regions C and D (Fig. 3)
. However, all of the 51 CpG sites within regions A to D were completely methylated in SNU-601. Interestingly, SNU-638 was heavily methylated in regions C and D (82% of 12 CpG sites and 60% of 14 CpG sites, respectively), which was similar to SNU-601 (90% and 81%, respectively). On the other hand, the methylation status of region B (30%) of SNU-638 was much different from that of SNU-601 (93%). These results may suggest that hypo-methylation in region A may not block transcriptional initiation of Cox-2 and that hyper-methylation of regions C and D is necessary but not sufficient to confer methylation-dependent repression of Cox-2 in the presence of a hypo-methylated region B.
|
. Treatment with 5-aza-CdR had little effect on Cox-2 expression and Cox-2 activity in SNU-484. However, it is noteworthy that the addition of 5-aza-CdR resulted in the superinduction of the mRNA and protein level of Cox-2 in SNU-601. Consistent with elevated Cox-2 expression, PGE2 production and Cox-2 enzymatic activity were also significantly higher after treatment with 5-aza-CdR, as compared with that achieved by DMSO treatment, in SNU-601 (Fig. 4, B and C)
. 5-aza-CdR-induced Cox-2 activity was completely abolished by NS-398 pretreatment, a selective inhibitor of Cox-2 (20)
. Similar to SNU-601, treatment of 5-aza-CdR elevated Cox-2 expression and enzymatic activity in SNU-638 with heterogeneous Cox-2 methylation status. Cox-2 mRNA was also detected after treatment with 5-aza-CdR in SNU-620 and SNU-719 (data not shown). Taken together, these results suggest that the lack of Cox-2 expression is attributable to a block of transcription caused by the methylation of the Cox-2 CpG island and that functionally active Cox-2 can be activated by 5-aza-CdR in SNU-601.
|
. SNU-601 was then monitored at various times after treatment with 5 µM 5-aza-CdR. Cox-2 mRNA and protein were easily detected between 48 and 72 h of treatment (Fig. 5B)
. Under demethylated conditions, the addition of IL-1ß caused a rapid and sustained up-regulation of Cox-2 mRNA, which was maximally increased 2-fold after 2 h and remained elevated for 6 h (Fig. 5C)
. This induction of Cox-2 mRNA was associated with an increase in Cox-2 protein, which significantly increased 4 h after stimulation with IL-1ß and lasted for 6 h. Therefore, demethylation by 5-aza-CdR resulted in the time-dependent reactivation of Cox-2 and restored responsiveness to IL-1ß in previously resistant SNU-601.
|
) into HCT-116 and SNU-601 and incubated the cells in a serum-starved state for 24 h to eliminate any interference by undefined substances within the serum. The relative Cox-2 promoter activities were determined by luciferase assay and were normalized to the ß-galactosidase activity. The results presented in Fig. 6B
show remarkable constitutively active Cox-2 promoter activities exhibited in SNU-601 despite serum starvation. Interestingly, the Cox-2 promoter activities of SNU-601 were greater (approximately 7-fold) than those of the HCT-116.
|
. The Cox-2 promoter activity was highest when the -327/+59 promoter construct was used. As the Cox-2 promoter was shortened, lower basal activities were realized. Thus, the -52/+59 construct exhibited an approximate 95% relative decrease in Cox-2 promoter activity compared with the -327/+59 construct. These results suggest that the constitutive Cox-2 promoter activities were derived using the cis-acting elements located in the Cox-2 promoter region, between nucleotides -327 and -53. In this region, three cis-acting elements (NF-
B, NF-IL6, and CRE) exist, which are known to be involved in cytokine-mediated promoter activity in the human Cox-2 gene (4
, 6
, 31
, 43)
. To precisely define which of these cis-acting elements is involved in the constitutively active Cox-2 promoter activity, transient transfections were performed in SNU-601 using site-specific mutant Cox-2 promoter constructs. As shown in Fig. 6D
, the introduction of ILM (-327/+59 construct in which the NF-IL6 site had been mutagenized) and CRM (mutated at the CRE site) construct significantly decreased basal Cox-2 promoter activity compared with the wild-type -327/+59 construct. However, the KBM construct (mutated at the NF-B site) had no effect on Cox-2 promoter activity. The Cox-2 promoter activity derived from the double mutant construct (KBM + ILM; mutated at both the NF-B and NF-IL6 sites) was similar to that of the ILM construct, demonstrating that the NF-B site is not involved in constitutive Cox-2 promoter activity in SNU-601. Finally, a triple mutant construct (Triple M; mutated at the NF-B, NF-IL6, and CRE sites) completely abolished basal Cox-2 promoter activity. These results suggest that constitutively active Cox-2 promoter activity was caused by using the NF-IL6 and CRE elements existing in the Cox-2 promoter region even without an added stimulus in SNU-601.
