JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M501675200 on March 8, 2005

J. Biol. Chem., Vol. 280, Issue 18, 18302-18310, May 6, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An addition or correction has been published
Right arrow All Versions of this Article:
280/18/18302    most recent
M501675200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Agoston, A. T.
Right arrow Articles by Nelson, W. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Agoston, A. T.
Right arrow Articles by Nelson, W. G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Increased Protein Stability Causes DNA Methyltransferase 1 Dysregulation in Breast Cancer*

Agoston T. Agoston, Pedram Argani, Srinivasan Yegnasubramanian, Angelo M. De Marzo, Mohammad Ali Ansari-Lari, Jessica L. Hicks, Nancy E. Davidson, and William G. Nelson{ddagger}

From the Sidney Kimmel Comprehensive Cancer Center, The Johns Hopkins University, School of Medicine, Baltimore, Maryland 21231-1000

Received for publication, February 14, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We report that DNA methyltransferase 1 (DNMT1) expression is dysregulated in breast cancer. The elevated protein levels are not a result of increased mRNA levels, but rather an increase in protein half-life. We found that DNMT1 protein levels were elevated in breast cancer tissues and in MCF-7 breast cancer cells relative to normal human mammary epithelial cells (HMECs) without a concomitant increase in DNMT1 mRNA or proliferative fraction. Although DNMT1 mRNA levels were properly S-phase-regulated in both cell types, DNMT1 protein levels did not follow S-phase fraction in MCF-7 cells. Rather, an increase in DNMT1 protein stability was found for MCF-7 cells relative to HMECs, and a destruction domain was mapped to the N-terminal 120 amino acids of DNMT1, which was required for its proper ubiquitination and degradation in HMECs. Furthermore, overexpression of DNMT1 with this deleted destruction domain in HMECs resulted in significantly increased genomic 5-methylcytosine levels relative to overexpression of the full-length protein. The regulation of DNMT1 destruction via this domain may be dysfunctional in cancer cells leading to subsequent cytosine hypermethylation in the genome.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Methylation of cytosine bases in CpG dinucleotides in gene promoters is recognized to trigger repression of DNA transcription (1). Normally, this methylation occurs in the promoter regions of silenced X-chromosome genes, imprinted genes, tissue-specific genes, and in parasitic DNA elements, and this epigenetic modification has been shown to recruit 5-methylcytosine-binding proteins and chromatin-modifying enzymes resulting in a repressive chromatin structure that is incompatible with gene expression (1, 2). Of the three known DNA methyltransferases with catalytic activity, DNMT1, DNMT3a, and DNMT3b, DNMT1 is the best studied. Its main function is thought to be propagation and maintenance of established methylation patterns, because it is expressed during S-phase (3), localizes to the replication fork via a proliferating cell nuclear antigen (PCNA)1 binding domain (4), and copies methylation patterns from the parent strand to the daughter strand (5).

Abnormal DNA methylation patterns have been found to be associated with tumorigenesis as an epigenetic alternative to gene inactivation by somatic mutation or deletion (1, 6). Several tumor suppressor genes (e.g. CDKN2A, APC, and RASSF1a) are hypermethylated in various cancer types, and a concomitant absence of gene expression has been shown to contribute to a neoplastic phenotype (7-9). Several studies have implicated DNMT1 in the tumorigenic process. When the enzyme has been overexpressed in mouse NIH3T3 cells, increased transformation and proliferation have been reported (10), and DNMT1 overexpression in human fibroblasts has resulted in methylation of endogenous CpG islands (11). In contrast, when RNA interference, antisense, and pharmacological inhibition have been used to antagonize DNMT1 function, demethylation and re-expression of various tumor suppressor genes and a decrease in transformation and proliferation have been seen (12-16). Furthermore, when a transgenic mouse with defective hypomorphic Dnmt1 alleles was crossed with the ApcMin/+ mouse prone to develop intestinal tumors, tumorigenesis is suppressed (17, 18). Finally, DNMT1 has been shown to be required for c-Fos transformation of rat fibroblasts (19).

DNMT1 may also act to silence tumor suppressor genes or to cause neoplastic transformation via a mechanism independent of its DNA methyltransferase activity, although there is less direct evidence for this. DNMT1 has been shown to interact with histone deacetylase 1 in co-immunoprecipitation experiments, leading to histone deacetylation, compaction of chromatin structure, and silencing of gene transcription in promoter-reporter transfection experiments (20). DNMT1 has also been shown to associate with retinoblastoma protein, a critical tumor suppressor gene involved in repressing transcription of S-phase-specific proteins required for DNA replication (20). Finally, DNMT1 binds to PCNA at the replication fork at the same domain that binds p21 (4), a cyclin-dependent kinase inhibitor that can cause p53-dependent and independent cell cycle arrest, raising the possibility that overexpression of DNMT1 could antagonize p21 function and vice versa.

Many studies have reported that increases in DNMT1 mRNA, DNMT1 protein, and DNA methyltransferase activity accompany carcinogenesis. Some have found an increase in DNMT1 mRNA in human tumor samples (21-23), but others have argued that there is no significant overexpression when differences in S-phase fraction are considered (24, 25). However, DNMT1 protein has been shown to be overexpressed disproportionate to S-phase fraction in estrogen receptor-negative cancer cell lines, including MCF-7 (26). Furthermore, DNMT1 mRNA levels are properly S-phase-regulated throughout the cell cycle in MCF-7 breast cancer cells following lovastatin-mediated G1 arrest and subsequent release (27).

Here, we report that, although DNMT1 mRNA levels are similarly regulated in normal human mammary epithelial cells (HMECs) and MCF-7 cells, DNMT1 protein levels are not because DNMT1 protein is not degraded in MCF-7 cells. Furthermore, these different DNMT1 protein levels in HMEC and MCF-7 cells appeared dependent on an N-terminal 120 amino acid destruction domain of DNMT1, which may be dysfunctional in cancer resulting in cytosine hypermethylation in the genome.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Immunohistochemistry of DNMT1 in Breast Cancer Tissues—Using an affinity-purified IgG preparation isolated from rat antiserum raised against a polypeptide (N-MADANSPPKPLSKPRTPRRS-C) derived from DNMT1 (28), we performed immunohistochemistry on formalin-fixed, paraffin-embedded tissue from two benign breast neoplasms (fibroadenomas), 42 primary in situ and/or invasive mammary carcinomas (31 ductal, 9 lobular, and 2 colloid), and 15 metastatic mammary carcinomas involving lung, brain, or bone. The percentage and the intensity of nuclear expression of DNMT1 and the proliferation marker Ki67 in surrounding benign breast tissue, benign lesions, in situ carcinoma, and invasive and metastatic carcinomas were assessed on serial sections.

