Targeting the ROS/PI3K/AKT/HIF‐1α/HK2 axis of breast cancer cells: Combined administration of Polydatin and 2‐Deoxy‐d‐glucose

Abstract It is well established that cancer cells depend upon aerobic glycolysis to provide the energy they need to survive and proliferate. However, anti‐glycolytic agents have yielded few positive results in human patients, in part due to dose‐limiting side effects. Here, we discovered the unexpected anti‐cancer efficacy of Polydatin (PD) combined with 2‐deoxy‐D‐glucose (2‐DG), which is a compound that inhibits glycolysis. We demonstrated in two breast cell lines (MCF‐7 and 4T1) that combination treatment with PD and 2‐DG induced cell apoptosis and inhibited cell proliferation, migration and invasion. Furthermore, we determined the mechanism of PD in synergy with 2‐DG, which decreased the intracellular reactive oxygen (ROS) levels and suppressed the PI3K/AKT pathway. In addition, the combined treatment inhibited the glycolytic phenotype through reducing the expression of HK2. HK2 deletion in breast cancer cells thus improved the anti‐cancer activity of 2‐DG. The combination treatment also resulted in significant tumour regression in the absence of significant morphologic changes in the heart, liver or kidney in vivo. In summary, our study demonstrates that PD synergised with 2‐DG to enhance its anti‐cancer efficacy by inhibiting the ROS/PI3K/AKT/HIF‐1α/HK2 signalling axis, providing a potential anti‐cancer strategy.

glycolysis, instead of oxidative phosphorylation. 7 Therefore, alteration of the manner of energy metabolism, a biochemical fingerprint of cancer cells, has been recognised as one of the ''hallmarks of cancer''. A series of cancer cell behaviours, such as infinite proliferation and metastasis, consume a large amount of energy. To acclimatise to this situation, cancer cells must elevate their glucose up-take due to glycolysis being less effective than oxidative phosphorylation in the adenosine triphosphate (ATP) yield. 8 Thus, targeting glycolysis is a promising therapeutic strategy for metastatic cancer. Emerging research suggests that an inhibition of glycolysis kills the malignant cells with a mild effect or even no effect on normal cells. 9,10 In fact, various types of drugs that inhibit glycolysis have been widely used in the treatment of malignant cancer.
2-deoxy-D-glucose (2-DG) is one of the most effective anti-glycolytic agents. It is phosphorylated by hexokinase (HK), which is the first rate-limiting enzyme of glycolysis and subsequently inhibits the pentose-phosphate pathway (NAPDH) and ATP generation. 9,11 2-DG can change the redox state of the cell, or the generation of free radicals and then disorder the cell cycle and induce apoptosis. 12 Previous studies have shown that 2-DG is an effective anti-cancer agent in cellular systems and in animal models. 13,14 However, 2-DG monotherapy has yielded few positive results in mouse xenografts and human patients, most likely due to dose-limiting side effects or the activation pro-survival pathways in cancer cells. 15,16 Recent studies in cancer chemotherapy are focused on a combination of two or more drugs. Polydatin (PD, 3,4'-5-trihydroxystilbene-3-β-D-glucopyranoside, shown in Figure 1A), is extracted from the roots of Polygonum cuspidatum Sieb and can be detected in grape, peanut, hop cones, red wines, hop pellets, cocoa-containing products, chocolate products and many daily diets, is widely applied in traditional Chinese therapy. 17 Previous studies demonstrated that PD has antioxidant, anti-inflammatory and anti-cancer activities and it is mainly used in cardiovascular, inflammatory, neurodegenerative, metabolic and age-related diseases. 18,19 PD exerts its anti-cancer effect in a variety of ways, such as the regulation of reactive oxygen species (ROS) 20 and inhibition of the PI3K/AKT pathway. 21 Interestingly, some studies indicated that PI3K/AKT inhibitors enhanced the therapeutic efficacy of 2-DG. 22,23 In addition, PD has received considerable attention for its beneficial effects on glucose and lipid regulation. 24 Thus, whether PD can promote the anti-cancer effects of 2-DG by regulating glucose metabolism, inhibiting the PI3K/AKT pathway or other mechanisms has aroused great interest. In the present study, we evaluated the anti-cancer effect of PD, 2-DG and their co-treatment in breast cancer cell lines (4T1 and MCF-7) and elucidated the underlying molecular mechanisms.

| Cell culture
The 4T1 Cells used in this study were obtained from the Chinese Academy of Sciences cell Bank (Shanghai, China) and MCF-7 cells were purchased from the American Tissue Culture Collection (ATCC, Rockville, MD). All cell lines were grown in RPMI 1640 medium with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin at 37°C with 5% CO2. Cells were cultured in T25 cell culture flasks. When the cell density reached more than 80%, the cells were subcultured into 96-well plates or 6-well plates for the corresponding experiments.

