- Email: [email protected]
Apoptotic effect of EGCG in HT-29 colon cancer cells via AMPK signal pathway
Cancer Letters 247 (2007) 115–121 www.elsevier.com/locate/canlet
Apoptotic effect of EGCG in HT-29 colon cancer cells via AMPK signal pathway Jin-Taek Hwang a, Joohun Ha a, In-Ja Park b, Seong-Kyu Lee c, Haing Woon Baik c, Young Min Kim d, Ock Jin Park b,* a
Department of Biochemistry and Molecular Biology, Medical Research Center for Bioreaction to Reactive Oxygen Species, Kyung Hee University College of Medicine, Seoul 130-701, South Korea b Department of Food and Nutrition, Hannam University, 133 Ojeong-dong Daedeok-gu, Daejeon 306-791, South Korea c Department of Biochemistry, College of Medicine, Eulji University, Daejeon 143-5, South Korea d Department of Biological Sciences, Hannam University, 133 Ojeong-dong Daedeok-gu, Daejeon 306-791, South Korea Received 26 July 2005; received in revised form 6 March 2006; accepted 27 March 2006
Abstract EGCG [(K)epigallocatechin-3-gallate], a green tea-derived polyphenol, has been shown to suppress cancer cell proliferation, and interfere with the several signaling pathways and induce apoptosis. Practically, there is emerging evidence that EGCG has a potential to increase the efficacy of chemotherapy in patients. We hypothesized that EGCG may exert cell cytotoxicity through modulating AMPK (AMP-activated protein kinase) followed by the decrease in COX-2 expression. EGCG treatment to colon cancer cells resulted in a strong activation of AMPK and an inhibition of COX-2 expression. The decreased COX-2 expression as well as prostaglandin E2 secretion by EGCG was completely abolished by inhibiting AMPK by an AMPK inhibitor, Compound C. Also, the activation of AMPK was accompanied with the reduction of VEGF (vascular endothelial growth factor) and glucose transporter, Glut-1 in EGCG-treated cancer cells. These findings support the regulatory role of AMPK in COX-2 expression in EGCG-treated cancer cells. Furthermore, we have found that reactive oxygen species (ROS) is an upstream signal of AMPK, and the combined treatment of EGCG and chemotherapeutic agents, 5-FU or Etoposide, exert a novel therapeutic effect on chemoresistant colon cancer cells. AMPK, a molecule of newly defined cancer target, was shown to control COX-2 in EGCG-treated colon cancer cells. q 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Epigallocatechin-3-gallate; AMP-activated protein kinase; Cyclooxygenase 2; Etoposide; 5-Flurouracil
1. Introduction Cancer chemoprevention by natural compounds has been investigated by many investigators, and among them bioactive flavonoid compounds, widely distributed in many beverages and food products, have been * Corresponding author. Tel.: C82 42 629 7493; fax: C82 42 629 7490. E-mail address: [email protected] (O.J. Park).
implicated in the prevention of human carcinogenesis possibly through their inhibitory activities of cell proliferation or survival [1]. (K)-Epigallocatechin-3gallate (EGCG), a major constituent of green tea, has been shown to exert the inhibitory effects on certain human cancers through the induction of several signal pathways [2–4]. Although the major and precise EGCG cancer preventive activities have not been elucidated yet, the several possibilities have been proposed. Importance of various cancer cell control including
0304-3835/$ - see front matter q 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2006.03.030
116
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
inhibiting tumor promotion enzymes such as cyclooxygenase-2 (COX-2) and lipoxygenase, blocking angiogenesis or induction of apoptosis by EGCG has been discussed [5]. AMP-activated protein kinase (AMPK) is a metabolic-sensing protein kinase, which plays an essential role as an energy-sensor mainly in ATP-deprived conditions [6]. Therefore, AMPK is known to play a major protective role under metabolic stressed conditions. In the activated states, AMPK down-regulates several anabolic enzymes and thus shuts down the ATP-consuming metabolic pathways [7]. Several reports have observed the strong pro-apoptotic potential of AMPK in activated conditions such as AICAR (AMPK activator)-treated cells or constitutively active AMPK mutants [8]. In the present study, we have investigated the role of AMP-activate protein kinase (AMPK) as well as COX-2 in EGCG-induced apoptotic states in human colon cancer cells. EGCG has shown to down-regulate COX-2 derived prostaglandin synthesis and the several compounds that inhibit COX-2 enzymes possess chemo-sensitizing and radio-sensitizing activities in vitro and in vivo [9]. We have observed the effects of EGCG on AMPK activation and COX-2 inhibition, and the interaction of these molecular events leading to apoptosis. Also, the possibility of EGCG rendering the efficacy in the combination chemotherapy with the anti-cancer agents in HT-29 colon cancer cells was explored.
PMSF, 1 mM sodium orthovanadate, 1 mM NaF, 1 mg/ml aprotinin, 1 mg/ml leupeptin, 1 mg/ml pepstatin) and subjected to the western blot analysis. 2.3. Prostaglandin enzyme ELISA assay The supernatants of control and treated cell cultures were collected by centrifuged at 2500!g for 5 min, and the sample was added to appropriate wells and PGE2 conjugate was pipetted into all wells except the blank well. Finally, monoclonal antibody against PGE2 was added to all wells except wells for blank and nonspecific binding test, followed by incubation for 18 h at 4 8C. The plates were washed four times and color reaction was developed by the addition of tetramethylbenzidine substrate. After 30 min of incubation at room temperature, the reaction was quenched by addition of 1 M sulfuric acid. Optical density was measured at 450 nm on an ELX800 reader (Bio-Tec Instruments, Inc., Winooski, Vermont). 2.4. Chromatin staining with Hoechst 33342 Apoptosis was observed by chromatin staining with Hoechst 33342, as previously described. Cells were incubated with each stimuli. After incubation the supernatant was discarded and cells were fixed with 3.5% formaldehyde in PBS for 30 min at room temperature, washed four times with PBS and exposed to Hoechst 33342 at 10 mM for 30 min at room temperature. Cell preparations were examined under ultraviolet illumination with a fluorescence microscope (Olympus Optical Co., Tokyo, Japan).
2. Material and method
2.5. Cell proliferation by MTT assay
2.1. Cell culture and reagents
Cells seeded on 96-well micro plates at 4000 cells/well were incubated with the test compounds for indicated time period. Respectively medium was removed and then incubated with 100 ml of MTT solution (2 mg/ml MTT in PBS) for 4 h. MTT is converted to a blue formazan. Absorbance was determined using an auto reader.
