The primary antibodies against Bim, ERK and phosphorylated ERK1/2 (pT202/Y204) were purchased from Bioworld Technology, Inc. (St. Louis Park, MN, USA). The primary antibody of goat anti-Trim2 was obtained from Sigma-Aldrich (Steinheim, Germany). Antibodies against AKT and phosphorylated AKT (pS473) were from Cell Signaling Technology (New England Biolabs, Hertfordshire, UK). The antibody against GPER was purchased from Abcam (Cambridge, MA, USA). β-actin antibody, goat antimouse IgG-HRP, and goat antirabbit IgG-HRP were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The reagents of GPER-specific agonists G1 and specific inhibitor G15 were purchased from Tocris Bioscience (St. Louis, MO, USA) and 4-hydroxytamoxifen (TAM), (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) from Sigma-Aldrich. Lipofectamine™ 2000 was purchased from Life Technologies (Carlsbad, CA, USA).

Human breast cancer cells (MCF-7) were routinely grown in RPMI-1640 (Gibco, Mulgrave VIC, Australia) containing 5% fetal bovine serum (FBS; Gibco), 10 μg/ml insulin, 100 IU/ml penicillin and 100 μg/ml streptomycin. The TAM-resistant cells (MCF-7R) (26) were derived from MCF-7 by continuous exposure to TAM (1 μM) diluted in 0.1% ethanol. The MCF-7R cells were then grown in RPMI-1640 medium with 5% FBS plus 100 nM TAM. Before all experiments, cells were switched to phenol red-free DMEM containing 0.5% charcoal-dextran-stripped FBS for 24 h, except where otherwise noted. Culture of MDA-MB-468, MDA-MB-231, BT549 and Hs578T cells were previously described (32).

The expression vector encoding Bim was constructed by inserting human Bim cDNA into pcDNA3.0 vector (Promega, Madison, WI, USA). The pcDNA-Bim and its control vector were transfected into TAM-resistant MCF-7R cell lines using Lipofectamine 2000 (Life Technologies). The siRNAs used in the present study were obtained from Shanghai GenePharma, Co., Ltd. (Shanghai, China). Bim (Bcl2-L11)-specific siRNA or control siRNA (100 nM) were transfected into MCF-7; Trim2-specific siRNA and control siRNA transfected into MCF-7R using Lipofectamine 2000 according to the manufacturer's instructions. The target sequences for Bim (Bcl2-L11) siRNA are 5′-GACAGAGCCACAAGGUAAUTT-3′ and 5′-AUUA CCUUGUGGCUCUGUCTT-3′. The control siRNA sequences are 5′-UUCUCCGAACGUGUCACGUTT-3′ (sense) and 5′-ACGUGACACGUUCGGAGAATT-3′ (antisense). The efficiency of gene knockdown was determined by qRT-PCR and western blot analysis.

Cell growth was determined by 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazoliumbromide (MTT) assay. The cells were plated at 5×10³ cells/well in 96-well microtiter plates. After incubation and treated with designed concentration of TAM for the specified time, MTT (5 mg/ml) was added to each well and incubated for 4 h. The absorbance was recorded on a digital spectrophotometer at a wavelength of 570 nm. The experiment was repeated three times.

A total of 77 breast cancer specimens and their matched primary tumor tissues (PTs) were obtained from patients with breast tumors resected at the First Affiliated Hospital of Chongqing Medical University under permission by the ethics committee of Chongqing Medical University (Chongqing, China). All of the patient details and exclusion criteria have been previously described (26). Immunohistochemistry staining was performed using an SP900 kit (Beijing Zhongshan Golden Bridge Biotechnology, Co., Ltd., Beijing, China) according to the manufacturer's protocol. Briefly, deparaffinized tissue sections of 4 μm thickness were heated for antigen retrieval at 95°C for 15 min in 10 mM citric acid buffer (pH 6.0). After treatment with 3% H₂O₂ for 10 min to quench endogenous peroxidase activity, the sections were blocked using goat serum and then incubated with primary antibodies targeting GPER and Bim at a 1:200 dilution at 4°C for 16 h. Following treatment of horseradish peroxidase-conjugated goat anti-rabbit IgG for 30 min at 37°C, sections were developed using diaminobenzidine (DAB) (Beijing Zhongshan Golden Bridge Biotechnology) and nuclei were counterstained with the Mayer's modified hematoxylin.

