SIRT1 is essential for oncogenic signaling by estrogen/estrogen receptor α in breast cancer

Abstract

The NAD-dependent histone deacetylase silent information regulator 1 (SIRT1) is overexpressed and catalytically activated in a number of human cancers, but recent studies have actually suggested that it may function as a tumor suppressor and metastasis inhibitor in vivo. In breast cancer, SIRT1 stabilization has been suggested to contribute to the oncogenic potential of the estrogen receptor α (ERα), but SIRT1 activity has also been associated with ERα deacetylation and inactivation. In this study, we show that SIRT1 is critical for estrogen to promote breast cancer. ERα physically interacted and functionally cooperated with SIRT1 in breast cancer cells. ERα also bound to the promoter for SIRT1 and increased its transcription. SIRT1 expression induced by ERα was sufficient to activate antioxidant and prosurvival genes in breast cancer cells, such as catalase and glutathione peroxidase, and to inactivate tumor suppressor genes such as cyclin G2 (CCNG2) and p53. Moreover, SIRT1 inactivation eliminated estrogen/ERα-induced cell growth and tumor development, triggering apoptosis. Taken together, these results indicated that SIRT1 is required for estrogen-induced breast cancer growth. Our findings imply that the combination of SIRT1 inhibitors and antiestrogen compounds may offer more effective treatment strategies for breast cancer.

Fig.1. Functional cooperation between E2-ERα and SIRT1

(A) Expression of ERα and SIRT1 mRNAs in ER-positive (ER⁺) breast tumors tissues and adjacent normal tissues as well as in ER-negative (ER⁻) breast tumor tissues and corresponding normal tissues. (B) ERα and SIRT1 mRNA levels were quantified by densitometry. ** p<0.01. (C) SIRT1 mRNA expression in human normal, ER⁺ and ER⁻ breast cancer cell lines. (D) SIRT1 mRNA expression in ZR75.1 cells, treated with or without E2 alone or in combination with TAM and ICI182740. (E) ERα and SIRT1 mRNA levels in ZR75.1 cells, cultured for several passages in regular and E2-free medium.

Fig.2. ERα binds to SIRT1 promoter and forms a complex

(A) SIRT1-pGL3 and SIRT1-pU3R2EGFP reporter constructs and potential ERα binding sites. (B) ZR75.1 cells were transfected with SIRT1-pU3R2EGFP reporter construct and then cultured in E2-free medium for 24 h. Cells were then treated with and without 17β-estradiol (E2), tamoxifen (TAM), 4-hydroxy tamoxifen (4-OH TAM) and ICI182780 for 24 h and the expression of GFP was monitored by epifluorescence. (C) ZR75.1 cells were transfected with SIRT1-pGL3 reporter construct and then cultured in E2-free medium for 24 h. Cells were then treated with and without E2, TAM, 4-OH TAM and ICI182780 for 24 h and luciferase activity was measured using the cell lysates. (D) ZR75.1 cells were transfected with both ER (ERα and ERβ) and ERR (ERRα, ERRβ and ERRγ) family members and Chromatin immunoprecipitation (ChIP) assays were carried out using antibodies specific for these proteins. Genomic DNA present in the immunoprecipitates was examined using the SIRT1 promoter-specific primers by PCR.

Fig.3. SIRT1 interacts with ERα

(A) SIRT1 was immunoprecipitated (IP) from ER⁻ (HMEC, MCF10A, MB231 and MB453) and ER⁺ (MCF7 and ZR75.1) cells using SIRT1 or ERα antibodies. Immunoblotting (IB) was done with SIRT1 or ERα antibodies. (B) Colocalization of SIRT1 and ERα in MCF-7 and ZR75.1 cells. Hoechst staining was used to locate the nucleus. The scale bars represent 20 µm. (C) ER⁻ MCF10A and MB231 cells were transfected with an expression construct of ERα. IP and IB were performed 48 h post-transfection with SIRT1 and ERα antibodies. (D) MCF10A cells were transfected with an ERα expression construct, and colocalization (arrows) was monitored with SIRT1 and ERα antibodies.

