Flavonoid GL-V9 induces apoptosis and inhibits glycolysis of breast cancer via disrupting GSK-3β-modulated mitochondrial binding of HKII

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original article

Flavonoid GL-V9 induces apoptosis and inhibits glycolysis of breast cancer via disrupting GSK-3β-modulated mitochondrial binding of HKII Yongjian Guoa,1, Libin Weib,1, Yuxin Zhoub, Na Lub, Xiaoqing Tanga, Zhiyu Lic, Xiaotang Wanga,∗ a

Department of Chemistry and Biochemistry, Florida International University, Miami, FL, USA State Key Laboratory of Natural Medicines, Jiangsu Key Laboratory of Carcinogenesis and Intervention, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing, 210009, People’s Republic of China c State Key Laboratory of Natural Medicines, Department of Natural Medicinal Chemistry, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing, 210009, People’s Republic of China b

A R T I C LE I N FO

A B S T R A C T

Keywords: GL-V9 HKII GSK-3β Mitochondria Apoptosis

Energy metabolism plays important roles in the growth and survival of cancer cells. Here, we find a newly synthesized flavonoid named GL-V9, which inhibits glycolysis and induces apoptosis of human breast cancer cell lines, and investigate the underlying mechanism. Results show that hexokinase II (HKII) plays important roles in the anticancer effects of GL-V9. GL-V9 not only downregulates the expression of HKII in MDA-MB-231 and MCF7 cells, but also induces dissociation of HKII from voltage-dependent anion channel (VDAC) in mitochondria, resulting in glycolytic inhibition and mitochondrial-mediated apoptosis. The dissociation of mitochondrial HKII is attributed to GSK-3β-induced phosphorylation of mitochondrial VDAC. Our in vivo experiments also show that GL-V9 significantly inhibits the growth of human breast cancer due to activation of GSK-3β and inactivation of AKT. Thus, GL-V9 induces cytotoxicity in breast cancer cells via disrupting the mitochondrial binding of HKII. Our works demonstrate the significance of metabolic regulators in cancer growth and offer a fresh insight into the molecular basis for the development of GL-V9 as a candidate for breast carcinoma treatment.

1. Introduction Breast cancer is the most common cancer in women worldwide (second most common cancer overall). In 2017, the number of new cases and deaths from breast cancer reached 252,710 and 40,610, respectively [1]. Depending on the type and degree of advancement, breast cancer is treated with surgery and then possibly chemotherapy and/or radiation therapy. Compared with the well-defined behavior of breast cancer, the molecular mechanisms of breast cancer initiation and progression remain poorly understood. The unique energy metabolism in cancer cells was first proposed by Warburg almost a century ago [2]. It is suggested that a shift in energy production from mitochondrial oxidative phosphorylation (OXPHOS) to aerobic glycolysis is a hallmark of cancer development. Aerobic glycolysis not only supports rapid growth, but also makes the cancer cells less dependent on oxygen availability and generates a suitable micro-environment for cancer progress [3]. Inhibition of glycolysis is a novel therapeutic focus in cancer managements. HK II mediates the first rate-controlling step of glycolysis, phosphorylating glucose to glucose-

6-phosphate. The high glycolytic rate is closely connected with the greatly increased expression of hexokinase II (HK II), and its mitochondrial binding with voltage-dependent anionic channel (VDAC) in malignant cells [4]. VDAC is a porin located in the outer mitochondrial membrane (OMM) and responsible for the exchanges of metabolites and ions between mitochondria and other cellular compartments [5]. It has been demonstrated that binding of HK II to VDAC inhibits apoptosis as a result of alterations in the permeability of OMM [6]. And the increased binding of HK II to VDAC is associated with drug resistance of cancer cells to chemotherapeutic agents [7]. Previous studies have suggested that activation of protein kinase B (PKB or AKT) enhances the binding of HK II to mitochondria by augmenting the uptake and metabolism of glucose, as well as affecting the phosphorylation of VDAC and/or HK II [8–10]. The ability of AKT to directly impact the binding of HK II to mitochondria may partially account for its antiapoptotic properties. Furthermore, AKT inhibits glycogen synthase kinase 3β (GSK-3β) by phosphorylating its Ser9. GSK3β can also localize to the mitochondria and is associated with mitochondrial dysfunction and apoptosis [11,12]. It has been reported

∗

Corresponding author. E-mail address: wangx@fiu.edu (X. Wang). 1 These authors contributed equally. https://doi.org/10.1016/j.freeradbiomed.2019.10.413 Received 30 July 2019; Received in revised form 23 October 2019; Accepted 23 October 2019 0891-5849/ © 2019 Published by Elsevier Inc.

Please cite this article as: Yongjian Guo, et al., Free Radical Biology and Medicine, https://doi.org/10.1016/j.freeradbiomed.2019.10.413

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

2.4. Annexin-V/PI double-staining assay

that GSK3β catalyzes the phosphorylation of VDAC, resulting in the detachment of HKII from VDAC and enhanced vulnerability of the mitochondria to pro-apoptotic conditions in cancer cells [13]. GL-V9 (5-hydroxy-8-methoxy-7-(4-(pyrrolidin-1-yl) butoxy)-4 Hchromen-4-one) is a newly synthesized flavonoid. Previous studies have demonstrated that GL-V9 inhibits tumor invasion and metastasis via downregulating the expression and activity of matrix metalloproteinase-2/9 in MDA-MB-231 and MCF7 cell lines in vitro [14]. It has also been observed that GL-V9 triggers mitochondria mediated apoptosis and reverses hypoxia–drug resistance in human hepatocellular carcinoma [15,16]. Here, we further investigate the anticancer activity and underlying mechanism of GL-V9 against breast cancer. Our results demonstrate that GL-V9 inhibits glycolysis and induces apoptosis of human breast cancer cells through GSK3β-regulated mitochondrial binding of HKII.

Cells were treated for 36 h with GL-V9 or w0gonin, then harvested and resuspended with PBS. Apoptotic cells were identified by double supravital staining with recombinant FITC (fluorescein isothiocyanate) -conjugated Annexin-V and PI, according to the manufacturer’s instructions of the Annexin V-FITC Apoptosis Detection kit (KeyGen, Nanjing, China). Apoptotic cell death was examined by FACSCalibur flow cytometry (Becton Dickinson, San Jose, CA). 2.5. Lactic acid production The generation of lactic acid in cell culture media was assayed using the Lactic Acid Detection kit (KeyGen, Nanjing, China) following the manufacturer’s instructions. The assay was monitored spectrophotometrically at 570 nm using the Thermo Scientific Varioskan Flash spectral scanning multimode reader (Thermo, Waltham, MA). The absorbance was normalized as follows: ODnormalized =ODmeasured/living cell numbertreated × living cell numbercontrol. The living cells were counted by trypan blue staining of collected cells.

