Inhibition of the Extracellular Signal-regulated Kinase/Mitogen-activated Protein Kinase Pathway Decreases DNA Methylation in Colon Cancer Cells*

The extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK-MAPK) pathway is a critical intermediary for cell proliferation, differentiation, and survival. In the human colon cancer cell line SW1116, treatment with the DNA methyltransferase 1 (DNMT1) inhibitor 5-aza-2′-deoxycytidine (5-aza-dC) or the ERK-MAPK inhibitors PD98059 or rottlerin, or transient transfection with the MAP/ERK kinase (MEK)1/2 small interfering RNA down-regulates DNMT1 and proliferating cell nuclear antigen levels. In this report, we found that drug treatment or small interfering RNA transfection of SW1116 cells induced promoter demethylation of the p16INK4A and p21WAF1 genes, which up-regulated their mRNA and protein expression levels. Flow cytometry revealed that rottlerin treatment induced cell cycle arrest at phase G1 (p < 0.05). Thus, the ERK-MAPK inhibitor treatment or siRNA-mediated knockdown of ERK-MAPK decreases DNA methylation via down-regulating DNMT1 expression and other unknown mediator(s) in SW1116 colon cancer cells.

Colon cancer is one of the most commonly occurring tumors and the leading cause of cancer death worldwide. Cells derived from the colon cancer cell line SW1116 are frequently used in molecular biological experiments (1)(2)(3). Mitogen-activated protein kinases (MAPKs) 3 are serine-threonine kinases that are activated by phosphorylation in response to a wide array of extracellular stimuli (4,5). The extracellular signal-regulated kinase (ERK)-MAPK pathway is a critical pathway for cell proliferation, differentiation, and survival. Activation of ERK1 and ERK2 (ERK1/2) in this pathway modulate a wide variety of cellular activities via the regulation of several transcription factors. The ability of the ERK-MAPK pathway to promote cell growth by activation of cyclin D is counterbalanced by the concomitant production of the cyclin-dependent kinase inhibitor p21 WAF1 . Moderate activation of the pathway leads to cell proliferation, whereas hyperactivation results in p21 WAF1 -mediated growth arrest. In addition, induction of the cell cycle inhibitory INK4 proteins, including p16 INK4A , is mediated by the Ras/Raf/MEK/ ERK pathway (6). Several lines of evidence have indicated that overexpression and activation of ERK-MAPK plays an important role in colon cancer progression, and that may be a useful molecular target for colon cancer therapy (7).
To date, 11 protein kinase C (PKC) isoenzymes have been identified (8). They have been implicated in the regulation of cell growth and differentiation, as well as in apoptosis and tumor progression (9). Although PKC-␦ is the most thoroughly studied member in the PKC subfamily, many questions remain to be answered about its substrate-specific activities (10).
Aberrant DNA methylation is now considered as an important epigenetic alteration that is intimately involved in the initiation and development of colon cancer in both mouse and human tumor model systems (11,12). In particular, hypermethylation in the promoters of tumor suppressor genes correlates with the loss of expression of these genes in cancer cell lines and primary tumors (13,14). In recent years, several reports have shown that two ERK-MAPK inhibitors, procainamide and hydralazine, produce hypomethylation in T cell lines (4,(15)(16)(17). These observations have led to efforts to use these drugs to demethylate tumor suppressor genes in cancer cells. However, the mechanism whereby these ERK-MAPK pathway inhibitors affect DNA methylation that silences gene transcription in cancer, particularly in human colon cancer, is poorly understood. Furthermore, the effect of the PKC-␦ inhibitor rottlerin (18) on tumor-related gene regulation and the cell cycle in human colon cancer cell lines is unknown.
The aim of this study was to further elucidate the etiological association of the ERK-MAPK pathway and methylation in human colon cancer, and to investigate the extent of the loss of MAP/ERK kinases (MEK) 1 and 2 and the effects of ERK signaling on cell proliferation and cycling. A colon cancer cell line, SW1116, was incubated with the MEK1/2 inhibitor PD98059 (2-(2Ј-amino-3Ј-methoxyphenyl)oxanaphthalen-4-one, a MEK inhibitor) and the PKC-␦ inhibitor rottlerin, and transiently transfected with the selective ablation of MEK subtypes by RNA interference (RNAi). The influence of these drugs or transfections on DNA methylation, subsequent expression of tumor suppressor genes, and the cell cycle were also assessed. We also investigated the mediating mechanisms of ERK-MAPK on DNA methylation.

