5-Aza-deoxycytidine induces selective degradation of DNA methyltransferase 1 by a proteasomal pathway that requires the KEN box, bromo-adjacent homology domain, and nuclear localization signal

Abstract

5-Azacytidine- and 5-aza-deoxycytidine (5-aza-CdR)-mediated reactivation of tumor suppressor genes silenced by promoter methylation has provided an alternate approach in cancer therapy. Despite the importance of epigenetic therapy, the mechanism of action of DNA-hypomethylating agents in vivo has not been completely elucidated. Here we report that among three functional DNA methyltransferases (DNMT1, DNMT3A, and DNMT3B), the maintenance methyltransferase, DNMT1, was rapidly degraded by the proteasomal pathway upon treatment of cells with these drugs. The 5-aza-CdR-induced degradation, which occurs in the nucleus, could be blocked by proteasomal inhibitors and required a functional ubiquitin-activating enzyme. The drug-induced degradation occurred even in the absence of DNA replication. Treatment of cells with other nucleoside analogs modified at C-5, 5-fluorodeoxyuridine and 5-fluorocytidine, did not induce the degradation of DNMT1. Mutation of cysteine at the catalytic site of Dnmt1 (involved in the formation of a covalent intermediate with cytidine in DNA) to serine (CS) did not impede 5-aza-CdR-induced degradation. Neither the wild type nor the catalytic site mutant of Dnmt3a or Dnmt3b was sensitive to 5-aza-CdR-mediated degradation. These results indicate that covalent bond formation between the enzyme and 5-aza-CdR-incorporated DNA is not essential for enzyme degradation. Mutation of the conserved KEN box, a targeting signal for proteasomal degradation, to AAA increased the basal level of Dnmt1 and blocked its degradation by 5-aza-CdR. Deletion of the catalytic domain increased the expression of Dnmt1 but did not confer resistance to 5-aza-CdR-induced degradation. Both the nuclear localization signal and the bromo-adjacent homology domain were essential for nuclear localization and for the 5-aza-CdR-mediated degradation of Dnmt1. Polyubiquitination of Dnmt1 in vivo and its stabilization upon treatment of cells with a proteasomal inhibitor indicate that the level of Dnmt1 is controlled by ubiquitin-dependent proteasomal degradation. Overexpression of the substrate recognition component, Cdh1 but not Cdc20, of APC (anaphase-promoting complex)/cyclosome ubiquitin ligase reduced the level of Dnmt1 in both untreated and 5-aza-CdR-treated cells. In contrast, the depletion of Cdh1 with small interfering RNA increased the basal level of DNMT1 that blocked 5-aza-CdR-induced degradation. Dnmt1 interacted with Cdh1 and colocalized in the nucleus at discrete foci. Both Dnmt1 and Cdh1 were phosphorylated in vivo, but only Cdh1 was significantly dephosphorylated upon 5-aza-CdR treatment, suggesting its involvement in initiating the proteasomal degradation of DNMT1. These results demonstrate a unique mechanism for the selective degradation of DNMT1, the maintenance DNA methyltransferase, by well-known DNA-hypomethylating agents.

FIG. 1.

DNMT1 is selectively degraded by 5-aza-CdR treatment of mammalian cells. (A) Dnmt1, Dnmt3a, and Dnmt3b protein levels in P1798 cell extracts. Western blot analysis of Dnmt1/3a/3b with WCEs from mouse lymphosarcoma cells (100 μg of protein) treated with 5-aza-CdR (2.5 μM) for various times with antibodies specific for these proteins. The blots were reprobed with β-tubulin to show equal loading of proteins. (B) DNMT1, DNMT3A, and DNMT3B protein levels in extracts from HeLa cells treated with various concentrations of 5-aza-CdR. Western blot analysis with WCEs (250 μg of protein) from HeLa cells. The smaller (70-kDa) subunit of Ku antigen was used as the control to determine equal loading of proteins. (C and D) The 5-aza-CdR concentration required to deplete DNMT1 is proportional to the endogenous DNMT1 level. Immunoblot analysis of DNMT1 with WCEs (200 μg of protein) from wild-type and adenomatous polyposis coli (APC)-mutated colon cancer cells treated with various concentrations of 5-aza-CdR. The blots were reprobed with β-tubulin to demonstrate comparable levels of proteins in the lanes. For quantitation, the levels in untreated cells were taken as 1. (E) Activation of MT-I, human MLH1, and MGMT genes after 5-aza-CdR treatment. Total RNA was isolated from P1798 cells treated with 5 μM 5-aza-CdR for 24 h. The expression of MT-I, human MLH1, and MGMT mRNAs was analyzed by RT-PCR. 18S RNA was amplified from all of the samples to normalize RNA input. (F) HeLa cells were either left untreated or treated with 10 μM flucytosine (5FC) or 5-fluorodeoxyuridine (5FU) for 12 h followed by treatment with 5-azaC-dR (5 μM) for an additional 12 h. Extracts were subjected to Western blot analysis.