2. Methods & materials 2.1. Reagents GL-V9 (5-hydroxy-8-methoxy-7-(4-(pyrrolidin-1-yl) butoxy)-4 Hchromen-4-one, C24H27NO5, MW 409.47, purity > 99%) is a new flavonoid derivative from wogonin (5,7-Dihydroxy-8-methoxy-2-phenyl4H-chromen-4-one, C16H12O5, MW 284.27, purity > 99%). GL-V9 was synthesize by Prof. Zhiyu Li in our lab [17]. Both GL-V9 and wogonin were dissolved in dimethyl sulfoxide (Sigma–Aldrich, St. Louis, MO, USA) to 100 mM as a stock solution, stored at -20 °C, and diluted with medium before each experiments. The final DMSO concentration did not exceed 0.1% throughout the study (without influences in cell growth and intracellular ROS level). 2-Deoxy-D-glucose (2-DG) was purchased from MCE (MedChemExpress, NJ, USA) and dissolved in distilled water to make a 1.0 M stock solution. LiCl was purchased from Sigma-Aldrich (Merck Life Science (Shanghai) Co., Ltd. Shanghai, China) and dissolved in distilled water to make a 100 mM stock solution. Human recombinant IGF-1 was lyophilized powder and purchased from Merck Millipore (Millipore, Ireland). The lyophilized recombinant human IGF-1 was reconstituted to 100 μg/mL using ddH2O as a stock solution and freezed at -80 °C. MK-2206 dihydrochloride (MK) was purchased from Santa Cruz (Santa Cruz Biotechnology Inc, CA) and dissolved in DMSO to make a 10 mM stock solution.

2.6. ATP Assessment Cellular ATP concentrations were measured with the ATP Bioluminescent Somatic Cell Assay Kit from Sigma. Cells were lysed in an ice-cold ATP releasing buffer. Using an ATP standard provided by Sigma, ATP concentrations were then determined with the kit, and normalized to protein concentrations. Aliquots of 100 μl were transferred to white 96-well assay plates. Following the addition of 100 μl luciferin and luciferase, luminescence was monitored on a luminometer Orion II (Berthold DS, Bleichstr, Pforzheim, Germany). 2.7. Glucose uptake assay Glucose was assayed using the Amplex Red Glucose Assay Kit (Invitrogen, Eugene, OR). After treatment, culture media was collected and diluted 1:4000 with buffer. The amount of glucose was detected using a fluorometer (Thermo, Waltham, MA) at Ex./Em. = 530 nm/ 590 nm according to the instructions of the assay kit. The absorbance was normalized as follows: ODnormalized =ODmeasured/living cell numbertreated × living cell numbercontrol. The living cells were counted by trypan blue staining of collected cells. Glucose uptake was determined by subtracting the amount of glucose in each sample from the total amount of glucose in the media (without cells).

2.2. Cell culture Human breast cancer cell lines MDA-MB-231 and MCF-7 were purchased from Cell Bank of Shanghai Institute of Biochemistry & Cell Biology, Chinese Academy of Sciences (Shanghai, China). All the cell lines were authenticated under STR (short tandem repeat) analysis by the suppliers. Then the cells were expanded and stored in liquid nitrogen upon receipt, and each aliquot was passaged for fewer than 25–30 times in our laboratory. MDA-MB-231 and MCF-7 cells were respectively cultured in PRMI-1640 and Dulbecco's Modified Eagle Medium (GIBCO, Invitrogen Corporation, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum (Sijiqing, Hangzhou, China), 100 U/ml penicillin G, and 100 μg/ml streptomycin at 37 °C with 5% CO2.

2.8. Measurements of oxygen consumption Cells were seeded at a density of 0.2 × 106 cells per well in a 96well BD Oxygen Biosensor System plate (BD Biosciences, San Jose, CA, USA), and culture media was added until a final volume of 200 μl per well was attained. After drug treatments, the plates were maintained at 37 °C in a humidified incubator with 5% CO2. Plates were scanned in a temperature-controlled (37 °C) plate reader (Thermo, Waltham, MA) with an excitation wavelength of 485 nm and an emission wavelength of 630 nm at predetermined times for a total of 72 h. The fluorescence traces in each well were normalized according to the signal in the airsaturated buffer. Slopes of fluorescence signal were calculated in the dynamic range of measurements to compare the respiratory rates of samples. Normalized relative fluorescence unit represents oxygen consumption.

2.3. Cell viability assays Cell viability was measured using 3-[4,5-dimethylthiazol-2-yl] -2,5diphenyltetrazolium bromide (the MTT assay). Cells (1 × 104) were treated with GL-V9 for 36 h at various concentrations (0~80 μM). Absorbance of the resulting formazan was measured spectrophotometrically at 570 nm using a Universal Microplate Reader EL800 (BIO-TEK instruments, Inc., Vermont, MA).

2.9. Mitochondrial membrane potential determination Following treatment, cells were harvested and resuspended with ice-cold PBS (2000 rpm × 5 min). Quantitative changes of mitochondrial membrane potential (MMP) at the early stage of cell apoptosis 2

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

weight ranging from 18 to 22 g were supplied by the Academy of Military Medical Sciences of the Chinese People’s Liberation Army (Certificate No. SYXK-(Su)2016-0010). The animals were kept at 22 ± 2 °C and 55–65% humidity in stainless steel cages under controlled lights (12 h light/day) and were fed with standard laboratory food and water. Animal care was conducted in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals published by the National Institute of Health, USA. This experiment was conducted in accordance with the guidelines issued by the State Food and Drug Administration (SFDA of China). Forty nude mice were inoculated subcutaneously with 1 × 107 MDA-MB-231 or MCF7 cells into the right axilla. After 12 days of growth, tumor sizes were determined using micrometer calipers. Mice with similar tumor volumes (mice with tumors that were too large or too small were eliminated) were randomly divided into the following five groups (6 mice per group): saline control, GL-V9 (20 mg/kg, i.v., every 2 days), wogonin (60 mg/kg, i.v., every 2 days), and Taxol (8 mg/ kg, i.v., twice a week). GL-V9 was prepared into lyophilized powder in the lab and dissolved in saline. and taxol injection were also diluted by saline. Tumor sizes were measured every 3 days using micrometer calipers and tumor volume was calculated using the following formula: TV (mm3) =d2 × D/2, where d and D were the shortest and the longest diameter, respectively. Mice were sacrificed on day 21, and tumor tissues were used for immunofluorescence and immunohistochemistry assays.

