- Email: [email protected]
β-Catenin pathway in breast cancer
Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
Available online at
ScienceDirect www.sciencedirect.com
Fructose-1,6-bisphosphatase is a novel regulator of Wnt/b-Catenin pathway in breast cancer Kaichun Lia , Mingzhen Yinga , Dan Fenga , Jie Dub , Shiyu Chenb , Bing Danb , Cuihua Wangb , Yajie Wanga,* a b
Department of Oncology, Changhai Hospital Affiliated to The Second Military Medical University, Shanghai 200433, PR China Department of Oncology, Tianyou Hospital Affiliated to Tongji University, Shanghai 200331, PR China
A R T I C L E I N F O
Article history: Received 29 August 2016 Received in revised form 10 October 2016 Accepted 17 October 2016 Keywords: FBP1 Oncomine GSEA Wnt/bbeta -Catenin pathway Breast cancer
A B S T R A C T
Fructose-1,6-bisphosphatase (FBP1), the rate-limiting enzyme in gluconeogenesis, is a tumor suppressor that frequently down-regulated in cancers, especially breast cancer. Here, we provide both supporting and contradicting evidences about the expression pattern and function of FBP1 in breast cancer. Data mining of Oncomine database showed that FBP1 is commonly up-regulated in tumor tissues compared with non-tumor tissues regardless of histological type. Analysis of a large-scale cohort derived from Kaplan-Meier Plotter showed that lower FBP1 expression associated with poor clinical outcome. Genetic silencing of FBP1 reduced aerobic glycolysis and the malignant potential of breast cancer cells. Gene set enrichment analysis (GSEA) of the expression profiles of breast cancer cells (n = 59) revealed that cells exhibiting high expression of FBP1 had a lower activity of Wnt/b-Catenin pathway. FBP1 downregulation enhanced the activity of Wnt/b-Catenin pathway and increased the level of its downstream targets, including c-Myc and MMP7. Collectively, our findings indicate that elevated FBP1 is a critical modulator in breast cancer progression by altering glucose metabolism and the activity of Wnt/ b-Catenin pathway. ã 2016 Elsevier Masson SAS. All rights reserved.
1. Introduction Breast cancer is the leading cause of cancer-related death in women through the world. Although the great improvement in diagnostic techniques and targeted therapies, the prognosis of breast cancer is still dissatisfy [1]. The considerable heterogeneity is a key factor that drives the malignant phenotypes of breast cancer [2]. Therefore, identifying the common characters in breast cancer across different subtypes is of great importance to develop potential therapeutic methods. Reprogrammed energy metabolism is an emerging hallmark of cancer cells [3]. The interconnectedness between glycolysis/ oxidative phosphorylation and anabolic gluconeogenesis drives the metabolic phenotype of cancer cells. Fructose-1,6-bisphosphatase (FBP), which catalyzes the hydrolysis of fructose-1,6bisphosphate to fructose-6-phosphate and inorganic phosphate, is a critical regulatory enzyme in gluconeogenesis [4]. Two isoforms
* Corresponding author at: Department of Oncology, Changhai Hospital Affiliated to The Second Military Medical University, No.168, Changhai Road, Shanghai 200433, PR China. E-mail address: [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.biopha.2016.10.050 0753-3322/ã 2016 Elsevier Masson SAS. All rights reserved.
are present in mammalian cells, FBP1 and FBP2. FBP1 is widely expressed in different tissues and FBP2 is specially expressed in muscle [5]. Recently, it has been reported that FBP1 is involved in the initiation and progression of multiple human cancers, such as hepatocellular carcinoma [6,7], renal carcinoma [8], lung cancer [9,10] and breast cancer [11]. FBP1 is frequently down-regulated and most reported as a tumor suppressor in cancer development. For example, ZEB1-mediated down-regulation of FBP1 in lung cancer predicts a poor prognosis and confers tumor cells with growth and invasion advantage [9,10]. And FBP1 is ubiquitously depleted and plays tumor-suppressive functions in clear cell renal cell carcinoma [8]. In breast cancer, the Snail-G9a-Dnmt1 complex mediates the depletion of FBP1 [11]. And notably, FBP1 is highly expressed in luminal subtype but nearly undetected in triplenegative breast cancer (TNBC). However, by data mining current available database, including Oncomine, Cancer Cell Line Encyclopedia database (CCLE) and The Cancer Genome Atlas database (TCGA), we found several contradictory evidence about the expression pattern of FBP1 in breast cancer. Furthermore, in this study, we also investigated the cellular functions and underlying mechanisms of FBP1 in breast cancer.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
2. Materials and methods 2.1. Cell culture, transfection and western blotting MCF7 and BT474 cell lines were obtained from American Type Culture Collection (ATCC, Manassas, Va., USA). Cells were cultured in DMEM medium (Invitrogen, Carlsbad, Calif., USA) supplemented with 10% fetal bovine serum (FBS) at 37 C with 5% CO2. For transfection, a total of 50 ng/ml siRNAs against FBP1 were transfected with RNAi MAX (Thermo Scientific, USA). To exclude unspecific effects, two efficient siRNAs were used in this study. The antisense sequences against FBP1 are 50 -AACATGTTCATAACCAGGTCG-30 (si-FBP1-1) and 50 -ATGTTGGAAGATCCATCAAGG-30 (si-FBP1-2). Western blotting was performed to determine to protein level of FBP-1 and b-actin. Antibodies used for FBP-1 (ab109732) and b-actin (ab8226) were purchased from Abcam. 2.2. Measurement of glucose and lactate level To detect the glucose consumption and lactate production in breast cancer cells upon FBP1 knockdown, the Glucose Colorimetric Assay Kit II (K686-100, BioVision) and Lactate Colorimetric/ Fluorometric Assay Kit (K607-100, BioVision) were used according to the manufacturer’s protocols, respectively. Specially, cell culture medium used in this experiment was phenol red-free. The glucose and lactate level were calculated using a standard calibration curve prepared under the same condition. 2.3. Cell proliferation and cell invasion assays For cell proliferation assay, 3000 cells were seeded in triplicate and MTT assay was performed according to the manufacturer’s protocols. For cell invasion assay, transwell model was used. Briefly, a total of 2 104 cells were seeded into the upper chamber of matrigel-coated filters (BD Bioscience, USA) and the lower chamber was filled with 800 ml of culture medium containing 5% FBS. After incubation for 48 h, cells remained on the upper chamber were removed. The invaded cells were fixed with 4% paraformaldehyde, stained with 0.5% crystal violet and counted in three random fields.