GPER scores were assigned as follows: the percentage of positive cells was categorized as 0 (negative staining in all cells), 1 (<1% cells stained), 2 (1–10% of cells stained), 3 (11–40% cells stained), 4 (41–70% cells stained) or 5 (71–100% cells stained), and staining intensity was categorized as 0 (negative), 1 (weak), 2 (moderate) or 3 (strong). Percentage and intensity scores were added to give total immunohistochemical scores, ranging from 0 to 8. Specimens that scored ≥2 were defined as GPER⁺.

The Bim expression was scored based on intensity (0–3) and extent (0, <10%, 1, 10–25%, 2, 26–50% and 3, >50%). The individual categories were multiplied to give a total immunohistochemical score ranging between 0 and 9. Samples that scored ≥3 were defined as positive immunohistochemical results.

Total RNA was extracted using TRizol^® reagent (Takara Biotechnology, Dalian, China) from MCF-7 and MCF-7R cells and reverse transcription was performed using the PrimeScript RT reagent kit (Takara Biotechnology) following the manufacturer's instruction. Quantitative real-time PCR was performed with SYBR Premix Ex Taq™ II (Takara Biotechnology). The specific primers for Bim are: 5′-CCTTTCTTGGCCCTT GTTCC-3′ (sense) and 5′-TTGTGGCTCTGTCTGTAGGG-3′ (antisense); the specific primers for Trim2: 5′-CACCAAGGA CAAAGACGGTG-3′ (sense) and 5′-ATCAGCGGATCGGAT CACTT-3′ (antisense). β-actin was used as internal control for normalizing different samples. The primer sequences for α-actin are: 5′-TGACGTGGACATCCGCAAAG-3′ (sense) and 5′-CTGGAAGGTGGACAGCGAGG-3′ (antisense). All experiments were performed at least three times.

The total protein was acquired using RIPA protein extraction buffer with protease inhibitor (Beyotime Institute of Biotechnology, Haimen, China). Cell lysates were electrophoresed with 10 or 12% SDS-PAGE, and the specific primary antibody targeting to each indicated protein was incubated with the membranes at 4°C overnight. The membranes were then incubated with the appropriate secondary antibody conjugated to horseradish peroxidase for 1 h and visualized by using the enhanced chemiluminescence imaging system (Amersham Pharmacia Biotech, Tokyo, Japan). The gray level of each band was quantified using the Quantity One 4.62 software (Bio-Rad Laboratories, Hercules, CA, USA). The results were expressed as fold change relative to the loading control (β-actin).

Cells were seeded into 6-wells plates and grown to 60% confluence. After cultured in FBS- and phenol-free medium for 24 h, the cells were then treated with or without TAM for another 24 h. At the end of the treatment, cells were washed with phosphate-buffered saline (PBS) twice and stained with 5 ml Annexin V-FITC and 5 ml propidium iodide following the manufacturer's instructions (Nanjing KeyGen Biotech, Co., Ltd., Nanjing, China). Apoptotic cells were determined using a BD FACScan flow cytometer (BD Biosciences, Mansfield, MA, USA).