FIG. 2.

5-Aza-CdR-induced degradation of DNMT1 remains unaffected in HeLa cells when DNA synthesis is blocked with aphidicolin. (A) BrdUrd incorporation is abolished in cells treated with aphidicolin. HeLa cells were treated with aphidicolin (20 μg/ml) for 24 h followed by treatment with 5-aza-C or 5-aza-CdR for an additional 12 h. Cells were incubated with BrdUrd (10 μM) for 2 h, washed, fixed, and stained with anti-BrdUrd antibody and eosin Y (to stain the cell body). (B) Thymidine incorporation is significantly inhibited in cells treated with aphidicolin. Cells were treated as described for panel A and incubated with [³H₁]thymidine for 2 h. Tritium incorporation into DNA was measured with a scintillation counter. (C and D) 5-Aza-CdR-induced degradation persists in cells treated with aphidicolin. Western blot analysis of Dnmt1 in extracts from cells treated with drugs as described for panel A. All quantitative results are the mean of three independent experiments.

FIG. 3.

5-Aza-CdR-induced degradation of DNMT1 occurs at the posttranslational level and can be blocked by proteasomal inhibitors. (A) Expression of DNMT1 mRNA is not reduced in cells upon treatment with 5-aza-CdR. The mRNA levels (copy numbers) of DNMTs normalized to 18S rRNA were analyzed by real time RT-PCR of HeLa cells treated with 5-aza-CdR for 24 h. The bar diagram shows the fold change in the mRNA levels for DNMT1, DNMT3A, and DNMT3B following treatment with various concentrations of 5-aza-CdR. The copy number for each DNMT normalized to 18S rRNA is presented below each bar. The results are the mean ± standard error of three independent experiments performed in triplicate. (B and C) 5-Aza-CdR-induced depletion is not due to inhibition of protein biosynthesis. Western blot analysis of WCEs from cells treated with cycloheximide (20 μg/ml) for 1 h before 5-aza-CdR (5 μM) exposure for 3 and 6 h. (D and E) Decreased half-life of Dnmt1 in cells treated with 5-aza-CdR. Cos-7 cells transfected with pcDnmt1-Flag were labeled with [³⁵S]methionine (for 2 h) at 36 h posttransfection, followed by a chase in complete medium containing an excess of unlabeled methionine for various times. Ectopic Dnmt1 was immunoprecipitated from WCEs (100 μg of protein) with anti-Flag antibody, separated by SDS-polyacrylamide gel electrophoresis, and subjected to autoradiography and phosphorimaging analysis. The level of a nonspecific polypeptide (NS) pulled down by anti-Flag antibody did not significantly change after cycloheximide treatment. (F and G) DNMT1 levels in WCEs from HeLa cells treated with various protease inhibitors 1 h before treatment with 5 μM 5-aza-CdR for 12 h. All quantitative results are the mean of three independent experiments.

FIG. 4.

5-Aza-CdR-induced degradation of DNMT1 occurs in the nucleus and can be blocked by a proteasomal inhibitor (lactacystin). (A) Western blot analysis of DNMT1 in nuclear and cytoplasmic fractions. Cells were treated with 20 μM lactacystin for 1 h prior to exposure to 5-aza-CdR (5 μM). Nuclear (100 μg of protein) and cytoplasmic (100 μg of protein) extracts from these cells were subjected to Western blot analysis with anti-DNMT1, anti-Ku70, and anti-GAPDH antibodies. (B) Immunofluorescence staining of DNMT1 and DNMT3A in HeLa cells. Cells grown on coverslips were either left untreated or treated with 5-aza-CdR (5 μM) for various times. The fixed cells were stained with mouse anti-DNMT1 monoclonal antibodies (Imgenex) and rabbit anti-DNMT3A polyclonal antibodies followed by TRITC-conjugated anti-mouse immunoglobulin G (IgG) (for DNMT1) and FITC-conjugated anti-rabbit IgG (for DNMT3A). Nuclei were stained with 4′,6′-diamidino-2-phenylindole (DAPI) and visualized under a fluorescence microscope (Nikon Eclipse 800). (C) Endogenous nuclear Dnmt1 (Cos-7) but not cytoplasmic ectopic mouse Dnmt1 is sensitive to 5-aza-CdR-induced degradation. Cos-7 cells expressing nuclear localization signal-deleted Dnmt1-Flag (ΔNLS) were stained with rabbit anti-DNMT1 antibody (for endogenous enzyme) and mouse anti-Flag monoclonal antibody M2 (for ectopic Dnmt1). FITC-conjugated anti-rabbit IgG and TRITC-conjugated anti-mouse IgG were used to detect the respective proteins.