The HT-29 human colon cancer line was purchased from ATCC (Gaithersburg, MD). Cells were cultured in RPMI1640 containing 10% FBS under normoxic conditions (20% O2, 5% CO2, 75% N2). (K)-Epigallocatechin-3-gallate, Etoposide, 5-fluorouracil, MTT, Hoechst 33342, DCFH, NAC and AICAR were obtained from Sigma. Celecoxib was purchased from Pharmacia Korea; Compound C was a gift from Merck Company. Prostaglandin enzyme assay kit was purchased from Amersham Pharmacia Biotech. The anti-phosphospecific antibodies that recognize phosphorylated ACCSer79 were from Cell Signaling Technology. Antibodies for COX-2, PARP, P53, and b-actin were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). 2.2. Protein extract and western blotting Cells were rinsed twice with ice-cold PBS and scraped with lysis buffer (50 mM Tris–HCl, pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM
2.6. RNA isolation and RT-PCR Total RNA was extracted from cells with Trizol Reagent (Life Technologies, Glasgow, United Kingdom) according to manufacturer instructions. We synthesized cDNA from 1 mg. Total RNA using a first strand cDNA synthesis kit (Amersham Pharmacia Biotech, Piscataway, New Jersey). The cDNA fragment was amplified by PCR using the following specific primers: VEGF: sense 5 0 -AGGAGGGCAGAATCATCACG-3 0 , anti-sense 5 0 -CAAGGCCCACAGGGATTTT CT-3 0 ; Glut-1: sense 5 0 -CGGGCCAAGAGTGTGAA-3 0 , antisense 5 0 -TGACGATACCGGAGCCAATG-3 0 ;
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
117
3. Results 3.1. EGCG and celecoxib inhibit proliferation of HT-29 colon cancer cells It has been reported that celecoxib, a selective COX2 inhibitor, inhibits cancer cell proliferation and EGCG can inhibit tumor cell proliferative or survival related proteins [10]. We investigated cancer cell inhibitory capacity of EGCG in comparison with celecoxib. As shown Fig. 1A, EGCG or celecoxib induced growth inhibition of HT-29 cells in a dose-dependent manner, and also EGCG 10 mM or celecoxib 50 mM have shown to inhibit the tumor cells growth for the indicated time period. EGCG and celecoxib have similarly effects on cancer cell death implying the possible similarity in blocking cell growth (Fig. 1B). 3.2. Effects of EGCG on the expression of VEGF and Glut-1 genes
Fig. 1. The effects of cell death on EGCG or celecoxib treated HT-29 colon cancer cells. Cells were treated with different concentrations of EGCG or celecoxib for 48 h (A) or cells were exposed to EGCG (100 mM) or celecoxib (50 mM) for the indicated time periods (B), and then cell viability was measured by MTT assay or Hoechst 33342 dye. The data are presented as the meanGSD.
Most of malignant cancer involves the alteration of survival genes such as glut-1 or VEGF, which play important roles in the chemo-resistance process. Our previous study has shown that the ability of genistein to induce apoptosis or decrease chemo-resistance in cancer cells was related to the expression of survival genes such as glut1 [11]. Therefore, in order to evaluate the effect of EGCG on survival gene expression, VEGF and Glut-1 gene expressions were determined using a semi-quantitative RT-PCR method. As shown Fig. 2, EGCG distinctively reduced mRNA levels of both Glut-1 and VEGF genes.
b-actin: sense 5 0 -GTGGGGGCGCCCAGGCACCA-3 0 , anti-sense 5 0 -CTCCTTAATGTCACGCACCA TTTC-3 0 ; PCR was initiated in a thermal cycle programmed at 95 8C for 5 min, 94 8C for 1 min, 58 8C for 1 min, 72 8C for 1 min, and amplified for 25 cycles. The amplified products were visualized on 1% agarose gels. 2.7. ROS measurement Cells were seeded on 12 well micro plate on cover glasses, after stimulated for indicated time period, respectively, cells were incubated with 10 mM of 2 0 ,7 0 -dichlorofluorescein diacetate (Sigma) for 30 min, and washed with PBS and fluorescence was measured with a fluorescence microscope.
Fig. 2. The effect of EGCG on Glut1 and VEGF mRNA levels. HT-29 cells exposed to different concentrations of EGCG for 12 h, and the mRNA levels of glut1, VEGF and b-actin genes were examined by RT-PCR using specific primers.
118
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
3.3. EGCG abrogates the basal COX-2 expression through AMPK phosphorylation while promoting the apoptotic protein expressions Several studies have indicated that overexpression of COX-2 is strongly related to a number of solid malignancies including colon cancers [12], and the ability of EGCG to modulate COX-2 expression along with apoptotic protein p53 expression and PARP cleavage can provide the involvement of EGCG in the control of COX-2 and apoptosis. EGCG abrogated
COX-2 expression in a dose-dependent manner, and the stimulation of p53 expression and PARP cleavage was also observed (Fig. 3A). We have investigated the connection between AMP-activated protein kinase (AMPK) and COX-2 expression by EGCG treatment. EGCG activated AMPK in a dose-dependent manner as shown by the increased phosphorylation of ACC (acetyl-CoA carboxylase) and the elevation of AMPK expression (Fig. 3A). The phosphorylation level of ACC serine 79, the best-characterized phosphorylation site of AMPK, designates AMPK activation. The level
Fig. 3. The effects of AMPK activator or inhibitor on EGCG INDUCED COX-2, phospho-ACC, PARP cleavage, p53 level or prostaglandINE2 productions. HT-29 cells were treated with EGCG for the indicated concentrations during 6 h (A), also cells were pretreated with compound C (10 mM) for 30 min, and exposed to EGCG (100 mM) for 6 h, and then COX-2, PARP cleavage, p53, phospho-ACC and AMPK level were determined by western blot analysis (B). Also independently HT-29 cell was pretreated with compound C (10 mM) for 30 min, then exposed to EGCG (100 mM) or celecoxib (50 mM) for 12 h, and amount of PGE2 present in the supernatant of the culture medium was measured by ELISA (C). Also cells were exposed to AICAR (1 mM) for 6 h, respectively, performed western blotting, PGE2 production tested, also apoptosis was examined with Hoechst 33342 dye (D).
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
119
reactive oxygen species, the cells were incubated EGCG in the presence of the antioxidant NAC, a broad ROS scavenger. NAC treatment almost completely scavenged the EGCG-generated ROS as well as AMPK activation (also Fig. 4). These results suggest that EGCG has a capacity to generate ROS, which is the upstream signal of AMPK acti-vation.
