Activation of p53 transcriptional activity by SMRT: a histone deacetylase 3-independent function of a transcriptional corepressor

Abstract

The silencing mediator of retinoic acid and thyroid hormone receptors (SMRT) is an established histone deacetylase 3 (HDAC3)-dependent transcriptional corepressor. Microarray analyses of MCF-7 cells transfected with control or SMRT small interfering RNA revealed SMRT regulation of genes involved in DNA damage responses, and the levels of the DNA damage marker γH2AX as well as poly(ADP-ribose) polymerase cleavage were elevated in SMRT-depleted cells treated with doxorubicin. A number of these genes are established p53 targets. SMRT knockdown decreased the activity of two p53-dependent reporter genes as well as the expression of p53 target genes, such as CDKN1A (which encodes p21). SMRT bound directly to p53 and was recruited to p53 binding sites within the p21 promoter. Depletion of GPS2 and TBL1, components of the SMRT corepressor complex, but not histone deacetylase 3 (HDAC3) decreased p21-luciferase activity. p53 bound to the SMRT deacetylase activation domain (DAD), which mediates HDAC3 binding and activation, and HDAC3 could attenuate p53 binding to the DAD region of SMRT. Moreover, an HDAC3 binding-deficient SMRT DAD mutant coactivated p53 transcriptional activity. Collectively, these data highlight a biological role for SMRT in mediating DNA damage responses and suggest a model where p53 binding to the DAD limits HDAC3 interaction with this coregulator, thereby facilitating SMRT coactivation of p53-dependent gene expression.

FIG 1

Profiling of SMRT-regulated gene expression in MCF-7 cells. (A) Scheme of siRNA-mediated depletion of SMRT and subsequent analysis of mRNA by microarray. MCF-7 cells were transfected with 20 pmol of control siRNA (siControl) or SMRT-specific siRNA (siSMRT). Cells were harvested 72 h after transfection, and the RNA was isolated and analyzed by an Affymetrix GeneChip microarray, followed by gene ontological analysis. (B) Heat map showing that SMRT depletion leads to decreased expression of genes involved in the DNA damage response. (C) Depletion of SMRT expression in MCF-7 cells leads to a decrease in BRCA1 and CHEK1 mRNAs. MCF-7 cells were transfected with 20 pmol of control siRNA (siCon) or SMRT-specific siRNA (siSMRT), and at 48 h thereafter, RNA was isolated and the levels of BRCA1 and CHEK1 mRNAs were quantitated by RT-qPCR and normalized to the signals obtained for 18S RNA. Values are the averages of four experiments ± SEMs. (D) Representative Western blot showing the efficiency of SMRT depletion by SMRT-specific siRNA in MCF-7 cells.

FIG 2

Depletion of SMRT expression compromises the DNA damage response in MCF-7 cells. (A) Western blot analysis showing the impact of SMRT depletion on doxorubicin (Dox)-induced DNA damage. MCF-7 cells were transfected with control (Con) or SMRT-specific siRNA. After 48 h, cells were treated with 0.1% ethanol (lanes V) or 1 nM estradiol (lanes E) for 4 h and subsequently treated with doxorubicin at the indicated concentrations for 16 h. Whole-cell lysates were resolved in SDS-polyacrylamide gels and immunoblotted for γH2AX and SMRT. (B) Saos-2 cells, which are deficient for p53, were analyzed as described for MCF-7 cells in the legend to panel A, except that cells were not treated with hormone. (C) Western blot analysis of PARP cleavage in SMRT-depleted versus control MCF-7 cells with and without doxorubicin treatment. For all blots, actin was used as a loading control.

FIG 3

SMRT regulates expression of p53 target genes in doxorubicin-treated MCF-7 cells. Cells were transfected with 20 pmol of control siRNA or SMRT-specific siRNA (siSMRT) and after 48 h were treated with vehicle (PBS; filled bars) or doxorubicin (1 μM; white bars) for 16 h. RNA was subsequently isolated, and the levels of mRNA for the p21 gene (A), BRCA1 (B), RAD51 (C), and CHEK1 (D) were quantitated by RT-qPCR and normalized to the signals obtained for 18S RNA. The changes in the mRNA levels of the indicated genes are represented relative to the levels of their respective control siRNAs in doxorubicin-treated cells. Values represent the means ± SEMs of data from five independent experiments. P values were determined by Student's t test. *, P ≤ 0.05; **, P ≤ 0.001; ***, P ≤ 0.0001.