For DNMT1 immunostaining, we used a modification the Catalyzed Signal Amplification system from DAKO (DAKO Corp., Carpentaria, CA), including all blocking steps. Briefly, antigen retrieval for DNMT1 was achieved by steam heating in DAKO Target Retrieval Buffer (DAKO) for 40 min. The primary anti-DNMT1 antibody was incubated for 45 min at room temperature. The biotinylated anti-rabbit IgG secondary antibody (Vector Laboratories, Burlingame, CA) was applied for 15 min at room temperature, and this was followed by application of the streptavidin-biotin horseradish peroxidase complex for 15 min at room temperature. The biotinyl tyramide was applied (diluted 1:10 in DAKO protein block) for 15 min at room temperature, and the reaction was developed by re-application of the streptavidin-biotin horseradish peroxidase complex using diaminobenzidine (brown staining) as the chromagen. Slides were counterstained with hematoxylin.

The anti-DNMT1 antibody dilution was optimized on formalin fixed, paraffin-embedded cell lines (1:2000) and fixed, paraffin-embedded tissue sections of human tonsil tissue, but staining was somewhat variable on fixed, paraffin-embedded breast biopsy clinical specimens apparently related to slight differences in fixation. Therefore, the dilution used for the breast tissue varied from 1:2000 to 1:8000 depending on the level of background stromal staining in the case. For staining optimization the HCT116 colon cancer cell line was used as a positive control and a derivative of this same cell line HCT116 DNMT1-/- (a generous gift from Dr. Bert Vogelstein) was used as a negative control. The HCT116 DNMT1-/- cells were derived from HCT116 parental cells after targeted disruption of both endogenous DNMT1 alleles (48). As an additional control, DNMT1 staining was completely abolished after preincubation with a 100-fold excess cognate peptide, but not with an unrelated peptide.

For Ki67 immunostaining, a representative formalin-fixed paraffin-embedded tissue block was chosen for labeling for each lesion. Unstained 4-µm sections were then cut from the paraffin block and deparaffinized by routine techniques. The slides were steamed for 20 min in sodium citrate buffer (diluted to 1x from 10x heat-induced epitope retrieval buffer, Ventana-Bio Tek solutions, Tucson, AZ). After cooling for 5 min, one slide was labeled with a 1:1000 dilution of a mouse monoclonal antibody to MIB 1 (clone Ki67, Dako) using the Bio Tek 1000 automated stainer (Ventana). Labeling was detected by adding biotinylated secondary antibodies, avidin-biotin complex, and 3,3'-diaminobenzidine. Sections were then counterstained with hematoxylin.

Cell Culture and Cell Cycle Synchronization—Human mammary epithelial cells (HMECs), prostate epithelial cells (PrECs), and normal human dermal fibroblasts (NHDFs) were obtained from Cambrex (East Rutherford, NJ) and grown in their respective media: MEGM, PrEGM, and FBM-2. MCF-7 breast cancer cells, LNCaP prostate cancer cells, and HCT116 colon cancer cells were obtained from ATCC (Manassas, VA) and grown w/10% fetal bovine serum in minimal essential medium, RPMI 1640, and McCoy's 5A media, respectively (Invitrogen). HMEC and MCF-7 cells were synchronized in G1-, early S-, and M-phases with 20 µM lovastatin for 24 h, 1 µg/ml aphidocolin for 24 h, and 100 ng/ml nocodazole for 16 h (Sigma), with 0.1% v/v Me2SO as a vehicle control. Cells synchronized with lovastatin were released by washing 3x with phosphate-buffered saline and adding growth medium with 2 mM mevalonate (Sigma). Cell cycle analysis was performed by propidium iodide flow cytometry (Guava PCA, Guava Technologies, Hayward, CA) after 70% ethanol fixation for 2 h, 100 µg/ml RNase treatment for 10 min, and staining of DNA content with 20 µg/ml propidium iodide in 0.1% Triton-X for 30 min (Sigma). S-phase fractions were determined by curve-fitting the histogram data with ModFIT (Verity Software House, Topsham, ME).

Quantitative Real-time RT-PCR, Ribonuclease Protection Assay, and Immunoblotting for DNMT1—Cells were harvested and washed once with phosphate-buffered saline. Cellular mRNA was prepared with the RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. Real-time RT-PCR was performed with the Quantitect Probe RT-PCR Kit (Qiagen) for 45 cycles of 15 s at 95 °C and 1 min at 60 °C, following the kit protocol. The primer/probe sequences for DNMT1 and the internal control TATA-binding protein were: DNMT1, GGTTCTTCCTCCTGGAGAATGTC, 6FAM5'-CCTTCAAGCGCTCCATGGTCCTGAA-3' BHQ1 (BIOSOURCE, Camarillo, CA), GGGCCACGCCGTACTG; TATA-binding protein, GTTCTGGGAAAATGGTGTGC, 6FAM5'-CCAAGAGTGAAGAACAGTCCAGACTGGC-3' BHQ1 (BIOSOURCE, Camarillo, CA), GCTGGAAAACCCAACTTCTG. Ribonuclease protection assay was performed using the Ribonuclease Protection Assay III kit (Ambion, Austin, TX). RNA probe was prepared with the MAXIScript T7 in vitro transcription kit (Ambion) using a cloned 542-bp DNMT1 fragment (nucleotides from 555-1097, NCBI RefSeq NM_001379 [GenBank] ) and the glyceraldehyde-3-phosphate dehydrogenase internal control template (Ambion). DNMT1 protein levels were determined by lysing cells in 2% SDS in phosphate-buffered saline at 95 °C for 30 min (viscosity reduction), immunoblotting onto nitrocellulose, and probing with anti-DNMT1 polyclonal antibody (described above), anti-PCNA monoclonal antibody (Ab-1, Calbiochem), anti-{beta}-actin monoclonal antibody (AC-15, Sigma), and anti-Histone H4 polyclonal antibody (Upstate Biotech, Waltham, MA).

Cycloheximide Translation Inhibition, Radiolabeled Pulse-chase Analysis of Protein Stability, and Immunoprecipitation of DNMT1—Cells were treated with 10 µg/ml cycloheximide and/or 10 µM MG132 (Sigma), harvested at various time points indicated, and immunoblotted for DNMT1 with total protein Ponceau S stain as a loading control. For pulse-chase analysis, HMEC and MCF-7 cells were radiolabeled with 100 µCi of [35S]methionine (Amersham Biosciences) in growth medium for 8 h, washed three times, and incubated in normal growth medium for the time periods indicated. Cells were lysed for 10 min on ice with 1% Triton-X in Tris-buffered saline with protease inhibitors (Protease Inhibitor Mixture Set III, Calbiochem) before centrifuging at 12,000 x g for 10 min and harvesting supernatant. Remaining incorporated radiolabel was determined by immunoprecipitation with 1 µg of anti-DNMT1 antibody and Protein A-agarose (Invitrogen), SDS-PAGE, autoradiography, and densitometry.

