Whey Protein Concentrate Renders MDA-MB-231 Cells Sensitive to Rapamycin by Altering Cellular Redox State and Activating GSK3β/mTOR Signaling

Whey protein concentrate (WPC) is an amino acid-rich supplement that has been shown to increase cellular antioxidant capacity. Mammalian target of rapamycin (mTOR) is a crucial regulator of signaling in mammalian cells, and serves as a therapeutic target for triple-negative breast cancer (TNBC). This study was designed to investigate the effect of combining WPC with rapamycin on MDA-MB-231 human breast cancer cells. These cells were found to be insensitive to rapamycin and exhibited higher glutathione (GSH) and reactive oxygen species levels than non-tumorigenic MCF-10A cells. However, for MDA-MB-231 cells, the half maximal inhibitory concentration of rapamycin was lower when this drug was administered in combination with WPC than when used alone. Furthermore, combining WPC with rapamycin depleted GSH levels and reduced Nrf2 nuclear accumulation. In addition, WPC activated GSK3β/mTOR signaling, and GSK3β appeared to be involved in the WPC-mediated Nrf2 reduction and mTOR activation. In conclusion, WPC induced rapamycin sensitivity in MDA-MB-231 cells by altering their redox state and activating GSK3β/mTOR signaling. These results not only suggest a novel therapeutic approach for breast cancer treatment, but also provide insight into the critical pathways affecting the resistance to mTOR inhibition observed in a subgroup of TNBC patients.

Reactive oxygen species (ROS) and cellular oxidative stress are associated with cancer 5,6 , and compared to normal cells, antioxidant capacity is upregulated in malignant cells to adapt to higher ROS levels 5 . Previous studies have reported that nuclear factor (erythroid-derived 2)-like 2 (Nrf2) responds to increased ROS levels by enhancing the expression of genes involved in maintaining the cellular redox balance, including those encoding glutamate-cysteine ligase (GCL) and glutathione reductase (GR), which are associated with glutathione (GSH) production and regeneration 7 . Furthermore, deregulation of Nrf2, such as that leading to its increased nuclear accumulation, reduces apoptosis and promotes drug resistance 8 .
Since mTOR and Nrf2 signaling are known drivers of human oncogenesis 9,10 , agents targeting these pathways have shown promise as treatments for breast cancer 11,12 . However, despite the establishment of this treatment strategy and its encouraging results, a proportion of patients with TNBC harbor tumors that are resistant to mTOR inhibitors 13 . Thus, identifying markers of mTOR inhibitor sensitivity or the development of combination therapy is urgently needed to improve TNBC response to mTOR inhibition.
Whey protein concentrate (WPC) is prepared in a special manner to preserve native forms of the cysteine-rich proteins in whey (serum albumin, lactoferrin, and α-lactalbumin) and functions as a GSH precursor in cells. We have shown previously that WPC supplementation increases antioxidant activity in human peripheral blood mononuclear cells and rats treated with high doses of alcohol 14,15 . Besides, we have demonstrated that WPC supplementation selectively depletes tumor GSH levels in 7,12-dimethylbenz[a]anthracene (DMBA)-induced mammary tumors in rats 16 . In addition, prior studies have reported that WPC ingestion promotes activation of the mTOR pathway 17 , and it was demonstrated that this supplement exerts its antioxidant effects through an Nrf2-dependent mechanism in endothelial cells 18 . Therefore, in the present work, we aimed to determine whether WPC can influence the susceptibility to mTOR inhibitors using MDA-MB-231 TNBC cells, a cell line that has been reported to be resistant 19,20 . These results might provide insight into the critical pathways that are involved in resistance to mTOR inhibition in TNBC, and could identify biomarkers of for the responsiveness to such inhibitors.