EXPERIMENTAL PROCEDURES
Cell Culture-The colon cancer-derived cell line SW1116 was maintained by serial passages in RPMI 1640 containing 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, and 100 g/ml streptomycin. Cultures were incubated at 37°C using standard tissue culture incubators as previously described (11). A total of 10 6 cells were plated per 100-mm dish.
Treatment with PD98059, Rottlerin, and 5-aza-dC-PD98059 and rottlerin were purchased from Promega and Sigma, respectively. The inhibitors were dissolved in Me 2 SO and used at final concentrations of 50 and 20 M, respectively. ERK-MAPK inhibitor treatments were performed on serumstarved SW1116 cells before stimulation with 50 ng/ml of phorbol myristate acetate for 15 min. To assess the effect of these inhibitors on methylation and gene expression, the cells were exposed to different concentrations, 2, 5, and 10 M 5-aza-dC (Sigma) for 24 and 72 h as the controls for ERK-MAPK inhibitor treatment. Cells treated with Me 2 SO for 24 h were used as a vehicle controls. DNA and RNA were harvested at various time points throughout the treatments.
RNAi and Transient Transfections-MEK1/2 expression was inhibited with reagents from commercial siRNA kits according to the manufacturer's instructions. The SignalSilence TM MEK1 kit and SignalSilence TM MEK2 kit (Cell Signaling Technology, Danvers, MA) included human MEK1 siRNA, MEK2 siRNA, control siRNA (Fluorescein Conjugate), and siRNA transfection reagent. Briefly, SW1116 were seeded in 12-well plates in medium containing 10% serum at a density that would allow cells to reach 50% confluence on the day of transfection. The transient transfections were performed using 100 nM of each siRNA duplex. For the double knock-down, 100 nM MEK1 and MEK2 siRNA were used. These siRNAs were complexed with transfection reagent, in serum-free and antibiotic-free culture fluid for 5 h at 37°C. After the transfection reagents were applied, cellular media were replaced with serum-containing maintenance media and the cells were incubated for 48 h. All experiments were repeated at least three times. The transfection efficiency was determined by fluorescein-conjugated nonspecific siRNA-transfected cells using a fluorescence microscope (Olympus, Japan). Cells were incubated for 5 min at room temperature with lysis buffer (K 2 HPO 4 , Triton), and lysates were assayed for luciferase activity. The protein concentrations were determined with a bicinchoninic acid (BCA) protein assay (Pierce), and luciferase activity was quantitated by luciferase assays (Promega) using a luminometer. The selective silencing of the appropriate MEK isozymes was confirmed by Western blotting with MEK1-and MEK2-specific antibodies.
Cell Viability Assays-The SW1116 cells were seeded onto 24-well plates with 2000 cells/well, cultured for 24 h, and transfected with the indicated siRNAs. The cells were then treated with the ERK-MAPK inhibitors or 5-aza-dC. Cell proliferation was measured using the methylthiazolyl blue tetrazolium bromide (MTT, (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenytetrazolium, Calbiochem) colorimetric dye assay in quadruplicate. At each time point, the SW1116 cells were incubated with 0.3 ml of MTT dye (1 mg/ml in serum-free media) for 2 h at 37°C. Adding 200 l of 0.04 N HCl in isopropyl alcohol produced the color, the absorbance of which was monitored at 570 nM.
