Role of apoptosis-related miRNAs in resveratrol-induced breast cancer cell death

Breast cancer is the most frequently diagnosed cancer in women, and one of the leading causes of cancer-related deaths worldwide. Recent evidences indicate that dietary agents such as resveratrol may inhibit cancer progression through modulation of microRNAs (miRNAs). We demonstrate that resveratrol regulates apoptotic and cell cycle machinery in breast cancer cells by modulating key tumor-suppressive miRNAs including miR-125b-5p, miR-200c-3p, miR-409-3p, miR-122-5p and miR-542-3p. Resveratrol-mediated miRNA modulation regulates key anti-apoptotic and cell cycle proteins including Bcl-2, X-linked inhibitor of apoptosis protein and CDKs, which are critical for its activity. Modulating miRNAs with mimics or inhibitors further validated a key role for miR-542-3p in MCF-7 and miR-122-5p in MDA-MB-231 breast cancer cell death in response to resveratrol. In conclusion, this study reveals novel miRNAs modulated by resveratrol that have a key role in breast cancer cell death.

Breast cancer is the second most common type of cancer in women, and the fifth most common cause of cancer-related deaths in the world. Nearly 200 000 women get diagnosed and about 40 000 die of breast cancer every year worldwide. 1 Prolonged use of chemotherapeutic drugs against breast cancer mostly renders the drug ineffective because of development of resistance against the therapeutic agents. Identifying alternative treatments is crucial to reduce the mortality rate related to breast cancer.
Cell cycle arrest and apoptosis are considered important for therapeutics targeting cancer cells. It is often observed that cancer cells have altered cell cycle machinery. During the transition of a normal cell to cancerous state, cyclin-dependent kinases (CDKs) that govern coordinated initiation, progression and completion of cell cycle are overexpressed causing uncontrolled abnormal cell growth. 2 Apoptosis or programmed cell death occurs naturally in all tissues to maintain tissue homeostasis and acts as a mechanism to eliminate unwanted cells. Cell division through the quiescence (G0) to the proliferative phases is controlled by the cell cycle. The DNA synthesis phase (S phase) and the mitosis (M phase) are separated by the G1 and G2 phases. Several drugs targeting the cell cycle have entered clinical trials and some of the wellknown drugs currently used exhibit their effects by targeting the cell cycle. Cell cycle arrest is known to cause apoptosis and cell death in human malignancies. 3,4 Apoptosis occurs via two controlled pathways: the extrinsic or death receptormediated pathway, which activates caspase-8; and the intrinsic or mitochondria-mediated pathway, which activates caspase-9. These caspases known as initiator caspases activate downstream effector caspases (caspase-3, -6, and -7), which induce cleavage of several key cellular proteins to activate cell death. Cancer therapies like chemotherapy and many anticancer drugs primarily act by inducing apoptosis.
Natural plant-derived compounds, including resveratrol have been reported to induce apoptosis and cell cycle arrest in tumor cells. [5][6][7] Resveratrol is a dietary agent found in a wide variety of plants like grapes, berries and peanuts and is known to have antioxidant and anti-inflammatory properties. It is emerging as a promising anticancer agent because of its chemopreventive and pro-apoptotic properties. [8][9][10][11] Resveratrol has been shown to have a crucial role in apoptosis induction in human breast cancer cells. 12,13 Moreover, studies show that several members of the mitogen-activated protein kinase signaling pathway are involved in this activation 14 and the intrinsic mitochondrial pathway, via activation of caspase-9 along with other key mediators calcium and calpain, is the major pathway involved in resveratrol-induced apoptosis. 15 MicroRNAs (miRNAs) are emerging as potential diagnostic, prognostic and therapeutic tools for breast cancer treatment. 16,17 MiRNAs are small non-coding single-stranded RNAs that negatively regulate gene expression by binding to mRNA and inhibiting translation. They control normal cell functions like cell cycle regulation, proliferation, differentiation and apoptosis. They have been implicated to have a critical role in the development and progression of various types of cancers including breast cancer. Owing to their significant and versatile roles, miRNAs are emerging as therapeutic tools for many cancers. Several miRNAs have been shown to be dysregulated in breast cancer tissues when compared with normal tissues. 18 Modulation of tumor-suppressive miRNA by natural chemopreventive agents such as resveratrol has been shown to induce cell death via apoptosis in various cancer cells including prostate cancer cells. 19 Interestingly, a link between resveratrol-induced apoptosis and miRNA modulation in relation to breast cancer has not been studied.
