Abstract
OBJECTIVE To explore the relationship between the methylation status of the promoter 5’CpG island region and the biological behavior of human colorectal cancer RKO cells in vitro.
METHODS RKO cells were treated with a selective DNA methyltransferase inhibitor-5-aza-2’-deoxycytidine (5-aza-CdR) for 72 h. Methylation-specific PCR (MSP), T-A cloning and DNA sequence analysis were used to determinate the 5’CpG island methylation status of the p16/CDKN2 tumor suppressor gene. Cell growth, morphological changes and apoptosis were analyzed by the MTT assay, flow cytometry, fluorescence staining and electron microscopy.
RESULTS The 5’CpG island of the p16/CDKN2 tumor suppressor gene in RKO cells was a typically hypermethylated. The DNA methyltransferase inhibitor (5-Aza-CdR) effectively reversed the hypermethylation status of the promoter region. With demethylation, RKO cell growth was suppressed, the cells doubling times were prolonged (P<0.01) and apoptosis was induced, which showed a relationship.
CONCLUSION A selective DNA methyltransferase (DNMT) inhibitor can inhibit proliferation by demethylation in 5’CpG islands, and may be a potential new therapy target for colorectal cancer.
keywords
Introduction
One of the most prominent features of carcinoma behavior is their unlimited proliferation and metastasis, resulting from activation of oncogenes and inactivation of tumor suppressor genes. DNA methylation is a major regulatory mechanism of tumor epigenetic control and it has an important role in gene transcription and splicing[1]. In recent years, it has been demonstrated that DNA promoter hypermethylation is associated with inactivition of several tumor suppessor genes[2-4], and these aberrant DNA methylations are always accompanied by a progressive dysregulation of DNA methyltransferase (DNMTs) expression[5]. In order to investigate the relationship between the methylation status of the promoter 5’ CpG island region and the biological behavior of human colorectal cancer cells, we treated RKO cells with a selective DNMTs inhibitor—5-aza-2’-deoxycytidine (5-Aza-CdR) to determine the effect of decreased methylation of an important cell cycle p16/CDKN2 gene, and to explore a potential therapy target for colorectal cancer.
Materials and Methods
Reagents
Dulbecco’s modified eagle medium (DMEM/HG) was a product of GIBCO; 5-Aza-CdR, hydroquinone and sodium bisulfite were from Sigma; 96-well plates were obtained from Costar; Wizard DNA Clean-up was from Promega. Primers were synthesized by Sangon. The QIAquick Gel Extraction Kit and Hot-star Taq DNA polymerase were from QIAGEN.
Cell culture and 5-Aza-CdR treatment
Human colorectal cancer RKO cells were maintained as monolayers in DMEM supplemented with 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and incubated at 37°C in a humidified incubator with 5% CO2 in air. The RKO cells were plated in growth medium for 24 h and then treated with 1×10-7, 5×10-7 and 1×10-6mol/L of 5-Aza-CdR for 72 h. Assays were conducted 9 days after initiation of the cultures.
Methylation-specific PCR (MSP)[6,7]
Genomic DNA was isolated from the RKO cell lines by proteinase K digestion and phenol/chloroform extraction. DNA (3 μg) in a volume of 50 μl was denatured by NaOH (final concentration, 0.2 μmol/L) for 10 min at 37°C. Thirty microliters of 10 μmol/L hydroquinone (Sigma) and 520 μl of 3 mol/L sodium bisulfite (Sigma) at pH5, both freshly prepared, were added and mixed, and the samples incubated under mineral oil at 50°C for 18 h. Modified DNA was purified using the Wizard DNA purification resin and washed with 80% isopropylalcohol (Promega), then eluted into 50 μl of water. Modification was completed by NaOH (final concentration, 0.3 mol/L) treatment for 5 min at room temperature, followed by ethanol precipitation. DNA was resuspended in 30 μl of water and used immediately, or stored at -20°C. Bisulfite-modified DNA was amplified with p16/CDKN2 gene methylated-specific (p16-M) and unmethylated-specific (p16-UM) primers respectively. Primers were placed near the transcriptional start site. P16-M and p16-UM primer pairs described as follows were from Shanghai Sangon Biological Engineering Technology and Service Co. Ltd. p16-M: 5’-TTA TTA GAG GGT GGG GCG GAT CGC-3’ (sense primer), 5’-CCA CCT AAA TCG ACC TCC GAC CG-3’ (antisense primer); p16-UM: 5’-TTA TTA GAG GGT GGG GTG GAT TGT-3’ (sense primer), 5’-CCA CCT AAA TCA ACC TCC AAC CA-3’ (antisense primer). Reactions were (68°C→55°C) ×13 cycles (-1°C→55°C) ×25 cycles, followed by a final 7 min extension at 72°C. Each product was directly loaded onto a 1.5% agarose gel, and stained with ethidium bromide.