Fig. 4. ROS generated by EGCG in HT-29 colon cancer cells. Cells were exposed to EGCG for 6 h, in the presence or absence of NAC (5 mM). After additional 30 min incubation in the presence of 10 mM DCFH-DA, and changes in fluorescence intensity were measured by fluorescence-activated cell scanning analysis. Also under the identical conditions, the phosphorylation level of ACC-Ser79 (P-ACC), were examined.
of COX-2 inhibition induced by EGCG was significantly abolished by an AMPK inhibitor Compound C. Also, the marked decrease in PARP cleavage was observed with the treatment of Compound C (Fig. 3B). EGCG treated cell abrogates basal PGE2 production, which was significantly increased by the treatment with EGCG and Compound C (Fig. 3C). These results indicate that AMPK may play a significant role in the regulation of COX-2, and PARP cleavage in EGCGtreated cancer cells. We further tested the role of AMPK in COX-2 regulation with AICAR (5-aminoimidazole-4-carboxamide-1-b-D-ribofuranoside), a cell permeable AMPK activator (Fig. 3D). The results have suggested that AMPK is required for the inhibition of COX-2 and thus reduction of PGE2 in cancerous cells treated with EGCG. 3.4. ROS is an upstream of AMPK activation treated with EGCG In previous report, it was found that triggering ROS could lead to AMPK activation [13], and therefore, we investigated whether EGCG treatment resulted in ROS increase. Measurement of H2O2 production using DCFH-DA in EGCG treated cells has shown that ROS are evidently elevated by EGCG (Fig. 4). To determine if the observed AMPK activation involved
Fig. 5. Synergistic apoptotic effects of combined with EGCG and various chemotherapy drugs in HT-29 colon cancer cells. HT-29 cells were pretreated with EGCG (100 mM) and exposed to 5-FU (50 mM) or Etoposide (50 mM) for the indicated time periods, and cell viability was determined by MTT assay (A). Under same conditions, cells underwent an additional 30 min incubation in the presence of 10 mM Hoechst 33342 and the fluorescence microscopic images are presented (B).
120
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
3.5. Novel perspective of EGCG in the combination chemotherapy Resistance to anticancer drugs in colon cancer is one of the most difficult obstacles to cancer chemotherapy [14]. We have investigated the efficacy of the combination therapy of EGCG along with anticancer drugs such as 5-FU (5-fluorouracil) or Etoposide. As shown in Fig. 5A, the treatment of HT-29 colon cancer cells with the combined 5-FU or Etoposide and EGCG markedly reduced tumor cell viability compared with 5FU or Etoposide alone. Also cell death induced by the combination of EGCG and chemotherapeutic agents revealed a dintictive apoptotic motif (Fig. 5B). 4. Discussion Cancer cells generally exhibit highly proliferative, migrative and matrix-invasive potentials often by modulating signaling molecules [15,16]. Promising anti-cancer agents with strong inhibitory properties of survival-related proteins or activating capacities of apoptotic proteins are scrutinized for cancer treatments, especially from natural origins, since natural compounds are considered safe as they are derived from commonly consumed foodstuff. In this study, we evaluated whether green tea polyphenol epigallocatechin-3-gallate (EGCG) inhibits HT-29 proliferation through the modulation of important signaling molecules involved in cell survival or apoptosis. We also have assessed a cancer preventive potential of EGCG in chemo-resistant HT-29 colon cancer cells. EGCG is a principle antioxidant derived from foodstuff. The extensive studies suggest that modulation of various signal pathways involved in tumor survival, growth, invasion or promotion is the important target for the development of cancerspecific therapeutic approaches with dietary factors like EGCG [17]. In present study, we demonstrated that a treatment with EGCG to chemoresistant HT-29 colon cancer cells attenuated cell proliferation comparable to a selective COX-2 inhibitor celecoxib. Several reports suggest that COX-2 inhibitors sensitize apoptotic process of chemoresistant cancer cells [18], and the various COX-2 inhibitors such as celecoxib, indomethacin, aspirin and NSAID, are primary candidates for colon cancer chemoprevention intervention trials [19]. The precise cellular and molecular events leading to sensitizing cancer cells by COX-2 inhibitors are not clear at present, nonetheless, many investigators have focused on the efficacy of naturally occurring COX-2
modulators in cancer prevention [20]. We have demonstrated that the efficacy of the combination treatment with EGCG and 5-FU or Etoposide in inhibiting cell growth of chemo-resistant cancer cell lines. HT-29 cells show special characteristics of highly proliferative and resistant to chemotherapeutic agents [21], and increased basal levels of proliferative proteins such as COX-2 [22]. The anti-tumor activity of EGCG was correlated with the inhibition of COX-2 through the activation of AMPK. EGCG treatment abrogated the increased basal level of COX-2 and furthermore, the extensive phosphorylation of AMPK by EGCG was observed. AMPK is known to be involved in the cellular homeostasis and it has been recently demonstrated that AMPK emerges as a pivot point between cell survivals or apoptosis [23]. Also LKB1, a tumor suppressor has been identified as an upstream kinase of AMPK [24], and taken together; the regulation of AMPK by EGCG evidently emerges as an important molecular target in anti-tumor control. Also a synthetic form of AMPK activator, AICAR has shown to be an inhibitor of tumor cell anabolism [25]. Our observation that green tea derived EGCG can activate AMPK and thus abrogate COX-2 and PGE2 production and further increase apoptosis in cancer cell lines suggest that AMPK is an important regulatory component of cancer therapy and COX-2 expression by EGCG. Several natural compounds for cancer treatment have been shown to cause increased cellular ROS generation [26], and one of the major effects is to generated increased intracellular ROS causing loss of mitochondria membrane permeability. Also ROS are one of the upstreams of AMPK activation signals, and implicated in the inhibition of COX-2. We have focused finding the role of AMPK in ROS-induced COX-2 inhibition in colon cancer cells. The results showed that EGCG significantly generated ROS, and this generated ROS could induce AMPK. The ROS scavenger NAC abolished ROS production completely as well as AMPK activation and apoptosis of cells treated EGCG. In conclusion, the present study demonstrates that green tea derived EGCG exhibits a variety of molecular events in HT29 colon cancer cells including inhibition of cell growth, induction of apoptosis and ROS generation, and inactivation of COX-2 expression and AMPK activation. We have identified ROS as one of the upstream regulator of AMPK, and the activation of AMPK as the key element in the regulation of COXexpression. Further investigation warrants elucidating the mechanism by which EGCG generates ROS and