DNMT1-FLAG/GFP Fusion Protein Constructs and Immunoprecipitation—Full-length DNMT1 (nucleotides from 238-5088, NCBI RefSeq NM_001379 [GenBank] ) and N-terminal deletions were PCR-amplified with a SalI site extension on the reverse primer (5'-GTCGACGCGGTACCCTTGGCAAAGCA) and the following forward primers: full-length 5'-ATGCCGGCGCGTACC; 120-aa N-terminal deletion 5'-ATGGCAGATGCCAACAGCC; 180-aa N-terminal deletion 5'-ATGGAAGAGTCTGAAAGAGCCAAATCG. These PCR products were cloned into the pCR2.1-TOPO plasmid of the TOPO TA cloning kit (Invitrogen). The EcoRI/SalI fragment was then subcloned into the EcoRI/SalI sites of the pEGFP-N1 Vector (Clontech, BD Biosciences, Palo Alto, CA). To generate the FLAG-tagged version, the AgeI/NotI fragment containing GFP was removed and replaced by a linker with the eight-amino acid FLAG sequence (DYKDDDDK-stop) flanked by AgeI/NotI sites. These constructs were transfected into cells with Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. GFP fluorescence analysis was performed with the BD-LSR flow cytometer (BD Biosciences), and FLAG epitopes were detected by immunoblotting with anti-FLAG antibody (Stratagene, La Jolla, CA). FLAG-tagged DNMT1 co-transfected with HA-tagged ubiquitin was also immunoprecipitated (as described above), immunoblotted, and probed with anti-HA antibodies (HA-7, Sigma).

5-Methylcytosine Level Determination by Mass Spectrometry—The overall 5-methylcytosine content (as a percentage of total cytosine content) in genomic DNA was determined by an high-performance liquid chromatography/mass spectrometry method. Briefly, 10 ng of genomic DNA was resuspended in 50 µl of high-performance liquid chromatography grade water and digested with 4 units of nuclease P1 (Sigma) at 65 °C for 10 min in a digestion buffer containing 0.04 mM DFAM, 3.25 mM NH4OAc, pH 5.0, 0.5 mM ZnCl2 in a final volume of 100 µl. 20 µl of 100 mM Trizma base, pH 8.5, was added, and this reaction was treated with 4 units of alkaline phosphatase at 37 °C for 1 h. Following incubation, 20 µl of 300 mM NH4OAc, pH 5.0, and 6 µl of 0.25 mM DFAM in 50 mM EDTA was added. Quantitation of 5-methylcytosine and cytosine was performed with an API 3000 LC/MS instrument (Applied Biosystems). Separation of free nucleotides formed after digestion and treatment with nuclease P1 and alkaline phosphatase was performed on a 250 x 2.00 mm, 5-µm C18 column. 15 µl of samples or standards (consisting of serial dilutions of 5-methylcytosine and cytosine maintained in a buffer identical to samples) was injected in triplicate along with 250 µl/min of a mobile phase profile consisting of 98% solution A (5 mM NH4OAc, 0.1% formic acid, pH ~ 3) and 2% solution B (90% acetonitrile) for 4 min, followed by a linear ramping to 60% solution A and 40% solution B in 1 min, then a ramping to 98% solution A and 2% solution B in 1 min, and finally maintaining this composition isocratically for the final 5 min. Cytosine was monitored in MRM mode with the ion pair 227/112, while 5-methylcytosine was monitored in MRM mode with the ion pair 242/126. The quantity of each analyte was calculated with reference to the standard dilution series, and the ratio of 5-methylcytosine to total cytosine (5-methylcytosine plus cytosine) was calculated.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
DNMT1 Protein Expression Was Discordant with Proliferative Fraction in Breast Cancer Tissues—Using an anti-DNMT1 antibody, we performed immunohistochemical staining analysis of DNMT1 expression using formalin-fixed, paraffin-embedded tissues from two benign breast neoplasms (fibroadenomas), 42 cases of primary in situ and/or invasive mammary carcinomas (31 ductal, 9 lobular, and 2 colloid), and 15 metastatic mammary carcinomas involving lung, brain, or bone. The fraction of nuclei expressing DNMT1 and the intensity of nuclear expression in surrounding benign breast tissue, benign lesions, in situ carcinoma, and invasive and metastatic carcinomas were assessed (Table I). A human colorectal carcinoma cell line with disrupted DNMT1 alleles, HCT116 DNMT1-/-, and its parental cell line, HCT116, were used as DNMT1-immunostaining controls (Fig. 1A).


View this table:
[in this window]
[in a new window]
  		TABLE I
Immunohistochemical analysis of DNMT1 protein expression in breast cancer tissues

The fraction of nuclei expressing DNMT1 and the intensity of nuclear expression in 41 normal terminal duct lobular units (normal), 2 benign breast tissues (fibroadenoma), 42 primary in situ and/or invasive mammary carcinomas (31 ductal, 9 lobular, and 2 colloid), and 15 metastatic mammary carcinomas involving lung, brain, or bone were assessed. The fraction of nuclei with positive Ki67 immunostaining was used as a measure of proliferative activity.

 
Of the 41 normal terminal duct lobular units that were examined, the mean fraction of nuclei expressing DNMT1 was 22% ± 19%, with the majority of cases showing weak or weak to moderate immunostaining intensity. A larger fraction of nuclei expressing DNMT1 was detected in in situ (n = 24; mean: 58% ± 22%), invasive (n = 35; mean: 67% ± 21%), and metastatic (n = 15; mean: 62% ± 23%) mammary carcinomas (Fig. 1, B-F). The intensity of anti-DNMT1 immunostaining in the majority of in situ, invasive, and metastatic mammary carcinomas was moderate or moderate to strong. However, in most of the neoplastic breast lesions, a significantly lower fraction of cells expressed the proliferation marker Ki67 than expressed DNMT1, suggesting that DNMT1 may be overexpressed in a manner discordant with proliferative fraction (Fig. 1, B-F, and Table I).

DNMT1 mRNA Levels Did Not Account for DNMT1 Protein Overexpression in MCF-7 Breast Cancer Cells Relative to HMECs—MCF-7 and HMECs cells were chosen as our tumornormal model system rather than human tissue samples for two reasons: 1) to control for S-phase fraction, which differs greatly in tumor versus normal tissue comparisons, but not by more than 2-fold in logarithmically growing cultured cells, and 2) to have a pharmacologically and genetically amenable system for study. DNMT1 mRNA levels, as measured by quantitative real-time RT-PCR (Fig. 2A) or ribonuclease protection assay (Fig. 2, B and C) were not more than 2-fold different between HMEC and MCF-7 cells, likely reflecting small differences in S-phase fraction (data not shown). However, DNMT1 protein levels were significantly different between HMEC and MCF-7 cells by immunoblot analysis (Fig. 2, D and E). An 8-fold dilution of MCF-7 total protein yielded DNMT1 levels comparable with HMECs, and densitometry revealed a 6-fold increase in DNMT1 levels in MCF-7 cells.