Results
Determination of cellular redox status. Increased GSH (Fig. 1a) and ROS (Fig. 1b,c) levels were noted in advanced breast cancer cell lines, including MDA-MB-231 and MDA-MB-468, compared to those in MCF-10A non-tumorigenic breast epithelial cells. Since GSH is a key cellular antioxidant, these results indicated that cancer cells have higher tolerance to ROS. WPC sensitizes MDA-MB-231 cells to rapamycin. To examine the sensitivity of each cell line to rapamycin, cells were treated with this drug at different concentrations for 72 hours, and cell viability was measured by standard MTT assays. As shown in Fig. 2a, growth of MDA-MB-468 cells was significantly inhibited by rapamycin, whereas MDA-MB-231 cells were resistant to the growth inhibitory effects of this drug, with observed half maximal inhibitory concentration (IC 50 ) values of 8.91 nM and >10 μM, respectively. Given these different responses to rapamycin, we chose the MDA-MB-231 line for further evaluation. In addition, cellular viability of MCF-10A, MDA-MB-468, and MDA-MB-231 cells was measured with WPC at concentrations of 1, 10, and 20 mg/mL. There were no significant differences (P > 0.05) in cell viability among the WPC-treated groups, despite increasing concentrations of WPC (Fig. S1a-c). Regarding viability of MDA-MB-231 cells, Fig. 2b shows the effect of 48-hour exposure to WPC (10 mg/mL) combined with different concentrations of rapamycin. Compared to treatment with rapamycin alone, the growth inhibitory effect of combined treatment resulted in a significant decrease in the IC 50 of rapamycin (from >10 μM to 2.54 μM, respectively). In vitro colony formation assays were then performed to further test the effect of WPC combined with rapamycin on cell growth. Figure 2c and d demonstrate that combined treatment led to a significant decrease in the number of colonies compared to that in the control and rapamycin-only groups. However, cells treated with WPC or rapamycin alone did not significantly differ from the control in this respect.
Effect of WPC, rapamycin, and their combination on cellular redox status. As shown in Fig. 3a, WPC (10 mg/mL) treatment selectively depleted GSH levels in MDA-MB-231 cells, but increased GSH levels in non-tumorigenic MCF-10A cells. Moreover, in MDA-MB-231 cells, rapamycin (1 μM) treatment did not alter GSH levels (Fig. 3b), but did significantly increase ROS concentrations (Fig. 3c,d) compared to those in the control. Furthermore, in comparison to those in the control, GSH levels were significantly lower following administration of WPC alone or in combination with rapamycin. However, WPC alone had no effect on ROS levels in Effect of WPC, rapamycin, and their combination on the expression of Nrf2, GCL, and GR in MDA-MB-231 cells. Since the intracellular GSH redox state is regulated by the Nrf2-GCL-GR-GSH signaling pathway, whole cell lysates were subjected to immunoblotting for Nrf2, the GCL catalytic subunit (GCLC), and GR to clarify the effects of WPC. As shown in Fig. 4, there were no significant differences (P > 0.05) in whole-cell Nrf2 protein expression between groups with WPC treatment alone or combined treatment, when compared to that in the control group. However, treatment with rapamycin alone increased whole-cell Nrf2 protein expression. Compared to the that in the control group, WPC alone and combined treatment resulted in reduced expression of GCLC and GR proteins. However, treatment with rapamycin alone had no effect on the expression of these proteins.

Rapamycin-induced nuclear accumulation of Nrf2 in MDA-MB-231 cells is reduced by WPC treatment.
