Introduction
Apoptosis and autophagy are two main modes of programmed cell death (1). Apoptosis is regulated by caspases and Bcl-2 family proteins and involved in multiple cellular signaling pathways (2). Autophagy is modulated as a survival mechanism induced in problematic environment to maintain cell integrity by nutrient starvation and/or metabolic stress (3). Recently, apoptosis and autophagy may be triggered by general upstream signaling to affect tumor cell development and therapy (4,5).
Apoptosis is characterized by cellular morphological membrane blebbing, cell shrinkage, nuclear fragmentation, chromatin and DNA condensation, and chromosomal DNA fragmentation (1,6). There are two major mechanisms involved in apoptotic cell death such as intrinsic (mitochondria-mediated) and extrinsic (death receptor) pathways (7–9). In the intrinsic pathway, the first regulatory stage is mediated by Bcl-2 family proteins. In apoptotic stimuli, the level of Bax (a pro-apoptotic protein) increased, following binding to Bcl-2 (a pro-survival protein) then releasing Bax/Bak molecules. Free Bax and Bak form oligomers leading to cytochrome c release from mitochondria to the cytoplasm. The released cytochrome c activates the caspase-9/-3 cascade to induce apoptosis. The second regulatory step of mitochondrial pathway is the formation of apoptosome. The mitochondria-related proteins (such as cytochrome c, Apaf-1 and pro-caspase-9) can be released to the cytoplasm to form apoptosome, which binds and cleaves initiator pro-caspase-9 and activates the caspase-9/-3 cascade to induce apoptosis (7,10).
Autophagy is an important response to the cellular environments and positively regulates cellular processes for survival or death during cell stress and damage, nutrient starvation, aging and pathogen infection (11,12). Morphological and biochemical modification involve autophagy-related proteins (Atgs) and related autophagy signal molecules (13). Once autophagic response is activated, the membrane nucleation is regulated by the signal cascade of phosphatidylinositol 3-kinase (PI3K) class III, Beclin-1, Rubicon and Atg14. Furthermore, Atg16L1-Atg12-Atg7-Atg5 complexes and microtubule-associated protein 1 light chain 3 type II (LC3-II) are required for autophagosome formation before the cytoplasm and phagophore of various organelles are sequestered (13,14). Autophagolysosome is from the autophagosome fusion with the lysosome to induce the degradation of the captured proteins or organelles. The autophagosomal marker LC3-II increases from the conversion of LC3-I during undergoing autophagic death (12,13). Thus, autophagy is developing as a new target for cancer therapy and chemoprevention (15), and induction of autophagic cell death is considered to possibly be one of the best strategies in cancer chemotherapy.
Resveratrol (3,4′,5-trihydroxy-trans-stilbene) is a polyphenolic compound and abundantly found in plants and foods (16–19). A wide spectrum of pharmacological bio-activities (antioxidant, anti-inflammatory, anti-atherosclerotic and antitumor properties) of resveratrol have been reported (16,20). Resveratrol was first found as a potential antitumor agent (21). A variety of biological effects of resveratrol include anti-proliferation, inhibition of angiogenesis and anti-metastatic activities, as well as inhibition of tumorigenesis (22,23). Although the antitumor effects of resveratrol have been found to modulate various steps of carcinogenesis and development, its underlying mechanism of anticancer effect regarding autophagy and apoptosis is unclear in resistant human oral cancer cells. The purpose of this study was to develop an understanding of the effects of resveratrol on cisplatin-resistant human oral cancer CAR cells and to further determine its therapeutic value and anticancer activity in treating oral cancer cells in vitro.
Materials and methods
Chemicals and reagents
β-actin antibody, cisplatin, 3-methyladenine (3-MA), monodansylcadaverine (MDC), and thiazolyl blue tetrazolium bromide (MTT) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Acridine orange (AO), LysoTracker Red DND-99, and trypsin-EDTA were purchased from Thermo Fisher Scientific (Waltham, MA, USA). The primary antibodies against cytochrome c, Apaf-1, AIF, Endo G, Bcl-2, and Bax and the goat anti-rabbit or anti-mouse IgG-horseradish peroxidase (HRP) secondary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The other primary antibodies used in this study were from Cell Signaling Technology (Danvers, MA, USA). Compound c and Z-VAD-FMK were from Merck Millipore (Billerica, MA, USA).