DNMT1 mRNA Levels but Not DNMT1 Protein Levels Were S-phase-regulated throughout the Cell Cycle in MCF-7—Cell cycle dependence of DNMT1 mRNA and DNMT1 protein levels was determined by pharmacological arrest of HMEC and MCF-7 cells in G1-, early S-, and M-phase by lovastatin, aphidocolin, and nocodazole, respectively, and the S-phase fractions were determined by propidium iodide staining of DNA content and flow cytometry (Fig. 3A). DNMT1 mRNA and DNMT1 protein levels were determined by quantitative real-time RT-PCR (Fig. 3B) and immunoblotting (Fig. 3, C and D), respectively. DNMT1 mRNA levels in both HMEC and MCF-7 cells followed S-phase fraction as did DNMT1 protein levels in HMECs; they were significantly reduced (~60%) after G1 arrest but increased relative to G1 when arrested in early S-phase and M-phase. In contrast, DNMT1 protein levels in MCF-7 cells did not follow S-phase fraction and did not show any significant decrease after G1 arrest. In addition, in a second experiment, HMEC and MCF-7 cells were released from lovastatin-mediated G1 arrest by treatment with mevalonate, and protein levels were quantified at various time points by immunoblotting and densitometry (Fig. 4, A and B). Again, DNMT1 protein levels in HMECs tracked tightly with S-phase fraction, whereas DNMT1 protein levels in MCF-7 cells had minimal S-phase regulation.

Proteasome-mediated DNMT1 Protein Degradation Was Impaired in MCF-7 Cells Relative to HMECs—DNMT1 protein stability was measured by two techniques. First, protein levels were measured by immunoblotting following cycloheximide-mediated translation inhibition (Fig. 5, A and B). Although HMECs exhibited a time-dependent degradation of DNMT1 protein levels (half-life =~6 h), no detectable degradation was observed in MCF-7 cells. Furthermore, degradation of DNMT1 protein following cycloheximide treatment was prevented by addition of a proteasome inhibitor (MG132), which also caused protein accumulation when used alone in both cell types (Fig. 5, C and D), suggesting that DNMT1 protein degradation was proteasome-mediated.



View larger version (104K):
[in this window]
[in a new window]
  		FIG. 1.
Immunohistochemistry of DNMT1 expression in breast cancer. Using an anti-DNMT1 antibody, we performed immunohistochemical staining using formalin-fixed, paraffin-embedded tissue from 2 benign breast neoplasms (fibroadenomas), 42 primary in situ and/or invasive mammary carcinomas (31 ductal, 9 lobular, and 2 colloid), and 15 metastatic mammary carcinomas involving lung, brain, or bone. A, DNMT1 immunostaining controls HCT116 DNMT1-/- (right) and its parental colorectal carcinoma cell line HCT116 DNMT1+/+ (left). B-E, DNMT1 (left) and Ki67 (right) immunostaining of ductal carcinoma in situ (B), lobular carcinoma in situ (C), invasive lobular carcinoma (D), and invasive ductal carcinoma (E). Note that the intensity and fraction of DNMT1 immunostaining in the invasive carcinoma in D and E is greater than that of the entrapped benign lobule, and exceeds the fraction of Ki67-positive immunostaining. F, immunostaining of DNMT1 in metastatic mammary carcinoma to bone (left) and lung (right).

 



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 2.
DNMT1 mRNA and DNMT1 protein expression in HMEC and MCF-7 cells. A, DNMT1 mRNA levels were determined by real-time RT-PCR relative to TATA-binding protein (TBP) internal control. B, DNMT1 mRNA levels in serially diluted mRNA samples were determined by ribonuclease protection assay and densitometric analysis (C) with glyceraldehyde-3-phosphate dehydrogenase as an internal control. D, DNMT1 protein levels were determined by immunoblotting and densitometric analysis (E) in serially diluted protein samples with B-actin and PCNA as controls. Error bars indicate ± S.D.

 
The second technique used to measure protein half-life was radiolabeled pulse-chase analysis. After incubation of both cell types with growth medium containing [35S]methionine, cells were washed and incubated for various time periods with unlabeled growth medium. DNMT1 protein was immunoprecipitated, and the remaining incorporated radioactivity was used as a measure of protein stability (Fig. 5, E and F). Although MCF-7 cells did not display any time-dependent change in radiolabeled DNMT1, HMECs showed a time-dependent disappearance of radiolabel (half-life < 6 h) that was prevented by MG132. These data support the hypothesis that DNMT1 degradation is proteasome-mediated, and that this degradation is impaired in MCF-7 cells relative to HMECs.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 3.
Cell cycle regulation of DNMT1 mRNA and DNMT1 protein levels. A, HMEC and MCF-7 cells were arrested by treatment with 20 µM lovastatin for 24 h (Lov, G1 arrest), 1 µg/ml aphidocolin for 24 h (Aph, early S-phase arrest), 100 ng/ml nocodazole for 16 h (Noc, M-phase arrest), and 0.1% v/v Me2SO for 24 h (Ctrl, vehicle control). S-phase fraction was determined by propidium iodide staining of DNA content and flow cytometry. B, DNMT1 mRNA levels were determined by real-time RT-PCR for HMEC and MCF-7 cells and normalized to TATA-binding protein internal control. C, DNMT1 protein levels in HMEC and MCF-7 cells were determined by immunoblotting and densitometry (D) with B-actin as a loading control. Error bars indicate ± S.D. of duplicate lanes for immunoblotting and triplicate samples for RT-PCR.

 



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 4.
DNMT1 protein levels versus S-phase fraction after G1 arrest and release. A, HMEC and MCF-7 cells were arrested in G1 with 20 µM lovastatin for 24 h and released with 2 mM mevalonate for the time periods indicated. DNMT1 protein levels were determined by immunoblotting and densitometry (B) with histone H4 and PCNA as loading controls, and S-phase fraction was determined by propidium iodide flow cytometry.