| Antibodies and reagents
Western blotting, immunofluorescence and immunohistochemistry were performed using the following antibodies: PI3K, phos-

| Cell viability and proliferation assay
CCK-8 was used to evaluate the viability and proliferation of breast cancer cell lines (4T1 and MCF-7). Cells were seeded in 96-well plates at a density of 5×10 4 cells/ml. After treatment, cells were continuously cultured with 10 μL of CCK8 in each well at 37°C for 2 hours. Cell proliferation was measured through absorbance (optical density) with a microplate reader (Bio-Rad Instruments, Hercules, CA) at 450 nm.

| Cell migration assay
A Wound Healing Assay was used to assess cell migration. As a previous study described the method, 25

| Flow cytometry apoptosis analysis
After treatment, cells were trypsinized and washed twice in cold phosphate-buffered saline. Cells were stained using the Annexin-V-PE apoptosis detection kit (BD, San Diego, CA) according to the manufacturer's protocol and were analysed by FACS (BD).

| Measurement of ROS production
The intracellular ROS level was measured by flow cytometry.
Briefly, cells were seeded at a density of 1 × 10 6 cells ml −1 into 6well plates and, after treatment as indicated, the intracellular ROS level was measured using the oxidative conversion of cell perme-

| RNA extraction and qPCR
Total RNA was extracted from collected cells using the TRIzol reagent (Invitrogen, USA) and reverse transcription to synthesise cDNA was performed using a reverse transcription kit (Takara, Japan). The primer sequences used are shown in Table 1. Quantitative real-time PCR mixture reagents contains 1 μL forward and reverse primers, 1 μL of cDNA, 10 μL SYBR qPCR Mix (Roche, Swiss) and 7 μL TE water and on a StepOne Plus™ Real-Time PCR System (BIO-RAD, USA). The target genes expression level was evaluated using the 2 −ΔΔCt comparative approach.

| Western blot analysis
The total protein of cells or tissues were extracted with a RIPA lysis solution involving phosphatase inhibitors (Vazyme, Nanjing, China) and then separated by SDS-PAGE and transferred to PVDF membrane. The membrane was incubated in blocking buffer (5% skim milk) and subsequently treated with primary antibody and secondary antibody. The protein expression levels were determined using the ECLPlus western blot Detection System.

| Histopathological assessment and immunohistochemistry
Tumour tissues and the major vital organs, such as the heart, liver and kidneys, of mice were isolated and fixed in 10% formalin. The tissues were dehydrated, paraffin embedded and then cut into 5-μm-thick sections for hematoxylin and eosin (H&E) staining.
Immunohistochemistry detection using the following primary antibodies: anti-VEGF, anti-Ki67, anti-HIF-α and anti-HK2 was performed on paraffin sections. The staining processes were performed according to standard methods. The sections were observed using an optical microscope (Olympus, Japan).  The symbols * and ** denote significant differences of P < 0.05 and P < 0.01, respectively

| Immunofluorescence assay
The breast cancer cells were inoculated into a 6-well-plate. After treatment, the cells were fixed with 4% (v/v) paraformaldehyde for 20 minutes, permeabilised with 0.2% Triton X-100 for 10 minutes and then blocked with 5% BSA for 1 hour, followed by incubated with the primary antibodies at 4°C overnight. After three washes in PBS, the cells were incubated with the FITC-labelled goat anti-rabbit

| TUNEL assay
The TdT-UTP nick end labelling (TUNEL) assay was performed using a

| Statistical analysis
All values are presented as the means ± SEM for three independent experiments. The intergroup differences were determined by a two-way ANOVA and Student's t test. A value of P ≤ 0.05 was considered statistically significant and P ≤ 0.01 was highly statistically significant. All results were analysed using GraphPad Prism 5.

| PD combined with 2-DG displays a potent anti-cancer activity in vitro
PD is known to inhibit proliferation and induce apoptosis in breast cancer and colorectal cancer. 26 The symbols * and ** denote significant differences of P < 0.05 and P < 0.01 versus the control group, respectively typical apoptotic proteins in cancer cells ( Figure 2H,I). The mRNA levels of Bax, Bcl-2, cIAP and XIAP were also detected by RT-PCR ( Figure 2J).
These results were consistent with the above conclusions. In summary, synergistic actions between the PD and 2-DG components contribute to an anti-tumour property in vitro.

| The PD and 2-DG combination treatment inhibits the PI3K/AKT prosurvival signalling pathway
To further understand the molecular mechanisms by which PD and 2-  Figure 3A,B). In addition, we also found that 2-DG co-treatment with LY294002 (Standard PI3K/ AKT Inhibitor) achieved these results ( Figure 3C,D). This indicates that the inhibition of the PI3K/AKT signalling pathway by PD curbs the 2-DG-induced prosurvival signalling. We further confirmed this result by immunofluorescence ( Figure 3E).