FIG 4

SMRT coactivates p53 transcriptional activity in MCF-7 cells. Cells were transfected with 20 pmol of control siRNA or SMRT-specific siRNA (siSMRT), prior to transfection with 1 μg of either the p21 (A) and MDM2 (B) luciferase reporter genes, and treated with vehicle (Veh; 0.1% ethanol) or 1 nM E2 for 24 h. Luciferase activity data are the means ± SEMs of three (p21) and two (MDM2) experiments. (C) MCF-7 cells were transfected with 500 ng of a Flag-tagged expression vector for SMRT or an empty-vector control (basal) along with 500 ng of the p21-Luc reporter gene. Cells were harvested 48 h after transfection, and luciferase activity was measured. The activity for SMRT is shown relative to that for the empty-vector control. Values represent the averages ± SEMs (n = 4). (D) MCF-7 cells were transfected with control siRNA or SMRT-specific siRNA, and 48 h later, the cells were treated with vehicle (PBS; without doxorubicin [−Dox]) or doxorubicin (+Dox; 1 μM) for 16 h. Whole-cell lysates were resolved in SDS-polyacrylamide gels, and the expression of the p21, p53, SMRT, and actin proteins was analyzed by Western blotting. (E) ZR75-1 cells were transfected with 30 pmol of control siRNA or SMRT-specific siRNA. After 48 h, the cells were treated with vehicle (PBS) or doxorubicin (1 μM) for 16 h, followed by RNA isolation and quantitation by RT-qPCR. The p21 mRNA levels were normalized to the signals obtained for 18S RNA. The changes in the mRNA levels are represented relative to the level of control siRNA in doxorubicin-treated cells. Values represent the means ± SEMs of data from two independent experiments. (F) Saos-2 cells were transfected with control siRNA (siC) or SMRT-specific siRNA (siS), and 48 h later, the cells were treated with vehicle (PBS without doxorubicin) or doxorubicin (+Dox; 0.5 μM) for 16 h. Whole-cell lysates were resolved by SDS-PAGE, and the expression of p21, BRCA1, SMRT, and actin proteins was analyzed by Western blotting. (G) MCF-7 cells were transfected with 500 ng of either the 14-3-3σ wild-type (WT) or mutant (MT) luciferase reporter genes, along with 500 ng pCR3.1 (basal) or full-length human SMRT plasmids. After 24 h, cells were treated with vehicle or 1 μM doxorubicin for 24 h. Luciferase activities are presented relative to those obtained for cells transfected with wild-type 14-3-3σ–Luc and SMRT and treated with doxorubicin. Values represent the averages ± SEMs (n = 3). RLU, relative light units.

FIG 5

SMRT associates with p53 binding sites on the p21 gene. (A) Schematic representation of the upstream region of the p21 gene, and the positions of the primers used in qPCR to target the 5′ and 3′ p53 binding and nonspecific sites are indicated by arrows. TSS, transcription start site. (B) Chromatin immunoprecipitation assays showing doxorubicin-induced recruitment of SMRT to p53 binding sites of the p21 gene promoter, as well as the upstream nonspecific (NS; p53 response element-deficient) site (inset). MCF-7 cells were treated with vehicle (Veh; PBS) or doxorubicin (Dox; 1 μM) for 6 h, followed by chromatin isolation and immunoprecipitation using antibodies for SMRT or p53 in parallel with the appropriate IgG negative control. Immunoprecipitated DNA was quantitated by qPCR using the indicated primers. Data represent the averages ± SEMs of three independent experiments.

FIG 6

SMRT binds to p53 both in vivo and in vitro. (A) MCF-7 cells were transfected with an expression vector for Flag epitope-tagged SMRT and treated with increasing concentrations of doxorubicin. Cell lysates were immunoprecipitated (IP) with p53 antibody or IgG, and immunoprecipitates were assessed by Western blotting with Flag antibody (top). Levels of SMRT (middle) and p53 (bottom) for input cell lysates were assessed by Western blotting with their respective antibodies. (B) HCT116 cells were treated with increasing doses of doxorubicin for 16 h, and total cell lysates were used for immunoprecipitation using antibodies for p53 or SMRT in parallel with the appropriate IgG negative control. Immunoprecipitated protein complexes were resolved by SDS-PAGE and analyzed by Western blotting using the indicated antibodies. (C) GST pulldown assays showing the interaction of p53 with full-length SMRTτ protein or its truncated mutants. Shown at the left are schematic representations of full-length SMRT and its truncated mutants. The bacterially expressed and purified GST-tagged p53 protein was incubated with in vitro-translated ³⁵S-labeled SMRTτ or its truncated mutants, and the p53-bound SMRT or deletion mutants were detected by autoradiography. Input represents 20% of the corresponding amount of in vitro-translated ³⁵S-labeled lysates. (D) Densitometric quantification of protein bands from panel C normalized to the amounts in the corresponding input lanes and represented relative to the amount of SMRTτ. Rel., relative.

FIG 7

The C-terminal regulatory domain of p53 is essential for SMRT-p53 interaction. (A) Schematic representations of full-length p53 and its truncated mutant derivatives used in the GST pulldown assay. TAD, transactivation domain; Pro, proline-rich domain; DBD, DNA binding domain; Tet, tetramerization; RD, regulatory domain. (B) The bacterially expressed and purified GST-tagged p53 wild-type protein or its truncated mutants were incubated with in vitro-translated ³⁵S-labeled SMRT-NT, followed by immunoprecipitation of the protein complexes using glutathione-agarose beads and SDS-PAGE. The SMRT-NT that bound to p53 or p53 mutants was detected by autoradiography. Input represents 20% of the amount of in vitro-translated ³⁵S-labeled SMRT-NT lysates. (C) Densitometric quantification of protein bands normalized to the amount in the corresponding input lanes and represented relative to the amount of the p53 wild type.