Real-time Reverse Transcriptase-PCR for mRNA Expression of Tumor Suppressor Genes-To evaluate the efficiency of treatment with MAPK inhibitors or the DNMT inhibitor, and transfection with MEK1/2 siRNA, mRNA transcription levels of DNMTS, p16 INK4A , and p21 WAF1 were detected using real time reverse transcriptase-PCR. Total RNA was isolated with a commercial kit (TRIzol) according to the manufacturer's instructions (Invitrogen). Reverse transcription reactions using 5 g of total RNA in a final reaction volume of 20 l were performed with Superscript II reverse transcriptase (Invitrogen). Relative quantitation data were obtained using the comparative C t method with the ABI PRISM 7700 Sequence Detection System (software version 1.6) according to the manufacturer's protocol. Primers and fluorogenic probes were provided by Shenyou Company (Shanghai). The sequences of the probes and the primers, as well as the PCR conditions, are shown in Table 1. Real time PCR was also performed with the primers and probe for ␤-actin to normalize each of the extracts for amplifiable human DNA. Results were expressed as the ratio of copies of target genes to ␤-actin. The C t values were measured, and the average C t of triplicate samples was calculated. Significant alteration of mRNA expression was defined as a 3-fold difference in the expression level after treatment relative to that before treatment (19).
MSP and Bisulfite Sequencing of p16 INK4A and p21 WAF1 Promoters-Methylation-specific PCR (MSP) was performed in CpG-rich regions of the p16 INK4A and p21 WAF1 gene promoters to monitor changes in the DNA methylation of tumor suppressor genes due to drug treatments or transfection. Bisulfite modification protocols were adapted as described by Xiong and Laird (20). Genomic DNA treated with bisulfite was amplified with p16 INK4A and p21 WAF1 promoter-specific primers ( Table 2). The primers were designed without CpG dinucleotides to enable both methylated and unmethylated alleles to be amplified. There were 100 ng of bisulfite-treated DNA, 0.1 mM dNTPs, 2.0 mM MgCl 2 , and 0.5 M primers in a 50-l PCR. The PCR product was directly loaded onto 3% agarose gels and electrophoresed. The gel was stained with ethidium bromide and directly visualized under UV illumination. Wild-type p16 INK4A and p21 WAF1 primers were used to verify that complete conversion of the DNA had occurred in the bisulfite reaction. A positive control for complete methylation was also amplified.
To further identify the methylation pattern within the CpG islands of the p16 INK4A and p21 WAF1 promoter in SW1116 cells, bisulfite sequencing was performed. DNA was treated with bisulfite as described above. Primer sequences of p16 INK4A and p21 WAF1 , annealing temperatures, and expected PCR product sizes are summarized in Table 2. PCR products were sequenced using ABI PRISM 3730 sequencer.
The densities of the Western blotting bands in each lane were normalized to the amount of total protein (for DNMT1, PCNA, or p21 WAF1 ) or total MAPKs as determined by the density of the ␤-actin band, or pan-MAPKs. For example, if the ␤-actin band was 30,000 units (pixels of brightness), then DNMT1 was normalized by the following calculation: 30,000/ (density of ␤-actin) ϫ (density of DNMT1). The Western blotting analysis was repeated at least three times (21).
Flow Cytometric (FCM) Detection of Cell Cycle Progression-Cell cycle analysis was carried out by FCM (22). In brief, ϳ1 ϫ 10 8 cells were removed from treated and mock-treated cultures at specified time points. The cells were washed twice with phosphate-buffered saline and fixed in ice-cold ethanol for 1 h. The samples were then concentrated by removing the ethanol and exposure to 1% (v/v) Triton X-100 (Sigma) and 0.01% RNase (mg/ml, Sigma) for 10 min at 37°C. Cellular DNA was stained in the dark with 0.05% propidium iodide for 20 min at 4°C. Cell cycle distributions were determined using a flow cytometer (model FACSCalibur, BD Biosciences). The data obtained from 10,000 cells was analyzed with the MultiCycle software package (Phoenix, San Diego, CA).
Statistics-Data were representative of at least three independent experiments performed in triplicate, and are presented as the mean Ϯ S.D. Comparisons between groups were made using the Student's paired t test. Relationships were analyzed by Fisher's exact test using SAS 6.12 software. A value of p Ͻ 0.05 was taken to indicate a significant difference between the mean values.

RESULTS
Both ERK-MAPK Inhibitors and siRNA Transfection Suppress Colon Cancer Cell Growth-We first examined the antiproliferative effects of ERK-MAPK inhibitors on the colon cancer cells. MTT assays showed that treatment of SW1116 cells with the PKC-␦ inhibitor rottlerin resulted in a significant inhibition of growth starting from 24 h at 20 mol/liter. Growth inhibition was also apparent at a low dose, 2.5 mol/liter, after 72 h (Fig. 1A). Dose-dependent studies using MTT tests further demonstrated that after 72 h of treatment, the MEK1/2 inhibitor PD98059 suppressed cell growth at intermediate concentrations (50 mol/liter) (Fig. 1B).