In this study, we investigated the anti-proliferative effects of miRNA modulation by resveratrol in breast cancer cells. We identified novel tumor-suppressive miRNAs differentially regulated by resveratrol in MCF-7 and MDA-MB-231 breast cancer cells that regulate apoptosis and cell cycle machinery. Furthermore, the rate-limiting miRNAs and their key target proteins that could serve as potential targets for tumor inhibition were identified. Elucidating the effect of resveratrolmediated miRNA modulation during breast cancer cell death may aid in better understanding of the underlying mechanisms that have a critical role in breast cancer.

Results
Caspase activation and apoptosis induction by resveratrol in breast cancer cells. We first characterized the apoptotic response to resveratrol treatment in MCF-7 and MDA-MB-231 human breast cancer cell lines. Cells were treated with various concentrations of resveratrol (0-300 μM) and apoptosis was determined after 24 h by Hoechst assay. Figures 1a and b show that resveratrol treatment caused a dose-dependent increase in apoptosis over control level, as indicated by increased nuclear fluorescence and chromatin condensation of the treated cells. Furthermore, caspase activity assays using specific enzyme substrates for caspase-8 and -9 showed a dose-dependent increase in caspase-8 and -9 activities in response to resveratrol in both MCF-7 and MDA-MB-231 cells (Figure 1c). The effect was more pronounced in MDA-MB-231 cells. Caspase activation and apoptosis induction was further confirmed by western blotting for various caspases (Figure 1d). Both cell lines were similarly treated with various concentrations of resveratrol for 24 h and analyzed for cell viability (Figures 1e and f). Resveratrol treatment significantly decreased cell viability in a dose-dependent manner in both MCF-7 and MDA-MB-231 cells. The IC 50 values were 162.08 ± 6.14 and 123.67 ± 18.11 μM as determined by the MTT assay.
Resveratrol regulates apoptosis-related miRNAs in breast cancer cells. MCF-7 and MDA-MB-231 cells were treated with resveratrol and 12 h post-treatment real-time PCR amplification of isolated RNA using human apoptosis miRNA array, which analyzes a panel of 84 miRNAs known to be associated with cellular apoptosis was performed. Data analysis using the Qiagen (Valencia, CA, USA) online data analysis software revealed many miRNAs to be regulated as a result of apoptotic effect of resveratrol treatment versus control samples (Supplementary Table 1). Interestingly, most miRNAs from the panel of tested miRNAs were found to be downregulated in both cell lines. Table 1 shows miRNAs regulated by more than twofold in response to resveratrol treatment and miR-409-3p (1.6-fold change), which was the only one upregulated in MCF-7 cells. Thirty-five miRNAs were found to be downregulated in MCF-7 cells by more than twofold, among which miR-542-3p (over eightfold) and miR-125b-5p (over fivefold) showed the largest changes. In MDA-MB-231 cells, of the upregulated miRNAs, miR-122-5p showed a 37.6-fold change. Twenty-five miRNAs were found to be downregulated by 42-fold, among which miR-542-3p (over 11-fold) and miR-200c-3p (over 8-fold) showed the largest changes.
Resveratrol-mediated miRNA regulate key apoptosis and cell cycle proteins. For both MCF-7 and MDA-MB-231 cell lines, miRNAs that were most regulated (maximum fold change) by resveratrol treatment were selected for further analysis. Bioinformatics analysis with the aid of miRNA databases revealed as well as predicted experimentally validated targets of the identified miRNA ( Table 2). Interestingly cyclins, caspases, Bcl-2 and many Bcl  Effect of resveratrol on breast cancer miRNAs. Modulation of breast cancer-related miRNAs by resveratrol in MCF-7 and MDA-MB-231 cells was investigated by quantitative realtime PCR using a human breast cancer array, which analyzes a panel of 84 miRNAs known to be associated with breast cancer (Supplementary Table 2). The data were analyzed using the miScript miRNA PCR Array Data Analysis portal (Qiagen), and a clustergram of the array data for each cell line was generated (Supplementary Figure 1). Breast cancer miRNAs regulated by more than twofolds in response to resveratrol treatment is shown in Figure 5 and Supplementary Table 3. Two miRNAs were upregulated in both MCF-7 and MDA-MB-231 cells by more than twofolds, of which miR-199a-5p was common to both. A total of 18 miRNAs in MCF-7 and 9 miRNAs in MDA-MB-231 cells were downregulated by more than twofolds with resveratrol treatment. Notable miRNAs that were common in both cell lines were miR-199a-5p, miR-125b-1-3p, miR-140-5p and miR-20a-5p. The expression profiles of miRNAs identified for each cell line were subjected to ingenuity pathways analysis (IPA), and a network map of potential proteins and pathways were generated using path design within the IPA core analysis tool for both MCF-7 and MDA-MB-231 cells (Figures 6a and b). IPA shows that several key apoptotic proteins already identified in this study form the central nodes of the network pathways predicted for both MCF-7 and MDA-MB-231 cells.
