Integration of estrogen and Wnt signaling circuits by the polycomb group protein EZH2 in breast cancer cells

Abstract

Essential for embryonic development, the polycomb group protein enhancer of zeste homolog 2 (EZH2) is overexpressed in breast and prostate cancers and is implicated in the growth and aggression of the tumors. The tumorigenic mechanism underlying EZH2 overexpression is largely unknown. It is believed that EZH2 exerts its biological activity as a transcription repressor. However, we report here that EZH2 functions in gene transcriptional activation in breast cancer cells. We show that EZH2 transactivates genes that are commonly targeted by estrogen and Wnt signaling pathways. We demonstrated that EZH2 physically interacts directly with estrogen receptor alpha and beta-catenin, thus connecting the estrogen and Wnt signaling circuitries, functionally enhances gene transactivation by estrogen and Wnt pathways, and phenotypically promotes cell cycle progression. In addition, we identified the transactivation activity of EZH2 in its two N-terminal domains and demonstrated that these structures serve as platforms to connect transcription factors and the Mediator complex. Our experiments indicated that EZH2 is a dual function transcription regulator with a dynamic activity, and we provide a mechanism for EZH2 in tumorigenesis.

FIG. 1.

EZH2 enhanced the transactivation of c-Myc and cyclin D1 promoters. (A) EZH2 dose- and time-dependent response of c-Myc (upper panel) and cyclin D1 (lower panel) promoters. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter, a Renilla construct, along with different amounts of an EZH2 expression construct. Eighteen hours after transfection, cells were treated with 100 nM E2 or ethanol (vehicle) for different times and harvested for a luciferase activity assay. Each bar represents the mean ± standard deviation (SD) for triplicate experiments. (B) EZH2 potentiation of c-Myc and cyclin D1 transactivation was dependent on estrogen receptors. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter along with different amounts of EZH2 expression construct. Eighteen hours after transfection, cells were treated with 100 nM E2, 1 mM ICI 182 780, or ethanol (vehicle) for different times and harvested for a luciferase activity assay. Each bar represents the mean ± SD for triplicate experiments. (C) EZH2 enhanced the transactivation of c-Myc and cyclin D1 promoters by SRC-1. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter along with the EZH2 expression construct and a SRC-1 expression construct. Eighteen hours after transfection, cells were treated with 100 nM E2 or ethanol (vehicle) for 24 h and harvested for a luciferase activity assay. Each bar represents the mean ± SD for triplicate experiments. (D) EZH2 enhanced the transactivation of c-Myc and cyclin D1 promoters by β-catenin. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter along with the EZH2 expression construct and a β-catenin expression construct. Forty-eight hours after transfection, cells were treated with E2 and harvested for a luciferase activity assay. Each bar represents the mean ± SD for triplicate experiments.

FIG. 2.

Transcriptional activation of endogenous genes by and functional specificity of EZH2. (A) EZH2 enhanced estrogen-stimulated mRNA expression of c-Myc and cyclin D1. MCF-7 cells were transfected with an empty vector or with an EZH2 expression plasmid. Sixty hours after transfection, the cells were switched to estrogen-deprived medium for 48 h. The cells were then left untreated or treated with 100 nM E2 for different times, as indicated, before the cells were collected for RNA extraction and for gene expression analysis by real-time RT-PCR. (B) EZH1 did not affect the transactivation of c-Myc promoters. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc, a Renilla construct, along with different amounts of an EZH1 expression construct. Eighteen hours after transfection, cells were treated with E2 or ethanol (vehicle) for different times and harvested for a luciferase activity assay. Each bar represents the mean ± standard deviation (SD) for triplicate experiments. (C) EZH2 did not affect androgen receptor-regulated gene transcription. LnCAP cells were transfected with a PSA-luc (prostate-specific antigen-luciferase) construct together with 100 ng or 500 ng of SRC-1 or 100 ng or 500 ng of EZH2 expression constructs as indicated. After growing the cells in the absence of steroids for 48 h, cells were treated with 100 nM dihydrotestosterone (DHT) for 12 h, and the reporter activity was measured. Each bar represents the mean ± SD for triplicate experiments.