Cell culture
CAR, a cisplatin-resistant cell line, was developed by exposing parental human tongue squamous cell carcinoma CAL 27 cell line (American Type Culture Collection, Manassas, VA, USA) to 10–80 µM of cisplatin, and CAR cells stably resistant to cisplatin as previously described (24,25). CAR cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 µg/ml streptomycin, and 2 mM L-glutamine (Thermo Fisher Scientific) in an incubator at 37°C with a humidified 5% CO2. Cisplatin-resistant CAR cells were continuously cultured in 80 µM cisplatin, unless otherwise indicated. Human normal gingival fibroblast (HGF) cells were a kind gift of Dr Tzong-Ming Shieh (Department of Dental Hygiene, China Medical University) and cultured in DMEM with 10% FBS, 100 U/ml penicillin and 100 µg/ml streptomycin as previously described (26).
Cell viability assay
CAR or HGF cells (1×104 cells/well) were plated in 96-well plates in 100 µl medium with or without 25, 50, 75, 100 and 200 µM of resveratrol for 48 h. CAR cells were individually pretreated with 10 mM 3-MA (a PI3K class III inhibitor), 10 µM compound c (an AMPK inhibitor) and 15 µM Z-VAD-FMK (a pan-caspase inhibitor) for 1 h, followed by treatment with or without 50 µM resveratrol for 48 h. After that, DMEM containing 500 µg/ml MTT was added, and medium was then aspirated from each well to detect viability as described elsewhere (27,28). Cell morphological examination of apoptotic characteristics and autophagic vacuoles was determined and photographed utilizing a phase-contrast microscope as previously described (29).
Observations for autophagic vacuoles and autophagy marker
CAR cells (5×104 cells/ml) were plated on sterile coverslips in tissue culture plates and treated with 50 µM resveratrol for 48 h. Cells were then individually stained with either 1 µg/ml AO, 1 µg/ml LysoTracker Red DND-99 or 100 µM MDC for 10 min as previously described (13,30). The autophagy marker LC3 was assayed using the Premo Autophagy Sensor LC3B-GFP (BacMam 2.0) kit (Thermo Fisher Scientific) as specified by the manufacturer. The occurrence of autophagic vacuoles, acidic vesicular organelles (AVOs) and LC3B-GFP were immediately monitored by fluorescence microscopy (Nikon, Melville, NY, USA).
TUNEL/DAPI staining
CAR cells (2×105 cells/well) into 12-well plates were incubated with 50 µM resveratrol. After exposure for 48 h, apoptotic DNA breaks were detected utilizing the In Situ Cell Death Detection kit, Fluorescein (Roche Diagnostics GmbH, Roche Applied Science, Mannheim, Germany) following the protocol provided by the manufacturer. Cells were counter-stained with 1 µg/ml 4′,6-diamidino-2-phe-nylindole (DAPI, Thermo Fisher Scientific) and photographed using a fluorescence microscope.
Western blotting
CAR cells (5×106 cells/T-75 flask) were exposed to resveratrol (0, 25, 50 and 100 µM) for indicated intervals of time. Whole-cell lysates were isolated, and quantification of the protein concentration was completed by Bradford assay (Bio-Rad Protein Assay, Bio-Rad Laboratories, Hercules, CA, USA) as previously described (31). Equal amounts of protein samples after boiling and the addition of denaturing sample buffer were separated using a 10% or 12% SDS-PAGE and further employed by immunoblotting as detailed by Chiang et al (29). All bands were normalized to the level of β-actin for each lane.
Quantitative RT-PCR
CAR cells (5×106 cells per T-75 flask) were treated with or without 50 µM resveratrol for 24 h. Cells were collected, and cell pellets were collected by centrifugation to extract total RNA by the Qiagen RNeasy Mini kit (Qiagen, Valencia, CA, USA). RNA purity was detected, and RNA sample was individually reverse-transcribed using the High Capacity cDNA Reverse Transcription kits (Applied Biosystems/Thermo Fisher Scientific). Quantitative PCR was assessed to amplify with 2X SYBR Green PCR Master Mix (Applied Biosystems/Thermo Fisher Scientific), and forward and reverse primers as listed in Table I were determined as previously reported (13,32). Applied Biosystems 7300 Real-time PCR System was employed at least in triplicate, and each value was performed in the comparative threshold cycles (CT) method to normalize to the housekeeping gene GAPDH.
Statistical analysis
All data are reported as the mean ± SEM of triplicate samples. The significant differences of data were subjected to Student's t-test for comparison of two groups as described in figure legends, and P-value <0.05 was considered to indicate a statistically significant difference.