Caspases are known to control the cell regulatory function that initiates the process of apoptosis and are an important feature of therapeutics that target cancer. 28 Our observation also confirms previous reports on the intrinsic mitochondrial pathway being the major pathway involved in resveratrolinduced apoptosis, via activation of caspase-9. 15 Resveratrol has also been reported to activate the extrinsic pathway of cell death, including FAS-dependent apoptosis in HL-60 pro-myelocytic leukemia cells 29,30 and TRAIL-induced apoptosis in prostate cancer cells. 31 Furthermore, resveratrol exerted its pro-apoptotic effects by downregulating expression of key anti-apoptotic proteins belonging to the inhibitor of apoptosis (IAPs) family of proteins Bcl-2 and XIAP (Figures 2a  and b) and also decreased expression of CDKs specific for G-phase arrest (Figures 2c and d). Cell cycle analysis by flow cytometry clearly showed a dose-dependent G1-arrest for both MCF-7 and MDA-MB-231 cells induced by resveratrol (Figures 2e and f). Cell cycle regulation upon treatment with resveratrol seems to be cell-type specificwhereas resveratrol induces G0/G1-arrest in prostate cancer cells and Resveratrol-mediated miRNAs in breast cancer R Venkatadri et al Several studies have focused on utilizing small molecule inhibitors and molecules to specifically target IAPs and CDKs and this has been proven as a successful strategy for cancer treatment. [37][38][39] We tested the effect of resveratrol on apoptosis-related miRNAs using human apoptosis miRNA array. MiRNAs are reported to have a pivotal role in breast cancer progression, functioning either as oncogenes or tumor-suppressor genes. MiRNAs have been shown to be dysregulated in breast cancer tissues as compared with normal tissues, 40 and therefore could be important targets in cancer treatment. Several miRNAs with altered expression patterns are identified in various types of cancers. 41,42 MiRNAs thus serve as therapeutic targets for cancer treatment because of their effect on multiple target genes and proteins. We report several miRNAs that were differentially regulated by resveratrol in MCF-7 and MDA-MB-231 cells (Table 1). Online databases such as TargetScan, miRBase and miRanda were used to identify major protein targets of the shortlisted miRNAs ( Table 2). [43][44][45][46] Two miRNAs in MCF-7 (miR-125b-5p and miR-542-3p) and two miRNAs in MDA-MB-231 (miR-200c-3p and miR-542-3p) were downregulated by over fivefolds. MiR-409-3p and miR-122-5p were the only upregulated miRNAs in MCF-7 and MDA-MB-231 cells, respectively. MiR-542-3p was downregulated in both the cell lines. Interestingly, all the shortlisted miRNAs were predicted to primarily target caspases, anti-apoptotic protein Bcl-2 and cell cycle-related cyclins ( Table 2). To investigate the role of miRNA modulation in resveratrolinduced apoptosis of breast cancer cells, one miRNA that was the most affected by resveratrol treatment for each cell line was selected for further validation. Based on their maximum fold changes, miR-542-3p for MCF-7 and miR-122-5p for MDA-MB-231 cells were selected. Chemically synthesized miRNA mimic and inhibitors were used to validate miRNAs and study their effects on tumor inhibition. 47,48 MCF-7 cells transfected with miR-542-3p mimic reversed the effect of resveratrol on apoptosis and the anti-apoptotic proteins Bcl-2 and XIAP (Figure 3). Recent studies have shown that downregulation of miR-542-3p is associated with cancer invasiveness and resistance to therapy. [49][50][51] Yoon et al. 52 reported that the observed tumor-suppressive function of this miRNA is by targeting the anti-apoptotic protein survivin. In our study, we found that resveratrol downregulates miR-542-3p, which is imperative for the observed tumor inhibitory effects. Individual miRNA can be controlled by other miRNA as part of miRNA clusters, which varies in each system. This observation is particularly interesting because it highlights complex interactions among miRNAs that are yet to be understood. MDA-MB-231 cells transfected with miR-122-5p inhibitor showed similar resistance to resveratrol-induced apoptosis and reversed its effect on anti-apoptotic proteins (Figure 4). MiR-122-5p has been reported to regulate tumorigenesis in hepatocellular carcinoma 53   modulators in the G1 phase in response to resveratrol treatment. As compared with data in response to direct resveratrol treatment presented in Figure 2, the experimental and culture conditions are different in Figures 3 and 4, where cells were transfected with miRNA mimics or inhibitors before treating with resveratrol. Therefore, the effect observed on apoptosis, viability, cell cycle arrest, Bcl-2 and XIAP proteins is less pronounced in cells, which were transfected before resveratrol treatment (Figures 3 and 4) as compared with direct treatment to resveratrol (Figure 2).