1145
reverse: 50 -GGCCCCATAAGGAGCTGAAT-30 ; GADPH forward: 50 CTGGGCTACACTGAGCACC-30 , GAPDH reverse: 50 -AAGTGGTCGTTGAGGGCAATG-30 . 2.6. Bioinformatic analysis Microarray expression data of FBP1 from multiple independent data sets were derived from the Oncomine repository (http:// www.oncomine.org/). And to examine the relative mRNA expression levels of FBP1 between normal and cancer samples in a variety of tissue types, the distributions of log2 median centered signal intensities were plotted using box plots. To determine the differential expression of FBP1 between breast cancer and normal tissues, a combined filter was used to display indicated datasets. The Cancer Type was defined as Breast Cancer and Data Type was mRNA, whereas Analysis Type was Cancer vs Normal Analysis. The mRNA expression of FBP1 in a repertoire of cancers located was analyzed by the online tool located in The Cancer Cell Line Encyclopedia database (CCLE, www.broadinstitute.org/ccle/home) [12]. All raw and processed data are available at the CCLE website and the website offers direct links to data visualization tools such as genepattern-based analysis tools for expression. By searching the gene symbol, FBP1, the expression level of FBP1 among different cancers was automatically generated by the CCLE analysis tool. Simultaneously, the processed data for FBP1 could be downloaded and the expression level of FBP1 in breast cancer cells could be extracted. The prognostic value of FBP1 in breast cancer was analyzed by an online survival analysis tool (http:// kmplot.com/analysis) and the precise method was reported previously [13]. Briefly, by entering the gene symbol, FBP1, the survival curve based on FBP1 expression in breast cancer could be automatically generated by the Kaplan Meier-plotter analysis tool. Specially, to examine the prognostic value of FBP1 in TNBC, the ER, PR and HER2 status were selected as negative. The transcriptomic difference between FBP-1 low group and FBP-1 high group was uncovered by Gene Set Enrichment Analysis (GSEA), which is also available in the CCLE online analysis tool (www.broadinstitute.org/ ccle/data/browseAnalyses). Based on FBP-1 expression level, the breast cancer cell lines were divided into FBP-1 low group or FBP-1 high group and subsequently subjected to GSEA. 2.7. Statistical analysis
2.4. Luciferase reporter assay The si-Control or si-FBP1 cells were seeded in 96-well plate and transfected with 1.0 mg of pTOP-Luc (b-catenin/TCF4 reporter plasmid) or pGL3-Luc (positive control plasmid) using lipofectamine 2000 (#11668027, Thermo Scientific, USA). After 24 h, cells were lysed and subjected to luciferase activity assay according to the manufacturer’s protocols (#TM040, Promega Corporation, Madison, WI, USA). Luciferase activity was normalized by total protein level of the samples. The ratio of luciferase activity was calculated and normalized to Renilla and each assay was done in triplicate. 2.5. Quantitative real-time PCR Total RNA was extracted from breast cancer cells using Trizol reagent (Takara, Dalian, China) and 500 ng RNA was used for synthesis of complementary DNA (cDNA) by reverse transcriptase (Takara, Japan). Quantitative real-time PCR was subsequently performed with SYBR Premix Ex Taq (Takara, Dalian, China) using ABIPrism-7500 Sequence Detector System (Applied Biosystems). GAPDH was used as the internal control for quantification. Primer sequences used in this study were listed as follows, FBP1 forward: 50 -CTACGCCAGGGACTTTGACC-30 , FBP1
Experiments were repeated at least twice. Data were expressed as the means SD. The statistical analysis was performed with GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA). P values were calculated based on two-tailed, unpaired Student’s ttests. P value less than 0.05 was considered statistically significant. 3. Results 3.1. Up-regulated FBP1 predicts a good prognosis in breast cancer Down-regulation of FBP1 has been reported in several different human malignancies, especially triple-negative breast cancer [11]. To fully uncover the expression profile of FBP1 in breast cancer, we firstly searched the Oncomine database and TCGA database (http:// cancergenome.nih.gov). To our surprise, FBP1 was consistently upregulated in tumor tissues compared with corresponding normal tissues across 11 independent cohorts derived from Oncomine database (Fig. 1A). Then by mining data deposited in TCGA, we found that FBP1 was highly expressed in breast cancer tissues irrespective of histological type, including ductal, lobular, mixed ductal and lobular, and mucinous breast cancer (Fig. 1B). Furthermore, data from the CCLE database showed that FBP1 was also highly expressed in breast cancer cell lines compared with the
1146
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
Fig. 1. Up-regulated FBP1 predicts a good prognosis in breast cancer. (A) The mRNA expression level of FBP1 in 11 independent datasets of breast cancer. (B) Box plot of FBP1 level in different subtypes of breast cancer. Data was derived from TCGA database. (C) The expression level of FBP1 in different human cancer cell lines. Data were derived from CCLE database. (D) The prognostic value of FBP1 expression in all subtypes of breast cancer. Data were analyzed in Kaplan Meier-plotter (http://kmplot.com/analysis/). (E) The prognostic value of FBP1 expression in triple-negative breast cancer.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
majority of other human cancer cells, especially originated from melanoma, osteosarcoma and glioma (Fig. 1C). These data above suggested that up-regulation of FBP1, to some extent, is a common event in breast cancer. To interrogate the prognostic value of FBP1 in breast cancer, Kaplan-Meier analysis was performed in a largescale dataset (http://kmplot.com/analysis/). The result showed that the higher expression of FBP1 predicted a good prognostic effect on progression-free survival of breast cancer patients, which is consistent with previous notion that FBP1 is a tumor suppressor in cancers (Fig. 1D). However, inconsistent with previous report in breast cancer, FBP1 did not predict a better prognosis in TNBC (Fig. 1E). 3.2. Genetic silencing of FBP1 promotes tumor progression of breast cancer in vitro Next, to determine the cellular functions of FBP1 in breast cancer, loss-of-function assay was performed. By searching CCLE database, two cell lines with higher FBP1 expression, MCF7 and BT474, were selected for further investigation (Fig. 2A). Two
1147
specific small interfere RNAs targeting FBP1 resulted in marked reduction in protein level of FBP1 in both MCF7 and BT474 cells (Fig. 2B). The regulatory functions of FBP1 in glycolysis have been reported in multiple cancers. To confirm the known suppressive role of FBP1 in glycolysis, we therefore detected two indictors of glycolysis, glucose consumption and lactate production upon genetic silencing of FBP1. Consistently, knockdown of FBP1 resulted in increased glucose consumption (Fig. 2C) and lactate production (Fig. 2D). Moreover, we evaluated the effects of FBP1 knockdown on cell proliferation and invasive potential of breast cancer. As shown in Fig. 2E, silencing of FBP1 significantly promoted the cell proliferation ratio of MCF7 and BT474 cells. Similarly, FBP1 knockdown increased the invasive capacity of MCF7 and BT474 cells as revealed by transwell assay (Fig. 2D). 3.3. FBP1 is negative regulator of wnt/b-Catenin pathway in breast cancer To illustrate the mechanism by which FBP1 inhibits tumor progression in breast cancer, Gene Set Enrichment Analysis (GSEA)
Fig. 2. Genetic silencing of FBP1 promotes tumor progression in vitro. (A) The mRNA level of FBP1 in breast cancer cell lines. (B) The silencing efficiency of FBP1 in MCF7 and BT474 cells was confirmed by Western Blotting. Glucose utilization (C) and lactate production (D) of in MCF7 and BT474 cells were measured upon silencing of FBP1. Cell proliferation (E) and invasive capacity (F) of MCF7 and BT474 cells were detected after knockdown of FBP1. *, p < 0.05; **, p < 0.01.