Our results indicated that RNAi-induced MEK1 or MEK2 deficiencies inhibited SW1116 cell growth, which declined maximally 24 h after transfection. This suppression lasted for another 12 h, and almost recovered at 48 h post-transfection. As shown in Fig. 1C, the degree of growth inhibition caused by siMEK1 was higher than that caused by siMEK2, whereas the growth inhibition was significantly different with both MEK1/2 deficient (siMEK1/2) at 24 h post-transfection (p Ͻ 0.01). PD98059 or Rottlerin Treatment Down-regulate Phosphorylation of ERK-MAPK-Cells were incubated with ERK-MAPK inhibitors, and cell lysates were analyzed for MAPKs phosphorylation by Western blot analysis. Total ERK1/2 protein levels were not decreased in PD98059-and rottlerin-treated lysates compared with the vehicle controls. However, PD98059 and rottlerin treatment did decrease the phorbol myristate acetatestimulated phosphorylation of MEK1/2 and ERK1/2 in SW1116 cells (Fig. 2, middle and upper panels). In contrast to the inhibition exerted by PD98059, rottlerin also significantly down-regulated the phosphorylation of PKC-␦ (Fig. 2, lower  panel).
The Depletion Effect of MEK1 and MEK2 by RNAi-We used siRNA to selectively reduce the expression levels of MEK1 and MEK2, which are major subunits of the ERK-MAPK pathway. With the high transfection efficiency of siRNA (data not shown), Western blot analysis revealed that the MEK1/2 protein levels in SW1116 cells transfected with the MEK siRNAs singly and together were reduced to 36 Ϯ 1.02, 34 Ϯ 0.98, and  9 Ϯ 0.76%, respectively. Control SW1116 cells transfected with nonsilencing siRNA in the three experiments produced normal MEK1 and MEK2 levels (Fig. 3A).
The effect of depleting MEK and ERK activities by suppressing MEK1/2 was assessed in Western blots using antibodies specifically recognizing phosphorylated (and thus activated) MEK1/2 and ERK1/2. Both knockdown experiments resulted in almost a total loss of pan-and phospho-MEK levels, whereas the phospho-ERK protein levels were analyzed as readouts for MEK activity upon MEK1/2 knockdown and the results showed that phospho-ERK was also descent (Fig. 3, B and C).
ERK-MAPK Inhibitors and siRNA Transfection Repress PCNA and DNMT1, but Not DNMT3A and DNMT3B-Western blotting was also performed to examine the effects of ERK-MAPK inhibitors on PCNA and DNMT1. As shown in Fig. 4, rottlerin (p Ͻ 0.01) and PD98059 (p Ͻ 0.05) significantly downregulated PCNA expression (Fig. 4), and also induced a decrease in DNMT1 protein expression (p Ͻ 0.05) that was similar to previous results obtained with cells treated by the DNMT inhibitor 5-aza-dC (p Ͻ 0.01) (Fig. 4).
Blocking the ERK-MAPK pathway inhibited SW1116 cell growth, as assessed by measuring PCNA and DNMT levels. PCNA is a cell proliferation marker gene, and DNMT1 plays an important role in maintaining DNA methylation patterns via complicated networks including signaling pathways and transcriptional factors relating to cell differentiation or carcinogenesis. Thus, we analyzed the effects of MEK1 and MEK2 depletion on PCNA and DNMT1 protein levels 24 h following cellular transfections. Upon depletion of either MEK1 or MEK2 alone, the PCNA in protein levels decreased along with cell viability and proliferation. Moreover, the protein expression of PCNA was significantly decreased in the double knockdown of MEK1/2. Unexpectedly, our data indicated that the protein of DNMT1 expression decreased concordantly with cell proliferation inhibition and PCNA (Fig. 4).