FIG. 5.

5-Aza-CdR-induced degradation of Dnmt1 occurs in ts20 cells only when E1 is active at a permissive temperature (34°C). (A) 5-Aza-CdR-induced degradation is blocked in E1 mutant cells at a nonpermissive temperature. ts20 cells grown at 34°C were incubated for 24 h at 39°C or 34°C followed by treatment with 5 μΜ 5-aza-CdR for various times. WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Dnmt1/3a/3b and anti-GAPDH antibodies. (B) Quantitative analysis of the data in panel A. Error bars indicate standard errors. (C) DNA methyltransferase activity is sensitive to 5-aza-CdR treatment only in cells with active E1. DNA methyltransferase activity was measured in nuclear extracts from ts20 cells by using poly(dI-dC) as a substrate. Cells were treated with ZLLL (25 μΜ) 1 h before the drug treatment. The numbers in the lower panel represent ³H₁ incorporation into each extract (100 μg of protein). The results are the mean ± standard error of three independent assays. (D) 5-Aza-CdR sensitivity of Dnmt1 is restored at a nonpermissive temperature in ts20 cells expressing wild-type Dnmt1. Western blot analysis of Dnmt1 in WCEs (100 μg of protein) from ts20 and H38-5 cells (ts20 cells transfected with wild-type E1) exposed to 34°C or 39°C before drug treatment. (E) Quantitative analysis of the data in panel D. All quantitative results are the mean ± standard error of three independent experiments. For quantitation, the levels of Dnmt1 in untreated cells were assigned a value of 1.

FIG. 6.

Mutation in the KEN box and deletions in the BAH domain or the NLS stabilize Dnmt1 against 5-aza-CdR-induced degradation, whereas mutation of cysteine in the PCQ motif cannot protect it from degradation. (A) Schematic representation of various mutants of mouse DNMT1 used in the present study. (B) Conserved KEN box in mammalian DNMT1. (C and D) The wild type and CS and catalytic domain deletion mutants are sensitive to 5-aza-CdR-induced degradation. Cells were cotransfected with 0.5 nmol of each of the Dnmt1 expression vectors along with GFP. Cells were distributed equally into three plates at 24 h posttransfection and treated with 5-aza-CdR (5 μM) for 6 and 12 h. WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Flag (for ectopic), anti-Dnmt1 (for endogenous), and anti-GFP antibodies. All quantitative results are the mean ± standard error of three independent experiments. (E) Dnmt3a and Dnmt3b (wild type and CS mutant) are resistant to degradation by 5-aza-CdR. Cos-7 cells were transfected with the expression vectors or the pcDNA3.1 vector alone followed by treatment with the inhibitor as described for panel B. WCEs (25 μg of protein) were subjected to Western blot analysis with antibodies against Dnmt3a and Dnmt3b. These proteins were not detected in extracts of Cos-7 cells transfected with the vector alone (data not shown). WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Dnmt1 antibody. (F) The wild type and CS and AAA mutants of Dnmt1 localize predominantly in the nucleus, whereas ΔNLS-Dnmt1 and ΔBAH-Dnmt1 localize exclusively in the cytoplasm. Cells transfected with the expression vectors were stained with anti-Flag antibody (to detect Dnmt1) and 4′,6′-diamidino-2-phenylindole (DAPI).

FIG. 6.