Results
Resveratrol exhibits cytotoxicity and suppresses CAR cell viability
Resveratrol concentration-dependently reduced viable CAR cells (Fig. 1A). The 50% maximal inhibitory concentration (IC50) of resveratrol for a 24-, 48- and 72-h treatments in CAR cells were 95.23±3.26, 73.23±2.29 and 51.62±3.36 µM, respectively. Resveratrol was also found to induce the formation of autophagic vacuoles and apoptotic bodies in CAR cells after 50 µM resveratrol challenge for 48 h (Fig. 1B). This finding implies that autophagic cell death and apoptotic effects play pivotal roles in resveratrol-treated CAR cells. Importantly, no cytotoxic effect or morphological trait change on resveratrol-treated normal HGF cells were observed (Fig. 1C and D), suggesting that resveratrol has low toxicity in normal oral cells. The IC50 value of HGF cells is over 100 µM resveratrol. Therefore, resveratrol reduced viability in CAR cells instead of normal HGF cells possibly through autophagic and apoptotic mechanisms.
Resveratrol elicits autophagic and apoptotic death in CAR cells
To further test the autophagic cell death caused by resveratrol, the formation of autophagosome vesicles in CAR cells was monitored. Resveratrol triggered the clear occurrence of AVOs within the cytoplasm in comparison to control using AO staining (Fig. 2A). The autophagic evidence was also found after resveratrol-treated CAR cells were individually stained with LysoTracker Red DND-99 and MDC, which are widely used fluorescent probes to preferentially accumulate in autophagic vacuoles. After 50 µM resveratrol exposure for 48 h, autophagic vacuoles were observed, and the fluorescent intensity of MDC staining was directly proportional to 50 µM resveratrol (Fig. 2A). The LC3 distribution of resveratrol-treated cells was also explored. A more punctate pattern of LC3-GFP was found in cells after resveratrol treatment (Fig. 2A). In addition, the occurrence of DNA fragmentation and condensation by TUNEL/DAPI staining was detected prior to 50 µM resveratrol exposure for 48 h. Resveratrol markedly provoked apoptotic DNA breaks (Fig. 2B), and quantitative data of TUNEL positive cells was concentration-dependent (Fig. 2C). Therefore, resveratrol-induced cell death was triggered by autophagic and apoptotic machineries in CAR cells.
Resveratrol regulates the AMPK, Akt and autophagy-related protein levels in CAR cells
The protein signals of AMPK and Akt pathway were clarified in resveratrol-treated cells. Resveratrol at 25, 50 and 100 µM for 24 h increased the protein levels of phosphorylated AMPKα on Thr172 and AMPKα, but it decreased the phosphorylation of Akt on Ser473 and mTOR on Ser2448 (Fig. 3A). Our findings indicate that resveratrol induced autophagic cell death through modulation of AMPK and Akt signaling. The key protein levels of autophagy marker were also monitored. Resveratrol at 25, 50 and 100 µM for 48 h increased the protein levels of Atg5, Atg7, Atg12, Atg14, Atg16L1, Beclin-1, PI3K class III and LC3-II, but it decreased Rubicon protein level (Fig. 3B). These data imply that resveratrol induced autophagy through interfering with the PI3K class III/Beclin-1/Atgs-associated signals in CAR cells.
Resveratrol stimulates autophagy-modulated mRNA levels in CAR cells
The autophagy-related gene was explored prior to resveratrol exposure for 24 h. Resveratrol enhanced the gene expression of Atg5 (Fig. 4A), Atg12 (Fig. 4B), Beclin-1 (Fig. 4C) and LC3-II (Fig. 4D) in CAR cells. Thus, the key autophagic effect on gene levels resulted from resveratrol challenge in CAR cells.
3-MA and compound c are against the autophagic effects on resveratrol-treated CAR cells
Cells were pretreated with 10 mM 3-MA (an inhibitor of PI3K class III) or 10 µM compound c (an AMPK inhibitor) and then exposed to 50 µM resveratrol for 48 h. The protein level of LC3-II and cell viability were monitored. Our data demonstrated that 3-MA and compound c inhibited the formation of LC3-II (Fig. 5A and C) and enhanced cell viability after resveratrol treatment (Fig. 5B and D) in comparison to resveratrol-treated sample, suggesting that resveratrol-caused autophagy was mediated through regulating AMPK and PI3K class III signaling in CAR cells.