In vitro methylation analysis (36
, 37)
was performed to examine whether constitutively active Cox-2 promoter activity was inhibited by the methylation of Cox-2 CpG island. pGL2-Basic control vector (a total 256 CpG sites are located in the vector sequences) and a series of Cox-2 promoter deletion constructs, -1432/+59, -327/+59, and -220/+59 (including 47, 19, and 12 CpG sites, respectively), were methylated with SssI (CpG) methylase and transiently transfected into SNU-601. CpG methylation of the pGL2-Basic control vector reduced the luciferase activity to 50%. However, compared with that of the unmethylated vector, CpG methylation of the Cox-2 promoter entirely abolished its activity (Fig. 6E)
. Taken together, our findings strongly suggest that the transcriptional silencing of Cox-2 expression was caused by hyper-methylation of Cox-2 CpG island in SNU-601.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It has been suggested recently by Toyota et al. (27)
that Cox-2 has an atypical CpG island that begins just after the transcriptional start site and that encompasses 350 bp near the exon 1-coding region of the Cox-2 gene. They also reported that methylation closest to its transcription initiation site is limited to neoplastic tissues and is associated with a loss of Cox-2 expression in colorectal cancer. However, in this study, we have shown that Cox-2 CpG islands are present more extensively than they expected. The Cox-2 promoter region from -590 to +186 (which contains 51 CpG sites with a G + C content of 0.578 and an observed/expected presence of CpG of 0.8) with respect to the transcriptional start site clearly meets the established criteria for a CpG island (Fig. 2A)
. To elucidate the importance of this CpG island, we compared the methylation status of this CpG island with Cox-2 expression in gastric carcinoma cell lines. By Southern and bisulfite-modification sequencing analysis, we have shown that three gastric cells, the Cox-2 expression of which was not observed, display hyper-methylation at this Cox-2 CpG island. The restoration of Cox-2 expression and IL-1ß sensitivity by 5-aza-CdR treatment (40
, 41)
provides a molecular mechanism that may explain the loss of Cox-2 expression in some gastric cell lines with methylation at the Cox-2 CpG island region. In addition, here we show for the first time that the reinduced Cox-2 protein retains its functionally active enzymatic activity, i.e., its cyclooxygenase and peroxidase activities (44)
, and thus metabolizes arachidonic acid into PGs (1)
, indicating that the demethylation of the Cox-2 CpG island can successfully recover its function. The significance of the methylation status of this CpG island is further supported by our observations that a Cox-2 promoter construct, which is inactive in normal cells without exogenous stimulation (4
, 5)
, is constitutively active in SNU-601 even under serum-depleted conditions. Especially, the NF-IL6 and CRE elements may be involved cooperatively in the transcriptional regulation of Cox-2 in SNU-601, even in the absence of any exogenous stimuli. A number of recent reports (45, 46, 47, 48)
also described a pivotal role of these sites in Cox-2 gene expression in a wide range of cells, in response to a variety of stimuli. On the other hand, Cox-2 promoter activity is almost completely abolished by the introduction of CpG-methylated Cox-2 promoter, which is similar to results obtained in investigations of p16INK4a (28)
and p14ARF (49)
. These results strongly suggest that if hyper-methylation of this Cox-2 CpG island did not exist, SNU-601 might have expressed high steady-state levels of Cox-2 mRNA and protein and elevated PGE2 biosynthesis (12)
. However, further evaluation is required to determine whether the relationship between the methylation status of this CpG island and the loss of Cox-2 gene expression is valid in primary gastric carcinoma tissues, because no methylation-positive case was detected among eight cases examined in this study (data not shown).