FIG. 3.

EZH2 transactivated c-Myc and cyclin D1 expression. (A) Map of cis-acting elements within c-Myc and cyclin D1 promoters which are responsible for EZH2 transactivation. The schematic represents the c-Myc promoter and cyclin D1 promoter, their deletion mutants, synthetic TCF binding sequence (GCTTTGATC), and mutated TCF binding sequence that were fused to a luciferase (Luc) reporter, shown on the left. MCF-7 cells were transfected with the indicated reporter plasmids with or without cotransfection of an EZH2 expression construct in the presence of estrogen. Each bar represents the mean ± standard deviation (SD) for triplicate experiments. (B) Regulation of the expression of endogenous c-Myc and cyclin D1 by EZH2. Left panel: enhancement of c-Myc and cyclin D1 expression by EZH2 overexpression. MCF-7 cells were grown in the absence of estrogen and transfected with the EZH2 expression construct or empty vector. Eighteen hours after transfection, the cells were treated with 100 nM of E2 for another 6 h, and the total proteins were extracted and examined by Western blotting analysis using antibodies against the indicated proteins. Middle panel: abolishment of c-Myc and cyclin D1 expression under EZH2 expression knockdown by RNAi. MCF-7 cells were grown in the presence of estrogen and transfected with the pSUPER-EZH2 siRNA (EZH2 siRNA 1) or pSILENCER-EZH2siRNA (EZH2 siRNA 2) construct. Forty-eight hours after transfection, total proteins were extracted and analyzed for the expression of the indicated proteins by Western blotting. Right panel: confirmation of β-catenin knockdown by RNAi in experiments described below. (C) Knockdown of expression of β-catenin by RNAi affected the transactivation of c-Myc and cyclin D1 promoters by both β-catenin and EZH2. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter, a Renilla construct, along with the EZH2 expression construct, the β-catenin expression construct, or a β-catenin siRNA construct. Twenty-four hours after transfection, cells were collected for a luciferase activity assay. Each bar represents the mean ± SD for triplicate experiments.

FIG. 4.

EZH2 interacted with ERα and β-catenin in vitro and in vivo. (A) Equal amounts of ³⁵S-labeled EZH2 (left panel) or SRC-1 or β-catenin (right panel) were used in GST pull-down experiments. Inputs represent 10% of fractions. (B) Coimmunoprecipitations for proteins that interacted with EZH2. MCF-7 cells were maintained in DMEM supplemented with 10% normal FBS for 72 h. Total proteins were then extracted, and coimmunoprecipitations (IP) were performed with antibodies against the indicated proteins. Inputs represent 10% of fractions. (C) Coimmunoprecipitations for proteins that interacted with EZH2. Upper panel: protein extracts from MCF-7 cells or LnCAP cells in the presence or absence of receptor ligands were immunoprecipitated with antibodies against ERs or AR. The immunoprecipitates were then immunoblotted with antibodies against EZH2 or SRC-1. Middle panel: FLAG-tagged EZH2 was expressed in MCF-7 cells and immunoprecipitated with anti-FLAG followed by immunoblotting with antibodies against the indicated proteins. Bottom panel: confirmation of ERα knockdown by Western blotting. (D) Effect of knockdown of expression of ERα, β-catenin, and EZH2 (left panel) or EED and SUZ12 (right panel) on the promoter activity of c-Myc and cyclin D1. MCF-7 cells were grown in the absence or presence of estrogen and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter, along with the indicated gene constructs. Forty-eight four hours after transfection, cells were collected for a luciferase activity assay. Each bar represents the mean ± standard deviation for triplicate experiments. (E) Confirmation of protein knockdown by Western blotting analysis (left panel) and coimmunoprecipitation with anti-ERα (middle panel) or anti-EED2 (right panel) antibody followed by immunoblotting with antibodies against the indicated proteins. Inputs represent 10% of fractions.