Fig.4. SIRT1 interaction requires an active E2-ERα complex

(A) ZR75.1 cells, treated with or without E2 (10 nM), E2 antagonists TAM (1µM), 4-OH TAM (1µM) and ICI182740 (1µM), SIRT1 inhibitors (NAD, 2 mM) and Sirtinol, 25 µM) or Type I and II HDAC inhibitor (TSA, 1 µM) were subjected to IP with an ERα antibody. The immunoprecipitates were used for IB with SIRT1 antibody. (B) ZR75.1 cells were transfected with pcDNA, SIRT1 and p300 expression constructs or treated with NAD⁺, Sirtinol and TSA. Nuclear extracts from these cells were used for IP with acetyl-lysine (Ac-Lys) antibody and the immunoprecipitates were immunoblotted with an ERα antibody. (C) ERα expression was analyzed in control and SIRT1 knockdown ZR75.1 cells. (D) Nuclear extracts from control and SIRT1 knockdown ZR75.1 cells were subjected to IP with Ac-Lys antibody. The immunoprecipitates were then used for immunoblotting with ERα and p53 antibodies.

Fig.5. SIRT1 is required for activation of antioxidant enzymes by E2/ERα

(A) DNA damage, with or without UV radiation, was quantified in control and SIRT1 knockdown ZR75.1 cells with or without E2 treatment. *** p<0.001; *p<0.05. (B) Cell cycle analysis was performed with cells described above, and the sub G0/1 cells were quantified using cell quest software. *** p<0.001. (C) The levels of malondialdehyde (MDA) were measured as a surrogate for lipid peroxidation in control and SIRT1 knockdown ZR75.1 cells with or without E2 treatment. *** p<0.001; *p<0.05. (D) Glutathione peroxidase (Gpx) and (E) superoxide dismutase (SOD) activities were measured in cell lysates. ***p<0.001; ** p<0.01; *p<0.05. (F) Gpx and SOD mRNA levels were analyzed in control and SIRT1 knockdown ZR75.1 cells.

Fig.6. SIRT1 is required for suppression of p53 by E2/ERα

(A) The expression levels of E2 target genes (c-Myc, cyclin D1, BMP7, survivin, p53 and cyclin G2) were analyzed by RT-PCR and western blot in control and SIRT1 knockdown ZR75.1 cells with or without E2 treatment. The red highlights identify the genes whose regulation by E2 is mediated through SIRT1. (B) The acetylation status of p53 and cyclin G2 was analyzed in control and SIRT1 knockdown ZR75.1 cells with or without E2 treatment. (C) ChIP assay was performed to examine the binding of ERα/SIRT1 to E2 target genes.

Fig.7. Loss of SIRT1 is associated with elimination of E2-induced cell survival and mammary tumorigenesis

(A) Control and SIRT1 knockdown ZR75.1 cells were treated with or without E2 for 2 weeks and the resulting colonies were quantified with Giemsa staining. (B) Seven days before tumor induction, female athymic nude mice (6 animals per group; 3 groups) were anesthetized and 3-mm pellets containing E2, 0.18 mg/21-day release, were implanted subcutaneously in the animal’s back. Another 18 animals were used as a control, without E2 treatment. One week after pellet implantation, ZR75.1-pLKO.1 and ZR75.1-SIRT1shRNA (two independent shRNA clones shRNA#2 and shRNA#3) cells (1×10⁷ cells in 100 µl PBS) were injected s.c. in the mammary fat pad of all animals. Tumor volume was used as a measure of tumor growth at 1, 2, 3 and 4 weeks after cell injections. We compared the tumor volume of pLKO.1 and SIRT1shRNA-induced tumor in the presence and absence of E2 and the statistical significance was calculated as ***p<0.001; ** p<0.01; *p<0.05*p<0.05 in each time point (C) Control and SIRT1 knockdown ZR75.1 cells were treated with E2, TAM, 4-OH TAM and ICI182780 for 48 h. Apoptotic cell death was analyzed by FACS. We compared the apoptotic cell death in control pLKO.1 and SIRT1shRNA cells treated with E2, TAM, 4-OH TAM and the statistical significance was calculated as ***p<0.001 (D). Proposed mechanism of E2-ERα and SIRT1 complex in regulation of tumor cell immortalization and tumor development.