Other workers and ourselves have demonstrated previously that the de novo methylation of only a small number of CpG sites in the promoter region can down-regulate the promoter activity of p16INK4a in bladder (28)
and gastric carcinomas (21)
. In the case of Cox-2, we were unable to identify this type of methylation "hot spot," which is a critical region capable of inhibiting the transcriptional initiation of Cox-2 by the methylation of only a small number of specific CpG sites. Dense methylation at the 3'-end of the Cox-2 CpG island (region C and D; Fig. 3
) is necessary but not sufficient for the complete inhibition of Cox-2 expression under hypo-methylation conditions at the 5'-end of the Cox-2 CpG island (region B).
What is the functional significance underlying the methylation-mediated transcriptional loss of Cox-2 in gastric carcinoma cells? To the best of our knowledge, this type of epigenetic gene inactivation is only restricted to tumor-suppressor genes during tumorigenesis (reviewed in Refs. 50 , 51 ). Moreover, accumulating evidence indicates that Cox-2 overexpression is associated with gastric cancer (10 , 12) , so that the loss of Cox-2 expression may provide a disadvantage for cancer cell survival, which may in turn detrimentally affect the development of gastric cancer. Interestingly, it was reported recently that the overexpression of Cox-2 is less frequent in colorectal (16) and gastric cancer cells (15) with MSI rather than in cells without MSI. Because there is a strong link between hyper-methylation of the hMLH1 gene and the MSI phenotype in gastric (17 , 18) and colorectal cancer (19 , 20) , this raises the possibility that Cox-2 CpG methylation is associated with the MSI phenotype caused by a defective MMR (52) . However, no significant correlation between Cox-2 methylation and the MSI phenotype was found in the gastric carcinoma cells examined in this study (53) , indicating that MSI may not be involved in the mechanisms responsible for Cox-2 methylation in gastric carcinoma cells. Although Yamamoto et al. (15) found that SNU-638 with the replication error phenotype+ did not express Cox-2 protein, the results of the present study show that low levels of Cox-2 mRNA and protein are expressed in SNU-638. The heterogeneous methylation status of the Cox-2 CpG island in SNU-638 may go some way toward explaining the difference between the above findings.
Recently, Trifan et al. (13)
reported that the overexpression of Cox-2 cDNA or mutant Cox-2 constructs suppressed cell cycle progression at the S-phase, with a concomitant increase in the G0/G1 population in human vascular endothelial cells. This implies that the overexpression of Cox-2 may induce G0/G1 growth arrest by an uncharacterized nonprostanoid-dependent signaling pathway. To elucidate the effects of Cox-2 overexpression on the cell cycle, in this study, SNU-601 was transiently transfected with wild-type Cox-2 cDNA (PHS23'UTR; Ref. 32
) or the two mutant Cox-2 cDNAs (S516M and S516Q, which possess only peroxidase activity, are without cyclooxygenase activity, and, therefore, mimic aspirin-treated Cox-2; Ref. 14
). Cell cycle distributions were then analyzed by DNA staining with propidium iodide followed by fluorescence-activated cell sorting analysis. Neither the wild-type Cox-2 cDNA nor the two mutant types had any effect on cell cycle progression in SNU-601 (data not shown). However, we were unable to exclude the possibility that this lack of responsiveness may result from the genetic and/or epigenetic alternation of other gene(s) that may contribute to the proliferation of SNU-601. Indeed, it is important to point out that the transcriptional initiation of p16INK4a (21
, 26)
, TIMP-3 (22)
, and TGF-ß type I receptor gene (23)
in SNU-601 is also blocked by hyper-methylation of the 5' CpG island in these genes. Thus, the functional significance of the down-regulation of Cox-2 during carcinogenesis merits further elucidation.