Furthermore, DNMT3A and DNMT3B are essential for de novo methylation. To clarify the effects of DNMT3A and DNMT3B with blocking the ERK-MAPK pathway, we examined their expressions by real time reverse transcriptase-PCR (Fig. 5). Our results indicated that ERK-MAPK inhibitors and siRNA transfection failed to affect the expression levels of DNMT3A and DNMT3B.
ERK-MAPK Inhibitors and siRNA Transfection Up-regulate p16 INK4A and p21 WAF1 -In this study, inhibiting the ERK-MAPK pathway induced cell cycle arrest. However, members of the cyclin-dependent kinase inhibitor family, including p21 WAF1 , p16 INK4A , and p27 KIP1 , inhibit a wide range of cyclin CDK complexes involved in G 1 and S phase progression. Thus, real time PCR was used to ascertain whether inhibiting or blocking the ERK-MAPK pathway exerts any effects on colon cancer-related tumor suppressor genes. The assay was also used to clarify whether PKC-␦ and MEK1/2 signaling would interact with the methylation-selective tumor-related genes. We found that both PD98059 and rottlerin increased p16 INK4A and p21 WAF1 transcriptional levels, which is consistent with data from samples treated with 5-aza-dC (Fig. 5). The results of MEK1/2 depletion also showed that p16 INK4A and p21 WAF1 transcriptional levels increased. Meanwhile, Western blot analysis indicated that PD98059, rottlerin, and MEK1/2 siRNA induced p21 WAF1 protein expression (Fig. 4).
ERK-MAPK Inhibitors Treatment or siRNA Transient Transfection Induce Demethylation of p16 INK4A and p21 WAF1 -To confirm that ERK-MAPK inhibition directly affect the methylation of p16 INK4A and p21 WAF1 promoters, the methylation status following treatments with ERK-MAPK inhibitors was determined using MSP. Bisulfite treatment converted cytosine residues in the genomic DNA to uracil, which were amplified as thymine during the subsequent PCR. As shown in Fig. 6, A and C, primers that specifically amplified either the methylated or unmethylated form of the p16 INK4As promoter produced 150-bp methylated bands from mock- . siRNA-mediated inhibition of MEK1/2 resulted in different MAPK activation characteristics. SW1116 cells were transiently transfected with the indicated siRNAs, and protein lysates were prepared at the indicated times after transfection. A, knockdown efficiency was assayed by Western blots using subtype-specific antibodies. The samples were obtained 24 h after transfection. B, to assay for MEK activation in siRNA-transfected cells, Western blots using anti-phospho-MEK, and anti-pan-MEK were employed. ␤-Actin was used for the loading control. C, to assay for ERK activation, Western blots using antiphospho-ERK and anti-pan-ERK were employed. ␤-Actin was used for the loading control. Quantification of the target protein bands relative to ␤-actin control is shown below the blots. treated SW1116 cells, indicating that the p16 INK4A promoter was methylated in these alleles. The methylated bands of the p16 INK4A gene from cells treated with 5-aza-dC (10 m) or ERK-MAPK inhibitors (50 M PD98059 and 20 M rottlerin), and MEK1/2 siRNA (100 nM) for 24 h were consistently less intense than the products of mock-treated cells. Thus, there was more 151-bp unmethylated product, than the corresponding methylated product, in 5-aza-dC or MAPK inhibitor-treated cells. These data indicated that untreated SW1116 cells were methylated at some CpG sites in most alleles. This finding also excluded the possibility that successful amplification could be attributed to incomplete bisulfite conversion.
We also examined the methylation state of the p21 WAF1 gene. As shown in Fig. 6, B and D, the methylated band was predominant in mock treated or transfection control cells. Similar to the results obtained with 5-aza-dC, rottlerin treatment and MEK1/2 depletion increased the unmethylated band and decreased the methylated band of the p21 WAF1 promoter. Thus, PD98059 inhibitors favored the demethylation of p21 WAF1 .
Sequencing analysis had the similar changes as MSP. As shown in Fig. 7, CpGs of p16 INK4A and p21 WAF1 promoters in mock treated SW1116 cells showed multiple cytosines due to the resistance of methylated cytosines within CpG dinucleotides to bisulfite conversion. However, using cells treated with 5-aza-dC or ERK-MAPK inhibitors (PD98059 and rottlerin), MEK1/2 siRNA transfection showed numerous thymidines due to sodium bisulfite conversion of unmethylated cytosines to uracil.