were measured by the Mitochondrial membrane potential Detection kit (KeyGen, Nanjing, China) with the lipophilic cation 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolcarbocyanine iodide (JC-1) and detected by FACSCalibur flow cytometry (Becton Dickinson, San Jose, CA). 2.10. Analysis of intracellular superoxide anions (O2•−) level Intracellular superoxide anions (O2•−) level was determined by DHE (S0063; Beyotime Institute of Biotechnology, Nantong, China), according to the manufacturer protocols. Cells were treated for 36 h, harvested and incubated with the dye, and were then detected on flow cytometer (Becton Dickinson Biosciences, Franklin Lakes, NJ). 2.11. Hydrogen peroxide (H2O2) assay The intracellular content of H2O2 was analyzed with a hydrogen peroxide assay kit (S0038; Beyotime Institute of Biotechnology, Nantong, China) according to the manufacturer protocol. In brief, cells were harvested, lysised and centrifuged at 12,000g for 5 min. Then, test tubes containing 50 μl of supernatants and 100 μl of test solution were placed at room temperature for 30 min, and were measured immediately at a wave length of 560 nm with a SynergyTM HT multimode reader (Bio-Tek, Winoosky, VT). Absorbance values were calibrated to a standard concentration curve to calculate the concentration of H2O2.

2.15. Immunohistochemistry 2.12. Preparation of mitochondrial extracts The expressions of GSK3β, p-GSK3β, AKT, p-AKT in the tumor tissue were assessed by the SP immunohistochemical method using a rabbitantihuman monoclonal antibody and an Ultra-Sensitive SP kit (kit 9710 MAIXIN, Maixin-Bio Co., Fujian, China). Tissue sections (4-mm thick) were placed onto treated slides (Vectabond, Vector Laboratories, Burlingame, CA, USA). Sections were heat-fixed, deparaffinized, and rehydrated through a graded alcohol series (100, 95, 85, 75%) to distilled water. Tissue sections were boiled in citrate buffer at high temperature for antigen retrieval, and treated with 3% hydrogen peroxide to block endogenous peroxidase activity. The slides were incubated with a protein-blocking agent (kit 9710 MAIXIN, Maixin-Bio Co., Fuzhou, Fujian) prior to the application of the primary antibody, and then incubated with the primary antibody at 4 °C overnight. The tissues were then incubated with the secondary biotinylated antispecies antibody and labeled using a modified staining procedure based on avidin–biotin complex immunoperoxidase following the instructions provided by the manufacturer of the Ultra-Sensitive SP kit.

The fractionation of the mitochondrial protein was extracted with the Mitochondria/Cytosol Fractionation Kit (BioViso, VA, USA). Briefly, 5 × 107 cells were collected by centrifugation at 600 × g for 5 min at 4 °C and washed with ice-cold PBS. Cells were resuspended with 1 ml of 1 × Cytosol Extraction Buffer Mix containing dithiothreitol and protease inhibitors and incubated on ice for 10 min. Then, cells were homogenized in an ice-cold grinder and the homogenate was transferred to a 1.5 ml microcentrifuge tube and centrifuged at 700 × g for 10 min at 4 °C. The supernatant was transferred to a fresh 1.5 ml microcentrifuge tube and centrifuged at 10 000 × g for 30 min at 4 °C. The supernatant was collected as cytosolic fraction and the pellet was resuspended in 0.1 ml Mitochondrial Extraction Buffer Mix containing DTT and protease inhibitors and then vortexed for 10 s and saved as mitochondrial fraction. 2.13. Immunoprecipitation

2.16. Statistical evaluation

For co-immunoprecipitation of VDAC complexes, VDAC were immunocaptured from mitochondrial extracts. Lysate of mitochondrial fraction (1 ml) containing 1.5 mg total protein was incubated, respectively, with 1 mg VDAC antibody and 20 ml protein A/G-conjugated beads (Santa Cruz Biotechnology, Santa Cruz, CA, USA) overnight. After four washes with TNES buffer, samples were centrifuged at 3000 × g for 2 min and resuspended in 20 ml SDS-sample buffer (0.5 M Tris–HCl, pH6.8, 20% glycerol, 2% SDS, 5% 2-mercaptoethanol, 4% bromophenol blue). For Western blot analysis, 10 μL samples were used. The immunocomplexes were analyzed by western blotting and probed with antibody against GSK3β or HKII. For co-immunoprecipitation of GSK3β complexes, GSK3β were immunocaptured from mitochondrial extracts. The same experimental procedures were conducted as described for VDAC complexes above. The immunocomplexes were analyzed by western blotting and probed with antibody against VDAC or HKII.

Data are presented as mean ± S.D. from triplicate parallel experiments unless otherwise indicated. Statistical analysis was performed using one-way ANOVA. Least Significant Difference test and Tukey’s HSD test were used for the one-way ANOVA analyses. 3. Results 3.1. GL-V9 has potent anticancer activity in breast cancer cells via inducing mitochondrial-mediated apoptosis As shown in Fig. 1A, GL-V9 was a novel synthetic flavonoid derived from the natural product wogonin that inhibited the growth and induced apoptosis of various tumor cells [18–20]. In previous studies, GLV9 inhibited the invasion of human breast carcinoma cells [14]. However, the growth inhibitory effect of GL-V9 in breast cancer cells has not been reported. As shown in Fig. 1B, compared with wogonin, GL-V9 had stronger capacity to inhibit the growth of MDA-MB-231 and MCF7 cells. IC50 values of 36 h GL-V9 treatment for MDA-MB-231 and

2.14. In vivo tumor growth assay Female athymic BALB/c nude mice (35–40 days old) with body 3

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 1. GL-V9 induces mitochondrial-mediated apoptosis in breast carcinoma. (A) Chemical structure of GL-V9 and Wogonin. i) GL-V9 (5-hydroxy-8-methoxy-7(4-(pyrrolidin-1-yl)butoxy)-4H-chromen-4-one,C 24H27NO5, MW409.47). ii) wogonin (5,7-Dihydroxy-8-methoxy-2-phenyl-4H-chromen-4-one, C16H12O5). (B–H) MDA-MB-231 and MCF-7 were treated, respectively, with GL-V9, or wogonin for 36 h. (B) Cell growth inhibition as assayed by MTT. Bars represent SD( ± ). (C) Nucleolus morphologic changes induced by wogonin observed under fluorescent microscope (400 × ). (D) Annexin-V/PI double-staining assay. Histograms of apoptosis rates were quantitated using both early and late apoptotic rates. (E) Western blot assays for the expressions of apoptosis-associated proteins. The changes of protein expressions were quantified. (F) The uptake of JC-1 analyzed by flow cytometry. The percentages of the loss of MMP were represented by histograms. (G) Mitochondrial and cytosolic fractions were isolated. Western blot assays were used to examine the expressions of cyt-c. The changes of protein expressions were quantified. (H) The intracellular level of O2·- was detected by flow cytometry. (I) The intracellular content of H2O2 was analyzed with a hydrogen peroxide assay kit. All the experiments were parallelly repeated in triplicate. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control of MDA-MB-231 cells, #p < 0.05 or ## < 0.01 versus non-treated control of MCF-7 cells.