Coimmunoprecipitation was performed as previously described (33,34). MCF-7 and MCF-7R cells were lysed in cell lysis buffer containing 50 mM HEPES (pH 7.2), 150 mM NaCl, 50 mM Tris-HCl, 1.0% Nonidet P-40, 1 mM EDTA containing protease inhibitors (1.0 g/ml aprotinin, 0.5 g/ml leupeptin, 1.0 g/ml pepstatin and 10 g/ml PMSF). The extracts were pre-cleared by rocking them at 4°C with washed protein G-agarose beads (Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The pre-cleared extracts were immunoprecipitated with 5 mg of with Bim antibody or Trim2 antibody and 30 ml of protein G-agarose for 12 h at 4°C. Equivalent amounts of immunoglobulin G (IgG) were used as the control. The beads were washed five times with lysis buffer and boiled in SDS sample buffer, and the released proteins were resolved by SDS-PAGE. The gel were transferred to nitrocellulose and western blotting was performed.

Statistical analysis was performed using the SPSS standard version 19.0 software (SPSS, Inc., Chicago, IL, USA). Data are represented as mean ± standard deviation from at least three independent determinations. The independent Student's t-test was calculated to compare the results between the two groups. P<0.05 was considered to be statistically significant.

A total of 114 breast cancer tissues were eligible for analysis according to our previous inclusion criteria; of these, 106 recurrent breast tumor samples (66 local and 40 distant metastases) were GPER positive identified by IHC staining. Among the 106 GPER⁺ specimens, GPER expression was increased in 73.58% (78/106), decreased in 5.66% (6/106) and unchanged in 20.76% (22/106) compared with the matched primary tumor tissues (PTs). All these GPER⁺ recurred tumor tissues (RTs) and the paired PTs were used to determine the expression of Bim (Bcl2-L11), an aberrant downregulated gene identified by cDNA-array in TAM-resistant breast cancer cells. Bim and GPER were shown to be located in the cell cytoplasm (Fig. 1A).

As shown in our reports, the mean IHC score for GPER is significantly increased in RTs (6.23±0.91) compared with that in PTs (3.46±1.07) (P=0.001). To understand whether Bim is involved in GPER-mediated TAM-resistance, Bim expression was scored in PTs and the paired RTs. Bim expression was decreased in 83.02%, 44/53), increased in 3.78% (4/106) and unchanged in 13.21% (14/106) of these 53 GPER⁺ tumors which relapsed during TAM treatment (Fig. 1B). The mean IHC score for Bim was 7.27±0.56 in PTs and 4.02±0.49 in RTs (Fig. 1B; P<0.0001). Among the 53 GPER⁺ RTs, Bim expression had a converse correlation with GPER expression (Fig. 1C; R²=0.54).

It has been reported that Bim plays an important role in cell apoptosis (35–37). To investigate the potential role of Bim protein in the process of TAM-induced apoptosis, we further determined the Bim expression in TAM-sensitive MCF-7 cells (MCF-7) and TAM-resistant MCF-7 cells (MCF-7R) by Q-PCR and western blot analyses. The transcription activity of Bim was not changed by Q-PCR. In contrast to their protein levels, high level of Bim in MCF-7 and obviously reduced Bim in MCF-7R were confirmed in western blot analyses (Fig. 2A and B). Moreover, the protein levels of Bim in MCF-7 were gradually increased in a TAM-dose dependent pattern (Fig. 2C, upper panel); however, the weak Bim protein had no response to TAM stimulation in MCF-7R (Fig. 2C, down panel). Consistent with these findings, more apoptotic cells by TAM were detected in MCF-7 cells than in MCF-7R cells; knockdown of Bim in MCF-7 cells significantly decreased the apoptosis cell ratio; however, ectopic expression of Bim in MCF-7R notably increased their sensitivity to TAM (Fig. 2D). In addition, similar cell survival results were acquired by MTT assays (Fig. 2E). These data suggest high level of Bim protein in MCF-7 is an effector to TAM treatment.