In this study, we demonstrate that the transcriptional silencing of Cox-2 is caused by the hyper-methylation of the Cox-2 CpG island and results in the inhibition of the binding of a certain transactivator that essentially uses the NF-IL6 and CRE elements in SNU-601. Although the biological functions associated with the presence of methylation-mediated transcriptional silencing of the Cox-2 gene are not clear, the DNA methylation of the Cox-2 CpG island is a new mechanism for the down-regulation of Cox-2 expression in gastric carcinoma cell lines.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by the Korean Ministry of Health and Welfare (HMP-99-M-03-0001; to Y-J. B.) through the Cancer Research Institute of Seoul National University College of Medicine and by 2001 BK21 Project for Medicine, Dentistry, and Pharmacy (to S. H. S., H-S. J., and Y-J. B.). 
2 To whom requests for reprints should be addressed, at Department of Internal Medicine, Seoul National University College of Medicine, Seoul 110-799, Korea. Phone: 82-2-760-2390; Fax: 82-2-762-9662; E-mail: bangyj{at}plaza.snu.ac.kr
3 The abbreviations used are: Cox, cyclooxygenase; PG, prostaglandin; PGH2, prostaglandin H2; IL, interleukin; NF-B, nuclear factor B; NF-IL6, NF for IL-6 expression site; CRE, cAMP response element; MSI, microsatellite instability; MMR, mismatch repair; 5-aza-CdR, 5-aza-2' deoxycytidine; HEL, human embryonic lung fibroblasts; FBS, fetal bovine serum; poly(A), polyadenylate.
Received 12/29/00. Accepted 4/ 3/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Evidence for post-transcriptional regulation. J. Biol. Chem., 269: 11769-11775, 1994.B site in the induction by lipopolysaccharide and suppression by dexamethasone of cyclooxygenase-2 in U937 cells. Biochem. Biophys. Res. Commun., 244: 143-148, 1998.[Medline]
716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell, 87: 803-809, 1996.[Medline]
. J. Biol. Chem., 275: 28028-28032, 2000.This article has been cited by other articles:
|
|
 |
|
  M. F.G. de Maat, C. J.H. van de Velde, N. Umetani, P. de Heer, H. Putter, A. Q. van Hoesel, G. A. Meijer, N. C. van Grieken, P. J.K. Kuppen, A. J. Bilchik, et al. Epigenetic Silencing of Cyclooxygenase-2 Affects Clinical Outcome in Gastric Cancer J. Clin. Oncol., November 1, 2007; 25(31): 4887 - 4894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Xue, Z. Lin, Y.-H. Cheng, C.-C. Huang, E. Marsh, P. Yin, M. P Milad, E. Confino, S. Reierstad, J. Innes, et al. Promoter Methylation Regulates Estrogen Receptor 2 in Human Endometrium and Endometriosis Biol Reprod, October 1, 2007; 77(4): 681 - 687. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Xue, Z. Lin, P. Yin, M. P. Milad, Y.-H. Cheng, E. Confino, S. Reierstad, and S. E. Bulun Transcriptional Activation of Steroidogenic Factor-1 by Hypomethylation of the 5' CpG Island in Endometriosis J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3261 - 3267. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Castells, A. Paya, C. Alenda, F. Rodriguez-Moranta, R. Agrelo, M. Andreu, V. Pinol, S. Castellvi-Bel, R. Jover, X. Llor, et al. Cyclooxygenase 2 expression in colorectal cancer with DNA mismatch repair deficiency. Clin. Cancer Res., March 15, 2006; 12(6): 1686 - 1692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Mitchell Unique Suppression of Prostaglandin H Synthase-2 Expression by Inhibition of Histone Deacetylation, Specifically in Human Amnion but Not Adjacent Choriodecidua Mol. Biol. Cell, January 1, 2006; 17(1): 549 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. Yim, H.-S. Jong, T. Y. Kim, H. H. Choi, S. G. Kim, S. H. Song, J. Kim, S.-G. Ko, J. W. Lee, T.-Y. Kim, et al. Cyclooxygenase-2 Inhibits Novel Ginseng Metabolite-Mediated Apoptosis Cancer Res., March 1, 2005; 65(5): 1952 - 1960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Soo, T. Putti, Q. Tao, B.-C. Goh, K.-H. Lee, L. Kwok-Seng, L. Tan, and W.