MCF7 cells were 14.90 ± 1.26 μM and 17.81 ± 2.08 μM, respectively. And for wogonin, the IC50 values were 109.22 ± 4.08 μM in MDA-MB231 cells and 98.53 ± 2.23 μM in MCF-7 cells. To investigate the apoptosis inducing effect of GL-V9, fluorescent microscope was employed. It is observed that untreated MDA-MB-231 and MCF7 cells stained equably with blue fluorescence, demonstrating steady chromatinic distribution in nucleolus. Upon treatment with 10 μM GL-V9, cells showed an early symptom of apoptosis and emitted bright fluorescence due to chromatin agglutination and nucleolus pyknosis. At concentrations of 20 and 30 μM GL-V9, cellular nucleus disintegrated and formed many nuclear fragments in both cells (Fig. 1C). Upon 30 μM wogonin, only MCF-7 cells showed nuclear fragments, instead MDA-MB-231 not. Then apoptosis (including early apoptosis and late apoaptosis) rates were tested by Annexin V/PI double staining assay. As shown in Fig. 1D, GL-V9 induced apoptosis of MDA-MB-231 and MCF7 cells in a concentration dependent manner. Compared with wogonin, GL-V9 was 3 times more efficient in promoting apoptosis in MDA-MB-231 and MCF7 cells. To further investigate the apoptosis inducing mechanisms of GL-V9, we measured the expressions of some key apoptotic proteins in MDA-

MB-231 and MCF7 cells. As shown in Fig. 1E, upon treatment with GLV9, caspase 3 was proteolytically activated, and the expressions of caspase 9, Bcl-2 and Bcl-xl were notably decreased, instead Bax was increased. And GL-V9 induced the loss of mitochondrial membrane potential (MMP) in the two breast cancer cell lines (Fig. 1F). Moreover, the amount of Cyt c decreased in mitochondria but increased in the cytosol following GL-V9 treatment (Fig. 1G). We also assayed the intracellular level of H2O2 and O2·- respectively. As shown in Fig. 1H and 1I, GL-V9 significantly increased the intracellular level of O2·-, instead did not affect H2O2 production. Little or much weaker effects on these mitochondrial apoptosis indexes were observed in the two cancer cell lines when cells were treated with 30 μM wogonin. These results suggested that GL-V9 was more effective than wogonin, inhibited growth and induced mitochondria-mediated apoptosis in human breast cancer cells. 3.2. GL-V9 suppresses aerobic glycolysis of human breast cancer cells Since GL-V9 induced mitochondrial dysfunction in breast cancer cells, it may influence other mitochondrial functions, especially energy 4

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 2. GL-V9 inhibits aerobic glycolysis of breast cancer cells. (A) Cell survival rates of MDA-MB-231 and MCF-7 cells treated with GL-V9 for 0–36 h were assayed by MTT. (B) Annexin-V/PI double-staining assay of MDA-MB-231 and MCF-7 cells treated by GL-V9 for 24 h. (C–F) MDA-MB-231 and MCF-7 cells were treated with GL-V9 for 24h. (C) Production of lactic acid was assayed by a Lactic Acid production Detection kit. (D) Glucose uptake was measured using the Amplex Red assay kit. (E) Quantification of ATP generation was detected by the luminometer Orion II. (F) Oxygen consumption was detected by BD Oxygen Biosensor System plate. All the experiments were parallelly repeated in triplicate. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control of MDA-MB-231 cells, # p < 0.05 or ##p < 0.01 versus non-treated control of MCF-7 cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

DG) was used as a positive control, which was a proven HKII inhibitor via preferentially interfering with glycolytic pathway and modulating mitochondrial HKII in human breast cancer cells [22]. 15 mM 2-DG for 36 h induced obvious apoptosis in MDA-MB-231 and MCF-7 cells (Fig. 3A and 3B). As shown in Fig. 3C and 3D, the levels of total and mitochondrial HKII were both decreased notably upon 20 and 30 μM GL-V9 treatment for 36 h. Little changes on VDAC was observed. Whereas 15 mM 2-DG only decreased mitochondrial HKII. Moreover, the mitochondrial localization of HKII was detected by immunofluorescence assay (Fig. 3E). In untreated cells, intense staining of HKII was detected and co-localized with mitochondrial marker TOMM20. After GL-V9 and 2-DG treatment, faint diffuse staining of HKII was detected in the cytosol, indicative of weaker co-localization of HKII with mitochondria. The subcellular localization of HKII was interdependent, to a large extent, on the interaction strength of HKII/VDAC [23]. Therefore, we detected the binding of HKII and VDAC in mitochondria by immunoprecipitation assay. Result showed that GL-V9 effectively suppressed the binding of mitochondrial HKII with VDAC (Fig. 3F).

metabolism. Energy exhaustion was a significant reason for the inhibition of rapidly growing cancer cells. Cancer cells show an increased dependence on glycolysis to meet their energy needs, regardless of whether they were well-oxygenated or not. Here, we investigated the influence of GL-V9 on the aerobic glycolysis in MDA-MB-231 and MCF7 cells. To exclude the influence of apoptotic death in metabolic evaluation, we performed a kinetic analysis of cell death and cell growth inhibition from 12 h to 36 h. As shown in Fig. 2A, no obvious cell death of MDA-MB-231 and MCF-7 cells were observed within 24 h treatment of GL-V9. After 24h treatment, cell survial ratios were rapidly decreased, especially upon 30 μM GL-V9. We further assayed the apoptosis rates induced by 24 h treatment of GL-V9. The result showed that the apoptosis rates were all less than 10% upon 10 μM, 20μM and 30 μM GL-V9 in both cell lines (Fig. 2B). Thus, we conducted metabolic evaluation upon 24 h GL-V9 treatment, which inhibited cell growth, but did not induce apoptotic cell death. Upon treatment of GL-V9 for 24 h, the lactate generation, glucose uptake and ATP production of MDA-MB-231 and MCF7 cells were all decreased (Fig. 2C, 2D and 2E). The degree of reduction in these glycolysis indexes strongly depended on the concentration of GL-V9. Finally, it is noticed that treatment of GL-V9 significantly promoted the oxygen consumption of human breast cancer cells between 12 h to 36 h (Fig. 2F). All these results indicated that GL-V9 suppressed aerobic glycolysis of breast cancer cells.