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
COX-2 inhibition through the activation of AMPK to unravel the molecular regulation of tumor cell control. Acknowledgements This work was supported by Hannam University Fund 2005. References [1] F.H. Sarkar, Y. Li, Soy isoflavones and cancer prevention, Cancer Invest. 21 (2003) 744–757. [2] S. Gupta, K. Hastak, F. Afaq, N. Ahmad, H. Mukhtar, Essential role of caspases in epigallocatechin-3-gallate-mediated inhibition of nuclear factor kappa B and induction of apoptosis, Oncogene 23 (14) (2004) 2507–2522. [3] C. Chen, G. Shen, V. Hebbar, R. Hu, E.D. Owuor, A.N. Kong, Epigallocatechin-3-gallate-induced stress signals in HT-29 human colon adenocarcinoma cells, Carcinogenesis 24 (2003) 1369–1378. [4] P. McLoughlin, M. Roengvoraphoj, C. Gissel, J. Hescheler, U. Certa, A. Sachinidis, Transcriptional responses to epigallocatechin-3 gallate in HT 29 colon carcinoma spheroids, Genes to Cells 9 (2004) 661–669. [5] S. Gately, The contributions of cyclooxygenase-2 to tumor angiogenesis, Cancer Metastasis Rev. 19 (2000) 19–27. [6] D.G. Hardie, D. Carling, M. Carlson, The AMP-activated/SNF1 protein kinase subfamily: metabolic sensors of the eukaryotic cell? Annu. Rev. Biochem. 67 (1998) 821–855. [7] B.E. Kemp, D. Stapleton, D.J. Campbell, Z.P. Chen, S. Murthy, M. Walter, B. Van Denderen, I.G. Jennings, T. Iseli, B.J. Michell, L.A. Witters, et al., AMP-activated protein kinase, super metabolic regulator, Biochem. Soc. Trans. 31 (2003) 162–168. [8] M. Garcia-Gil, R. Pesi, S. Perna, S. Allegrini, M. Giannecchini, M. Camici, et al., 5 0 -Aminoimidazole-4-carboxamide riboside induces apoptosis in human neuroblastoma cells, Neuroscience 117 (4) (2003) 811–820. [9] T.W. Davis, N. Hunter, O.C. Trifan, L. Milas, L. Masferrer, COX-2 Inhibitors as radiosensitizing agents for cancer therapy, Am. J. Clin. Oncol. 26 (2003) S58–S61. [10] A.H. Conney, Enzyme induction and dietary chemicals as approaches to cancer chemoprevention: the Seventh DeWitt S. Goodman Lecture, Cancer Res. 63 (2003) 7005–7031. [11] J.T. Hwang, J.H. Ha, O.J. Park, Combination of 5-fluorouracil and genistein induces apoptosis synergistically in chemoresistant cancer cells through the modulation of AMPK and COX-2 signaling pathways, Biochem. Biophy. Res. Commun. 332 (2005) 433–440. [12] M.V. Swamy, C.R. Herzog, M.V. Swamy, C.R. Rao, Inhibition of COX-2 in colon cancer cell lines by celecoxib increases the nuclear localization of active p53, Cancer Res. 63 (17) (2003) 5239–5242.
121
[13] J.T. Hwang, M.Y. Lee, S.N. Jung, H.J. Lee, I. Kang, S.S. Kim, et al., AMP-activated protein kinase activity is required for vanadate-induced hypoxia-inducible factor 1alpha expression in DU145 cells, Carcinogenes 25 (12) (2004) 2497–2507. [14] A. Sobrero, D. Kerr, B. Glimelius, E. Van Cutsem, G. Milano, D.M. Pritchard, et al., New directions in the treatment of colorectal cancer: a look to the future, Eur. J. Cancer 36 (5) (2000) 559–566. [15] C.J. Sherr, Principles of tumor suppression, Cell 116 (2) (2004) 235–246. [16] M.Z. Fang, Y. Wang, N. Ai, Z. Hou, Y. Sun, H. Lu, et al., Tea polyphenol (K)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines, Cancer Res. 63 (2003) 7563–7570. [17] S. Valcic, B.N. Timmermann, D.S. Alberts, G.A. Wachter, M. Krutzsch, J. Wymer, et al., Inhibitory effect of six green tea catechins and caffeine on the growth of four selected human tumor cell lines, Anticancer Drug 7 (1996) 461–468. [18] K. Hasegawa, Y. Ohashi, K. Ishikawa, A. Yasue, R. Kato, Y. Achiwa, et al., Expression of cyclooxygenase-2 in uterine endometrial cancer and anti-tumor effects of a selective COX-2 inhibitor, Int. J. Oncol. 26 (5) (2005) 1419–1428. [19] T.E. Attar, H. Lin, Prostaglandin synthesis by squamous carcinoma cells of head and neck, and its inhibition by nonsteroidal anti-inflammatory drugs, J. Oral Pathol. 16 (1987) 483–487. [20] F. Ye, J. Wu, T. Dunn, J. Yi, X. Tong, D. Zhang, Inhibition of cyclooxygenase-2 activity in head and neck cancer cells by genistein, Cancer Lett. 211 (1) (2004) 39–46. [21] F. Fan, J.S. Wey, M.F. McCarty, A. Belcheva, W. Liu, T.W. Bauer, et al., Expression and function of vascular endothelial growth factor receptor-1 on human colorectal cancer cells, Oncogene 24 (16) (2005) 2647–2653. [22] G. Calviello, F. Di Nicuolo, S. Gragnoli, E. Piccioni, S. Serini, N. Maggiano, et al., nK3 PUFAs reduce VEGF expression in human colon cancer cells modulating the COX-2/PGE2 induced ERK-1 and -2 and HIF-1alpha induction pathway, Carcinogenesis 25 (12) (2004) 2303–2310. [23] D.G. Hardie, New roles for the LKB1/AMPK pathway, Curr. Opin. Cell Biol. 17 (2) (2005) 167–173. [24] R.J. Shaw, M. Kosmatka, N. Bardeesy, R.L. Hurley, L.A. Witters, R.A. DePinho, et al., The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress, Proc. Nat. Acad. Sci. USA 101 (10) (2004) 3329–3335. [25] J.V. Swinnen, A. Beckers, K. Brusselmans, S. Organe, J. Segers, L. Timmermans, et al., Mimicry of a cellular low energy status blocks tumor cell anabolism and suppresses the malignant phenotype, Cancer Res. 65 (6) (2005) 2441–2448. [26] W.H. Chang, J.J. Liu, C.H. Chen, T.S. Huang, F.J. Lu, Growth inhibition and induction of apoptosis in MCF-7 breast cancer cells by fermented soy milk, Nutr. Cancer 43 (2) (2002) 214–226.