FIG. 5.

(A) Recruitment of ERα, β-catenin, and EZH2 on c-Myc promoters. MCF-7 cells were grown in the absence of estrogen for at least 3 days and left untreated or treated with 100 nM of E2 for 45 min. ChIP assays and ChIP/re-IP experiments were performed using specific antibodies against ERα, β-catenin, and EZH2 by real-time quantitative PCR analysis. The results are means ± standard deviations (SD) from three independent experiments. (B) Confirmation of the silenced expression of the indicated proteins by Western blotting. (C) Recruitment of ERα, β-catenin, and EZH2 on the indicated gene promoters in MCF-7 cells with (middle panel) or without (upper panel) β-catenin knockdown by real-time quantitative PCR analysis. The results are means ± SD from three independent experiments. (C) ChIP analysis of the methylation status of lysine 27 (K27) and lysine 4 (K4) of histone H3 under the indicated experimental conditions.

FIG. 6.

The functional domains of EZH2 that are involved in transactivation of the c-Myc and cyclin D1 promoters. (A) Schematic diagrams of EZH2 deletion mutants. (B) Transactivation activity of EZH2 deletion mutants on the c-Myc or cyclin D1 promoter. MCF-7 cells were grown in DMEM supplemented with 10% normal FBS and were cotransfected with c-Myc-Luc or cyclin D1-Luc reporter, a Renilla construct, along with the EZH2 deletion constructs. Eighteen hours after transfection, cells were collected for a luciferase activity assay. Each bar represents the mean ± standard deviation (SD) for triplicate experiments. (C) GST pull-down experiments for interaction between ERα (upper panel) or β-catenin (lower panel) and the EZH2 deletion mutants. (D) Direct interaction between EZH2 and TRAP220/DRIP205 as detected by GST pull-down experiments (upper panel) and by coimmunoprecipitation with primary antibodies against EZH2 and then blotting with antibodies against TRAP220/DRIP205 (lower panel). Inputs represent 10% of fractions. (E) TRAP220/DRIP205 potentiated EZH2-enhanced transactivation of the c-Myc promoter. MCF-7 cells were grown in the absence of estrogen and were cotransfected with c-Myc-Luc reporter along with EZH2 or a TRAP220/DRIP205 expression construct or with a TRAP220/DRIP205 siRNA construct. Eighteen hours after transfection, cells were treated with 100 nM E2 or ethanol (vehicle) for 16 h and harvested for a luciferase activity assay. Each bar represents the mean ± SD for triplicate experiments. Protein expression was confirmed by Western blotting (right panel).

FIG. 7.

(A) EZH2 promotes cell cycle progression. Serum-starved MCF-7 cells were transfected with an empty vector or the wild-type (wt) or SET domain-deleted (ΔSET) EZH2 expression construct or an EZH2 siRNA construct, plus c-Myc, cyclin D1, or cyclin D2 expression constructs. Twenty-four hours after transfection, cells were treated with estrogen for another 16 h and collected for cell cycle profile analysis by cell flow cytometry. The percentages of cells in G₁, S, and G₂/M phases are shown. (B) EZH2 promoted the G₁/S transition of MCF-7 cells. MCF-7 cells were transfected with an EZH2 expression construct or an EZH2 siRNA construct. Twenty-four hours after transfection, cells were switched to normal medium (serum) or treated with 1 mM ICI 182 780 (arrested) for 24 h before adding E2 for the indicated times. The cells were then collected for cell cycle profile analysis by cell flow cytometry. The percentages of cells in G₁, S, and G₂/M phases under different experimental conditions are shown. (C) Western blotting analysis of protein expression under the indicated experimental conditions. (D) Model for EZH2 involvement in transactivation. See the text for details.