1148
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
was performed in the expression profiles of breast cancer cell lines. Based on FBP1 expression in CCLE dataset, a total of 59 breast cancer cell lines was divided into two group, the lower FBP1 group (Class + 1, n = 31) and the higher FBP1 group (Class + 2, n = 28). The difference of FBP1 expression between two groups was significant (Fig. 3A). GSEA showed that genes in Wnt/b-Catenin signaling were prominently enriched in the lower FBP1 group, indicating that FBP1 is a potential suppressor of Wnt/b-Catenin signaling (Fig. 3B). To test this hypothesis, luciferase reporter assay was performed. As indicated in Fig. 3C, FBP1 knockdown remarkably promoted the activity of Wnt/b-Catenin signaling in both MCF7 and BT474 cells. Consistent with this result, the two downstream targets, c-Myc and MMP7, were also increased in FBP1-silenced cells (Fig. 3D). Notably, data from TCGA database also confirmed the inverse relationship between FBP1 and c-Myc or MMP7 (Fig. 3E). Collectively, these data above suggested that the tumorsuppressive roles of FBP1 might be mediated by inactivation of Wnt/b-Catenin pathway. 4. Discussion In the present study, we provide several novel insights into the FBP1 and breast cancer. First, FBP1 is highly expressed in breast cancer. Indeed, down-regulation of FBP1 in cancers has been reported in all currently published articles. And several regulators of its repression have been revealed, including DNA hypermethylation [6], ZEB1 [9], NF-kB [14] and Snail [11]. However, in this study, by a large-scale data mining, we confirmed that FBP1
was overexpressed, at least to some extent, in breast cancer. Previous report in breast cancer has described that FBP1 was preferentially depleted in TNBC and certainly expressed at comparable expression in luminal breast cancer [11]. As TNBC accounts for 10–20% of all invasive breast cancers, we therefore proposed that the majority of other invasive mammary carcinomas are highly expressed with FBP1 [15]. Then by kaplan-Meier analysis, we found that FBP1 predicted a favorable prognosis in breast cancer, which is consistent with the observations in other cancers. Specially, in TNBC, no correlation between FBP1 and patients’ survival time was found, indicating the prognostic value of FBP1 in breast cancer may be molecular type-dependent. Second, by loss-of-function study, we further confirm the tumor-suppressive role of FBP1 in breast cancer. FBP1 knockdown has been associated with enhanced glycolysis, which is a common metabolic phenotype in cancer cells [8,9,11,16]. Consistently, knockdown of FBP1 results in decreased glucose consumption and lactate production in breast cancer cells. Apart from implications in glucose metabolism, FBP1 also functions as a tumor suppressor through induction of cell cycle arrest and inhibition of cancer stem cell-like property. In line with this notion, silencing of FBP1 significantly inhibited the cell proliferation rate and invasive capacity of breast cancer cells. Third, our data indicates that FBP1 is a novel negative regulator of Wnt/b-Catenin signaling. Previously, two critical mechanisms of FBP1 in cancers were revealed. In renal cancer, FBP1 restrains tumor progression in an enzyme activity-independent manner, but by suppressing the hypoxia-inducible factor (HIF) function through
Fig. 3. FBP1 is negative regulator of Wnt/b-Catenin pathway in breast cancer. (A) The expression level of FBP1 in lower group (n = 31, Class + 1) and high group (n = 28, Class + 2). (B) Genes in Wnt/b-Catenin signaling identified by GSEA were enriched in FBP1 lower group. (C) Luciferase activity assay was performed in MCF7 and BT474 cells at 48 h after transfected with specific FBP1 siRNAs. (D) The mRNA level of c-Myc and MMP7 was measured at 48 h after knockdown of FBP1. (E) Correlation between FBP1 expression and c-Myc or MMP7 was analyzed by the expression data from TCGA database. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
direct interaction with the inhibitory domain of HIF [8]. In breast cancer, FBP1 enhances the interaction between b-Catenin and Tcell factor and then reprograms glucose metabolism, which ultimately increases tumorigenecity [11]. In this study, by performing GSEA in a total of 59 breast cancer cells, we revealed that FBP is closely associated with the activity of Wnt/b-Catenin signaling. This data not only supported, but also amplified the findings reported previously in breast cancer. However, due to insufficient data to analyze the difference among different types of breast cancer, especially TNBC, we cannot fully rule out the possibility of other factors that contribute to the malignant phenotypes among different subtypes of breast cancer. Furthermore, we showed that c-Myc and MMP7, two crucial downstream factors of Wnt/b-Catenin signaling were negatively correlated the level of FBP1. As c-Myc is also a transcription factor involved in multiple cellular functions [17–19], especially its regulatory role in glycolysis, we hypothesized that the inhibitory effect of FBP1 in glycolysis might be mediated by down-regulation of c-Myc. And MMP7 is a potential effector in FBP1-mediated inhibition of invasive potential [20]. However, the precise mechanism by which how FBP1 regulates Wnt/b-Catenin signaling warrant further investigation. In conclusion, our data, as a proof of principle, suggest that FBP1 is frequently up-regulated in most breast cancer and predicts a good prognosis. FBP1 acts a tumor suppressor by inactivating Wnt/ b-Catenin signaling. Thus, it will be of great value to develop possible application of FBP1 in the treatment of several subtypes of breast cancer. Conflict of interest The authors declare that there is no conflict of interests.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Acknowledgement
[16]
We thank Shanghai Promoters Information Technology Co., Ltd for technical assistance.