 
DNMT1 Dysregulation Was Observed in LNCaP and HCT116 Cancer Cells Relative to Normal PrEC and NHDF Cells—Investigation of DNMT1 dysregulation was extended to four other cell types: LNCaP, a prostate cancer cell line; HCT116, a colon cancer cell line; normal prostate epithelial cells (PrECs); and normal human dermal fibroblasts (NHDFs). DNMT1 protein stability was analyzed by immunoblotting (Fig. 6, A and B) following 24-h lovastatin treatment or 6-h cycloheximide treatment (with or without proteasome inhibition by MG132). Normal PrECs and NHDFs exhibited an "HMEC-like" pattern of DNMT1 protein regulation, having a significant decrease after lovastatin and cycloheximide treatment. Furthermore, this degradation was prevented by proteasome inhibition. LNCaP and HCT116 cells, however, showed a minimal decrease in DNMT1 protein levels following lovastatin or cycloheximide treatment, similar to MCF-7 cells. Therefore, DNMT1 dysregulation, the absence of cell cycle-specific proteasome-mediated degradation, may be a general characteristic of many cancer cell types.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 5.
DNMT1 protein stability in HMEC and MCF-7 cells. A, HMEC and MCF-7 cells were treated with 10 µg/ml cycloheximide (CHX) for the time periods indicated, and DNMT1 protein levels were determined by immunoblotting and densitometry (B) with total protein staining (Ponceau S) as an internal control. C, HMEC and MCF-7 cells were treated with 10 µM MG132 and/or 10 µg/ml cycloheximide for 6 h, and DNMT1 protein levels were determined by immunoblotting and densitometry (D) with B-actin as a loading control. E, HMEC and MCF-7 cells were radiolabeled with 100 µCi of [35S]methionine in growth medium for 8 h, washed three times, and incubated in normal growth medium for the time periods indicated. Remaining incorporated radiolabel was determined by immunoprecipitation of DNMT1, SDS-PAGE, autoradiography, and densitometry (F). Total protein from the lysate was used as a labeling control. Error bars are ± S.D. of duplicate lanes.

 
Differential DNMT1 Ubiquitination and Degradation in HMECs and MCF-7 Was Dependent on an N-terminal 120-Amino Acid Destruction Domain—To create a genetically amenable model system of study, we created a fusion protein consisting of DNMT1 fused to an 8-amino acid FLAG tag, or the N terminus of green fluorescent protein (GFP) (Fig. 7A). Deletion analysis was used to map the domain in DNMT1 that was responsible for the differential protein stability in HMEC and MCF-7 cells. Since the N terminus of a protein often has a large influence on its stability, we started with N-terminal deletion constructs of our DNMT1 fusion proteins. The first deletion construct eliminated the N-terminal 120 amino acids but not the PCNA binding domain (PBD) or the nuclear localization signal (NLS). The second deletion eliminated the N-terminal 180 amino acids, including the PBD and NLS. These DNMT1-FLAG/GFP constructs (along with unfused GFP control) were transfected into both HMEC and MCF-7 cells, and the DNMT1-FLAG fusion protein levels were determined by immunoblotting with an anti-FLAG antibody (Fig. 7, B and D). DNMT1-GFP fluorescence was monitored by fluorescence microscopy to confirm proper nuclear localization and expression levels (Fig. 8, A-D) and by flow cytometry for quantitative analysis of fusion protein levels (Fig. 8E). As expected, the full-length and 120-amino acid deletion constructs, but not the PBD/NLS-deleted constructs, displayed proper nuclear localization of the fusion protein.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 6.
DNMT1 dysregulation in LNCaP and HCT116 cancer cells relative to normal PrECs and NHDFs. A, human prostate epithelial cells (PrECs), normal human dermal fibroblasts (NHDFs), human prostate adenocarcinoma LNCaP cells, and human colorectal cancer HCT116 cells were treated with 20 µM lovastatin for 24 h (Lov), or 10 µg/ml cycloheximide for 6 h (lanes C) and/or 10 µM MG132 for 6 h (lanes M). DNMT1 protein levels were determined by immunoblotting and densitometry (B) with B-actin as a loading control. Error bars are ± S.D. of duplicate lanes.

 
Although the DNMT1-FLAG full-length protein levels were quite disparate in HMEC and MCF-7 cells, deletion of the N-terminal 120 amino acids significantly increased the protein accumulation in HMECs to the level of MCF-7 while having little effect on the protein levels in MCF-7 cells (Fig. 7, B and C). This result was visibly confirmed in the DNMT1-GFP fluorescence experiments (Fig. 8, A-D), and flow cytometry (Fig. 8E) revealed a 3-fold difference in protein accumulation in HMECs versus MCF-7 cells (7.5-versus 2.5-fold accumulation) when the N-terminal 120 amino acids were deleted. However, deletion of the next 60 amino acids (including the PBD and NLS) did not seem to have a further effect on accumulation.

FLAG-tagged DNMT1 constructs were also co-transfected with hemagglutinin (HA)-tagged ubiquitin, immunoprecipitated with anti-FLAG antibodies, immunoblotted, and probed with anti-HA antibodies to determine the presence of ubiquitin modifications. In HMECs, the full-length DNMT1-FLAG protein had significantly more ubiquitination than the N-terminal 120-amino acid deletion mutant, and this increased ubiquitination correlated with the destruction of the protein (Fig. 7, C and E). However, in MCF-7 cells, the full-length protein did not have any detectable increase in ubiquitin modification relative to the N-terminal deletion mutant. These data suggest that the N-terminal 120 amino acids comprise a destruction domain that is responsible for the differential DNMT1 protein levels between HMEC and MCF-7 cells, and that proper ubiquitination and degradation via this destruction domain may be dysfunctional in MCF-7 cells.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 7.
DNMT1-FLAG fusion protein expression. A, schematic of DNMT1-FLAG/GFP fusion protein constructs with 120-aa and 180-aa N-terminal deletions. B, FLAG-tagged DNMT1 full-length and 120-aa N-terminal deletion constructs were transfected into HMEC and MCF-7 cells, and protein levels were determined by anti-FLAG immunoblotting and densitometry (D) with B-actin as a loading control. C, FLAG-tagged DNMT1 full-length and 120-aa N-terminal deletion constructs were co-transfected into HMEC and MCF-7 cells with HA-tagged ubiquitin, immunoprecipitated with anti-FLAG antibodies, immunoblotted, probed with anti-HA antibodies, and quantified by densitometry (E). Error bars are ± S.D. of duplicate lanes.