Apoptotic effect of EGCG in HT-29 colon cancer cells via AMPK signal pathway Jin-Taek Hwang a, Joohun Ha a, In-Ja Park b, Seong-Kyu Lee c, Haing Woon Baik c, Young Min Kim d, Ock Jin Park b,* a
Department of Biochemistry and Molecular Biology, Medical Research Center for Bioreaction to Reactive Oxygen Species, Kyung Hee University College of Medicine, Seoul 130-701, South Korea b Department of Food and Nutrition, Hannam University, 133 Ojeong-dong Daedeok-gu, Daejeon 306-791, South Korea c Department of Biochemistry, College of Medicine, Eulji University, Daejeon 143-5, South Korea d Department of Biological Sciences, Hannam University, 133 Ojeong-dong Daedeok-gu, Daejeon 306-791, South Korea Received 26 July 2005; received in revised form 6 March 2006; accepted 27 March 2006
Abstract EGCG [(K)epigallocatechin-3-gallate], a green tea-derived polyphenol, has been shown to suppress cancer cell proliferation, and interfere with the several signaling pathways and induce apoptosis. Practically, there is emerging evidence that EGCG has a potential to increase the efficacy of chemotherapy in patients. We hypothesized that EGCG may exert cell cytotoxicity through modulating AMPK (AMP-activated protein kinase) followed by the decrease in COX-2 expression. EGCG treatment to colon cancer cells resulted in a strong activation of AMPK and an inhibition of COX-2 expression. The decreased COX-2 expression as well as prostaglandin E2 secretion by EGCG was completely abolished by inhibiting AMPK by an AMPK inhibitor, Compound C. Also, the activation of AMPK was accompanied with the reduction of VEGF (vascular endothelial growth factor) and glucose transporter, Glut-1 in EGCG-treated cancer cells. These findings support the regulatory role of AMPK in COX-2 expression in EGCG-treated cancer cells. Furthermore, we have found that reactive oxygen species (ROS) is an upstream signal of AMPK, and the combined treatment of EGCG and chemotherapeutic agents, 5-FU or Etoposide, exert a novel therapeutic effect on chemoresistant colon cancer cells. AMPK, a molecule of newly defined cancer target, was shown to control COX-2 in EGCG-treated colon cancer cells. q 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Epigallocatechin-3-gallate; AMP-activated protein kinase; Cyclooxygenase 2; Etoposide; 5-Flurouracil
1. Introduction Cancer chemoprevention by natural compounds has been investigated by many investigators, and among them bioactive flavonoid compounds, widely distributed in many beverages and food products, have been * Corresponding author. Tel.: C82 42 629 7493; fax: C82 42 629 7490. E-mail address: [email protected] (O.J. Park).
implicated in the prevention of human carcinogenesis possibly through their inhibitory activities of cell proliferation or survival [1]. (K)-Epigallocatechin-3gallate (EGCG), a major constituent of green tea, has been shown to exert the inhibitory effects on certain human cancers through the induction of several signal pathways [2–4]. Although the major and precise EGCG cancer preventive activities have not been elucidated yet, the several possibilities have been proposed. Importance of various cancer cell control including
0304-3835/$ - see front matter q 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2006.03.030
116
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
inhibiting tumor promotion enzymes such as cyclooxygenase-2 (COX-2) and lipoxygenase, blocking angiogenesis or induction of apoptosis by EGCG has been discussed [5]. AMP-activated protein kinase (AMPK) is a metabolic-sensing protein kinase, which plays an essential role as an energy-sensor mainly in ATP-deprived conditions [6]. Therefore, AMPK is known to play a major protective role under metabolic stressed conditions. In the activated states, AMPK down-regulates several anabolic enzymes and thus shuts down the ATP-consuming metabolic pathways [7]. Several reports have observed the strong pro-apoptotic potential of AMPK in activated conditions such as AICAR (AMPK activator)-treated cells or constitutively active AMPK mutants [8]. In the present study, we have investigated the role of AMP-activate protein kinase (AMPK) as well as COX-2 in EGCG-induced apoptotic states in human colon cancer cells. EGCG has shown to down-regulate COX-2 derived prostaglandin synthesis and the several compounds that inhibit COX-2 enzymes possess chemo-sensitizing and radio-sensitizing activities in vitro and in vivo [9]. We have observed the effects of EGCG on AMPK activation and COX-2 inhibition, and the interaction of these molecular events leading to apoptosis. Also, the possibility of EGCG rendering the efficacy in the combination chemotherapy with the anti-cancer agents in HT-29 colon cancer cells was explored.
PMSF, 1 mM sodium orthovanadate, 1 mM NaF, 1 mg/ml aprotinin, 1 mg/ml leupeptin, 1 mg/ml pepstatin) and subjected to the western blot analysis. 2.3. Prostaglandin enzyme ELISA assay The supernatants of control and treated cell cultures were collected by centrifuged at 2500!g for 5 min, and the sample was added to appropriate wells and PGE2 conjugate was pipetted into all wells except the blank well. Finally, monoclonal antibody against PGE2 was added to all wells except wells for blank and nonspecific binding test, followed by incubation for 18 h at 4 8C. The plates were washed four times and color reaction was developed by the addition of tetramethylbenzidine substrate. After 30 min of incubation at room temperature, the reaction was quenched by addition of 1 M sulfuric acid. Optical density was measured at 450 nm on an ELX800 reader (Bio-Tec Instruments, Inc., Winooski, Vermont). 2.4. Chromatin staining with Hoechst 33342 Apoptosis was observed by chromatin staining with Hoechst 33342, as previously described. Cells were incubated with each stimuli. After incubation the supernatant was discarded and cells were fixed with 3.5% formaldehyde in PBS for 30 min at room temperature, washed four times with PBS and exposed to Hoechst 33342 at 10 mM for 30 min at room temperature. Cell preparations were examined under ultraviolet illumination with a fluorescence microscope (Olympus Optical Co., Tokyo, Japan).
2. Material and method
2.5. Cell proliferation by MTT assay
2.1. Cell culture and reagents
Cells seeded on 96-well micro plates at 4000 cells/well were incubated with the test compounds for indicated time period. Respectively medium was removed and then incubated with 100 ml of MTT solution (2 mg/ml MTT in PBS) for 4 h. MTT is converted to a blue formazan. Absorbance was determined using an auto reader.
The HT-29 human colon cancer line was purchased from ATCC (Gaithersburg, MD). Cells were cultured in RPMI1640 containing 10% FBS under normoxic conditions (20% O2, 5% CO2, 75% N2). (K)-Epigallocatechin-3-gallate, Etoposide, 5-fluorouracil, MTT, Hoechst 33342, DCFH, NAC and AICAR were obtained from Sigma. Celecoxib was purchased from Pharmacia Korea; Compound C was a gift from Merck Company. Prostaglandin enzyme assay kit was purchased from Amersham Pharmacia Biotech. The anti-phosphospecific antibodies that recognize phosphorylated ACCSer79 were from Cell Signaling Technology. Antibodies for COX-2, PARP, P53, and b-actin were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). 2.2. Protein extract and western blotting Cells were rinsed twice with ice-cold PBS and scraped with lysis buffer (50 mM Tris–HCl, pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM
2.6. RNA isolation and RT-PCR Total RNA was extracted from cells with Trizol Reagent (Life Technologies, Glasgow, United Kingdom) according to manufacturer instructions. We synthesized cDNA from 1 mg. Total RNA using a first strand cDNA synthesis kit (Amersham Pharmacia Biotech, Piscataway, New Jersey). The cDNA fragment was amplified by PCR using the following specific primers: VEGF: sense 5 0 -AGGAGGGCAGAATCATCACG-3 0 , anti-sense 5 0 -CAAGGCCCACAGGGATTTT CT-3 0 ; Glut-1: sense 5 0 -CGGGCCAAGAGTGTGAA-3 0 , antisense 5 0 -TGACGATACCGGAGCCAATG-3 0 ;
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
117
3. Results 3.1. EGCG and celecoxib inhibit proliferation of HT-29 colon cancer cells It has been reported that celecoxib, a selective COX2 inhibitor, inhibits cancer cell proliferation and EGCG can inhibit tumor cell proliferative or survival related proteins [10]. We investigated cancer cell inhibitory capacity of EGCG in comparison with celecoxib. As shown Fig. 1A, EGCG or celecoxib induced growth inhibition of HT-29 cells in a dose-dependent manner, and also EGCG 10 mM or celecoxib 50 mM have shown to inhibit the tumor cells growth for the indicated time period. EGCG and celecoxib have similarly effects on cancer cell death implying the possible similarity in blocking cell growth (Fig. 1B). 3.2. Effects of EGCG on the expression of VEGF and Glut-1 genes
Fig. 1. The effects of cell death on EGCG or celecoxib treated HT-29 colon cancer cells. Cells were treated with different concentrations of EGCG or celecoxib for 48 h (A) or cells were exposed to EGCG (100 mM) or celecoxib (50 mM) for the indicated time periods (B), and then cell viability was measured by MTT assay or Hoechst 33342 dye. The data are presented as the meanGSD.
Most of malignant cancer involves the alteration of survival genes such as glut-1 or VEGF, which play important roles in the chemo-resistance process. Our previous study has shown that the ability of genistein to induce apoptosis or decrease chemo-resistance in cancer cells was related to the expression of survival genes such as glut1 [11]. Therefore, in order to evaluate the effect of EGCG on survival gene expression, VEGF and Glut-1 gene expressions were determined using a semi-quantitative RT-PCR method. As shown Fig. 2, EGCG distinctively reduced mRNA levels of both Glut-1 and VEGF genes.
b-actin: sense 5 0 -GTGGGGGCGCCCAGGCACCA-3 0 , anti-sense 5 0 -CTCCTTAATGTCACGCACCA TTTC-3 0 ; PCR was initiated in a thermal cycle programmed at 95 8C for 5 min, 94 8C for 1 min, 58 8C for 1 min, 72 8C for 1 min, and amplified for 25 cycles. The amplified products were visualized on 1% agarose gels. 2.7. ROS measurement Cells were seeded on 12 well micro plate on cover glasses, after stimulated for indicated time period, respectively, cells were incubated with 10 mM of 2 0 ,7 0 -dichlorofluorescein diacetate (Sigma) for 30 min, and washed with PBS and fluorescence was measured with a fluorescence microscope.
Fig. 2. The effect of EGCG on Glut1 and VEGF mRNA levels. HT-29 cells exposed to different concentrations of EGCG for 12 h, and the mRNA levels of glut1, VEGF and b-actin genes were examined by RT-PCR using specific primers.
118
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
3.3. EGCG abrogates the basal COX-2 expression through AMPK phosphorylation while promoting the apoptotic protein expressions Several studies have indicated that overexpression of COX-2 is strongly related to a number of solid malignancies including colon cancers [12], and the ability of EGCG to modulate COX-2 expression along with apoptotic protein p53 expression and PARP cleavage can provide the involvement of EGCG in the control of COX-2 and apoptosis. EGCG abrogated
COX-2 expression in a dose-dependent manner, and the stimulation of p53 expression and PARP cleavage was also observed (Fig. 3A). We have investigated the connection between AMP-activated protein kinase (AMPK) and COX-2 expression by EGCG treatment. EGCG activated AMPK in a dose-dependent manner as shown by the increased phosphorylation of ACC (acetyl-CoA carboxylase) and the elevation of AMPK expression (Fig. 3A). The phosphorylation level of ACC serine 79, the best-characterized phosphorylation site of AMPK, designates AMPK activation. The level
Fig. 3. The effects of AMPK activator or inhibitor on EGCG INDUCED COX-2, phospho-ACC, PARP cleavage, p53 level or prostaglandINE2 productions. HT-29 cells were treated with EGCG for the indicated concentrations during 6 h (A), also cells were pretreated with compound C (10 mM) for 30 min, and exposed to EGCG (100 mM) for 6 h, and then COX-2, PARP cleavage, p53, phospho-ACC and AMPK level were determined by western blot analysis (B). Also independently HT-29 cell was pretreated with compound C (10 mM) for 30 min, then exposed to EGCG (100 mM) or celecoxib (50 mM) for 12 h, and amount of PGE2 present in the supernatant of the culture medium was measured by ELISA (C). Also cells were exposed to AICAR (1 mM) for 6 h, respectively, performed western blotting, PGE2 production tested, also apoptosis was examined with Hoechst 33342 dye (D).
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
119
reactive oxygen species, the cells were incubated EGCG in the presence of the antioxidant NAC, a broad ROS scavenger. NAC treatment almost completely scavenged the EGCG-generated ROS as well as AMPK activation (also Fig. 4). These results suggest that EGCG has a capacity to generate ROS, which is the upstream signal of AMPK acti-vation.
Fig. 4. ROS generated by EGCG in HT-29 colon cancer cells. Cells were exposed to EGCG for 6 h, in the presence or absence of NAC (5 mM). After additional 30 min incubation in the presence of 10 mM DCFH-DA, and changes in fluorescence intensity were measured by fluorescence-activated cell scanning analysis. Also under the identical conditions, the phosphorylation level of ACC-Ser79 (P-ACC), were examined.
of COX-2 inhibition induced by EGCG was significantly abolished by an AMPK inhibitor Compound C. Also, the marked decrease in PARP cleavage was observed with the treatment of Compound C (Fig. 3B). EGCG treated cell abrogates basal PGE2 production, which was significantly increased by the treatment with EGCG and Compound C (Fig. 3C). These results indicate that AMPK may play a significant role in the regulation of COX-2, and PARP cleavage in EGCGtreated cancer cells. We further tested the role of AMPK in COX-2 regulation with AICAR (5-aminoimidazole-4-carboxamide-1-b-D-ribofuranoside), a cell permeable AMPK activator (Fig. 3D). The results have suggested that AMPK is required for the inhibition of COX-2 and thus reduction of PGE2 in cancerous cells treated with EGCG. 3.4. ROS is an upstream of AMPK activation treated with EGCG In previous report, it was found that triggering ROS could lead to AMPK activation [13], and therefore, we investigated whether EGCG treatment resulted in ROS increase. Measurement of H2O2 production using DCFH-DA in EGCG treated cells has shown that ROS are evidently elevated by EGCG (Fig. 4). To determine if the observed AMPK activation involved
Fig. 5. Synergistic apoptotic effects of combined with EGCG and various chemotherapy drugs in HT-29 colon cancer cells. HT-29 cells were pretreated with EGCG (100 mM) and exposed to 5-FU (50 mM) or Etoposide (50 mM) for the indicated time periods, and cell viability was determined by MTT assay (A). Under same conditions, cells underwent an additional 30 min incubation in the presence of 10 mM Hoechst 33342 and the fluorescence microscopic images are presented (B).
120
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
3.5. Novel perspective of EGCG in the combination chemotherapy Resistance to anticancer drugs in colon cancer is one of the most difficult obstacles to cancer chemotherapy [14]. We have investigated the efficacy of the combination therapy of EGCG along with anticancer drugs such as 5-FU (5-fluorouracil) or Etoposide. As shown in Fig. 5A, the treatment of HT-29 colon cancer cells with the combined 5-FU or Etoposide and EGCG markedly reduced tumor cell viability compared with 5FU or Etoposide alone. Also cell death induced by the combination of EGCG and chemotherapeutic agents revealed a dintictive apoptotic motif (Fig. 5B). 4. Discussion Cancer cells generally exhibit highly proliferative, migrative and matrix-invasive potentials often by modulating signaling molecules [15,16]. Promising anti-cancer agents with strong inhibitory properties of survival-related proteins or activating capacities of apoptotic proteins are scrutinized for cancer treatments, especially from natural origins, since natural compounds are considered safe as they are derived from commonly consumed foodstuff. In this study, we evaluated whether green tea polyphenol epigallocatechin-3-gallate (EGCG) inhibits HT-29 proliferation through the modulation of important signaling molecules involved in cell survival or apoptosis. We also have assessed a cancer preventive potential of EGCG in chemo-resistant HT-29 colon cancer cells. EGCG is a principle antioxidant derived from foodstuff. The extensive studies suggest that modulation of various signal pathways involved in tumor survival, growth, invasion or promotion is the important target for the development of cancerspecific therapeutic approaches with dietary factors like EGCG [17]. In present study, we demonstrated that a treatment with EGCG to chemoresistant HT-29 colon cancer cells attenuated cell proliferation comparable to a selective COX-2 inhibitor celecoxib. Several reports suggest that COX-2 inhibitors sensitize apoptotic process of chemoresistant cancer cells [18], and the various COX-2 inhibitors such as celecoxib, indomethacin, aspirin and NSAID, are primary candidates for colon cancer chemoprevention intervention trials [19]. The precise cellular and molecular events leading to sensitizing cancer cells by COX-2 inhibitors are not clear at present, nonetheless, many investigators have focused on the efficacy of naturally occurring COX-2
modulators in cancer prevention [20]. We have demonstrated that the efficacy of the combination treatment with EGCG and 5-FU or Etoposide in inhibiting cell growth of chemo-resistant cancer cell lines. HT-29 cells show special characteristics of highly proliferative and resistant to chemotherapeutic agents [21], and increased basal levels of proliferative proteins such as COX-2 [22]. The anti-tumor activity of EGCG was correlated with the inhibition of COX-2 through the activation of AMPK. EGCG treatment abrogated the increased basal level of COX-2 and furthermore, the extensive phosphorylation of AMPK by EGCG was observed. AMPK is known to be involved in the cellular homeostasis and it has been recently demonstrated that AMPK emerges as a pivot point between cell survivals or apoptosis [23]. Also LKB1, a tumor suppressor has been identified as an upstream kinase of AMPK [24], and taken together; the regulation of AMPK by EGCG evidently emerges as an important molecular target in anti-tumor control. Also a synthetic form of AMPK activator, AICAR has shown to be an inhibitor of tumor cell anabolism [25]. Our observation that green tea derived EGCG can activate AMPK and thus abrogate COX-2 and PGE2 production and further increase apoptosis in cancer cell lines suggest that AMPK is an important regulatory component of cancer therapy and COX-2 expression by EGCG. Several natural compounds for cancer treatment have been shown to cause increased cellular ROS generation [26], and one of the major effects is to generated increased intracellular ROS causing loss of mitochondria membrane permeability. Also ROS are one of the upstreams of AMPK activation signals, and implicated in the inhibition of COX-2. We have focused finding the role of AMPK in ROS-induced COX-2 inhibition in colon cancer cells. The results showed that EGCG significantly generated ROS, and this generated ROS could induce AMPK. The ROS scavenger NAC abolished ROS production completely as well as AMPK activation and apoptosis of cells treated EGCG. In conclusion, the present study demonstrates that green tea derived EGCG exhibits a variety of molecular events in HT29 colon cancer cells including inhibition of cell growth, induction of apoptosis and ROS generation, and inactivation of COX-2 expression and AMPK activation. We have identified ROS as one of the upstream regulator of AMPK, and the activation of AMPK as the key element in the regulation of COXexpression. Further investigation warrants elucidating the mechanism by which EGCG generates ROS and
J.-T. Hwang et al. / Cancer Letters 247 (2007) 115–121
COX-2 inhibition through the activation of AMPK to unravel the molecular regulation of tumor cell control. Acknowledgements This work was supported by Hannam University Fund 2005. References [1] F.H. Sarkar, Y. Li, Soy isoflavones and cancer prevention, Cancer Invest. 21 (2003) 744–757. [2] S. Gupta, K. Hastak, F. Afaq, N. Ahmad, H. Mukhtar, Essential role of caspases in epigallocatechin-3-gallate-mediated inhibition of nuclear factor kappa B and induction of apoptosis, Oncogene 23 (14) (2004) 2507–2522. [3] C. Chen, G. Shen, V. Hebbar, R. Hu, E.D. Owuor, A.N. Kong, Epigallocatechin-3-gallate-induced stress signals in HT-29 human colon adenocarcinoma cells, Carcinogenesis 24 (2003) 1369–1378. [4] P. McLoughlin, M. Roengvoraphoj, C. Gissel, J. Hescheler, U. Certa, A. Sachinidis, Transcriptional responses to epigallocatechin-3 gallate in HT 29 colon carcinoma spheroids, Genes to Cells 9 (2004) 661–669. [5] S. Gately, The contributions of cyclooxygenase-2 to tumor angiogenesis, Cancer Metastasis Rev. 19 (2000) 19–27. [6] D.G. Hardie, D. Carling, M. Carlson, The AMP-activated/SNF1 protein kinase subfamily: metabolic sensors of the eukaryotic cell? Annu. Rev. Biochem. 67 (1998) 821–855. [7] B.E. Kemp, D. Stapleton, D.J. Campbell, Z.P. Chen, S. Murthy, M. Walter, B. Van Denderen, I.G. Jennings, T. Iseli, B.J. Michell, L.A. Witters, et al., AMP-activated protein kinase, super metabolic regulator, Biochem. Soc. Trans. 31 (2003) 162–168. [8] M. Garcia-Gil, R. Pesi, S. Perna, S. Allegrini, M. Giannecchini, M. Camici, et al., 5 0 -Aminoimidazole-4-carboxamide riboside induces apoptosis in human neuroblastoma cells, Neuroscience 117 (4) (2003) 811–820. [9] T.W. Davis, N. Hunter, O.C. Trifan, L. Milas, L. Masferrer, COX-2 Inhibitors as radiosensitizing agents for cancer therapy, Am. J. Clin. Oncol. 26 (2003) S58–S61. [10] A.H. Conney, Enzyme induction and dietary chemicals as approaches to cancer chemoprevention: the Seventh DeWitt S. Goodman Lecture, Cancer Res. 63 (2003) 7005–7031. [11] J.T. Hwang, J.H. Ha, O.J. Park, Combination of 5-fluorouracil and genistein induces apoptosis synergistically in chemoresistant cancer cells through the modulation of AMPK and COX-2 signaling pathways, Biochem. Biophy. Res. Commun. 332 (2005) 433–440. [12] M.V. Swamy, C.R. Herzog, M.V. Swamy, C.R. Rao, Inhibition of COX-2 in colon cancer cell lines by celecoxib increases the nuclear localization of active p53, Cancer Res. 63 (17) (2003) 5239–5242.
121
[13] J.T. Hwang, M.Y. Lee, S.N. Jung, H.J. Lee, I. Kang, S.S. Kim, et al., AMP-activated protein kinase activity is required for vanadate-induced hypoxia-inducible factor 1alpha expression in DU145 cells, Carcinogenes 25 (12) (2004) 2497–2507. [14] A. Sobrero, D. Kerr, B. Glimelius, E. Van Cutsem, G. Milano, D.M. Pritchard, et al., New directions in the treatment of colorectal cancer: a look to the future, Eur. J. Cancer 36 (5) (2000) 559–566. [15] C.J. Sherr, Principles of tumor suppression, Cell 116 (2) (2004) 235–246. [16] M.Z. Fang, Y. Wang, N. Ai, Z. Hou, Y. Sun, H. Lu, et al., Tea polyphenol (K)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines, Cancer Res. 63 (2003) 7563–7570. [17] S. Valcic, B.N. Timmermann, D.S. Alberts, G.A. Wachter, M. Krutzsch, J. Wymer, et al., Inhibitory effect of six green tea catechins and caffeine on the growth of four selected human tumor cell lines, Anticancer Drug 7 (1996) 461–468. [18] K. Hasegawa, Y. Ohashi, K. Ishikawa, A. Yasue, R. Kato, Y. Achiwa, et al., Expression of cyclooxygenase-2 in uterine endometrial cancer and anti-tumor effects of a selective COX-2 inhibitor, Int. J. Oncol. 26 (5) (2005) 1419–1428. [19] T.E. Attar, H. Lin, Prostaglandin synthesis by squamous carcinoma cells of head and neck, and its inhibition by nonsteroidal anti-inflammatory drugs, J. Oral Pathol. 16 (1987) 483–487. [20] F. Ye, J. Wu, T. Dunn, J. Yi, X. Tong, D. Zhang, Inhibition of cyclooxygenase-2 activity in head and neck cancer cells by genistein, Cancer Lett. 211 (1) (2004) 39–46. [21] F. Fan, J.S. Wey, M.F. McCarty, A. Belcheva, W. Liu, T.W. Bauer, et al., Expression and function of vascular endothelial growth factor receptor-1 on human colorectal cancer cells, Oncogene 24 (16) (2005) 2647–2653. [22] G. Calviello, F. Di Nicuolo, S. Gragnoli, E. Piccioni, S. Serini, N. Maggiano, et al., nK3 PUFAs reduce VEGF expression in human colon cancer cells modulating the COX-2/PGE2 induced ERK-1 and -2 and HIF-1alpha induction pathway, Carcinogenesis 25 (12) (2004) 2303–2310. [23] D.G. Hardie, New roles for the LKB1/AMPK pathway, Curr. Opin. Cell Biol. 17 (2) (2005) 167–173. [24] R.J. Shaw, M. Kosmatka, N. Bardeesy, R.L. Hurley, L.A. Witters, R.A. DePinho, et al., The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress, Proc. Nat. Acad. Sci. USA 101 (10) (2004) 3329–3335. [25] J.V. Swinnen, A. Beckers, K. Brusselmans, S. Organe, J. Segers, L. Timmermans, et al., Mimicry of a cellular low energy status blocks tumor cell anabolism and suppresses the malignant phenotype, Cancer Res. 65 (6) (2005) 2441–2448. [26] W.H. Chang, J.J. Liu, C.H. Chen, T.S. Huang, F.J. Lu, Growth inhibition and induction of apoptosis in MCF-7 breast cancer cells by fermented soy milk, Nutr. Cancer 43 (2) (2002) 214–226.