[17]
Reference:
[18]
[1] C. DeSantis, J. Ma, L. Bryan, A. Jemal, Breast cancer statistics 2013, CA. Cancer J. Clin. 64 (2014) 52–62. [2] F. Beca, K. Polyak, Intratumor heterogeneity in Breast cancer, Adv. Exp. Med. Biol. 882 (2016) 169–189. [3] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144 (2011) 646–674. [4] M.S. Jurica, A. Mesecar, P.J. Heath, W. Shi, T. Nowak, B.L. Stoddard, The allosteric regulation of pyruvate kinase by fructose-1,6-bisphosphate, Structure 6 (1998) 195–210. [5] A. Marin-Hernandez, S. Rodriguez-Enriquez, P.A. Vital-Gonzalez, F.L. FloresRodriguez, M. Macias-Silva, M. Sosa-Garrocho, R. Moreno-Sanchez,
[19]
[20]
1149
Determining and understanding the control of glycolysis in fast-growth tumor cells. Flux control by an over-expressed but strongly product-inhibited hexokinase, FEBS J. 273 (2006) 1975–1988. M. Chen, J. Zhang, N. Li, Z. Qian, M. Zhu, Q. Li, J. Zheng, X. Wang, G. Shi, Promoter hypermethylation mediated downregulation of FBP1 in human hepatocellular carcinoma and colon cancer, PLoS One 6 (2011) e25564. H. Hirata, K. Sugimachi, H. Komatsu, M. Ueda, T. Masuda, R. Uchi, S. Sakimura, S. Nambara, T. Saito, Y. Shinden, T. Iguchi, H. Eguchi, S. Ito, K. Terashima, K. Sakamoto, M. Hirakawa, H. Honda, K. Mimori, Decreased expression of fructose-1,6-bisphosphatase associates with glucose metabolism and tumor progression in hepatocellular carcinoma, Cancer Res. 76 (2016) 3265–3276. B. Li, B. Qiu, D.S. Lee, Z.E. Walton, J.D. Ochocki, L.K. Mathew, A. Mancuso, T.P. Gade, B. Keith, I. Nissim, M.C. Simon, Fructose-1,6-bisphosphatase opposes renal carcinoma progression, Nature 513 (2014) 251–255. J. Zhang, J. Wang, H. Xing, Q. Li, Q. Zhao, J. Li, Down-regulation of FBP1 by ZEB1mediated repression confers to growth and invasion in lung cancer cells, Mol. Cell. Biochem. 411 (2016) 331–340. H. Sheng, L. Ying, L. Zheng, D. Zhang, C. Zhu, J. Wu, J. Feng, D. Su, Down expression of FBP1 is a negative prognostic factor for non-small-cell lung cancer, Cancer Invest. 33 (2015) 197–204. C. Dong, T. Yuan, Y. Wu, Y. Wang, T.W. Fan, S. Miriyala, Y. Lin, J. Yao, J. Shi, T. Kang, P. Lorkiewicz, D. St Clair, M.C. Hung, B.M. Evers, B.P. Zhou, Loss of FBP1 by snail-mediated repression provides metabolic advantages in basal-like breast cancer, Cancer Cell 23 (2013) 316–331. J. Barretina, G. Caponigro, N. Stransky, K. Venkatesan, A.A. Margolin, S. Kim, C.J. Wilson, J. Lehar, G.V. Kryukov, D. Sonkin, A. Reddy, M. Liu, L. Murray, M.F. Berger, J.E. Monahan, P. Morais, J. Meltzer, A. Korejwa, J. Jane-Valbuena, F.A. Mapa, J. Thibault, E. Bric-Furlong, P. Raman, A. Shipway, I.H. Engels, J. Cheng, G. K. Yu, J. Yu, P. Aspesi Jr., M. de Silva, K. Jagtap, M.D. Jones, L. Wang, C. Hatton, E. Palescandolo, S. Gupta, S. Mahan, C. Sougnez, R.C. Onofrio, T. Liefeld, L. MacConaill, W. Winckler, M. Reich, N. Li, J.P. Mesirov, S.B. Gabriel, G. Getz, K. Ardlie, V. Chan, V.E. Myer, B.L. Weber, J. Porter, M. Warmuth, P. Finan, J.L. Harris, M. Meyerson, T.R. Golub, M.P. Morrissey, W.R. Sellers, R. Schlegel, L.A. Garraway, The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity, Nature 483 (2012) 603–607. B. Gyorffy, A. Lanczky, Z. Szallasi, Implementing an online tool for genomewide validation of survival-associated biomarkers in ovarian-cancer using microarray data from 1287 patients, Endocr. Relat. Cancer 19 (2012) 197–208. X. Liu, X. Wang, J. Zhang, E.K. Lam, V.Y. Shin, A.S. Cheng, J. Yu, F.K. Chan, J.J. Sung, H.C. Jin, Warburg effect revisited: an epigenetic link between glycolysis and gastric carcinogenesis, Oncogene 29 (2010) 442–450. P. Kumar, R. Aggarwal, An overview of triple-negative breast cancer, Arch. Gynecol. Obstet. 293 (2016) 247–269. Y. Zhu, M. Shi, H. Chen, J. Gu, J. Zhang, B. Shen, X. Deng, J. Xie, X. Zhan, C. Peng, NPM1 activates metabolic changes by inhibiting FBP1 while promoting the tumorigenicity of pancreatic cancer cells, Oncotarget 6 (2015) 21443–21451. S. Matkar, P. Sharma, S. Gao, B. Gurung, B.W. Katona, J. Liao, A.B. Muhammad, X. C. Kong, L. Wang, G. Jin, C.V. Dang, X. Hua, An epigenetic pathway regulates sensitivity of breast cancer cells to HER2 inhibition via FOXO/c-Myc axis, Cancer Cell 28 (2015) 472–485. M.H. Cho, J.H. Park, H.J. Choi, M.K. Park, H.Y. Won, Y.J. Park, C.H. Lee, S.H. Oh, Y.S. Song, H.S. Kim, Y.H. Oh, J.Y. Lee, G. Kong, DOT1L cooperates with the c-Mycp300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression, Nat. Commun. 6 (2015) 7821. Z. Chen, Y. Wang, C. Warden, S. Chen, Cross-talk between ER and HER2 regulates c-MYC-mediated glutamine metabolism in aromatase inhibitor resistant breast cancer cells, J. Steroid Biochem. Mol. Biol. 149 (2015) 118–127. N. Dey, B. Young, M. Abramovitz, M. Bouzyk, B. Barwick, P. De, B. Leyland-Jones, Differential activation of Wnt-beta-catenin pathway in triple negative breast cancer increases MMP7 in a PTEN dependent manner, PLoS One 8 (2013) e77425.