 
Overexpression of DNMT1 with a Deleted N-terminal Destruction Domain in HMECs Caused Increased Genomic 5-Methylcytosine Levels Relative to Overexpression of the Full-length Protein—FLAG-tagged DNMT1 constructs were overexpressed in HMECs for 72 h with GFP overexpression as a negative control, and genomic 5-methylcytosine levels were determined relative to unmethylated cytosine by high-performance liquid chromatography/mass spectrometry. Full-length DNMT1-FLAG and GFP overexpression did not differ significantly from each other in 5-methylcytosine levels, but the overexpression of the N-terminal 120- and 180-amino acid deletion mutants caused significantly increased 5-methylcytosine levels relative to full-length protein (Fig. 9). However, when the CG island of the tumor suppressor gene CDKN2A was studied by bisulfite sequencing analysis, no extensive CG methylation was found above the detection limit of 6% of alleles (data not shown). These data suggest that expression of the dysregulated DNMT1 isoforms, although insufficient to methylate this particular tumor suppressor gene, can nonetheless cause genomic cytosine hypermethylation.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have demonstrated that DNMT1 protein is overexpressed and discordant with proliferative fraction in a variety of breast cancers relative to histologically normal tissue, in MCF-7 breast cancer cells relative to normal HMECs, and in colon cancer HCT116 cells and prostate cancer LNCaP cells relative to normal PrECs and NHDFs. These findings are in agreement with several studies of many different cancer types suggesting that this dysregulation of DNMT1 protein levels may be a general feature of cancer cells and a significant contributor to carcinogenesis (28-30). Normal colonic tissue was found to express DNMT1 in concordance with the S-phase marker Ki67, whereas many adenocarcinomas were shown to overexpress the protein in excess of S-phase fraction (28). In the stomach, DNMT1 overexpression, disproportionate with proliferative activity, was detected in 72% of gastric cancers and correlated with significantly poorer tumor differentiation and hypermethylation at multiple CpG islands (29). Finally, in the bladder, transitional cell carcinomas and dysplastic urothelium were reported to have significantly increased DNMT1 protein levels, which correlated with histological grade (30). Even more strikingly, however, DNMT1 protein was overexpressed in non-cancerous urothelium from patients with bladder cancer before any significant increase in proliferative index, possibly implicating DNMT1 in the development of premalignant lesions (30).



View larger version (57K):
[in this window]
[in a new window]
  		FIG. 8.
DNMT1-GFP fusion protein expression. A-D, GFP-tagged DNMT1 full-length, 120-aa, and 180-aa N-terminal deletion constructs were transfected into HMEC and MCF-7 cells, and protein levels were determined by direct GFP fluorescence and flow cytometry (E). 4',6-Diamidino-2-phenylindole nucleic acid stain was used as a control for cell number. Flow cytometry fluorescence analysis was performed on a gated population of cells, excluding untransfected background cells, yielding the fraction of total cells positive for GFP fluorescence and their mean fluorescence intensity. Total GFP fluorescence was calculated from the product of the fraction of positive cells and their mean fluorescence intensity.

 
We have found a significant increase in DNMT1 protein stability by radiolabeled pulse-chase analysis and cycloheximide translation inhibition experiments, providing a mechanism for DNMT1 overexpression in MCF-7 cells relative to HMECs without concomitantly increased DNMT1 mRNA levels. This differential protein accumulation was dependent on the N-terminal 120 amino acids, which seemed to function normally as a destruction domain in HMECs causing ubiquitination and proteasome-mediated degradation. Interestingly, an oocyte-specific DNMT1 isoform, DNMT1o, which is translated from a downstream start codon from an alternatively spliced transcript, thus lacking the N-terminal 118 amino acids (31), has been shown to be significantly stabilized in vivo relative to the full-length somatic DNMT1 by cycloheximide translation experiments (32). This stabilization results in an accumulation of ~5-fold of protein levels, similar to our findings in MCF-7 breast cancer cells relative to HMECs. These data support a hypothesis that DNMT1 protein in MCF-7 cells may abnormally behave as a DNMT1o isoform, whether due to a cis dysfunction (e.g. improper post-translational modifications) or trans dysfunction (e.g. mutated/absent regulatory factors that bind the N terminus).



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 9.
5-Methylcytosine levels in HMECs following overexpression of DNMT1 N-terminal deletion mutants. FLAG-tagged DNMT1 full-length, 120-aa, and 180-aa N-terminal deletion mutants were overexpressed in HMECs for 72 h with GFP transfection as a negative control. Genomic DNA was isolated and nuclease-digested, and 5-methylcytosine levels relative to unmethylated cytosine were determined by high-performance liquid chromatography/mass spectrometry. Error bars are ± S.D. of triplicate measurements.

 
A DNA methyltransferase-associated protein, DMAP1, has been found to interact with the N-terminal 120 amino acids of DNMT1 through a yeast two-hybrid assay (33). DMAP1 also interacts with TSG101, a putatively inactive ubiquitin ligase that has been associated with breast cancer tumorigenesis by some but not all studies (34-37). Regardless of its role in the pathogenesis of breast cancer, TSG101 may play an important role in either regulating ubiquitination or actually conjugating ubiquitin itself (38-41), and this may influence the stability of DNMT1. DMAP1 has also been shown to interact with a protein called Daxx by yeast two-hybrid assays, and co-immunoprecipitation experiments have shown these two proteins form a complex of which DNMT1 is a member (42). Daxx overexpression was reported to protect DMAP1 from proteasome-mediated degradation, and it is quite possible that Daxx may stabilize not only DMAP1, but DNMT1 as well. Further work will determine whether this interaction with DMAP1 is necessary for the proper regulation of DNMT1 in normal cells.

We have shown that overexpression of the non-degradable DNMT1 N-terminal deletion mutants can cause genomic cytosine hypermethylation that is not seen with the full-length protein. Although overexpression of these isoforms was not sufficient to cause extensive methylation the CDKN2A tumor suppressor gene, many reasons can account for this including the limited ability of bisulfite sequencing analysis to detect low copy methylated alleles (below 6%), the limited duration of this experiment (72 h), and the lack of cell replication following transfection, which may be necessary to allow selection and outgrowth of the hypermethylated variants. Remarkably, the increased levels of genomic 5-methylcytosine at 6.0% and 6.5% observed were quite significant considering that HMECs have a baseline cytosine methylation of ~5% and an upper limit of about 10% as estimated by SssI methylase treatment of human white blood cell DNA, which should methylate all genomic CG sites.2 Furthermore, given that the transfection efficiency of HMECs is ~30%, the total 5-methylcytosine levels we observed with the dysregulated isoform transfections were likely to be even higher in the sub-population of transfected cells. It is also noteworthy that the highest genomic 5-methylcytosine levels achieved was with transfection of the DNMT1 N-terminal deletion lacking the PCNA binding domain, which may implicate not only cell cycle dysregulation of DNMT1 in causing cytosine hypermethylation, but mislocalization as well.

Future studies will further examine the contribution of overexpressed DNMT1 to mammary carcinogenesis and whether this "DNMT1o-like" phenotype is physiologically relevant. Some of the first studies implicating DNMT1 overexpression in transformation in NIH3T3 cells and fibroblasts serendipitously used the earliest cloned DNMT1 sequence, which lacked the true start codon of the full-length DNMT1 protein, resulting in the translation of the DNMT1o-like isoform. Stable transfection and overexpression of this DNMT1o-like protein led to methylation of endogenous CpG islands in genes such as endoplasmic reticulum and ECAD, known to be frequently methylated in breast tumors (8, 11), suggesting that DNMT1o-like activity may be relevant to cancer development. Whether breast cancers, which show an overexpression of DNMT1, also have the highest incidence of methylation abnormalities in these tumor suppressor genes has not been tested.

Finally, although few clinical trials of DNA methyltransferase inhibitors have been performed for solid tumors, some studies have shown a modest antitumor activity of 5-azacytidine in breast cancer ranging from 6.4% to 63% (43-45). In addition, it has been shown that the toxicity of 5-azacytidine, and perhaps its anti-tumor activity, is at least partially mediated through its ability to covalently trap DNMT1 on the DNA duplex rather than simply through demethylation (46). This study also predicted that tumors overexpressing DNA methyltransferase should be more susceptible to 5-azacytidine, and it will be interesting to examine whether tumors overexpressing DNMT1 protein may have greater sensitivity to DNA methyltransferase inhibitors in a manner analogous to the correlation between topoisomerase overexpression and sensitivity to topoisomerase inhibitors (47).


   FOOTNOTES
 
* This work was supported by National Institutes of Health Specialized Program of Research Excellence breast cancer Grant CA88843 and by NCI, National Institutes of Health Grants CA58236 and CA70196. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed: Cancer Research Bldg. 151, 1650 Orleans St., Baltimore, MD 21231-1000. Tel.: 410-614-1661; Fax: 410-502-9817; E-mail: bnelson{at}jhmi.edu.

1 The abbreviations used are: PCNA, proliferating cell nuclear antigen; HMEC, human mammary epithelial cell; PrEC, prostate epithelial cell; NHDF, normal human dermal fibroblast; RT, reverse transcription; GFP, green fluorescent protein; aa, amino acid(s); HA, hemagglutinin; PBD, PCNA binding domain; NLS, nuclear localization signal. Back

2 S. Yegnasubramanian, personal communication, 02/08/2005. Back


   ACKNOWLEDGMENTS
 
We thank Leslie Meszler of the Johns Hopkins Cell Imaging Facility for expertise with flow cytometry, Byron Lee for the generous gift of cloned DNMT1, Robert O'Meally for the development and assistance with mass spectrometry analysis, and Bert Vogelstein for the generous gift of HCT116 DNMT1-/- cells.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Jones, P. A., and Laird, P. W. (1999) Nat. Genet. 21, 163-167[CrossRef][Medline] [Order article via Infotrieve]
  2. Jones, P. A., and Baylin, S. B. (2002) Nat. Rev. Genet. 3, 415-428[Medline] [Order article via Infotrieve]
  3. Leonhardt, H., Page, A. W., Weier, H. U., and Bestor, T. H. (1992) Cell 71, 865-873[CrossRef][Medline] [Order article via Infotrieve]
  4. Chuang, L. S., Ian, H. I., Koh, T. W., Ng, H. H., Xu, G., and Li, B. F. (1997) Science 277, 1996-2000[Abstract/Free Full Text]
  5. Bestor, T. H. (2000) Hum. Mol. Genet. 9, 2395-2402[Abstract/Free Full Text]
  6. Laird, P. W., and Jaenisch, R. (1996) Annu. Rev. Genet. 30, 441-464[CrossRef][Medline] [Order article via Infotrieve]
  7. Esteller, M., Corn, P. G., Baylin, S. B., and Herman, J. G. (2001) Cancer Res. 61, 3225-3229[Abstract/Free Full Text]
  8. Yang, X., Yan, L., and Davidson, N. E. (2001) Endocr. Relat. Cancer 8, 115-127[Abstract]
  9. Yegnasubramanian, S., Kowalski, J., Gonzalgo, M. L., Zahurak, M., Piantadosi, S., Walsh, P. C., Bova, G. S., De Marzo, A. M., Isaacs, W. B., and Nelson, W. G. (2004) Cancer Res. 64, 1975-1986[Abstract/Free Full Text]
  10. Wu, J., Issa, J. P., Herman, J., Bassett, D. E., Jr., Nelkin, B. D., and Baylin, S. B. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 8891-8895[Abstract/Free Full Text]
  11. Vertino, P. M., Yen, R. W., Gao, J., and Baylin, S. B. (1996) Mol. Cell. Biol. 16, 4555-4565[Abstract]
  12. Robert, M. F., Morin, S., Beaulieu, N., Gauthier, F., Chute, I. C., Barsalou, A., and MacLeod, A. R. (2003) Nat. Genet. 33, 61-65[CrossRef][Medline] [Order article via Infotrieve]
  13. Kautiainen, T. L., and Jones, P. A. (1986) J. Biol. Chem. 261, 1594-1598[Abstract/Free Full Text]
  14. Bender, C. M., Pao, M. M., and Jones, P. A. (1998) Cancer Res. 58, 95-101[Abstract/Free Full Text]
  15. Belinsky, S. A., Klinge, D. M., Stidley, C. A., Issa, J. P., Herman, J. G., March, T. H., and Baylin, S. B. (2003) Cancer Res. 63, 7089-7093[Abstract/Free Full Text]
  16. Ferguson, A. T., Vertino, P. M., Spitzner, J. R., Baylin, S. B., Muller, M. T., and Davidson, N. E. (1997) J. Biol. Chem. 272, 32260-32266[Abstract/Free Full Text]
  17. Eads, C. A., Nickel, A. E., and Laird, P. W. (2002) Cancer Res. 62, 1296-1299[Abstract/Free Full Text]
  18. Laird, P. W., Jackson-Grusby, L., Fazeli, A., Dickinson, S. L., Jung, W. E., Li, E., Weinberg, R. A., and Jaenisch, R. (1995) Cell 81, 197-205[CrossRef][Medline] [Order article via Infotrieve]
  19. Bakin, A. V., and Curran, T. (1999) Science 283, 387-390[Abstract/Free Full Text]
  20. Robertson, K. D., Ait-Si-Ali, S., Yokochi, T., Wade, P. A., Jones, P. L., and Wolffe, A. P. (2000) Nat. Genet. 25, 338-342[CrossRef][Medline] [Order article via Infotrieve]
  21. el-Deiry, W. S., Nelkin, B. D., Celano, P., Yen, R. W., Falco, J. P., Hamilton, S. R., and Baylin, S. B. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 3470-3474[Abstract/Free Full Text]
  22. Issa, J. P., Vertino, P. M., Wu, J., Sazawal, S., Celano, P., Nelkin, B. D., Hamilton, S. R., and Baylin, S. B. (1993) J. Natl. Cancer Inst. 85, 1235-1240[Abstract/Free Full Text]
  23. Belinsky, S. A., Nikula, K. J., Baylin, S. B., and Issa, J. P. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 4045-4050[Abstract/Free Full Text]
  24. Lee, P. J., Washer, L. L., Law, D. J., Boland, C. R., Horon, I. L., and Feinberg, A. P. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 10366-10370[Abstract/Free Full Text]
  25. Eads, C. A., Danenberg, K. D., Kawakami, K., Saltz, L. B., Danenberg, P. V., and Laird, P. W. (1999) Cancer Res. 59, 2302-2306[Abstract/Free Full Text]
  26. Nass, S. J., Ferguson, A. T., El-Ashry, D., Nelson, W. G., and Davidson, N. E. (1999) Oncogene 18, 7453-7461[CrossRef][Medline] [Order article via Infotrieve]
  27. Robertson, K. D., Keyomarsi, K., Gonzales, F. A., Velicescu, M., and Jones, P. A. (2000) Nucleic Acids Res. 28, 2108-2113[Abstract/Free Full Text]
  28. De Marzo, A. M., Marchi, V. L., Yang, E. S., Veeraswamy, R., Lin, X., and Nelson, W. G. (1999) Cancer Res. 59, 3855-3860[Abstract/Free Full Text]
  29. Etoh, T., Kanai, Y., Ushijima, S., Nakagawa, T., Nakanishi, Y., Sasako, M., Kitano, S., and Hirohashi, S. (2004) Am. J. Pathol. 164, 689-699[Abstract/Free Full Text]
  30. Nakagawa, T., Kanai, Y., Saito, Y., Kitamura, T., Kakizoe, T., and Hirohashi, S. (2003) J. Urol. 170, 2463-2466[CrossRef][Medline] [Order article via Infotrieve]
  31. Gaudet, F., Talbot, D., Leonhardt, H., and Jaenisch, R. (1998) J. Biol. Chem. 273, 32725-32729[Abstract/Free Full Text]
  32. Ding, F., and Chaillet, J. R. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 14861-14866[Abstract/Free Full Text]
  33. Rountree, M. R., Bachman, K. E., and Baylin, S. B. (2000) Nat. Genet. 25, 269-277[CrossRef][Medline] [Order article via Infotrieve]
  34. Zhong, Q., Chen, Y., Jones, D., and Lee, W. H. (1998) Cancer Res. 58, 2699-2702[Abstract/Free Full Text]
  35. Lo, Y. F., Chen, T. C., Chen, S. C., and Chao, C. C. (2000) Breast Cancer Res. Treat 60, 259-266[CrossRef][Medline] [Order article via Infotrieve]
  36. Trink, B., Pai, S. I., Spunt, S. L., Raman, V., Cairns, P., Jen, J., Gabrielson, E., Sukumar, S., and Sidransky, D. (1998) Oncogene 16, 2815-2818[CrossRef][Medline] [Order article via Infotrieve]
  37. Carney, M. E., Maxwell, G. L., Lancaster, J. M., Gumbs, C., Marks, J., Berchuck, A., and Futreal, P. A. (1998) J. Soc. Gynecol. Investig. 5, 281-285[Medline] [Order article via Infotrieve]
  38. Burgdorf, S., Leister, P., and Scheidtmann, K. H. (2004) J. Biol. Chem. 279, 17524-17534[Abstract/Free Full Text]
  39. Lu, Q., Hope, L. W., Brasch, M., Reinhard, C., and Cohen, S. N. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 7626-7631[Abstract/Free Full Text]
  40. Myers, E. L., and Allen, J. F. (2002) J. Virol. 76, 11226-11235[Abstract/Free Full Text]
  41. Ponting, C. P., Cai, Y. D., and Bork, P. (1997) J. Mol. Med. 75, 467-469[CrossRef][Medline] [Order article via Infotrieve]
  42. Muromoto, R., Sugiyama, K., Takachi, A., Imoto, S., Sato, N., Yamamoto, T., Oritani, K., Shimoda, K., and Matsuda, T. (2004) J. Immunol. 172, 2985-2993[Abstract/Free Full Text]
  43. Bellet, R. E., Mastrangelo, M. J., Engstrom, P. F., Strawitz, J. G., Weiss, A. J., and Yarbro, J. W. (1974) Cancer Chemother. Rep. 58, 217-222[Medline] [Order article via Infotrieve]
  44. Cunningham, T. J., Nemoto, T., Rosner, D., Knight, E., Taylor, S., Rosenbaum, C., Horton, J., and Dao, T. (1974) Cancer Chemother. Rep. 58, 677-681[Medline] [Order article via Infotrieve]
  45. Weiss, A. J., Metter, G. E., Nealon, T. F., Keanan, J. P., Ramirez, G., Swaiminathan, A., Fletcher, W. S., Moss, S. E., and Manthei, R. W. (1977) Cancer Treat Rep. 61, 55-58[Medline] [Order article via Infotrieve]
  46. Juttermann, R., Li, E., and Jaenisch, R. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 11797-11801
  47. Cardoso, F., Durbecq, V., Larsimont, D., Paesmans, M., Leroy, J. Y., Rouas, G., Sotiriou, C., Renard, N., Richard, V., Piccart, M. J., and Di Leo, A. (2004) Int. J. Oncol. 24, 201-209[Medline] [Order article via Infotrieve]
  48. Rhee, I., Jair, K. W., Yen, R. W., Lengauer, C., Herman, J. G., Kinzler, K. W., Vogelstein, B., Baylin, S. B., and Schuebel, K. E. (2000) Nature 404, 1003-1007[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Mol Cancer ResHome page
Q. Zhou, A. T. Agoston, P. Atadja, W. G. Nelson, and N. E. Davidson
Inhibition of Histone Deacetylases Promotes Ubiquitin-Dependent Proteasomal Degradation of DNA Methyltransferase 1 in Human Breast Cancer Cells
Mol. Cancer Res., May 1, 2008; 6(5): 873 - 883.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
A. T. Agoston, P. Argani, A. M. De Marzo, J. L. Hicks, and W. G. Nelson
Retinoblastoma Pathway Dysregulation Causes DNA Methyltransferase 1 Overexpression in Cancer via MAD2-Mediated Inhibition of the Anaphase-Promoting Complex
Am. J. Pathol., May 1, 2007; 170(5): 1585 - 1593.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
A. Miremadi, M. Z. Oestergaard, P. D.P. Pharoah, and C. Caldas
Cancer genetics of epigenetic genes
Hum. Mol. Genet., April 15, 2007; 16(R1): R28 - R49.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
J. Torrisani, A. Unterberger, S. R. Tendulkar, K. Shikimi, and M. Szyf
AUF1 Cell Cycle Variations Define Genomic DNA Methylation by Regulation of DNMT1 mRNA Stability
Mol. Cell. Biol., January 1, 2007; 27(1): 395 - 410.
[Abstract] [Full Text] [PDF]

Home page