Since there were no significant differences in whole-cell Nrf2 protein expression among groups, we next evaluated the cellular translocation of Nrf2. As indicated in Fig. 5a and b, rapamycin treatment induced Nrf2 nuclear accumulation; however, WPC administration decreased the presence of Nrf2 in the nucleus, and inhibited the effect of rapamycin on Nrf2 translocation. WPC activates the GSK3/mTOR pathway. We next assessed the combined effects of WPC and rapamycin on mTORC1 in MDA-MB-231 cells, since this complex is the direct target of rapamycin and is activated by nutritional factors. The level of phosphorylated (p)-mTOR (Ser2448) is a marker of mTORC1 activation. As shown in Fig. 6, the p-mTOR/mTOR ratio was approximately 2.2-fold higher in the WPC-only group compared to that in the control group. In contrast, compared to that in the control, the p-mTOR/mTOR ratio was found to be reduced with combined treatment. In addition, the p-p70S6K/p70S6K and the p-4E-BP1/4E-BP1 ratios were 1.4-fold and 1.5-fold higher, respectively, in the WPC-only group than the control group, whereas this effect was negated by the administration of rapamycin together with WPC. Strikingly, relative to that in the control, WPC upregulated p-GSK3β/GSK3β, both alone (1.32-fold) and with combined treatment (2.2-fold).

WPC reduces Nrf2 nuclear accumulation and induces mTORC1 activation via GSK3β activity.
To determine the mechanism underlying the WPC-mediated reduction in Nrf2 nuclear accumulation, we downregulated GSK3β protein expression using siRNA specific to the corresponding mRNA. Cells were transfected with non-specific (scrambled) siRNA or GSK3β siRNA, and were then treated with WPC. Protein expression was then detected by immunoblotting (Fig. 7a). GSK3β was inhibited, as expected, and this was found to result in increased nuclear Nrf2, compared to cytosolic levels (Fig. 7b,c). Moreover, cells treated with GSK3β siRNA exhibited a reduced p-mTOR/mTOR ratio in response to WPC treatment (Fig. 7d,e).

Discussion
Although many studies have investigated molecular markers such as VEGF 21 , EGFR 22 , and mTOR 23 as drug targets for TNBC, a greater understanding of the molecular basis of this disease is required for the development of effective treatments. In the present study, we demonstrated that WPC treatment renders MDA-MB-231 cells susceptible to rapamycin by (i) altering the cellular redox state and (ii) activating GSK3β/mTORC1 signaling.  Basal-like TNBC cells were reported to be highly sensitive to the mTOR inhibitor everolimus, whereas a subgroup of TNBC cells, comprising lines characterized as stem-like cancer cells, are less sensitive to this drug 24 . Given that the cellular redox state affects the regulation of signal transduction pathways and the development of drug resistance, this could be used to dictate therapeutic strategies for TNBC 25 . Many studies have shown that redox mechanisms regulate mTORC1 activity 26 . For example, Sarbassov et al. demonstrated that cysteine oxidants such as phenylarsine oxide and diamide activate mTORC1 in vivo and in vitro 27 .
In the current work, we first observed that GSH and ROS levels were elevated in MDA-MB-231 cells compared to those in non-tumorigenic MCF-10A cells. Our previous results indicated an imbalance in GSH redox status in breast cancer patients. The high GSH levels observed in breast cancer tissues might be associated with enhanced cell proliferation and resistance to oxidative stress, representing a selective growth advantage for tumor cells over their normal counterparts 28 . In addition, Panis et al. reported that compared to adjacent normal breast tissue, tumors are sites of elevated oxidative and nitrosative stress, with concomitant augmented antioxidant capacity 29 .