Resveratrol modulates caspase-3, caspase-9 and apoptotic-related protein levels in CAR cells
The levels of caspase-3, caspase-9 and apoptotic-related protein molecules were investigated. Resveratrol at 25, 50 and 100 µM for 48 h increased the protein levels of caspase-3/-9 (Fig. 6A), cytochrome c, Apaf-1, AIF, Endo G, Bax and Bad (Fig. 6B), while it decreased the protein level of Bcl-2 and phosphorylation of Bad on Ser136. Our findings reveal that resveratrol induced apoptotic CAR cell death through a mitochondria-dependent pathway.
Resveratrol elicits the caspase-3/-9 activities and protective effect of pan-caspase inhibitor (Z-VAD-FMK) against resveratrol-induced caspase-dependent apoptosis on CAR cells
To further confirm the activity of caspase-3/-9 in resveratrol-treated cells, the caspase activities were detected. Cells were pretreated with 15 µM Z-VAD-FMK and then exposed to 50 µM resveratrol for 48 h. The protein levels of cleaved caspase-9, cleaved caspase-3 and cell viability were checked. Our data showed that resveratrol at 25, 50 and 100 µM for 48 h increased the activities of caspase-9 (Fig. 7A) and caspase-3 (Fig. 7B). Furthermore, Z-VAD-FMK inhibited the formation of cleaved caspase-9 and cleaved caspase-3 (Fig. 7C), as well as enhanced the viability of resveratrol-treated cells (Fig. 7D). Our results suggest that the intrinsic mechanism was required for resveratrol-induced apoptosis in CAR cells.
Discussion
Resveratrol has been shown to have a variety of biological actions and many health benefits (18,19). It is well documented that resveratrol is an effective anti-oxidant with anti-neoplastic potential for cell cycle arrest and apoptosis (16,33–36). Furthermore, resveratrol has been shown to induce autophagic cell death in ovarian (33,34) and colon (35,36) cancer cells. Many studies showed that resveratrol-induced tumor cell death may be mediated through autophagy and apoptosis (34,37,38). Our study is first to demonstrate that resveratrol triggered cell autophagy and apoptosis in cisplatin-resistant human oral cancer CAR cells.
In the present study, our results revealed that the resveratrol induced the formation of autophagic vesicle, AVOs and LC3B occurrence (Fig. 2A) in CAR cells. LC3 is a hallmark of autophagic membrane to monitor early autophagosome formation (37). It is important to note that CAR cell morphology prior to resveratrol treatment showed strictly apoptotic action, and intensive accumulation of autophagic vacuoles was also detected. It started before pronounced hallmarks of apoptosis appeared, suggesting that the autophagic pathway switched to the apoptotic machinery (Fig. 2). Noteworthy, resveratrol was reported to induce both autophagy and apoptosis in human bladder and breast cancer cells (39,40). Moreover, resveratrol also induced autophagy followed by apoptosis in human colorectal DLD1 cancer cells (41). Thus, our study is in agreement with previous findings (39–41).
In addition, morphology of resveratrol-treated cells could be affected by both apoptosis and presumably defective autophagy. Resveratrol also simultaneously enhanced the protein level of autophagy-associated proteins (Atg complexes, Beclin-1, PI3K class III and LC3-II) (Fig. 3B), and mRNA expression of autophagic genes Atg5, Atg12, Beclin-1 and LC3-II (Fig. 4) in CAR cells. 3-MA (an inhibitor of PI3K class III) has been shown to potently inhibit autophagy-dependent protein degradation and to suppress the formation of autophagosomes (42). Importantly, our results showed that 3-MA inhibited the autophagic vesicle formation and LC3-II protein expression induced by resveratrol (Fig. 5A). Autophagy (programmed cell death type II) plays a double role in carcinogenesis, while apoptosis is well known as programmed cell death type I (11,12). Cells undergoing autophagic cell death display distinct morphological features (1,43). In contrast, autophagy is early degradation of organelles whereas preservation of nucleus until late stages.
Moreover, autophagy serves as a key mechanism of cell survival and is able to represent an adaptive mechanism to survive drug-induced cellular stress in tumor cells (11,12). It has been reported that autophagy precedes apoptosis in resveratrol-treated cervical cancer HeLa and Cx cells (44), suggesting that resveratrol might lead to apoptotic cell death through autophagy. Resveratrol induced autophagy in HT-29 and COLO 201 cells, and the magnitude of resveratrol-induced apoptosis and autophagy was different upon colon cancer cell lines (45). In the present study, CAR cells were used as an experimental model on cellular mechanisms. Therefore, we demonstrated that resveratrol may act as a modulator of autophagy in CAR cells.