Mutation in the KEN box and deletions in the BAH domain or the NLS stabilize Dnmt1 against 5-aza-CdR-induced degradation, whereas mutation of cysteine in the PCQ motif cannot protect it from degradation. (A) Schematic representation of various mutants of mouse DNMT1 used in the present study. (B) Conserved KEN box in mammalian DNMT1. (C and D) The wild type and CS and catalytic domain deletion mutants are sensitive to 5-aza-CdR-induced degradation. Cells were cotransfected with 0.5 nmol of each of the Dnmt1 expression vectors along with GFP. Cells were distributed equally into three plates at 24 h posttransfection and treated with 5-aza-CdR (5 μM) for 6 and 12 h. WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Flag (for ectopic), anti-Dnmt1 (for endogenous), and anti-GFP antibodies. All quantitative results are the mean ± standard error of three independent experiments. (E) Dnmt3a and Dnmt3b (wild type and CS mutant) are resistant to degradation by 5-aza-CdR. Cos-7 cells were transfected with the expression vectors or the pcDNA3.1 vector alone followed by treatment with the inhibitor as described for panel B. WCEs (25 μg of protein) were subjected to Western blot analysis with antibodies against Dnmt3a and Dnmt3b. These proteins were not detected in extracts of Cos-7 cells transfected with the vector alone (data not shown). WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Dnmt1 antibody. (F) The wild type and CS and AAA mutants of Dnmt1 localize predominantly in the nucleus, whereas ΔNLS-Dnmt1 and ΔBAH-Dnmt1 localize exclusively in the cytoplasm. Cells transfected with the expression vectors were stained with anti-Flag antibody (to detect Dnmt1) and 4′,6′-diamidino-2-phenylindole (DAPI).

FIG. 7.

Dnmt1 is ubiquitinated in vivo and interacts with Cdh1, the substrate recognition subunit of APC/C^Cdh1 (E3 ligase) which is involved in its degradation and which is dephosphorylated upon 5-aza-CdR treatment. (A) Dnmt1 is ubiquitinated in vivo. Cos-7 cells were transfected with wild-type pcDnmt1-Flag, HA-ubiquitin, or both. At 36 h posttransfection, cells were treated with lactacystin (20 μM) for 4 h. WCEs (250 μg of protein) from these cells were immunoprecipitated with anti-Flag or anti-HA antibodies and subjected to immunoblot analysis with both antibodies. (B and C) Overexpression of Cdh1 decreases basal Dnmt1 levels and enhances 5-aza-CdR-induced degradation. Cos-7 cells were transfected with wild-type pcDnmt1-Flag and increasing amounts of pcCdh1, pcCdc20, or pCMV-HA (empty vector). After 24 h, cells were split into two; 12 h later, cells were either left untreated or treated with 5-aza-CdR for an additional 6 h. WCEs (100 μg of protein) were subjected to Western blot analysis with anti-Flag, anti-HA, or anti-Ku70 antibodies. (D and E) DNMT1 levels are elevated in cells (untreated or 5-aza-CdR treated) depleted of Cdh1 by RNA interference. HeLa cells were transfected with Cdh1 siRNA or nonspecific siRNA (NS) followed by 5-aza-CdR exposure for 6 h (see Materials and Methods for details). Western blot analysis (100 μg of protein) was performed with anti-DNMT1, anti-Cdh1, or anti-Cdc20 antibodies. (F) Dnmt1 associates with Cdh1. Cells were cotransfected with Dnmt1-Flag, pcCdh1-HA, or pcCdc20-HA. After 32 h, cells were treated with lactacystin (20 μM) for 4 h. Cell extracts prepared in TNN buffer were subjected to immunoprecipitation with anti-Flag or anti-HA antibodies followed by Western blot analysis with both antibodies. (G and H) Phosphorylation of Cdh1 decreases whereas that of Dnmt1 is not significantly altered in cell treated with 5-aza-CdR. Cos-7 cells were transfected with pcCdc20, pcCdh1, or pcDnmt1-Flag. After 36 h, cells were either left untreated or treated with 5-aza-CdR (5 μM) followed by labeling with ³²P-labeled orthophosphate (1 mCi/ml) for 2 h. WCEs (250 μg of protein) were immunoprecipitated with either anti-HA (for Cdc20 and Cdh1) or anti-Flag (for Dnmt1) antibodies. Precipitated proteins were transferred to a nitrocellulose membrane, which was subjected to phosphorimaging analysis followed by Western blot analysis to measure specific protein levels. The ³²P signal in each band was quantified by using ImageQuant software (Molecular Dynamics). The quantitative results are the mean of three independent experiments.