Finally, the correlation between promoter methylation status and reactivation of expression of p16 INK4A and p21 WAF1 genes treated with ERK-MAPK inhibitors or transfected with MEK1/2 siRNA was investigated. We found that increased p16 INK4A and p21 WAF1 mRNA expression were significantly associated with promoter demethylation.
ERK-MAPK Inhibition Arrest the Cell Cycle-To further study the signaling pathways involved in ERK-MAPK inhibition-induced cell cycle arrest, the effects of PD98059, rottlerin, and MEK1/2 siRNA on p16 INK4A , or p21 WAF1 -mediated G 1 phase arrest, were evaluated. As illustrated in Fig. 8, pretreatment of SW1116 cells with rottlerin and MEK1/2 siRNA blocked the cell cycle in the G 1 phase (p Ͻ 0.05). Interestingly, PD98059 caused another cell cycle perturbation, S-phase arrest (23) (p Ͻ 0.05). This observation is consistent with up-regulation of p16 INK4A and p21 WAF1 mRNA, suggesting that the ERK-MAPK pathways are involved in this process and rottlerin partially induced G 1 arrest. Overall, we consistently observed a relationship between p16 INK4A and p21 WAF1 expression and cell cycle arrest in PD98059-or rottlerin-treated cells, or MEK1/2 RNAi-transfected cells.

DISCUSSION
Deregulation of the cell cycle commonly contributes to tumorigenesis in human colon cancer. The ERK-MAPK pathway is necessary for cell proliferation (7,24), and activation of this signaling pathway has been shown to be important in intestinal epithelial differentiation (25). There is also a growing body of evidence suggesting that activation of the  ERK-MAPK pathway is involved in pathogenesis, progression, and oncogenic behavior of human colon cancer (7,26).
Regarding the relationship between the ERK-MAPK signaling pathway and DNA methylation, Richardson and colleagues (15) have suggested that DNMT is decreased by inhibiting signaling through the ras-MAPK pathway in human T cells. The T cells of lupus patients show decreased DNMT mRNA and enzyme activity, resulting in DNA hypomethylation (4). In contrast, relatively little is known about mechanisms that the inhibition of MAPK use to induce the expression of tumor suppressor genes, and no data are yet available on the relationship among ERK-MAPK, DNMT1, and DNA methylation in human cancer cells, including those of colon cancer.
Herein, we demonstrate that imbalanced MEK1/2 signaling leads to changes in the amplitude and duration of ERK-MAPK activity. Consequently, p16 INK4A and p21 WAF1 levels are altered, and cell cycle progression is perturbed. siMEKs deplete these target proteins almost completely, but the residual unaffected MEK1/2 molecules remain functional,  retaining a low level of constant ERK activation that is considered baseline.
Methylation of 5Ј-CpG islands in tumor suppressor genes and elevated DNMT levels were also prevalent features of human neoplasia. In the present study, we found that hindering ERK-MAPK activity by inhibiting or blocking MEK1/2 via RNAi transfection down-regulated DNMT1 mRNA and protein expression levels in a similar fashion as the DNMT inhibitor in SW1116 cells.
ERK-MAPK inhibitors, particularly rottlerin, resulted in the demethylation of p16 INK4A and p21 WAF1 . Moreover, PD98059, rottlerin, and MEK siRNA increased the expression of p16 INK4A and p21 WAF1 . Taken together, these experiments provide direct evidence linking endogenous ERK activation to DNA methylation in a colon cancer cell line. We believe that these differences are because the MEK inhibitor and siMEKs induced the expression of p16 INK4A and p21 WAF1 , arrested cell cycle progression, and inhibited cell proliferation following a decreased activity of the DNMT inhibitor gene promoter.
The findings of several studies have suggested that PKC-␦ is involved in regulation of cell growth and differentiation, and that it plays a role in apoptosis and tumor development (27). However, the effects of PKC-␦ on tumor suppressor gene expression are poorly understood. Little is known about the biological function of PKC-␦ as it relates to methylation. The data from this report show that rottlerin, a PKC-␦-specific inhibitor, decreases DNMT1 expression and increases p16 INK4A and p21 WAF1 expression, whereas PD98059 up-regulate significantly p16 INK4A and partially up-regulate p21 WAF1 when compared with vehicle controls. These results provide further support for the hypothesis that PKC-␦, and possibly also MEK, activation plays an important role in the hypermethylation of tumor-suppressor genes and the pathogenesis of colon cancer. Comparison of rottlerin-treated to mock treated cells further indicated that rottlerin treatment produced a notable partial reversal of the transformed phenotype.
Cell differentiation requires the proper balance of activation and inhibition of the ERK-MAPK signaling molecules. FCM analysis showed that with rottlerin treatment, MEK2 or MEK1/2 siRNA induced cell cycle arrest in the G 1 phase. The findings that the ERK-MAPK signal transduction pathway inhibitors affect cell cycle regulation in the SW1116 cell line are consistent with the view that ERK-MAPK signaling may regulate DNA methylation. These results are also the first indication that rottlerin plays a role in p16 INK4A -or p21 WAF1 -induced cell cycle arrest. Interestingly, PD98059 and MEK1 siRNA arrest cell cycle in the S phase but not G 1 phase in the SW1116 cell line. In fact, PD98059 inhibits MEK1 only but not including MEK2 (28,29) and the loss of MEK2 or ERK2 caused cell cycle arrest at G 1 (30). PD98059 affects cell transition from the S to G 2 phase in undifferentiated Caco-2 (31,32), and induce cell cycle arrest of the G 1 phase in human colon cancer cell line SW620 but not in HCT116 (32). The transient transfection of MEK1/2 siRNAs is sufficient to provide the control of cell cycle entry via ERK response genes p21 WAF1 and p16 INK4A , but fails to induce a sustained expression of p21 WAF1 and growth arrest.
The question of how ERK-MAPK inhibitors or MEK1/2 siRNA down-regulate DNA methylation merits further atten-tion. DNMT1 and other mediator(s) are involved in this process. In previous reports, PCNA has been shown to interact with p21 WAF1 (33,34) and PCNA labeling indices were significantly higher in methylated prostate cancer samples than in unmethylated samples (p ϭ 0.048) (35). Moreover, the observation that PCNA can bind to DNMT1 is consistent with the view that maintenance of the methylation pattern in the genome is also dependent upon PCNA and the relative levels of p21 WAF1 and DNMT1 (36). Accordingly, PCNA can only be co-immunoprecipitated with DNMT1 from transformed cells. In addition, when DNA damage occurs, p21 WAF1 colocalizes with PCNA at repair sites and can effectively prevent methylation of damaged DNA by precluding DNMT1 interaction with PCNA, but only in non-transformed cells (36,37). Furthermore, Williams and colleagues (38) demonstrated that complete loss of caveolin-1 results in PCNA overexpression that is accompanied by p21 WAF1 down-regulation and basal hyperactivation of ERK1/2. However, there is no sufficient evidence that indicated that PCNA is involved in the process of inhibiting ERK-MAPK pathway decreased DNA methylation.
Data from the present experiment, however, showed that ERK-MAPK inhibitor treatment or MEK1/2 siRNA transfection induce the re-expression of p21 WAF1 through decreasing DNMT1 and demethylation. Recent reports from several laboratories have revealed that ERK pathway signaling can increase p21 WAF1 protein levels (39). In contrast, MAPK suppresses p21 WAF1 expression in SW480 (40) and COLO-205 cells (41), but induces S phase arrest in DLD-1 cells (42). These observations suggest that ERK pathway signaling may have differing effects in different cell types, and thus that the consequences of activating the ERK-MAPK pathway may vary with cell type. It has become clear that there are complicated interactions among ERK-MAPK pathway constituents, DNA methylation, and the expression of p21 WAF1 in SW1116 cells.
In light of the present data considered in the context of previous findings, we can hypothesize that DNMT1 but not DNMT3A or DNMT3B expression is decreased by rottlerin, PD98059 treatment, and MEK1/2 depletion due to an ERK-MAPK pathway-dependent inhibition of other mediator(s) (unknown mediators might be include PCNA). Ultimately these findings may provide a novel strategy for colon cancer therapy.