Available online at
ScienceDirect www.sciencedirect.com
Fructose-1,6-bisphosphatase is a novel regulator of Wnt/b-Catenin pathway in breast cancer Kaichun Lia , Mingzhen Yinga , Dan Fenga , Jie Dub , Shiyu Chenb , Bing Danb , Cuihua Wangb , Yajie Wanga,* a b
Department of Oncology, Changhai Hospital Affiliated to The Second Military Medical University, Shanghai 200433, PR China Department of Oncology, Tianyou Hospital Affiliated to Tongji University, Shanghai 200331, PR China
A R T I C L E I N F O
Article history: Received 29 August 2016 Received in revised form 10 October 2016 Accepted 17 October 2016 Keywords: FBP1 Oncomine GSEA Wnt/bbeta -Catenin pathway Breast cancer
A B S T R A C T
Fructose-1,6-bisphosphatase (FBP1), the rate-limiting enzyme in gluconeogenesis, is a tumor suppressor that frequently down-regulated in cancers, especially breast cancer. Here, we provide both supporting and contradicting evidences about the expression pattern and function of FBP1 in breast cancer. Data mining of Oncomine database showed that FBP1 is commonly up-regulated in tumor tissues compared with non-tumor tissues regardless of histological type. Analysis of a large-scale cohort derived from Kaplan-Meier Plotter showed that lower FBP1 expression associated with poor clinical outcome. Genetic silencing of FBP1 reduced aerobic glycolysis and the malignant potential of breast cancer cells. Gene set enrichment analysis (GSEA) of the expression profiles of breast cancer cells (n = 59) revealed that cells exhibiting high expression of FBP1 had a lower activity of Wnt/b-Catenin pathway. FBP1 downregulation enhanced the activity of Wnt/b-Catenin pathway and increased the level of its downstream targets, including c-Myc and MMP7. Collectively, our findings indicate that elevated FBP1 is a critical modulator in breast cancer progression by altering glucose metabolism and the activity of Wnt/ b-Catenin pathway. ã 2016 Elsevier Masson SAS. All rights reserved.
1. Introduction Breast cancer is the leading cause of cancer-related death in women through the world. Although the great improvement in diagnostic techniques and targeted therapies, the prognosis of breast cancer is still dissatisfy [1]. The considerable heterogeneity is a key factor that drives the malignant phenotypes of breast cancer [2]. Therefore, identifying the common characters in breast cancer across different subtypes is of great importance to develop potential therapeutic methods. Reprogrammed energy metabolism is an emerging hallmark of cancer cells [3]. The interconnectedness between glycolysis/ oxidative phosphorylation and anabolic gluconeogenesis drives the metabolic phenotype of cancer cells. Fructose-1,6-bisphosphatase (FBP), which catalyzes the hydrolysis of fructose-1,6bisphosphate to fructose-6-phosphate and inorganic phosphate, is a critical regulatory enzyme in gluconeogenesis [4]. Two isoforms
* Corresponding author at: Department of Oncology, Changhai Hospital Affiliated to The Second Military Medical University, No.168, Changhai Road, Shanghai 200433, PR China. E-mail address: [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.biopha.2016.10.050 0753-3322/ã 2016 Elsevier Masson SAS. All rights reserved.
are present in mammalian cells, FBP1 and FBP2. FBP1 is widely expressed in different tissues and FBP2 is specially expressed in muscle [5]. Recently, it has been reported that FBP1 is involved in the initiation and progression of multiple human cancers, such as hepatocellular carcinoma [6,7], renal carcinoma [8], lung cancer [9,10] and breast cancer [11]. FBP1 is frequently down-regulated and most reported as a tumor suppressor in cancer development. For example, ZEB1-mediated down-regulation of FBP1 in lung cancer predicts a poor prognosis and confers tumor cells with growth and invasion advantage [9,10]. And FBP1 is ubiquitously depleted and plays tumor-suppressive functions in clear cell renal cell carcinoma [8]. In breast cancer, the Snail-G9a-Dnmt1 complex mediates the depletion of FBP1 [11]. And notably, FBP1 is highly expressed in luminal subtype but nearly undetected in triplenegative breast cancer (TNBC). However, by data mining current available database, including Oncomine, Cancer Cell Line Encyclopedia database (CCLE) and The Cancer Genome Atlas database (TCGA), we found several contradictory evidence about the expression pattern of FBP1 in breast cancer. Furthermore, in this study, we also investigated the cellular functions and underlying mechanisms of FBP1 in breast cancer.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
2. Materials and methods 2.1. Cell culture, transfection and western blotting MCF7 and BT474 cell lines were obtained from American Type Culture Collection (ATCC, Manassas, Va., USA). Cells were cultured in DMEM medium (Invitrogen, Carlsbad, Calif., USA) supplemented with 10% fetal bovine serum (FBS) at 37 C with 5% CO2. For transfection, a total of 50 ng/ml siRNAs against FBP1 were transfected with RNAi MAX (Thermo Scientific, USA). To exclude unspecific effects, two efficient siRNAs were used in this study. The antisense sequences against FBP1 are 50 -AACATGTTCATAACCAGGTCG-30 (si-FBP1-1) and 50 -ATGTTGGAAGATCCATCAAGG-30 (si-FBP1-2). Western blotting was performed to determine to protein level of FBP-1 and b-actin. Antibodies used for FBP-1 (ab109732) and b-actin (ab8226) were purchased from Abcam. 2.2. Measurement of glucose and lactate level To detect the glucose consumption and lactate production in breast cancer cells upon FBP1 knockdown, the Glucose Colorimetric Assay Kit II (K686-100, BioVision) and Lactate Colorimetric/ Fluorometric Assay Kit (K607-100, BioVision) were used according to the manufacturer’s protocols, respectively. Specially, cell culture medium used in this experiment was phenol red-free. The glucose and lactate level were calculated using a standard calibration curve prepared under the same condition. 2.3. Cell proliferation and cell invasion assays For cell proliferation assay, 3000 cells were seeded in triplicate and MTT assay was performed according to the manufacturer’s protocols. For cell invasion assay, transwell model was used. Briefly, a total of 2 104 cells were seeded into the upper chamber of matrigel-coated filters (BD Bioscience, USA) and the lower chamber was filled with 800 ml of culture medium containing 5% FBS. After incubation for 48 h, cells remained on the upper chamber were removed. The invaded cells were fixed with 4% paraformaldehyde, stained with 0.5% crystal violet and counted in three random fields.