3.4. The GSK-3β-mediated phosphorylation of VDAC is involved in GL-V9induced dissociation of mitochondrial HK II The interaction of hexokinase with mitochondria is regulated to facilitate apoptotic effects [5]. We used LiCl (a GSK-3β inhibitor which can induce the autophosphorylation of GSK-3β) to investigate the roles of GSK-3β in the binding of HKII to mitochondria. As shown in Fig. 4A, LiCl inhibited the interaction between HKII and VDAC in both MDAMB-231 and MCF7 cells. Although both total and mitochondrial GSK-3β levels were increased upon GL-V9 treatment, phosphorylated GSK-3β (GSK-3β(s9)) was decreased as well(Fig. 4B). These results implicated that GSK-3β was involved in GL-V9-induced dissociation of HK II from VDAC. Further studies showed that GSK-3β could bind with VDAC in mitochondria instead of HKII (Fig. 4C). GL-V9 promoted the binding of mitochondrial GSK-3β with VDAC, and phosphorylation of GSK-3β

3.3. The dissociation of HK II from mitochondria accounts for GL-V9induced mitochondrial dysfunction and apoptosis The data above demonstrated that GL-V9 could induce mitochondria-mediated apoptosis and inhibit glycolysis. Whether there were some relationships between the apoptosis-inducing effect and glycolysis modulation? It has been reported that HK II plays a significant role in both glycolysis and mitochondrial homeostasis [21]. Therefore, we investigated the influence of GL-V9 on mitochondrial HKII in human breast cancer cell lines MDA-MB-231 and MCF7. 2-deoxy-D-glucose (25

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 3. GL-V9 suppresses HK II binding to VDAC in mitochondria. MDA-MB-231 and MCF-7 were treated with GL-V9 or 2-DG(15 mM) for 36 h, respectively. (A) Nucleolus morphologic changes induced by 2-DG. (B) Annexin-V/PI double-staining assay of MDA-MB-231 and MCF-7 cells treated with 2-DG. (C) The protein expressions of total HKII and VDAC. (D)The levels of HK II in mitochondrial and cytosolic fractions. (E) Immunofluorescence using antibodies specific to HKII and mitochondrial marker TOMM20. (F) HK II was immunoprecipitated using VDAC antibody. Western blot assays were performed for HK II and VDAC. All the experiments were parallelly repeated in triplicate. The changes of protein levels were quantified. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control of MDA-MB-231 cells, #p < 0.05 or ##p < 0.01 versus non-treated control of MCF-7 cells.

GL-V9 played important roles in the activation of GSK-3β and the dissociation of HKII from mitochondria.

induced by LiCl reversed the effects of GL-V9 (Fig. 4D). All these results suggested that GL-V9 promoted the binding of GSK-3β with VDAC in mitochondria, facilitating the phosphorylation of VDAC and consequently resulting in the dissociation of HK II from VDCA.

3.6. GL-V9 inhibits the growth of breast tumor in vivo via activating GSK-3β and inactivating AKT

3.5. GL-V9 activates GSK-3β in mitochondria via inhibiting AKT To verify the results from our in vitro studies, we investigated the anticancer effects of GL-V9 on the growth of tumor xenograft models with human MDA-MB-231 or MCF-7 cell lines. As shown in Fig. 6A–6C, 20mg/kg GL-V9 significantly inhibited the growth of tumor in nude mice. Both 20mg/kg GL-V9 and 60mg/kg wogonin were as effective as positive drug Taxol (8mg/kg) for tumor inoculated with MDA-MB231 cells. In mice inoculated with MCF-7 cells, GL-V9 had similar inhibitory effect with Taxol, but more effective than wogonin. It is worth mentioning that GL-V9 treatments had little influence on body weight of the animals (Fig. 6D). Hematoxylin and eosin (HE) staining showed that GL-V9-treated group displayed stronger infiltration and fewer vessels than the control group (Fig. 7A, Table1&2). Immunohistochemistry (IHC) assay showed that GL-V9 inhibited the phosphorylation of GSK-3β and the activation of AKT in breast tumor tissue (Fig. 7B), corroborating with our in vitro

It has been reported that AKT inhibits the activity of GSK-3β in human neuroblastoma cells by changing its phosphorylation state (phosphorylating Ser9) [24]. Therefore, we investigated the influence of AKT on GSK-3β in human breast cancer cells. As shown in Fig. 5A, in human breast cancer cells, insulin-like growth factor 1 (IGF-1, an AKT activator) promoted the phosphorylation of GSK-3β,whereas MK-2206 (MK, an AKT inhibitor) inhibited the phosphorylation of GSK-3β, in good agreement with literature report [25]. And IGF-1 blocked the binding of GSK-3β with VDAC in mitochondria (Fig. 5B). GL-V9 not only decreased the protein expression of AKT, but also suppressed its activation, no matter whether GL-V9 induced apoptosis (36 h treatment) or not (24 h treatment) (Fig. 5C). In the presence of IGF, GL-V9enhanced binding of GSK-3β to VDAC in mitochondria was disturbed (Fig. 5D). These results indicated that inactivation of AKT induced by 6

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 4. GSK-3β binds with and phosphorylates VDAC, which is responsible for GL-V9-induced dissociation of HKII from mitochondria. (A) Mitochondria were isolated after 15mM LiCl treatment for 36 h, and HK II was immunoprecipitated using VDAC antibody. Western blot assays were performed for HK II and VDAC. (B) The expressions of GSK-3β and p-GSK-3β in total proteins and mitochondrial fraction were assayed by Western blot. (C) Mitochondria were isolated, and HK II as well as VDAC were respectively immunoprecipitated using GSK-3β antibody. Western blot assays were performed for HK II, VDAC and GSK-3β. (D) Mitochondria were isolated upon 20 μM GL-V9 with/without 15mM LiCl treatment. VDAC was immunoprecipitated using GSK-3β antibody. Western blot assays were performed for VDAC and GSK3β. All the experiments were parallelly repeated in triplicate. The changes of protein levels were quantified. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control of MDA-MB-231 cells, #p < 0.05 or ##p < 0.01 versus non-treated control of MCF-7 cells.

studies.

wogonin. GL-V9 induced apoptosis and inhibited glycolysis of human breast cancer cells through promoting the dissociation of HK II from the mitochondria. The phosphorylation of VDAC by GSK-3β was responsible for GL-V9-induced dissociation of mitochondrial HK II. Moreover, activation of GSK-3β by GL-V9 was resulted from the direct effect of GL-V9 as well as the inactivation of AKT.

4. Discussion In this work, we investigated the anticancer mechanisms of a newly synthesized flavonoid GL-V9, a derivative of a natural product 7

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

(caption on next page)

8

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 5. GL-V9 suppresses the expression and activity of AKT, resulting in the activation of GSK-3β. (A) MDA-MB-231 and MCF-7 treated with AKT activator (IGF-1, 100ng/mL) or inhibitor (MK, 0.5 μM) for 36 h. Western blot assays for AKT, p-AKT, GSK-3β and p-GSK-3β. (B) Mitochondria were isolated upon 100ng/mL IGF-1treatment, and GSK-3β was immunoprecipitated using VDAC antibody. Western blot assays were performed for GSK-3β and VDAC. (C) The expressions of AKT and p-AKT in total proteins upon GL-V9 treatment were assayed by Western blot. (D) Mitochondria were isolated upon 20 μM GL-V9 treatment with/without 100ng/ mL IGF-1, and GSK-3β was immunoprecipitated using VDCA antibody. Western blot were performed for VDAC and GSK-3β. All the experiments were parallelly repeated in triplicate. The changes of protein levels were quantified. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control of MDA-MB-231 cells, # p < 0.05 or ##p < 0.01 versus non-treated control of MCF-7 cells.

binding domain and its putative BH4 domain, which is proposed to compete with the binding of VDAC to Bcl-xL (an anti-apoptotic member of the BCL-2 family proteins) [35]. Thus, HKII binding displaces Bcl-xL from VDAC and thereby facilitates its direct interaction with proapoptotic BCL-2 family members, such as Bax and its homolog Bak, and protects against OMM permeabilization. Alternatively, if HK II is detached, it may prevent Bcl-xL binding to Bax, resulting in the release of cytochrome c and other apoptosis enhancing proteins. AKT is an key oncogene protein kinase that controls various cellular functions and can exert antiapoptotic effects [36]. It is reported that AKT phosphorylates HKII at threonine 473 directly, promoting the binding of HKII with VDAC and the observed anti-apoptotic effects [10]. When AKT is inhibited, GSK-3β is dephosphorylated and activated. A large proportion of GSK-3β is found in the cytosol, but pools are also found in the nucleus and mitochondria. Many studies demonstrated that GSK-3β was implicated in inducing mitochondrial dysfunction. DNA damages can induce the interaction between GSK-3β and p53 at the mitochondria, whereas inhibition of GSK-3β blocks caspase activation and suppresses apoptosis [37]. In our studies, we found that GSK-3β could bind with VDAC in mitochondria, but not bind with HKII. GSK-3β phosphorylate VDAC on Thr51 region, which is highly conserved in mitochondrial porins and is in close proximity to Glu72, a residue critical to the binding of VDAC with HK II [38]. GL-V9 could inactivate AKT and promote the interaction of GSK-3β with VDAC, resulting in the disassociation of HKII from mitochondria and consequently, the inhibition of aerobic glycolysis and the induction of apoptosis. Our in vivo experiments further confirmed the effects of GL-

Metabolism reprograming is a unique feature of cancer cells [26]. Studies of the metabolic changes in cancers provide rationale for and insight about cancer therapy. It has long been proposed that most cancer cells satisfy their energy needs through a high rate of aerobic glycolysis (a low energy yielding process) followed by lactic acid fermentation, instead of mitochondrial oxidative phosphorylation (the most efficient energy yielding process), even in the presence of abundant oxygen [27,28]. In order to meet the rapid proliferation of tumor cells, glycolysis not only provides the necessary building materials for cell components, but also reduces the threat of oxidative stress [29]. Some rate-limiting enzymes of glycolysis, such as HK and pyruvate kinase (PK), in addition to ensuring glycolytic reactions, have other biological functions including transcriptional regulation and apoptosis induction [30,31]. Some inhibitors or regulators of glycolytic enzymes have been developed to extract cancer cells’ “sweet tooth” [32]. The recent approval of the first tumor metabolism regulating drug enasidenib by FDA served as a new gateway to cancer therapy. Another regulator methyl jasmonate can interfered with interactions between HK and VDAC and thus display anticancer activity [33]. Here, we found that GL-V9 was a potential cancer metabolism regulator. Our results demonstrated that disassociation of HKII from VDAC induced by GL-V9 was responsible for the disruption of mitochondria and suppression of glycolysis in human breast cancer cell. In malignant tumors, the level of HK II expression was 10 fold higher than normal tissues [34]. Stable binding of HK II to mitochondria promotes aerobic glycolysis, but impairs mitochondrial respiration [21]. HKII binds to VDAC through both its N-terminal membrane

Fig. 6. GL-V9 inhibited the growth of transplanted human breast tumor. Nude mice inoculated with MDA-MB-231 or MCF-7 cells were treated with saline control, GL-V9 (20mg/kg, i.v.), wogonin (60mg/kg, i.v.) and Taxol (8mg/kg, i.v.) (n=8 for each group). (A)Tumor weight, (B) Tumor inhibitory rates, (C)tumor volume, and (D) animal weight. Bars represent SD, *p < 0.05 or **p < 0.01 versus non-treated control group inoculated with MDA-MB-231 cells, #p < 0.05 or ## p < 0.01 versus non-treated control group inoculated with MCF-7 cells. 9

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

Fig. 7. The influences of GL-V9 in expression and activity of AKT and GSK3β in the transplanted human breast tumor tissue. (A) Histological analysis were photographed ( × 100). (B) Proteins expressions in breast tumor tissue were assessed by immunohistochemistry.

metabolic reactions. Notably, investigations have showed that upregulated HKII expression are associated with drug resistance phenotypes in breast cancer cells [41,42]. Therefore, HKII may be a genetic modifier of drug resistance that can be therapeutically exploited. And a better understanding of the functional differences of HKII in cancer tissue and normal tissues will undoubtedly contribute to better cancer therapies. In summary, the present studies demonstrate that GL-V9 induces apoptosis and inhibits aerobic glycolysis by blocking the mitochondrial binding of HKII in human breast cancer cells. The inactivation of AKT and the phosphorylation of VDAC by GSK-3β are responsible for the disassociation of HKII from mitochondria. This work provides mechanistic evidence that GL-V9 is a potent candidate for breast cancer treatment and a potential adjunctive agent for tumor chemotherapy.