-S. Hsieh Overexpression of Cyclooxygenase-2 in Nasopharyngeal Carcinoma and Association With Epidermal Growth Factor Receptor Expression Arch Otolaryngol Head Neck Surg, February 1, 2005; 131(2): 147 - 152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. J. Chang, M. S. Wu, J. T. Lin, B. S. Sheu, T. Muta, H. Inoue, and C.-C. Chen Induction of Cyclooxygenase-2 Overexpression in Human Gastric Epithelial Cells by Helicobacter pylori Involves TLR2/TLR9 and c-Src-Dependent Nuclear Factor-{kappa}B Activation Mol. Pharmacol., December 1, 2004; 66(6): 1465 - 1477. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sato, N. Maehara, and M. Goggins Gene Expression Profiling of Tumor-Stromal Interactions between Pancreatic Cancer Cells and Stromal Fibroblasts Cancer Res., October 1, 2004; 64(19): 6950 - 6956. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Jang, H. J. Kang, J. R. Kim, and C. H. Yang Non-steroidal anti-inflammatory drug activated gene (NAG-1) expression is closely related to death receptor-4 and -5 induction, which may explain sulindac sulfide induced gastric cancer cell apoptosis Carcinogenesis, October 1, 2004; 25(10): 1853 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Muller, L. Ivarsson, H. Schrocksnadel, H. Fiegl, A. Widschwendter, G. Goebel, S. Kilga-Nogler, H. Philadelphy, W. Gutter, C. Marth, et al. DNA Methylation Changes in Sera of Women in Early Pregnancy Are Similar to Those in Advanced Breast Cancer Patients Clin. Chem., June 1, 2004; 50(6): 1065 - 1068. [Full Text] [PDF]
|
|
|
|
|
|
G.-C. Hur, S. J. Cho, C.-H. Kim, M. K. Kim, S. I. Bae, S. Y. Nam, J.-W. Park, W. H. Kim, and B. L. Lee Manganese Superoxide Dismutase Expression Correlates with Chemosensitivity in Human Gastric Cancer Cell Lines Clin. Cancer Res., November 15, 2003; 9(15): 5768 - 5775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Yuan, R. Z. Luo, S. Fujii, L. Wang, W. Hu, M. Andreeff, Y. Pan, M. Kadota, M. Oshimura, A. A. Sahin, et al. Aberrant Methylation and Silencing of ARHI, an Imprinted Tumor Suppressor Gene in which the Function Is Lost in Breast Cancers Cancer Res., July 15, 2003; 63(14): 4174 - 4180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Maihofner, M. P. Charalambous, U. Bhambra, T. Lightfoot, G. Geisslinger, N. J. Gooderham, and The Colorectal Cancer Group Expression of cyclooxygenase-2 parallels expression of interleukin-1beta, interleukin-6 and NF-kappaB in human colorectal cancer Carcinogenesis, April 1, 2003; 24(4): 665 - 671. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Satoh, T. Nakagawachi, H. Nakadate, Y. Kaneko, Z. Masaki, T. Mukai, and H. Soejima Significant Reduction of WT1 Gene Expression, Possibly Due to Epigenetic Alteration in Wilms' Tumor J. Biochem., March 1, 2003; 133(3): 303 - 308. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, unmethylated cytosines; •, methylated cytosines.
) or mutated sites (
) are labeled accordingly. As described under "Materials and Methods," cells were transfected with Cox-2 promoter constructs or pGL2-Basic control vector together with pSV-ß galactosidase vector. After transfection, cells were maintained in a serum-starved state for 24 h before cell lysates were analyzed. Expression was assessed as luciferase activity and was normalized to the ß-galactosidase activity. B, Cox-2 promoter activity of SNU-601 was compared with that of HCT-116. Under the same conditions, SNU-601 was transfected with a series of Cox-2 promoter deletions (C) or mutant constructs (D). E, complete inhibition of Cox-2 promoter activity by in vitro CpG methylation. The Cox-2 promoter constructs (-1432/+59, -327/+59, and -220/+59) and pGL2-Basic control vector were methylated in vitro with SssI (CpG) methylase. Methylated (
) constructs (2 µg) were transiently transfected into SNU-601. Luciferase activity represents data that have been normalized with ß-galactosidase activity. Columns, means; bars, SD (n = 5 for B and D; and n = 3 for C and E). (*, P < 0.01 comparing