1145
reverse: 50 -GGCCCCATAAGGAGCTGAAT-30 ; GADPH forward: 50 CTGGGCTACACTGAGCACC-30 , GAPDH reverse: 50 -AAGTGGTCGTTGAGGGCAATG-30 . 2.6. Bioinformatic analysis Microarray expression data of FBP1 from multiple independent data sets were derived from the Oncomine repository (http:// www.oncomine.org/). And to examine the relative mRNA expression levels of FBP1 between normal and cancer samples in a variety of tissue types, the distributions of log2 median centered signal intensities were plotted using box plots. To determine the differential expression of FBP1 between breast cancer and normal tissues, a combined filter was used to display indicated datasets. The Cancer Type was defined as Breast Cancer and Data Type was mRNA, whereas Analysis Type was Cancer vs Normal Analysis. The mRNA expression of FBP1 in a repertoire of cancers located was analyzed by the online tool located in The Cancer Cell Line Encyclopedia database (CCLE, www.broadinstitute.org/ccle/home) [12]. All raw and processed data are available at the CCLE website and the website offers direct links to data visualization tools such as genepattern-based analysis tools for expression. By searching the gene symbol, FBP1, the expression level of FBP1 among different cancers was automatically generated by the CCLE analysis tool. Simultaneously, the processed data for FBP1 could be downloaded and the expression level of FBP1 in breast cancer cells could be extracted. The prognostic value of FBP1 in breast cancer was analyzed by an online survival analysis tool (http:// kmplot.com/analysis) and the precise method was reported previously [13]. Briefly, by entering the gene symbol, FBP1, the survival curve based on FBP1 expression in breast cancer could be automatically generated by the Kaplan Meier-plotter analysis tool. Specially, to examine the prognostic value of FBP1 in TNBC, the ER, PR and HER2 status were selected as negative. The transcriptomic difference between FBP-1 low group and FBP-1 high group was uncovered by Gene Set Enrichment Analysis (GSEA), which is also available in the CCLE online analysis tool (www.broadinstitute.org/ ccle/data/browseAnalyses). Based on FBP-1 expression level, the breast cancer cell lines were divided into FBP-1 low group or FBP-1 high group and subsequently subjected to GSEA. 2.7. Statistical analysis
2.4. Luciferase reporter assay The si-Control or si-FBP1 cells were seeded in 96-well plate and transfected with 1.0 mg of pTOP-Luc (b-catenin/TCF4 reporter plasmid) or pGL3-Luc (positive control plasmid) using lipofectamine 2000 (#11668027, Thermo Scientific, USA). After 24 h, cells were lysed and subjected to luciferase activity assay according to the manufacturer’s protocols (#TM040, Promega Corporation, Madison, WI, USA). Luciferase activity was normalized by total protein level of the samples. The ratio of luciferase activity was calculated and normalized to Renilla and each assay was done in triplicate. 2.5. Quantitative real-time PCR Total RNA was extracted from breast cancer cells using Trizol reagent (Takara, Dalian, China) and 500 ng RNA was used for synthesis of complementary DNA (cDNA) by reverse transcriptase (Takara, Japan). Quantitative real-time PCR was subsequently performed with SYBR Premix Ex Taq (Takara, Dalian, China) using ABIPrism-7500 Sequence Detector System (Applied Biosystems). GAPDH was used as the internal control for quantification. Primer sequences used in this study were listed as follows, FBP1 forward: 50 -CTACGCCAGGGACTTTGACC-30 , FBP1
Experiments were repeated at least twice. Data were expressed as the means SD. The statistical analysis was performed with GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA). P values were calculated based on two-tailed, unpaired Student’s ttests. P value less than 0.05 was considered statistically significant. 3. Results 3.1. Up-regulated FBP1 predicts a good prognosis in breast cancer Down-regulation of FBP1 has been reported in several different human malignancies, especially triple-negative breast cancer [11]. To fully uncover the expression profile of FBP1 in breast cancer, we firstly searched the Oncomine database and TCGA database (http:// cancergenome.nih.gov). To our surprise, FBP1 was consistently upregulated in tumor tissues compared with corresponding normal tissues across 11 independent cohorts derived from Oncomine database (Fig. 1A). Then by mining data deposited in TCGA, we found that FBP1 was highly expressed in breast cancer tissues irrespective of histological type, including ductal, lobular, mixed ductal and lobular, and mucinous breast cancer (Fig. 1B). Furthermore, data from the CCLE database showed that FBP1 was also highly expressed in breast cancer cell lines compared with the
1146
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
Fig. 1. Up-regulated FBP1 predicts a good prognosis in breast cancer. (A) The mRNA expression level of FBP1 in 11 independent datasets of breast cancer. (B) Box plot of FBP1 level in different subtypes of breast cancer. Data was derived from TCGA database. (C) The expression level of FBP1 in different human cancer cell lines. Data were derived from CCLE database. (D) The prognostic value of FBP1 expression in all subtypes of breast cancer. Data were analyzed in Kaplan Meier-plotter (http://kmplot.com/analysis/). (E) The prognostic value of FBP1 expression in triple-negative breast cancer.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
majority of other human cancer cells, especially originated from melanoma, osteosarcoma and glioma (Fig. 1C). These data above suggested that up-regulation of FBP1, to some extent, is a common event in breast cancer. To interrogate the prognostic value of FBP1 in breast cancer, Kaplan-Meier analysis was performed in a largescale dataset (http://kmplot.com/analysis/). The result showed that the higher expression of FBP1 predicted a good prognostic effect on progression-free survival of breast cancer patients, which is consistent with previous notion that FBP1 is a tumor suppressor in cancers (Fig. 1D). However, inconsistent with previous report in breast cancer, FBP1 did not predict a better prognosis in TNBC (Fig. 1E). 3.2. Genetic silencing of FBP1 promotes tumor progression of breast cancer in vitro Next, to determine the cellular functions of FBP1 in breast cancer, loss-of-function assay was performed. By searching CCLE database, two cell lines with higher FBP1 expression, MCF7 and BT474, were selected for further investigation (Fig. 2A). Two
1147
specific small interfere RNAs targeting FBP1 resulted in marked reduction in protein level of FBP1 in both MCF7 and BT474 cells (Fig. 2B). The regulatory functions of FBP1 in glycolysis have been reported in multiple cancers. To confirm the known suppressive role of FBP1 in glycolysis, we therefore detected two indictors of glycolysis, glucose consumption and lactate production upon genetic silencing of FBP1. Consistently, knockdown of FBP1 resulted in increased glucose consumption (Fig. 2C) and lactate production (Fig. 2D). Moreover, we evaluated the effects of FBP1 knockdown on cell proliferation and invasive potential of breast cancer. As shown in Fig. 2E, silencing of FBP1 significantly promoted the cell proliferation ratio of MCF7 and BT474 cells. Similarly, FBP1 knockdown increased the invasive capacity of MCF7 and BT474 cells as revealed by transwell assay (Fig. 2D). 3.3. FBP1 is negative regulator of wnt/b-Catenin pathway in breast cancer To illustrate the mechanism by which FBP1 inhibits tumor progression in breast cancer, Gene Set Enrichment Analysis (GSEA)
Fig. 2. Genetic silencing of FBP1 promotes tumor progression in vitro. (A) The mRNA level of FBP1 in breast cancer cell lines. (B) The silencing efficiency of FBP1 in MCF7 and BT474 cells was confirmed by Western Blotting. Glucose utilization (C) and lactate production (D) of in MCF7 and BT474 cells were measured upon silencing of FBP1. Cell proliferation (E) and invasive capacity (F) of MCF7 and BT474 cells were detected after knockdown of FBP1. *, p < 0.05; **, p < 0.01.