WPC constitutes a rich source of bioavailable cysteine that can be used for GSH synthesis, and contains all nine essential amino acids. It is used to promote muscle protein synthesis and as a nutritional supplement for chemotherapy patients 30 . The PTEN-positive cell line MDA-MB-231 has been reported to be resistant to mTOR inhibition; however, we observed here that WPC treatment rendered these cells sensitive to rapamycin. Furthermore, WPC selectively altered GSH levels, affecting those in MDA-MB-231 cells, but not those in non-tumorigenic MCF-10A cells. This might be explained by the fact that GSH synthesis is regulated by negative feedback. Since GSH levels are much higher in tumor cells than in normal cells, it is easier to reach the threshold at which GSH synthesis is inhibited by negative feedback in the former than in the latter 31,32 . In addition, we found that treatment with 1 μM rapamycin, with or without WPC, induces ROS production in MDA-MB-231 cells. The relationship between rapamycin and ROS remains disputed, with some studies suggesting that rapamycin reduces cellular ROS levels 33 and others claiming the opposite 34 . Such discrepancies might be explained by the important effect of rapamycin concentration 35,36 . In addition, there were no obvious changes in the viability of cells with WPC treatment. It is notable that in the present work, WPC selectively affected GSH levels but did not change cellular ROS content and viability, implying that a synergistic antitumor effect might be achieved by decreasing GSH levels in tandem with WPC treatment 37 . Nrf2 regulates the inducible antioxidant program in response to cellular stress 38 . Here, we revealed that rapamycin treatment induces the nuclear accumulation of Nrf2, consistent with the observations of other researchers. It has also been reported that TORC1 inhibition by rapamycin increases transcription of the Nrf2 gene 39,40 . Notably, combining rapamycin with WPC negated the promotion of Nrf2 nuclear accumulation by the rapamycin treatment in the present study. Moreover, we observed reduced expression of GCLC and GR proteins when cells were treated with WPC. Consistent with the observations of our previous in vivo study, we found that WPC supplementation reduced the nuclear accumulation of Nrf2 in tumor tissue 16 . Kerasioti et al. indicated that whey protein exerts its antioxidant activity through an Nrf2-dependent mechanism in endothelial cells; however, it was cell type dependent 18 .
It has been suggested that p-p70S6K low /p-4E-BP1 high cancer cells might be intrinsically resistant to mTORC1 inhibitors 41 . Of note, we found here that WPC treatment reversed this phenotype by enhancing phosphorylation of mTOR (Ser2448), p70S6K (Thr389), and 4E-BP1. Furthermore, we revealed that WPC upregulated GSK3β expression, consistent with a study by Stretton et al., in which GSK3β activity was shown to be required for the support of mTORC1 signaling in response to nutrient availability 4 . It has also been recognized that amino acid treatment promotes mTORC1 activation 42,43 . Our investigation suggests a combined therapy that might be beneficial for the subgroup of TNBC patients with tumors resistant to mTOR inhibitors.
Finally, in the present study, we inhibited GSK3β activity using siRNA, the results of which suggested that GSK3β is required for the reduction in Nrf2 nuclear accumulation mediated by WPC treatment. This is consistent with a report that activation and direct inhibition of GSK3β leads to nuclear exclusion and accumulation of Nrf2, respectively 44 . In addition, GSK3β was found to be associated with WPC-induced mTORC1 activation. Taken together, our results indicate that WPC leads to the depletion of GSH by reducing Nrf2 nuclear accumulation and activating GSK3β/mTOR signaling, rendering MDA-MB-231 cells susceptible to rapamycin. In conclusion, we explored the result of combining antioxidant and mTOR inhibitory effects, and clarified the underlying mechanism for the first time (Fig. 8). Our findings might serve as a reference for the development of such combination therapies for the treatment of breast cancer, and particularly TNBC.

Materials and Methods
Reagents and antibodies. The WPC (Immunocal, Immunotec Research Ltd., Quebec, Canada) contained about 90% whey protein isolate, < 1.5% lactose, < 0.5% fat, and <5.0% moisture, with a solubility index of 99% at pH 4.6. WPC was freshly prepared before each experiment. Preparation of WPC. Preparation of WPC was the same as in our previous study 15 . We followed the instructions of Immunocal to design the concentrations of WPC and WPC was freshly prepared before each experiment. Briefly, 50 mg of WPC was dissolved in a fresh 5 mL serum-free medium and centrifuged at 12,000 × g for 10 minutes at room temperature. The supernatant was filtered with a 0.2 μm filter. Therefore, concentrations of 10 mg/ mL of WPC were used in this experiment.