FIG 8

Effect of HDAC3, GPS2, and TBL1 depletion on p21 mRNA levels. MCF-7 cells were transfected with 20 pmol of control siRNA or siRNA specific for GPS2 (A), TBL1 (B), or HDAC3 (C). After 48 h, the cells were treated with vehicle (PBS; filled bars) or doxorubicin (0.5 μM; white bars), and p21 mRNA levels were quantitated by RT-qPCR and normalized to the signals obtained for 18S RNA. The changes in the p21 mRNA levels were represented relative to the level of control siRNA in doxorubicin-treated cells. Values represent the means ± SEMs of data from four (GPS2 and HDAC3) and three (TBL1) independent experiments. (D) Effect of HDAC3 depletion on p53 levels. MCF-7 cells were transfected with 20 pmol of control or HDAC-specific siRNA, followed by treatment with vehicle or doxorubicin for 24 h. Cells were lysed, and proteins were resolved in SDS-polyacrylamide gels and immunoblotted with p53 and actin (loading control) antibodies. (E) Chromatin immunoprecipitation assay showing doxorubicin-induced recruitment of p53 at the p21 gene promoter relative to that at enhancer sites. MCF-7 cells were treated with vehicle (PBS) or doxorubicin (1 μM) for 6 h, followed by chromatin isolation and immunoprecipitation using p53 antibody in parallel with the IgG negative control. Immunoprecipitated DNA was quantitated by qPCR using the indicated primers. Data represent averages ± SEMs of three independent experiments. The changes in the recruitment levels were represented relative to the amount of p53 recruited to the 5′ enhancer site following doxorubicin treatment. (F) Effect of NCoR depletion in MCF-7 cells on p21 mRNA levels assessed by RT-qPCR. Cells were transfected with control siRNA or NCoR siRNA, and p21 mRNA levels were quantitated as described for panels A to C. P values were determined by Student's t test. *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.0001.

FIG 9

The SMRT DAD binds to p53. (A) Bacterially expressed GST-tagged p53 protein was incubated with the in vitro-translated, ³⁵S-labeled N terminus of SMRT (SMRT-NTΔ36-254), its deletion mutants, or the SMRT DAD fragment (residues 255 to 480) alone. Shown at the right are schematic representations of the SMRT deletion mutants and the DAD fragment. The protein complexes were immunoprecipitated by glutathione-agarose beads and resolved by SDS-PAGE, and the bound SMRT mutants or the DAD fragment was detected by autoradiography. Input represents 20% of the corresponding amount of the in vitro-translated, ³⁵S-labeled lysates. (B) Competition between p53 and HDAC3 for binding to the SMRT DAD. Gal4-SMRT (DAD) was expressed in HCT116 cells and immobilized onto anti-Gal4 agarose beads to serve as bait. To this, in vitro-translated ³⁵S-labeled p53 was added with or without in vitro-translated, ³⁵S-labeled HDAC3. The protein complexes bound to agarose beads were resolved by SDS-PAGE, and the bound HDAC3 and p53 proteins were detected by autoradiography. Input represents 1 or 0.1% of the amount of in vitro-translated ³⁵S-labeled lysates for HDAC3 and p53, respectively. (C) HCT116 cells were transfected with 1 μg expression vectors for control (pCR3.1) or Flag epitope-tagged SMRTτ or SMRT DADm and treated for 16 h with vehicle (−) or 0.5 μM doxorubicin (+). Cell lysates were subjected to immunoprecipitation using antibodies for Flag (lanes F) or an IgG negative control (lanes I). Immunoprecipitated protein complexes were resolved by SDS-PAGE and analyzed by Western blotting (WB) using p53 antibody (top). Relative input levels of p53 (middle) and actin (bottom) in the cell lysates were assessed by Western blot analyses. (D) HCT116 cells were transfected with 250 ng expression vectors for SMRTτ, SMRTΔ36-254, and SMRT DADm or an empty-vector control along with 1 μg of vector containing the p21-Luc reporter gene. Cells were harvested at 48 h after transfection, and luciferase activity was measured. The activity for the SMRT mutants is shown relative to that of SMRTτ. Values represent the averages ± SEMs (n = 2).

FIG 10

Proposed model for SMRT coactivation of p53 transcriptional activity. Model for SMRT coactivation of p53 transcriptional activity. In response to DNA damage, p53 expression is stabilized and p53 binds to the regulatory regions of its target genes. The C terminus of p53 binds to the DAD region of SMRT, effectively limiting the ability of HDAC3 to bind to p53-bound SMRT and repress p53 target genes. Instead, a SMRT complex containing GPS2, TBL1, and, potentially, CBP/p300 is able to achieve an open chromatin structure with acetylated (Ac) histones that promotes expression of p53 target genes.