1148
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
was performed in the expression profiles of breast cancer cell lines. Based on FBP1 expression in CCLE dataset, a total of 59 breast cancer cell lines was divided into two group, the lower FBP1 group (Class + 1, n = 31) and the higher FBP1 group (Class + 2, n = 28). The difference of FBP1 expression between two groups was significant (Fig. 3A). GSEA showed that genes in Wnt/b-Catenin signaling were prominently enriched in the lower FBP1 group, indicating that FBP1 is a potential suppressor of Wnt/b-Catenin signaling (Fig. 3B). To test this hypothesis, luciferase reporter assay was performed. As indicated in Fig. 3C, FBP1 knockdown remarkably promoted the activity of Wnt/b-Catenin signaling in both MCF7 and BT474 cells. Consistent with this result, the two downstream targets, c-Myc and MMP7, were also increased in FBP1-silenced cells (Fig. 3D). Notably, data from TCGA database also confirmed the inverse relationship between FBP1 and c-Myc or MMP7 (Fig. 3E). Collectively, these data above suggested that the tumorsuppressive roles of FBP1 might be mediated by inactivation of Wnt/b-Catenin pathway. 4. Discussion In the present study, we provide several novel insights into the FBP1 and breast cancer. First, FBP1 is highly expressed in breast cancer. Indeed, down-regulation of FBP1 in cancers has been reported in all currently published articles. And several regulators of its repression have been revealed, including DNA hypermethylation [6], ZEB1 [9], NF-kB [14] and Snail [11]. However, in this study, by a large-scale data mining, we confirmed that FBP1
was overexpressed, at least to some extent, in breast cancer. Previous report in breast cancer has described that FBP1 was preferentially depleted in TNBC and certainly expressed at comparable expression in luminal breast cancer [11]. As TNBC accounts for 10–20% of all invasive breast cancers, we therefore proposed that the majority of other invasive mammary carcinomas are highly expressed with FBP1 [15]. Then by kaplan-Meier analysis, we found that FBP1 predicted a favorable prognosis in breast cancer, which is consistent with the observations in other cancers. Specially, in TNBC, no correlation between FBP1 and patients’ survival time was found, indicating the prognostic value of FBP1 in breast cancer may be molecular type-dependent. Second, by loss-of-function study, we further confirm the tumor-suppressive role of FBP1 in breast cancer. FBP1 knockdown has been associated with enhanced glycolysis, which is a common metabolic phenotype in cancer cells [8,9,11,16]. Consistently, knockdown of FBP1 results in decreased glucose consumption and lactate production in breast cancer cells. Apart from implications in glucose metabolism, FBP1 also functions as a tumor suppressor through induction of cell cycle arrest and inhibition of cancer stem cell-like property. In line with this notion, silencing of FBP1 significantly inhibited the cell proliferation rate and invasive capacity of breast cancer cells. Third, our data indicates that FBP1 is a novel negative regulator of Wnt/b-Catenin signaling. Previously, two critical mechanisms of FBP1 in cancers were revealed. In renal cancer, FBP1 restrains tumor progression in an enzyme activity-independent manner, but by suppressing the hypoxia-inducible factor (HIF) function through
Fig. 3. FBP1 is negative regulator of Wnt/b-Catenin pathway in breast cancer. (A) The expression level of FBP1 in lower group (n = 31, Class + 1) and high group (n = 28, Class + 2). (B) Genes in Wnt/b-Catenin signaling identified by GSEA were enriched in FBP1 lower group. (C) Luciferase activity assay was performed in MCF7 and BT474 cells at 48 h after transfected with specific FBP1 siRNAs. (D) The mRNA level of c-Myc and MMP7 was measured at 48 h after knockdown of FBP1. (E) Correlation between FBP1 expression and c-Myc or MMP7 was analyzed by the expression data from TCGA database. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
K. Li et al. / Biomedicine & Pharmacotherapy 84 (2016) 1144–1149
direct interaction with the inhibitory domain of HIF [8]. In breast cancer, FBP1 enhances the interaction between b-Catenin and Tcell factor and then reprograms glucose metabolism, which ultimately increases tumorigenecity [11]. In this study, by performing GSEA in a total of 59 breast cancer cells, we revealed that FBP is closely associated with the activity of Wnt/b-Catenin signaling. This data not only supported, but also amplified the findings reported previously in breast cancer. However, due to insufficient data to analyze the difference among different types of breast cancer, especially TNBC, we cannot fully rule out the possibility of other factors that contribute to the malignant phenotypes among different subtypes of breast cancer. Furthermore, we showed that c-Myc and MMP7, two crucial downstream factors of Wnt/b-Catenin signaling were negatively correlated the level of FBP1. As c-Myc is also a transcription factor involved in multiple cellular functions [17–19], especially its regulatory role in glycolysis, we hypothesized that the inhibitory effect of FBP1 in glycolysis might be mediated by down-regulation of c-Myc. And MMP7 is a potential effector in FBP1-mediated inhibition of invasive potential [20]. However, the precise mechanism by which how FBP1 regulates Wnt/b-Catenin signaling warrant further investigation. In conclusion, our data, as a proof of principle, suggest that FBP1 is frequently up-regulated in most breast cancer and predicts a good prognosis. FBP1 acts a tumor suppressor by inactivating Wnt/ b-Catenin signaling. Thus, it will be of great value to develop possible application of FBP1 in the treatment of several subtypes of breast cancer. Conflict of interest The authors declare that there is no conflict of interests.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Acknowledgement
[16]
We thank Shanghai Promoters Information Technology Co., Ltd for technical assistance.
[17]
Reference:
[18]
[1] C. DeSantis, J. Ma, L. Bryan, A. Jemal, Breast cancer statistics 2013, CA. Cancer J. Clin. 64 (2014) 52–62. [2] F. Beca, K. Polyak, Intratumor heterogeneity in Breast cancer, Adv. Exp. Med. Biol. 882 (2016) 169–189. [3] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144 (2011) 646–674. [4] M.S. Jurica, A. Mesecar, P.J. Heath, W. Shi, T. Nowak, B.L. Stoddard, The allosteric regulation of pyruvate kinase by fructose-1,6-bisphosphate, Structure 6 (1998) 195–210. [5] A. Marin-Hernandez, S. Rodriguez-Enriquez, P.A. Vital-Gonzalez, F.L. FloresRodriguez, M. Macias-Silva, M. Sosa-Garrocho, R. Moreno-Sanchez,
[19]
[20]
1149
Determining and understanding the control of glycolysis in fast-growth tumor cells. Flux control by an over-expressed but strongly product-inhibited hexokinase, FEBS J. 273 (2006) 1975–1988. M. Chen, J. Zhang, N. Li, Z. Qian, M. Zhu, Q. Li, J. Zheng, X. Wang, G. Shi, Promoter hypermethylation mediated downregulation of FBP1 in human hepatocellular carcinoma and colon cancer, PLoS One 6 (2011) e25564. H. Hirata, K. Sugimachi, H. Komatsu, M. Ueda, T. Masuda, R. Uchi, S. Sakimura, S. Nambara, T. Saito, Y. Shinden, T. Iguchi, H. Eguchi, S. Ito, K. Terashima, K. Sakamoto, M. Hirakawa, H. Honda, K. Mimori, Decreased expression of fructose-1,6-bisphosphatase associates with glucose metabolism and tumor progression in hepatocellular carcinoma, Cancer Res. 76 (2016) 3265–3276. B. Li, B. Qiu, D.S. Lee, Z.E. Walton, J.D. Ochocki, L.K. Mathew, A. Mancuso, T.P. Gade, B. Keith, I. Nissim, M.C. Simon, Fructose-1,6-bisphosphatase opposes renal carcinoma progression, Nature 513 (2014) 251–255. J. Zhang, J. Wang, H. Xing, Q. Li, Q. Zhao, J. Li, Down-regulation of FBP1 by ZEB1mediated repression confers to growth and invasion in lung cancer cells, Mol. Cell. Biochem. 411 (2016) 331–340. H. Sheng, L. Ying, L. Zheng, D. Zhang, C. Zhu, J. Wu, J. Feng, D. Su, Down expression of FBP1 is a negative prognostic factor for non-small-cell lung cancer, Cancer Invest. 33 (2015) 197–204. C. Dong, T. Yuan, Y. Wu, Y. Wang, T.W. Fan, S. Miriyala, Y. Lin, J. Yao, J. Shi, T. Kang, P. Lorkiewicz, D. St Clair, M.C. Hung, B.M. Evers, B.P. Zhou, Loss of FBP1 by snail-mediated repression provides metabolic advantages in basal-like breast cancer, Cancer Cell 23 (2013) 316–331. J. Barretina, G. Caponigro, N. Stransky, K. Venkatesan, A.A. Margolin, S. Kim, C.J. Wilson, J. Lehar, G.V. Kryukov, D. Sonkin, A. Reddy, M. Liu, L. Murray, M.F. Berger, J.E. Monahan, P. Morais, J. Meltzer, A. Korejwa, J. Jane-Valbuena, F.A. Mapa, J. Thibault, E. Bric-Furlong, P. Raman, A. Shipway, I.H. Engels, J. Cheng, G. K. Yu, J. Yu, P. Aspesi Jr., M. de Silva, K. Jagtap, M.D. Jones, L. Wang, C. Hatton, E. Palescandolo, S. Gupta, S. Mahan, C. Sougnez, R.C. Onofrio, T. Liefeld, L. MacConaill, W. Winckler, M. Reich, N. Li, J.P. Mesirov, S.B. Gabriel, G. Getz, K. Ardlie, V. Chan, V.E. Myer, B.L. Weber, J. Porter, M. Warmuth, P. Finan, J.L. Harris, M. Meyerson, T.R. Golub, M.P. Morrissey, W.R. Sellers, R. Schlegel, L.A. Garraway, The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity, Nature 483 (2012) 603–607. B. Gyorffy, A. Lanczky, Z. Szallasi, Implementing an online tool for genomewide validation of survival-associated biomarkers in ovarian-cancer using microarray data from 1287 patients, Endocr. Relat. Cancer 19 (2012) 197–208. X. Liu, X. Wang, J. Zhang, E.K. Lam, V.Y. Shin, A.S. Cheng, J. Yu, F.K. Chan, J.J. Sung, H.C. Jin, Warburg effect revisited: an epigenetic link between glycolysis and gastric carcinogenesis, Oncogene 29 (2010) 442–450. P. Kumar, R. Aggarwal, An overview of triple-negative breast cancer, Arch. Gynecol. Obstet. 293 (2016) 247–269. Y. Zhu, M. Shi, H. Chen, J. Gu, J. Zhang, B. Shen, X. Deng, J. Xie, X. Zhan, C. Peng, NPM1 activates metabolic changes by inhibiting FBP1 while promoting the tumorigenicity of pancreatic cancer cells, Oncotarget 6 (2015) 21443–21451. S. Matkar, P. Sharma, S. Gao, B. Gurung, B.W. Katona, J. Liao, A.B. Muhammad, X. C. Kong, L. Wang, G. Jin, C.V. Dang, X. Hua, An epigenetic pathway regulates sensitivity of breast cancer cells to HER2 inhibition via FOXO/c-Myc axis, Cancer Cell 28 (2015) 472–485. M.H. Cho, J.H. Park, H.J. Choi, M.K. Park, H.Y. Won, Y.J. Park, C.H. Lee, S.H. Oh, Y.S. Song, H.S. Kim, Y.H. Oh, J.Y. Lee, G. Kong, DOT1L cooperates with the c-Mycp300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression, Nat. Commun. 6 (2015) 7821. Z. Chen, Y. Wang, C. Warden, S. Chen, Cross-talk between ER and HER2 regulates c-MYC-mediated glutamine metabolism in aromatase inhibitor resistant breast cancer cells, J. Steroid Biochem. Mol. Biol. 149 (2015) 118–127. N. Dey, B. Young, M. Abramovitz, M. Bouzyk, B. Barwick, P. De, B. Leyland-Jones, Differential activation of Wnt-beta-catenin pathway in triple negative breast cancer increases MMP7 in a PTEN dependent manner, PLoS One 8 (2013) e77425.