Table 1 Pathology microscopic examination table for the xenograft tissue of MDA-MB231 cells. Treated Groups

Necrosis

Stroma

control GL-V9 Taxol Wognin

1 0.5 1 1

Vessel 1 0.5 0.5 1

Hemorrhage 0.5 0.5 0.5 1

Infiltration

Score

1 – – –

3.5 1.5 2 3

The tumor tissue were assessed by histological analysis. The tags in the table represent the severity: Minor "0.5", Mild "1", Annual "2", Severe "3", Extremely severe "4", and Normal "-". Table 2 Pathology microscopic examination table for the xenograft tissue of MCF7 cells. Treated Groups

Necrosis

Stroma

control GL-V9 Taxol Wognin

1 1 1 2

Vessel 1 0.5 1 0.5

Hemorrhage 0.5 – – –

Infiltration

Score

1 0.5 – –

3.5 2 2 2.5

Declaration of competing interest The authors declare no conflict of interest. Acknowledgement This work was supported by the National Science & Technology Major Project (No. 2018ZX09711001-005-023), Social Development Project of Jiangsu Provincial Science and Technology Department (BE2018711), and the National Natural Science Foundation of China (No. 81873051).

The tumor tissue were assessed by histological analysis. The tags in the table represent the severity: Minor "0.5", Mild "1", Annual "2", Severe "3", Extremely severe "4", and Normal "-"..

References

V9 in the inhibition of the expression and activity of AKT, as well as the activation of GSK-3β. Importantly, GL-V9 had no significant influence on body weight and regular physiological activities of mice. Furthermore, GL-V9 treated mice did not show symptoms of bone marrow suppression, the major side effect of taxol when applied at a dosage of comparable therapeutic effect [34]. Both our results and some previous reports have showed that the mitochondrial binding of HKII promotes cell survival [39,40]. The interaction between HKII and mitochondria plays a more important role in regulating apoptotic responsiveness of cells than merely catalyzing

[1] C.E. DeSantis, J. Ma, A. Goding Sauer, L.A. Newman, A. Jemal, Breast cancer statistics, 2017, racial disparity in mortality by state, CA A Cancer J. Clin. 67 (2017) 439–448. [2] O. Warburg, On the origin of cancer cells, Science 123 (1956) 309–314. [3] B. Zhivotovsky, S. Orrenius, The Warburg Effect returns to the cancer stage, Semin. Cancer Biol. 19 (2009) 1–3. [4] E. Bustamante, P.L. Pedersen, High aerobic glycolysis of rat hepatoma cells in culture: role of mitochondrial hexokinase, Proc. Natl. Acad. Sci. U. S. A. 74 (1977) 3735–3739. [5] J.G. Pastorino, J.B. Hoek, Regulation of hexokinase binding to VDAC, J. Bioenerg. Biomembr. 40 (2008) 171–182.

10

Free Radical Biology and Medicine xxx (xxxx) xxx–xxx

Y. Guo, et al.

[23] F. Chiara, D. Castellaro, O. Marin, V. Petronilli, W.S. Brusilow, M. Juhaszova, S.J. Sollott, M. Forte, P. Bernardi, A. Rasola, Hexokinase II detachment from mitochondria triggers apoptosis through the permeability transition pore independent of voltage-dependent anion channels, PLoS One 3 (2008) e1852. [24] J.E. Lee, J.S. Kang, Y.W. Ki, S.H. Lee, S.J. Lee, K.S. Lee, H.C. Koh, Akt/GSK3beta signaling is involved in fipronil-induced apoptotic cell death of human neuroblastoma SH-SY5Y cells, Toxicol. Lett. 202 (2011) 133–141. [25] X. Ma, Y. Bai, IGF-1 activates the P13K/AKT signaling pathway via upregulation of secretory clusterin, Mol. Med. Rep. 6 (2012) 1433–1437. [26] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144 (2011) 646–674. [27] M.V. Liberti, J.W. Locasale, The Warburg effect: how does it benefit cancer cells? Trends Biochem. Sci. 41 (2016) 211–218. [28] M.G. Vander Heiden, L.C. Cantley, C.B. Thompson, Understanding the Warburg effect: the metabolic requirements of cell proliferation, Science 324 (2009) 1029–1033. [29] D. Pathania, M. Millard, N. Neamati, Opportunities in discovery and delivery of anticancer drugs targeting mitochondria and cancer cell metabolism, Adv. Drug Deliv. Rev. 61 (2009) 1250–1275. [30] S.P. Mathupala, Y.H. Ko, P.L. Pedersen, Hexokinase II: cancer's double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria, Oncogene 25 (2006) 4777–4786. [31] R.B. Hamanaka, N.S. Chandel, Cell biology. Warburg effect and redox balance, Science 334 (2011) 1219–1220. [32] C. Rosano, Molecular model of hexokinase binding to the outer mitochondrial membrane porin (VDAC1): implication for the design of new cancer therapies, Mitochondrion 11 (2011) 513–519. [33] N. Goldin, L. Arzoine, A. Heyfets, A. Israelson, Z. Zaslavsky, T. Bravman, V. Bronner, A. Notcovich, V. Shoshan-Barmatz, E. Flescher, Methyl jasmonate binds to and detaches mitochondria-bound hexokinase, Oncogene 27 (2008) 4636–4643. [34] K.C. Patra, Q. Wang, P.T. Bhaskar, L. Miller, Z. Wang, W. Wheaton, N. Chandel, M. Laakso, W.J. Muller, E.L. Allen, A.K. Jha, G.A. Smolen, M.F. Clasquin, B. Robey, N. Hay, Hexokinase 2 is required for tumor initiation and maintenance and its systemic deletion is therapeutic in mouse models of cancer, Cancer Cell 24 (2013) 213–228. [35] J.G. Pastorino, J.B. Hoek, Hexokinase II: the integration of energy metabolism and control of apoptosis, Curr. Med. Chem. 10 (2003) 1535–1551. [36] J.A. Fresno Vara, E. Casado, J. de Castro, P. Cejas, C. Belda-Iniesta, M. GonzalezBaron, PI3K/Akt signalling pathway and cancer, Cancer Treat Rev. 30 (2004) 193–204. [37] P. Watcharasit, G.N. Bijur, L. Song, J. Zhu, X. Chen, R.S. Jope, Glycogen synthase kinase-3beta (GSK3beta) binds to and promotes the actions of p53, J. Biol. Chem. 278 (2003) 48872–48879. [38] R.B. Robey, N. Hay, Mitochondrial hexokinases, novel mediators of the antiapoptotic effects of growth factors and Akt, Oncogene 25 (2006) 4683–4696. [39] K. Gottlob, N. Majewski, S. Kennedy, E. Kandel, R.B. Robey, N. Hay, Inhibition of early apoptotic events by Akt/PKB is dependent on the first committed step of glycolysis and mitochondrial hexokinase, Genes Dev. 15 (2001) 1406–1418. [40] N. Majewski, V. Nogueira, P. Bhaskar, P.E. Coy, J.E. Skeen, K. Gottlob, N.S. Chandel, C.B. Thompson, R.B. Robey, N. Hay, Hexokinase-mitochondria interaction mediated by Akt is required to inhibit apoptosis in the presence or absence of Bax and Bak, Mol. Cell 16 (2004) 819–830. [41] O. Kaplan, G. Navon, R.C. Lyon, P.J. Faustino, E.J. Straka, J.S. Cohen, Effects of 2deoxyglucose on drug-sensitive and drug-resistant human breast cancer cells: toxicity and magnetic resonance spectroscopy studies of metabolism, Cancer Res. 50 (1990) 544–551. [42] K. Komurov, J.T. Tseng, M. Muller, E.G. Seviour, T.J. Moss, L. Yang, D. Nagrath, P.T. Ram, The glucose-deprivation network counteracts lapatinib-induced toxicity in resistant ErbB2-positive breast cancer cells, Mol. Syst. Biol. 8 (2012) 596.