Determination of GSH. The 100 μL cell suspension, as indicated above, in an amber microcentrifuge tube was mixed with 0.1% MPA in a 1:2 volumetric ratio, and then it was centrifuged at 12,000 × g for 10 minutes after being adequately mixed by a vortex mixer. The supernatant was assayed on a capillary electrophoresis analyzer (P/ACE MDQ, Beckman Coulter) after being filtered with a 0.2 μm syringe set. The analysis was performed at a constant temperature (28 °C) with 300 mM borate running buffer (pH 7.8) and a UV absorbance detector set to 200 nm.  Cell viability assays. Cell viability was assessed by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) dye conversion at 570 nm following the manufacturer's instructions. Briefly, cells were seeded 5,000 per well in a 96-well flat bottom plate and grown for 24 hours. Cells were then treated for 48 hours or 72 hours with indicated concentrations of rapamycin and WPC in the presence of 10% FBS. 20 μl of MTT (5 mg/ mL in PBS) was then added to each well and incubated 2 hours at 37 °C. The formazan in viable cells were dissolved with 100 μl of dimethyl sulfoxide and determined by reading optical densities in a microplate reader (Dynatech Laboratories, Chantilly, VA, USA).
Colonies formation assay. MDA-MB-231 cells were seeded in 6-well plate at a density of 1,500 cells per well duplicate. These cells were cultured with respective treatment including WPC (2 mg/mL), rapamycin (200 nM), and their combination. The medium was replaced every 2-3 days until day 14. Colonies were stained with 0.2% (w/v) crystal violet in buffered formalin for 10 minutes. The number of colonies was counted.
Preparation of cytosolic and nuclear extracts. Cytosolic and nuclear extracts were isolated according to Schreiber et al. 45 with some modify cations. After rinsing with cold PBS, cells were resuspended in cold lysis buffer A (20 mM HEPES pH 8.0, 1 mM EDTA, 1.5 mM MgCl 2 , 10 mM KCl, 1 mM DTT, 1 mM sodium orthovanadate, 1 mM NaF, 1 mM PMSF, 0.5 mg/mL benzamidine, 0.1 mg/ml leupeptin, and 1.2 mg /mL aprotinin). They were incubated on ice for 15 min, after which 7.5 μL of 10% NP-40 detergent was added. Afterwards the cells were intensively stirred in a vortex mixer for 10 s. The homogenate was centrifuged at 15,000 × g for 50 seconds, and the supernatant was used as the cytosolic extract. The nuclear pellet was resuspended in cold extraction buffer B (20 mM HEPES pH 8.0), 1 mM EDTA, 1.5 mM MgCl 2 , 10 mM KCl, 1 mM DTT, 1 mM sodium orthovanadate, 1 mM NaF, 1 mM PMSF, 0.5 mg/mL benzamidine, 0.1 mg/mL leupeptin, 1.2 mg/mL aprotinin, and 20% glycerol). All protein fractions were stored at −70 °C until used.
Western blotting. Cells were seeded, cultured, and treated with WPC (10 mg/mL), rapamycin (1 μM), and their combination for 48 hours. The proteins were separated by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membrane. The membranes were blocked with 5% milk for 1 hour and incubated with targeted primary antibodies overnight at 4 °C, and then incubated with HRP-conjugated secondary antibodies at room temperature for 1 hour. Finally, the signals were enhanced by ECL (Millipore, Temecula, CA, USA) and exposed to X-ray film for autoradiogram. The intensity of the bands was quantified by densitometry and the results were evaluated relative to the expression of the constitutively expressed protein β-actin.
Small interfering RNA (siRNA). GSK-3β was silenced using siRNA (Santa Cruz Biotechnology, sc-35527) and a nonspecific scrambled siRNA (sc-37007) used as the negative control. Cells at 80% confluency were transfected with 2 μg of siRNA/well in 6-well plates using lipofectamine 2000 (Invitrogen). After they had been transfected, the cells were incubated for 18 hours in DMEM supplemented with 10% FBS at 37 °C, then treated with WPC or rapamycin for additional 48 hours.