T-A cloning and DNA sequence analysis
The MSP products were separated on a agarose gel, and the bands containing p16-M and p16-UM cDNA were cut off and placed into the QIAquick spin column. The cDNA was purified and ligated into pGEM-T Easy Vecter according to the manufacturer’s instructions (Promega). The ligation reaction used T4 DNA Ligase 1 μl, 2×Ligation Buffer (LB) 5 μl, pGEM-T easy 1 μl, MSP cDNA 1 μl, dd H2O 2 μl and the reaction was incubated 2~4 h at room temperature. The products were transfected into competent E.coli. DH5α, 50 μl and heat-shocked for 60 s in a water bath at exactly 42°C. Then the tubes were immediately returned to ice for 2 min and gently mixed with 200 μl LB/IPTG and 4 μl/X-Gal 16 μl, after which the plate was incubated overnight at 37°C. The recombinant white clone was screened and generated in LB overnight at 300 rpm 37°C. The recombinant plasmid DNA was isolated and further sequenced using an ABI PRISM 377 DNA Sequencer.
Assay of cell proliferation and doubling time
Colorimetric MTT assays were used to observe cell proliferation. The RKO cells at an exponential phase were seeded at 1×104/ml density in 96-well plates with 200 μl cell suspension per well. Each group included 3 duplicate wells and 7 plates. Every day 20 μl MTT (5 mg/ml) was added to each well and the cells cultured at 37°C for 4 h. The supernate was discarded and 150 μl dimethyl sulfoxide (DMSO) added for coloration. After the crystals were dissolved by vibration for 10 min, the absorbance (A) value was read on an enzymelabeled Minireader (Opsys MD) at a test wavelength of 570 nm to obtain an average value. Cell doubling time (TD) was calculated using the following equation: TD= t × log2/(logNt-logNo).
Morphological study using fluorescence microscopy
RKO cells grown on glass slides were treated with 0, 1×10-7, 5×10-7 and 1×10-6mol/L 5-Aza-CdR for 72 h followed by washing with phosphate-buffered saline (PBS). The cells were stained with 100 mg/ml aeridine orange (AO) and visualized immediately under a fluorescence microscope, at the peak excitation wave length of 520 nm.
Transmission electron microscopic observation
Cells were harvested after 5-Aza-CdR treatment and washed 3 times with PBS, fixed in 4°C 2.5% glutar-aldehyde, and then postfixed in osmium tetroxide and embedded. Ultrathin sections were prepared, stained with uranyl acetate and lead citrate, and examined under a transmission electron microscope (JEM-200X JEOL) to observe the morphological changes.
Flow cytometric analysis
The suspended single cell solutions were subjected to treatment with 5-Aza-CdR at different concentrations (0, 1×10-7, 5×10-7 and 1×10-6mol/L) for 72 h and then harvested. Each group was cultured in triplicate. The cells were washed with PBS 3 times, fixed with 75% ethanol at 4°C overnight, washed with PBS again, treated with 200 μl 1% RNAse at 37°C for 30 min and stained with 50 μg/L PI for 30 min in darkness. Apoptotic cells were assayed using a FACS Calibur Becton Dickinson Flow Cyctometer at 488 nm.
Statistical analysis
All data were expressed as the mean±standard deviation. Statistical analysis was performed by the t test using software SPSS10.0 for windows. P<0.05 was considered significant.
Results
Detection of the methylation status of the promoter CpG island region of the p16/CDKN2 gene
As shown in Fig.1, the DNA amplification product (lane 1: 243 bp) in untreated RKO cells was only by the methylation-specific primer (p16-M) PCR, and no DNA product (lane 2) was amplified by the unmethylation-specific primer (p16-UM) PCR. When RKO cells were treated with different concentrations of 5-Aza-CdR, DNA products were all amplified by p16-M and p16-UM (lane 3~8). The higher the concentration of 5-Aza-CdR, the more/less of p16-UM/p16-M PCR product appeared. There was an apparent concentration-response relationship.
The electrophoresis map of the DNA methylation status of RKO cells by MSP. M: DNA marker DL2000; lane 1, 3, 5, 7: 0, 0.1, 0.5, 1.0 umol/L 5-Aza-CdR p16-M; lane 2, 4, 6, 8: 0, 0.1, 0.5, 1.0 umol/L 5-Aza-CdR p16-UM respectively.
On the basis of the MSP, the T-A cloning and DNA sequence were further analyzed. Sequencing results showed that all cytosines in untreated RKO cells remained as C (cytosine), while all cytosines in treated RKO cells had been converted to T (thymidine). It indicated that all the Cs in the former were methylated, and all the Cs in the latter were demethylated (Fig.2).
DNA alleic fragment sequence in 5-Aza-CdR-untreated and -treated RKO cells. All cytosines in 5-Aza-CdRuntreated RKO cells (upper) remain as C, while all CpG cytosines in 5-Aza-CdR-treated RKO cells (bottom) had been converted to T. A: DNA sequence of untreated (control) RKO cells. B: DNA sequence of treated RKO cells.
Effects of demethylation of the CpG island on cell growth and doubling times
RKO cell growth was significantly inhibited and showed a concentration-dependent growth curve after treatment with 1×10-7, 5×10-7, and 1×10-6 mol/L 5-Aza-CdR (Fig.3). The doubling times (TD) of the cells treated with 5-Aza-CdR were 25.9±2.4, 28.6± 2.9, and 32.5±3.0 h respectively, prolonging the TD 17.4±2.0 h of the untreated (control) group (P<0.05 vs. 1×10-7 mol/L 5-Aza-CdR; P<0.01 vs. 5×10-7 mol/L and 1×10-6 mol/L 5-Aza-CdR respectively).
Growth curve of RKO colorectal cancer cells after treatment with 5-Aza-CdR.
Morphological observation
Under the fluorescence microscope, the untreated (control) RKO cells stained by acridine orange (AO) exhibited a green fluorescence. The nuclei were homogeneous and round, with a smooth unclear membrane, free of both chromatin condensation and fragmentation. However, smaller nuclei, increased density of nuclear chromatin, fragmentation of nuclei and apoptotic body formation were identified in 5-Aza-CdR-treated RKO cells (Fig.4).
Fluorescence microscopy for apoptosis induced by 5-Aza-CdR for 72 h. A: Untreated (control); B, C and D, 5-Aza-CdR treated: 0.1, 0.5 and 1.0 μmol/L.
Ultrastructural observation by transmission electron microscopy
Under transmission electron microscopy, the untreated (control) cells showed round nuclei, regular shapes, homogeneous chromatin and integrated nuclear membranes. However, the nuclear membrane of demethylated cells became irregular and smaller, the chromatin was fragmented and located along the nuclear edges, or formed irregularly shaped crescents at the nuclear edges, or became condensed, or fragmented, or formed apoptotic bodies (Fig.5).
Transmission electron ultrastructural changes after 72 h in apoptoticinduced RKO cells. A: untreated (control ); B: treated with 5-Aza-CdR.
Detection of apoptotic cells by flow cytometry
Flow cytometric analysis showed that all the RKO cells treated with different levels of 5-Aza-CdR appeared in a hypodiploid peak (apoptotic peak) on the histogram before the G0/G1 transition. The apoptotic rates were 3.36±0.25%, 6.72±0.61% and 16.85± 1.46% respectively, which were higher than that in control cells (1.28±0.22%, P<0.01 vs. 5×10-7 mol/L or 1×10-6 mol/L 5-Aza-CdR). The apoptotic forming ability increased gradually with increasing levels of 5-Aza-CdR treatment, and showed a positive relationship (Table 1).
The influence of 5-Aza-CdR on the apoptotic rate and cell cycle distribution in RKO cells (%, 
, n=3).
Discussion
DNA methylation is a covalent modification of the fifth carbon position of the pyrimidine ring of cytosines in CpG dinucleotides, and DNA methyltranferase (DNMTs) catalyzes the transfer of a methyl residue from S-adenosyl-methionine (SAM) to a cytosine of a CpG sequence[8]. Recently, some studies have shown that almost 50% of human cancers were found to have a mutation and/or a deletion of a tumor suppressor gene. In some tumors there was no evidence of mutatious in the coding region, but these tumor suppressor genes were frequently made inactive by CpG hypermethylation[9-11]. So promoter a typical hypermethylation is indicated to be an important mechnism in silencing tumor suppressor genes.
In our study, we first analyzed the effect of 5-Aza-CdR on the methylation status of the promoter CpG island region for the p16/CDKN2 gene in RKO cells. Up to the present, the main methods of methylation analysis include: restriction enzymes, restriction enzyme-PCR, Southern blots, MSP, DHPLC, COBRA, DNA sequence analysis etc., in which DNA sequencing is the most reliable. MSP and DNA sequencing were combined to show that all cytosines (C) in treated RKO cells had been converted to thymidines (T) and all cytosines in untreated (control) RKO cells remain as cytosines. It was shown that 5-Aza-CdR had an effective demethylation role and could reverse the methylation status of the promoter CpG island region of p16/CDKN2 in RKO cells. By bisulfite modification, all cytosines were converted to uracils (U) and further matched with thymidines during amplification, but those 5-methylcytosines (5-mC) were resistant to this modification and remain as cytosines(C)[12]. This altered DNA can be amplified by methylation-specific and unmethylation-specific primers, then be sequenced, providing detailed information of the methylation status within the amplified region of all CpG sites. The results revealed that the p16/CDKN2 tumor suppressor gene in colorectal cancer RKO cells correlated with the a typical hypermethylation in the promoter CpG island region. 5-Aza-CdR can lead to 5-methylcytosine demethylation in a CpG island by inhibiting DNMTs in a concentration-dependent manner. Combining the two methods, we specifically assessed the methylation status of CpG islands by taking advantage of the sequence differences resulting from bisulfite modification.
It has been demonstrated that apoptosis is a programmed-cell death (PCD) process which is controlled by apoptotic-associated genes after receiving exterior-signal stimulation. PCD has an important biological function, and its function is not only restricted to embryogenesis, immunosurveillance, and stabilization, but also to cellular proliferation, carcinogenesis and development[13]. In our study, the RKO cells after 5-Aza-CdR treatment exhibited characteristics of apoptosis including nuclear membrane shrinkage, condensation, fragmentation of nuclear chromatin adjacent to the nuclear membrane, and the formation of an overall apoptotic body appearance.
With an increase in 5-Aza-CdR concentration, the apoptotic rate was significantly enhanced and showed a concentration-dependent response. However, the exact mechanism of how apoptosis may be regulated by DNA demethylation in tumor cells remains unclear. Whether it is related to demethylation of the promoter region in apoptotic-related genes (bcl-2, bax, c-myc etc.) is unknown, so further experiments to define the relationship should be conducted.
Proliferation and metastasis of carcinoma cells are fundamental activities in the development of human malignant tumors. The cellular proliferation activity is an important factor in assessing the tumor’s biological behavior, and the tumor characteristic influence the effect of chemotherapy, radiotherapy and other treatment modalities on the patient’s prognosis. Our studies showed that the cytosine demethylation of the DNA promoter CpG island region can significantly inhibit the growth of RKO cells, increase the cell doubling-time, and cause a concentration–dependent inhibitory effect. It is purposed that the anti-tumor action might have a cause and effect relationship between the promoter region demethylation of the tumor suppressor genes and reactivation of transcription and expression of genes such as the p16/CDKN2 gene[14]. In addition, it may be associated with cell apoptosis after demethylation.
Global DNA hypomethylation and regional CpG island hypermethylation have been shown to coexist during oncogenic transformation[15]. DNA hypomethylation may increase expression of oncogenes, but alternatively, DNA hypermethylation may help to silence tumor suppressor genes. So further studies will be needed to understand the important role of epigenetics in carcinogenesis and development of human cancer. Furthermore, the opposed characteristics of promoter hypermethylation in gene silencing suggest clinical significant therapeutic alternatives designed to suppress abnormal hypermethylation.
Acknowledgements
We are very grateful to Mr. Dajun Deng, professor of the Clinical Oncology Hospital, Beijing University for providing the human colorectal RKO cell lines.
Footnotes
This study was supported by grants from the National Key Basic Research Project Foundation (No.G1998051200 ) and the Science and Technology Development Foundation of Zhejiang Province (No.011110541).
- Received August 14, 2007.
- Accepted December 21, 2007.
- Copyright © 2008 by Tianjin Medical University Cancer Institute & Hospital and Springer
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.