Resveratrol and other polyphenols exhibit antioxidant properties (16–18). The formation of autophagic vacuoles in resveratrol-treated CAR cells was blocked by 3-MA to inhibit the sequestration step of macroautophagy, i.e. autophagosome formation (data not shown). An increased level of LC3-II was caused by resveratrol treatment (Fig. 3B) in CAR cells. This indicated that the conversion of LC3-I to LC3-II, typically observed after induction of autophagy, reflecting increased number of autophagosomes due to an elevated activity of autophagic process (37,41). The elevated amount of LC3-I in CAR cells after a 24-h treatment suggests that its processed LC3-II form is still required. Furthermore, accumulation of autophagic vacuoles in resveratrol-treated cells was observed by detecting LC3-II positive CAR cells (Fig. 2A).
In this study, our data demonstrated that resveratrol induced autophagic cell death through inhibition of Akt and enhancement of AMPK signaling in CAR cells (Fig. 3A). The abnormal activation of Akt/mTOR signaling cascade that stimulates tumor cell growth, proliferation, survival and resistance to drug-induced apoptosis (46). Currently, autophagy regulation is mediated through PI3K/Akt/mTOR signal pathway (11,12). AMPK and Akt serine/threonine kinase [protein kinase B (PKB)] are vital to be involved in the induction of cell autophagy and apoptosis (47). Regulation of the Akt signaling is essential for developing therapeutic inhibitors for tumor cells (13). Zhao et al (48) showed that Akt inhibitor (MK-2206) induces autophagy in human nasopharyngeal cancer NPC cells in vitro. It is also reported that the polyphenolic compound plumbagin triggers autophagic and apoptotic cell death through suppressing Akt pathway in human non-small cell lung cancer cells (49). AMPK activation promotes autophagy by directly activating Ulk1 through direct phosphorylation of Ser317 and Ser777 because a key energy sensor regulates cellular metabolism to maintain energy homeostasis (50,51).
We suggest herein that AMPK activation by resveratrol directly phosphorylates Ulk1 on Ser317 to mediate cell autophagy, which also contributed to CAR cell apoptosis. As previously mentioned, multiple antitumor agents, including vincristine, taxol and doxorubicin, activate AMPK to promote cancer cell death (52). Din et al (53) showed that aspirin elicits autophagy cell death through activation of AMPK and inhibition of mTOR activation in colorectal cancer cells in vivo and in vitro. Furthermore, Liu et al (54) also demonstrated that AMPK activator AA005 induces mTORC1 inhibition and autophagy cell death in colon cancer cells. These results are similar to our findings presented here. Hence, resveratrol exerts the effects of autophagy and apoptosis by altering the activation of Akt and AMPK signaling.
Apoptosis is associated with the disruption of mitochondrial membrane potential (ΔΨm), and the mitochondrion integrate cell death signals and autophagy activation (6,55). Apoptosis and autophagic cell death are interconnected through mitochondrial permeability transition (16), and the activation of caspase-3 and increased release of cytochrome c contribute to resveratrol-triggered apoptosis. Our data indicated a rapid collapse of ΔΨm after resveratrol challenge at an early stage (data not shown). The loss of mitochondrial membrane polarization seems to be an intracellular signal that may be responsible for the induction of autophagy and apoptosis (10). Resveratrol led to phosphatidylserine externalization as well as disruption of the mitochondrial membrane potential, activation of caspase-9/-3 and DNA degradation (Figs. 2B, 6A and 7). Z-VAD-FMK (a pan-caspase inhibitor) has been shown to potently inhibit apoptosis-dependent caspase activity (32). Resveratrol-induced cell death was effectively inhibited by Z-VAD-FMK, indicating the involvement of the mitochondrial apoptotic pathway (Fig. 8). These findings provide evidence for induction of apoptosis in CAR cells after resveratrol challenge.
In summary, our findings suggest that treatment of CAR cells with resveratrol caused reduction rather than enhancement of autophagic degradation, which in turn leads to cell death. Autophagy is initiated in resveratrol-treated CAR cells, but removal of autophagic vacuoles is probably inefficient. Further studies are required to fully explain this phenomenon. Our studies provide new insight into the molecular mechanism of action of resveratrol and proposes its potential therapeutic use. Understanding the underlying pathways affected by resveratrol in CAR cells may help to identify new targets for effective cancer therapy or predict potential adverse drug effects.