[6] M. Hashimoto, J.E. Wilson, Membrane potential-dependent conformational changes in mitochondrially bound hexokinase of brain, Arch. Biochem. Biophys. 384 (2000) 163–173. [7] S. Abu-Hamad, H. Zaid, A. Israelson, E. Nahon, V. Shoshan-Barmatz, Hexokinase-I protection against apoptotic cell death is mediated via interaction with the voltagedependent anion channel-1: mapping the site of binding, J. Biol. Chem. 283 (2008) 13482–13490. [8] R.L. Elstrom, D.E. Bauer, M. Buzzai, R. Karnauskas, M.H. Harris, D.R. Plas, H. Zhuang, R.M. Cinalli, A. Alavi, C.M. Rudin, C.B. Thompson, Akt stimulates aerobic glycolysis in cancer cells, Cancer Res. 64 (2004) 3892–3899. [9] L. Sun, S. Shukair, T.J. Naik, F. Moazed, H. Ardehali, Glucose phosphorylation and mitochondrial binding are required for the protective effects of hexokinases I and II, Mol. Cell. Biol. 28 (2008) 1007–1017. [10] S. Miyamoto, A.N. Murphy, J.H. Brown, Akt mediates mitochondrial protection in cardiomyocytes through phosphorylation of mitochondrial hexokinase-II, Cell Death Differ. 15 (2008) 521–529. [11] P. Macanas-Pirard, N.S. Yaacob, P.C. Lee, J.C. Holder, R.H. Hinton, G.E. Kass, Glycogen synthase kinase-3 mediates acetaminophen-induced apoptosis in human hepatoma cells, J. Pharmacol. Exp. Ther. 313 (2005) 780–789. [12] M. Nishihara, T. Miura, T. Miki, M. Tanno, T. Yano, K. Naitoh, K. Ohori, H. Hotta, Y. Terashima, K. Shimamoto, Modulation of the mitochondrial permeability transition pore complex in GSK-3beta-mediated myocardial protection, J. Mol. Cell. Cardiol. 43 (2007) 564–570. [13] J.G. Pastorino, J.B. Hoek, N. Shulga, Activation of glycogen synthase kinase 3beta disrupts the binding of hexokinase II to mitochondria by phosphorylating voltagedependent anion channel and potentiates chemotherapy-induced cytotoxicity, Cancer Res. 65 (2005) 10545–10554. [14] L. Li, P. Chen, Y. Ling, X. Song, Z. Lu, Q. He, Z. Li, N. Lu, Q. Guo, Inhibitory effects of GL-V9 on the invasion of human breast carcinoma cells by downregulating the expression and activity of matrix metalloproteinase-2/9, Eur. J. Pharm. Sci. 43 (2011) 393–399. [15] L. Li, N. Lu, Q. Dai, L. Wei, Q. Zhao, Z. Li, Q. He, Y. Dai, Q. Guo, GL-V9, a newly synthetic flavonoid derivative, induces mitochondrial-mediated apoptosis and G2/ M cell cycle arrest in human hepatocellular carcinoma HepG2 cells, Eur. J. Pharmacol. 670 (2011) 13–21. [16] L. Zhao, W. Li, Y. Zhou, Y. Zhang, S. Huang, X. Xu, Z. Li, Q. Guo, The overexpression and nuclear translocation of Trx-1 during hypoxia confers on HepG2 cells resistance to DDP, and GL-V9 reverses the resistance by suppressing the Trx-1/Ref-1 axis, Free Radic. Biol. Med. 82 (2015) 29–41. [17] Y. Zhao, Y. Sun, Y. Ding, X. Wang, Y. Zhou, W. Li, S. Huang, Z. Li, L. Kong, Q. Guo, N. Lu, GL-V9, a new synthetic flavonoid derivative, ameliorates DSS-induced colitis against oxidative stress by up-regulating Trx-1 expression via activation of AMPK/ FOXO3a pathway, Oncotarget 6 (2015) 26291–26307. [18] L. Wei, N. Lu, Q. Dai, J. Rong, Y. Chen, Z. Li, Q. You, Q. Guo, Different apoptotic effects of wogonin via induction of H(2)O(2) generation and Ca(2+) overload in malignant hepatoma and normal hepatic cells, J. Cell. Biochem. 111 (2010) 1629–1641. [19] L. He, N. Lu, Q. Dai, Y. Zhao, L. Zhao, H. Wang, Z. Li, Q. You, Q. Guo, Wogonin induced G1 cell cycle arrest by regulating Wnt/beta-catenin signaling pathway and inactivating CDK8 in human colorectal cancer carcinoma cells, Toxicology 312 (2013) 36–47. [20] X. Xu, Y. Zhang, W. Li, H. Miao, H. Zhang, Y. Zhou, Z. Li, Q. You, L. Zhao, Q. Guo, Wogonin reverses multi-drug resistance of human myelogenous leukemia K562/ A02 cells via downregulation of MRP1 expression by inhibiting Nrf2/ARE signaling pathway, Biochem. Pharmacol. 92 (2014) 220–234. [21] N. Shulga, R. Wilson-Smith, J.G. Pastorino, Sirtuin-3 deacetylation of cyclophilin D induces dissociation of hexokinase II from the mitochondria, J. Cell Sci. 123 (2010) 894–902. [22] S. Ganapathy-Kanniappan, J.F. Geschwind, Tumor glycolysis as a target for cancer therapy: progress and prospects, Mol. Cancer 12 (2013) 152.

11