We also performed miRNA expression profiling using Breast Cancer arrays to identify key miRNAs known to be associated with breast cancer progression in response to resveratrol treatment ( Figure 5 and Supplementary Table 3). Several miRNAs were modulated in MCF-7 and MDA-MB-231 cells in response to resveratrol treatment, of which four miRNAs associated with breast cancers including miR-199a-5p, miR-125b-1-3p, miR-140-5p and miR-20a-5p were common in both cell lines. These miRNAs have been widely reported to have a role in cancer initiation, progression and inhibition. MiR-199a-5p has been reported to suppress liver cancer by inhibiting glucose metabolism, regulate autophagic response in breast cancer and function as an oncogene in the development of gastric cancer. [55][56][57] The miR-125b, miR-140 and miR-17/20 clusters are also reported to inhibit breast cancer progression. [58][59][60] Validation of these miRNAs warrants further investigation and is beyond the scope of this study, which focuses on cell death and cell cycle pathways. Interestingly, miR-125b-1-3p is the only miRNA observed to be downregulated by resveratrol in both the apoptotic and breast cancer arrays. This provides extensive scope for further investigations as downregulation of miR-125b has been observed in head and neck cancer development and breast cancer progression. 61,62 Overall, our findings demonstrate that resveratrol-induced apoptosis in MCF-7 and MDA-MB-231 cells and showed a dose-dependent effect on cell viability. Resveratrol mediated its apoptotic effect via caspase activation and downregulation of anti-apoptotic proteins including XIAP and Bcl-2. Interestingly, resveratrol-modulated apoptotic miRNAs had a key role in mediating its effect on breast cancer cell death. Overall, our study reveals a novel role for resveratrol in inhibiting breast cancer progression by modulating tumorsuppressive miRNAs and thereby affecting cell cycle and apoptosis. We report novel miRNAs including miR-542-3p and miR-122-5p to have a key role in resveratrol-mediated effects on MCF-7 and MDA-MB-231 breast cancer cells, respectively. Evaluating the potential of resveratrol to modulate epigenetic events associated with breast cancer could facilitate in the discovery and development of novel therapeutic strategies against this disease.

Materials and Methods
Chemicals and reagents. Resveratrol was obtained from Sigma-Aldrich (St. Louis, MO, USA). A 100 mM stock was prepared in ethanol and stored at 4˚C until further use. All antibodies including caspase-8, caspase-9, caspase-3, Bcl-2, XIAP, HRP-conjugated anti-rabbit IgG and anti-mouse IgG antibodies was purchased from Cell Signaling Technology (Danvers, MA, USA). β-Actin antibody, Hoechst 33342 and propidium iodide was purchased from Sigma-Aldrich.
Cell culture. Human breast adenocarcinoma cells MCF-7 and MDA-MB-231 were obtained from American Type Culture Collection (Manassas, VA, USA). Cells were cultured in Dulbecco's Modified Eagle medium (Thermo Scientific, Waltham, MA, USA) supplemented with 10% FBS, 2 mM L-glutamine, 100 U/ml penicillin and 100 mg/ml streptomycin in a 5% CO 2 environment at 37°C.
Apoptosis assay. Apoptosis was determined by Hoechst 33342 DNA fragmentation assay. Briefly, cells were incubated with 10 μg/ml Hoechst 33342 nuclear stain (Life Technologies, Carlsbad, CA, USA) for 30 min at 37°C and percentage of cells having intensely condensed chromatin and/or fragmented nuclei by fluorescence microscopy (EVOS All-in-one digital inverted fluorescence microscope, Thermo Fisher Scientific, Waltham, MA, USA) were scored. From random fields, nuclei were analyzed for each sample. The apoptotic index was calculated as apoptotic nuclei/total nuclei x 100 (%) using ImageJ software (Java image processing, NIH, Bethesda, MD, USA).
Caspase assay. Caspase activity was determined by fluorometric assay using the enzyme substrate IETD-AMC for caspase-8 and LEHD-AMC for caspase-9, which are specifically cleaved by the respective enzymes at the Asp residue to release the fluorescent group, AMC. Cell extracts containing 50 μg of protein were incubated with 100 mM HEPES containing 10% sucrose, 10 mM dithiothreitol, 0.   on a native agarose gel by analyzing the 18S and 28S ribosomal RNA bands and by measuring purity at 260 and 280 (data not shown). (ii) miRNA amplification: cDNA synthesis was performed using the miScript II RT kit (Qiagen) as per the manufacturer's instructions. Real-time PCR amplification of the cDNA was performed using miScript SYBR Green PCR kit (Qiagen) for known apoptosis-related miRNA using Human Apoptosis miRNA PCR arrays (Qiagen). cDNA conversion and real-time PCR were performed as per the manufacturer's instructions. (iii) PCR array data analysis: data sets for each sample were collected at the same threshold and baseline levels to maintain consistency between samples for further analysis. The PCR array data were analyzed using the web-based data analysis software offered by Qiagen (http://pcrdataanalysis.sabiosciences. Figure 6 Path design analysis of breast cancer-related miRNAs. Network maps of important biomarkers predicted to have a key role in breast cancer in response to resveratrol treatment were generated using IPA. Potential protein and miRNA interactions were visualized using the Path Design Suite for both (a) MCF-7 and (b) MDA-MB-231 cells. Dotted lines connect genes that are known to be affected indirectly and solid lines connect genes that are known to be directly affected by resveratrol treatment com/mirna/arrayanalysis.php). A twofold change cut-off was selected to identify miRNA that were significantly altered because of resveratrol treatment. The accession number and sequence of the identified miRNA were obtained using the miRBase (www.mirbase.org) database. miRWalk (http://www.umm. uni-heidelberg.de/apps/zmf/mirwalk/index.html) was used to obtain a list of predicted and validated miRNA targets from apoptosis-related pathways using the following pathway (database): Apoptosis (KEGG), Caspase (Biocarta), Death (Biocarta) and Mitochondria (Biocarta).
Transient transfection. MCF-7 and MDA-MB-231 cells were seeded in sixwell culture plates. The cells were transfected in complete medium with miRNA mimic, miRNA inhibitor or control (AllStars Hs Cell Death Control siRNA) (Qiagen) using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) transfecting agent according to the manufacturer's protocol. Briefly, lipofectamine (15 μl) was added to miRNA mimic or miRNA inhibitor (100 nM) diluted with 50 μl serum-free medium. This complex was incubated for 15 min and then added to the cells. After 40 h, the medium was discarded and the cells were washed with PBS and treated with resveratrol in serum-free medium. Gating strategy. Acquired samples were initially screened with SSC-H versus FSC-H density plot in linear scale and the gates were set to exclude debris. Another density plot was generated with the cells in the set gates and analyzed in PE-H versus PE-A for doublet discrimination. Singlet cells were gated, exported as FCS files and analyzed using ModFit 4.0.5 LT cell cycle analysis software (Verity Software House, Topsham, ME, USA).
Western blotting. After specific treatments, cell lysates were resolved on a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. The protein concentration was determined using a bicinchoninic acid protein assay kit (Pierce Biotechnology, Rockford, IL, USA), and equal amount of protein was loaded per sample. The membrane was blocked with TBS-T (0.1% Tween-20 in TBS) containing 5% dry milk, and incubated with primary antibody overnight at 4 o C. After three washes with TBS-T, the membrane was incubated with HRP-conjugated secondary antibody for 1 h and then washed with TBS-T. The immune complexes were detected by chemiluminescence (Supersignal West Femto; Pierce Biotechnology) using MyECL Imager (Thermo Scientific), and quantified by using ImageJ (NIH, Image analysis using Java) digitizing software. Mean densitometry data from independent experiments were normalized to the control.
Path design analysis. The IPA platform was used to discern global interaction networks associated with miRNA that were differentially regulated in MCF-7 and MDA-MB-231 cells. The miRNA data were uploaded into the software, and Network analysis was performed to identify important regulatory proteins associated with regulation of the miRNA. These protein pathways and miRNA were mapped out using the Path Design tool available to aid in visualization of interactions across various cellular compartments.
Statistical analysis. Representative data from three or more independent experiments are shown as mean value ± S.E.M. Statistical analysis was performed with two-way analysis of variance to identify differences between groups using GraphPad Prism Software (San Diego, CA, USA) and P values o0.05 considered significant.