| PD inhibits ROS generation, a critical regulatory factor of the PI3K/ AKT pathway
Further studies were performed to determine the upstream regulators of the PI3K/AKT pathway and its downstream targets. As

| The synergistic anti-cancer effect of PD and 2-DG is through the suppression of HK2 in breast cancer cells
A typical hallmark of malignant cells is their capacity to perform glycolysis at a higher rate. 7 The structural features of 2-DG prompted us to evaluate the glycolytic phenotype in PD and 2-DG co-treated breast cancer cells. The results indicated that the first step rate limiting enzyme, glycolysis hexokinase (HK2), which is overexpressed in malignant cells, 34 had significantly decreased protein levels following co-treatment with PD and 2-DG ( Figure 5A,B). In further studies we examined other markers of glycolysis, such as glucose transporter 1 (GLUT1) and Lactate dehydrogenase A (LDH-A) ( Figure 5A,B), and the results were consistent with the changes of HK2. In addition, we noticed an aberrant overexpression of hypoxia-inducible factor 1α (HIF1α) in many different cancers and HK-2 is one of its downstream transcriptional targets. 35,36 In this study, we also found that a combination of PD and 2-DG inhibited the protein expression of HIFIα in 4T1 and MCF-7 cells (Figure 5A,B). In addition, we also found that targeting HK2 with siRNA can block breast cancer cell survival and  Figures   5C5-7). Overall, the PD and 2-DG combination inhibits glycolytic metabolism by suppressing HIF1α/HK2 in breast cancer cells.

| The effects of PD and 2-DG on tumour growth: murine syngeneic tumour models
To

| D ISCUSS I ON
The pivotal role of glycolysis in cancer onset and progression has been recently recognised. 37,38 Therefore, correcting the level of glycolysis of malignant cancer cells is one of the emerging research hotspots in cancer chemotherapy. However, the agents that specifically inhibit glycolytic metabolism have not yielded the expected effect in cancer patients, most likely due to cancer cells developing resistance to the agents or to dose-limiting side effects. 11,16 Combinational chemotherapies have been widely used to minimise acquired resistance. 39 In this study, we demonstrate that the synergy between PD The glucose analogue, 2-DG, which is phosphorylated by hexokinase, competes with 6-phosphate glucose (6-P-G) to inhibit glycolysis, resulting in cancer cell death due to cell energy ATP deprivation. 9,11 Polydatin, also named piceid (PD), has many biomedical properties, such as anti-platelet aggregation, antioxidative, anti-cancer, anti-inflammatory and immune-regulating functions. 40 In this study, we found that 2-DG and PD both inhibited cell viability in 4T1 and MCF-7 ( Figure 1) and that PD combined with 2-DG obviously induced apoptosis and led to a reduction in cell proliferation and migration in 4T1 and MCF-7 compared to the individual treatment groups ( Figure 2). However, previous studies have shown that 2-DG induces the phosphorylation of AKT and the efficacy of clinical trials has previously been limited by the systemic toxicity. 41,42 The activation of AKT is commonly observed in a variety of cancers and seems to be intricately associated with aerobic glycolysis, proliferation and invasiveness. 43,44 In this study, we found that the mechanism of action of the treatment combination is that PD blocks the activation of the PI3K/ AKT pathway by 2-DG ( Figure 3).
Under physiological conditions, various redox systems ensure that ROS are appropriately utilised to accomplish specific functions, such as signalling and protein regulation. 45 It is generally accepted that ROS are tumour suppressors that are implicated in tumorigenesis, progression and survival phenotypes. 45 was also found to be down-regulated by ROS in cancer cells. 47,48 ROS is a double-edged sword and the influence of ROS within cancer cells, whether they are favourable or harmful, may depend on several factors, such as cell type, stimulus, duration, specificity and levels of ROS.
The overexpression of the oncogene HIF-1 in cancer cells results in their adaptation to a microenvironment that is hypoxic compared to normal cells, thereby avoiding apoptosis. The mechanism of action of HIF-1 includes a switch from oxidative phosphorylation to glycolysis, increasing glycogen synthesis and a switch from glucose to glutamine as the major substrate for fatty acid synthesis. 49 It is worth noting that the phosphorylation of AKT results in the upregulation of HIF1 in cancer cells. 50 Consistent with these results, we found that PD and 2-DG co-treatment increased HIF-1α by inhibit the PI3K/AKT pathway in breast cancer cell lines ( Figure 5). The glycolysis rate strongly depends on the upregulated expression and activity of glucose transporters (Gluts), such as Glut1, which is regulated by HIF-1α and is considered the main overexpressed Glut, with a 10-12-fold higher expression in tumours than in normal cells. 49,51 In this study, the expression levels of Glut 1, LDH-A and HK2 were detected to assess the glycolysis rate and the results indicated that PD and 2-DG co-treatment inhibited glycolysis in breast cancer cell lines. In addition, HK2 is the direct target of HIF-1α in HCC cells. 52 In this study, this is evidenced by the silencing of HK2 by siRNA enhancing the inhibitory effect of combination treatment on glycolysis and growth ( Figure 5).
In line with our in vitro data, PD and 2-DG co-treatment potently suppressed tumour growth in murine syngeneic breast tumour models. Collectively, we first determined the synergistic anti-cancer effects of PD and 2-DG coutreatment and found that the effect was mediated by the ROS/PI3K/AKT/HIF1α/HK2 and glycolysis pathways in breast cancer. This study is the first to indicate the efficient combination treatment of PD and 2-DG, providing a potential treatment strategy for breast cancer patients.

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests.