PMC

H3K4me1/2 demethylation and Pol II release at RPRM promoter. (A) ChIP assays for H3K4me1 and H3K4me2 were performed in MCF7 cells. Recruitment to different positions in the RPRM gene, the pS2 promoter, and the ERα binding site ER3440 used as a control () is shown. Data are represented as % input and are an average of two independent experiments ± SEM. (B) ChIP assays for H3K4me1 and H3K4me2 were performed in MCF7 cells after treatment with vehicle (veh) or E2 for 45 min. The recruitment to different positions in the RPRM gene, the pS2 promoter, and the ERα binding site ER3440 used as a control () is shown. Data are represented as the log₂ E2 fold change compared to vehicle control and are an average of two independent experiments ± SEM. *, P < 0.05, t test comparing recruitment in the presence of E2 versus vehicle. (C) ChIP assays for RNA Pol II were performed in MCF7 cells after treatment with vehicle or E2 for 45 min. Data are represented as the log₂ E2 fold change compared to vehicle control and are an average of three independent experiments ± SEM. *, P < 0.05, t test comparing recruitment in the presence of E2 versus vehicle.

Identification of estrogen-repressed genes in breast cancer cell lines. qRT-PCR was used to test E2-mediated repression of candidate genes, with pS2 and CCNG2 serving as controls. The relative mRNA expression is depicted as ligand-mediated fold change compared to vehicle control. This fold change is reflected in the color intensity, as shown in the color scale, and E2-mediated repression for each gene shown is significant for at least the time point shown. (The raw relative mRNA expression data are shown in Fig. S1 in the supplemental material.) MCF7 (A), ZR-75-1 (B), and T47D (C) breast cancer cells were treated for 4, 8, or 24 h with either vehicle, E2, or a combination of E2 and ICI 182,780. The data are an average from three replicates.

Involvement of p53 in E2 repression of RPRM. (A) ChIP assays for p53 and IgG (negative control) in MCF7 cells after treatment with vehicle or E2 for 45 min. The recruitment of p53 to the RPRM +0.1-kb promoter is shown. Data are representative of three independent experiments. (B) MCF7 cells were transfected with nonspecific siRNA (siNS) or siRNA against p53 followed by either a vehicle (V) or an E2 (E) treatment for 12 h. The knockdown of p53 is shown in the left panel, and its effect on RPRM expression is shown in the right panel. qRT-PCR was used to calculate relative mRNA expression as fold change compared to the vehicle control. The data are an average of three replicates ± SEM. *, P < 0.05, t test comparing E2 repression in the presence of siNS and sip53.

HDAC7 is necessary for repression of RPRM and other E2-repressed genes. (A) RPRM qRT-PCR was performed using RNA from MCF7 cells transfected with nonspecific siRNA (siNS) or siRNA against HDACs 1 to 10 and treated with vehicle (V) or E2 (12 h). (B) HDAC7 ChIP in MCF7 cells treated with vehicle (veh) or E2 for 5, 15, 45, and 60 min. Data are represented as the log₂ E2 fold change compared to vehicle control and are an average of two independent experiments ± SEM. (C) HDAC7 ChIP in MCF7 cells treated with vehicle, E2, or 4-OH-tamoxifen for 15 min. Nonfunctional ERE was used as negative control. *, P < 0.05, t test comparing recruitment in the presence of E2 versus vehicle. (D) HDAC7 ChIP in MCF7 cells, transfected with siNS (ns) or ERα (ER) siRNA, followed by 45 min of treatment with vehicle or E2. *, P < 0.05, t test comparing recruitment in the presence of siNS and ERα siRNA. (E) The E2-mediated repression of candidate genes was tested using RNA from MCF7 cells transfected with siNS or HDAC7 siRNA. The data are an average of three replicates ± SD and are representative of two independent experiments. *, P < 0.05, t test.

ERα is recruited to an E2-responsive RPRM promoter. (A) Simplified model showing consensus sites and positions of primers used in subsequent ChIP assays in the RPRM gene. (B) ChIP assays for ERα in MCF7 cells after treatment with vehicle (veh) or E2 for 45 min. The recruitment of ERα to different positions in the RPRM gene as well as to the negative control region, NFERE, and the pS2 promoter is shown. Data are represented as the log₂ E2 fold change compared to vehicle control and are an average of four independent experiments ± SEM. A t test analysis was performed in which E2 treatment groups were compared to the vehicle group. *, P < 0.05 (C) ChIP assays for ERα in MCF7 cells after treatment with E2 for various time points, as indicated. The recruitment of ERα to different positions in the RPRM gene is shown. Data are represented as the log₂ E2 fold change compared to vehicle control and are an average of two independent experiments. (D) MCF7 cells were transfected with the RPRM promoter and treated with vehicle, E2, or ICI for 24 h, following which luciferease reporter assays were performed. The data represent relative luminescence units (RLU) that are an average of three replicates ± SD and are representative of at least five independent experiments. *, P < 0.05, t test.

FoxA1 is essential for E2 repression of RPRM. (A) Protein expression levels of FoxA1 and β-actin in ERα-negative and -positive cell lines, as indicated, were measured by immunoblotting. (B) RPRM qRT-PCR using RNA from MCF7 cells that were transfected with siNS or FoxA1 siRNA and treated with vehicle (V) or E2 (12 h). The inset shows protein levels of FoxA1 and β-actin, as measured by a Western blot. The data are an average of three replicates ± SD and are representative of three independent experiments. *, P < 0.05, t test. (C) FoxA1 ChIP in MCF7 cells. (Left panel) Data are represented as percent input and are an average of four independent experiments ± SEM. *, P < 0.05, t test comparing recruitment relative to +4.0-kb RPRM control recruitment. (Right panel) Cells were treated with vehicle (veh) or E2 for 45 min. FoxA1 recruitment is represented as the log₂ E2 fold change compared to vehicle control and is an average of three independent experiments ± SEM. *, P < 0.05, t test comparing recruitment in the presence of E2 versus vehicle. (D) FoxA1 ChIP in MCF7 cells, transfected with siNS (ns), or ERα siRNA, followed by 45 min of treatment with vehicle or E2. Data are presented and analyzed as in the right panel in panel C. (E) Endogenous HDAC7 and FoxA1 were coimmunoprecipitated in MCF7 cells, using antibodies as indicated. Five percent input was loaded on the gel.

HDAC7 interacts with ERα and represses its transcriptional induction activity independent of HDAC7's deacetylase activity. (A) Protein expression levels of HDAC7, ERα, and β-actin in cell lines, as indicated, were measured by immunoblotting. (B) Normalized expression data for HDAC7 in ERα-positive and -negative tumors in various studies as listed on the graph obtained from the Oncomine Cancer Profiling Database. (C, upper panel) 293T cells were transfected with ERα and Flag-HDAC7, treated with vehicle or E2 (16 h), lysed, and immunoprecipitated (IP) and immunoblotted as indicated. (C, lower panel) Coimmunoprecipitation of endogenous ERα and HDAC7 in Ly2 breast cancer cells, using antibodies as indicated. Five percent (top panel) or 1% (bottom panel) input was loaded on the gel. (D) MCF7 cells were transfected with ERE-Tk-Luc and increasing amounts of Flag-HDAC7 (0, 50, 100, and 250 ng) or SAFB1 (250 ng [used as a positive control]) () and treated with vehicle or E2. The data represent the average of three replicates ± SD and are representative of five independent experiments. (E) MCF7 cells were transfected with ERE-Tk-Luc and pcDNA3.1 or Flag-HDAC7 constructs and treated as indicated. The data represent averages of three replicates ± SD and are representative of three independent experiments. A t test analysis was performed in which the E2 treatment group for each HDAC7 transfection was compared to the E2 treatment group in the absence of HDAC7 transfection. *, P < 0.05. V, vehicle. (F) ChIP assays for H3K9ac were performed in MCF7 cells after treatment with vehicle (veh) or E2 for 45 min. The recruitment to different positions in the RPRM gene and pS2 promoter is shown. Data are represented as the log₂ E2 fold change compared to the vehicle control and are an average of two independent experiments ± SEM. *, P < 0.05, t test comparing recruitment in absence and presence of E2.

RPRM, a cell cycle inhibitor, is a primary and direct ERα target gene. (A to I) qRT-PCR was used to calculate relative RPRM mRNA expression as ligand-mediated fold change compared to the vehicle control. The data are an average of three replicates ± standard error of the mean (SEM). (A) RPRM knockdown increases S-phase entry of breast cancer cells. MCF7 cells were transfected with either nonspecific siRNA (siNS) or RPRM siRNA (siRPRM), and qPCR was used to calculate relative mRNA expression. The data are an average from three replicates ± SEM. siNS or RPRM siRNA-transfected MCF7 cells were treated with either vehicle or E2 for 16 h, and the percentage of cells in each phase of the cell cycle (G₀/G₁, S, and G₂/M) was determined using fluorescence-activated cell sorter (FACS) analysis. The data are an average of three replicates in four independent experiments. (B) MCF7 cells were transfected with either nonspecific siRNA (siNS) or ERα siRNA (siER) followed by either vehicle or E2 treatment for 12 h. The inset shows protein levels of ERα and β-actin as measured by immunoblotting. WB, Western blotting. (C) ERα-negative HCC1143 breast cancer cells were treated with vehicle or E2 for 12 h, and RNA was measured by qPCR. (D) MCF7 cells were treated with E2 for 1, 2, 3, 4, and 12 h. (E) MCF7 cells were treated with different doses of E2, as indicated, for 8 h. The data are represented as relative mRNA expression of the different doses compared to the E2 dose of 10⁻¹² M. (F) MCF7 cells were treated with E2, 4-OH-tamoxifen (tam), and ICI 182,780, for 16 h, and RNA was isolated for qRT-PCR. veh, vehicle. (G) MCF7 cells were treated with either vehicle, E2, or a combination of vehicle or E2 and cycloheximide (CHX) for 4 and 8 h. (H) MCF7 cells were treated with either actinomycin D (ActD) or a combination of ActD and E2 for 4 and 8 h. The data are an average of three replicates ± standard deviations (SD). (I) Nuclear run-on assays were performed in MCF7 cells treated with E2 for 0.5 h and 7 h, as previously described (). Transcript levels of pS2, CCNG2, and RPRM were determined by qRT-PCR. The error bars represent the SEM from three determinations. In panels B, F, and G, asterisks indicate P < 0.05 by t test.