To further confirm the role of Bim on apoptosis in response to TAM in MCF-7, specific siRNAs against Bim (Fig. 3A) were employed. Knockdown of Bim in MCF-7 attenuated the activated PARP and cleaved caspase-3 in exposure to TAM treatment (Fig. 3B, left panel). Overexpression of Bim in MCF-7R increased the levels of activated PARP and cleaved caspase-3 under treatment of TAM (Fig. 3B, right panel). Consistently, the percentage of apoptosis cells was 40.5% in MCF-7 and 6.75% in Bim-knocked down MCF-7 cells, whereas the apoptotic cells were 6.8% in MCF-7R and 44.6% in MCF-7R with ectopic Bim under TAM treatment (Fig. 3C). These data indicate that Bim plays a critical role in TAM-induced apoptosis in MCF-7 cells through activation of PARP and caspase-3.

To understand why Bim was decreased in MCF-7R, bioinformatics was applied. It was found that a tripartite motif protein 2 (TRIM2) upregulated in MCF-7R cells (Fig. 4A and B) may interact with Bim. TRIM2 has been reported to be function as the E3 ligase to enhance Bim ubiquitinylation (23,24). Indeed, the binding of TRIM2 and Bim in MCF-7R was confirmed using co-immunoprecipitation analysis (Fig. 4C). Knockdown of Trim2 using specific siRNA in MCF-7R, the levels of Bim were notably increased and its expression could be regulated by TAM stimulation (Fig. 4D). Upregulation of Bim expression by knocking down Trim2 in MCF-7R rescued its potential to TAM sensitivity, suggesting the TRIM2-mediated ubiquitinylation degradation of Bim may contribute to TAM-resistance of MCF-7.

Previous reports have revealed an activated GPER in TAM-resistant breast cancer cells (2,26), and the above data also display an adverse relationship between Bim and GPER in TAM-resistant breast tumors. The expression of Bim, TRIM and GPER were further determined in ER-negative breast cancer cells and TAM-resistant MCF-7R cells. As shown in Fig. 5A, lower levels of Bim, and higher levels of TRIM2 and GPER were detected in most of ER-breast cancer cells and MCF-7R cells compared with TAM-sensitive MCF-7 cells. Treating cells with TAM, G1 (the GPER-specific agonist) or TAM combined with G15 (the GPER-specific antagonist), Bim could be stimulated by TAM, but not by G1 in MCF-7 cells (Fig. 5B and C, left panel). However, the levels of Bim were further decreased and TRIM2 expression was enhanced by TAM and G1 in MCF-7R cells (Fig. 5B and C, right panel), suggesting that activation of GPER may govern the levels of TRIM2 and Bim. In contrast to their protein levels, the transcription activity of Bim was changed under the treatments of TAM in MCF-7R cells (Fig. 5D, left panel); however, mRNA expression of Trim2 were obviously increased under the treatment of TAM and G1, but attenuated by G15 (Fig. 5D, right panel). These data suggest that activation of GPER is necessary for TRIM2 expression.

It has been previously demonstrated that TAM activates the MAPK/ERK and PI3K/AKT pathways through GPER/EGFR signaling axis in MCF-7R cells (2,26–29). To investigate whether these path ways are involved in regulating Bim protein levels, MCF-7R cells were treated with TAM alone and TAM combined with U0126 (the MAPK/ERK inhibitor) or LY294002 (the PI3K inhibitor). The protein levels of Bim were notably increased in MCF-7R treated with TAM combined with U0126, not LY294002 (Fig. 6A), indicating that MAPK/ERK signaling pathway may involve in regulation of Bim protein levels. Treating cells with TAM alone or TAM combined with specific inhibitor, mRNA expression of Bim was not altered (Fig. 6B). However, mRNA expression of Trim2 was significantly reduced after inhibiting MAPK/ERK activation in MCF-7R cells (Fig. 6C). This led to decrease in the interaction between TRIM2 and Bim under the treatment of U0126 in MCF-7R (Fig. 6D). Thus, GPER and its downstream MAPK/ERK signaling pathway are involved in regulation of Trim2 expression and affect the binding between TRIM2 and Bim resulting in reduced Bim in TAM-resistant breast cancer cells.