Introduction
Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) and protein kinase B (AKT) are involved in major pro-survival signaling pathways in cancer cells. While excessive proliferation and survival of aberrant cells is suppressed in normal homeostasis, in tumor cells, this is overridden by the upregulation of antiapoptotic proteins and other cell death-associated factors. In particular, AKT regulates cell survival, metabolism, invasion and metastasis, and is also pivotal role in antiapoptotic and antiproliferative signaling in response to essential cytokines. AKT-dependent cell survival is mediated through the induction of antiapoptotic gene expression (1), including members of the forkhead and B-cell lymphoma (BCL)-2 family, as well as the nuclear factor (NF)-κB. It is known that the BCL-2 family of proteins regulates apoptosis through modulating mitochondrial permeability (2). In addition, AKT-dependent degradation of the inhibitor of NF-κB leads to NF-κB translocation from the cytoplasm to the nucleus where it promotes cell survival by regulating the expression levels of target genes. AKT can also inhibit p53, which is a tumor suppressor protein that induces apoptosis, senescence and cell cycle arrest (3).
Consequently, drugs designed to target AKT have become an important and viable approach in the development of cancer therapeutics with a number of AKT inhibitors already being used in clinical trials. Additionally, it has become increasingly apparent that inhibiting AKT via the inhibition of upstream signals can be equally as significant as direct targeting of this kinase. More specifically, tumor necrosis factor receptor-associated factor (TRAF)6, which is an E3 ubiquitin ligase, has been shown to ubiquitylate and activate AKT (4), and it also serves an essential role in regulating the NF-κB and mitogen-activated protein kinase signaling cascades. TRAF6 consists of a conserved C-terminal domain that regulates interactions with upstream signaling proteins, and an N-terminal domain that comprises the core of the ubiquitin ligase catalytic domain (5). In the presence of E2 ubiquitin-conjugating enzymes, TRAF6 catalyzes the transfer of polyubiquitin chains to the substrates (6). Notably, TRAF6 mediates K63-directed ubiquitylation of AKT, and this type of modification is associated with functional changes in the target substrate in contrast to classical K48-directed ubiquitylation, which leads to substrate degradation.
Additionally, in the context of insulin-like growth factor-1 stimulation, overexpression of TRAF6 enhances AKT phosphorylation and activation. By contrast, AKT phosphorylation at Thr-308 and Ser-473 was inhibited in TRAF6-deficient mouse embryonic fibroblasts (4). TRAF6 is recruited by the phosphorylated discoidina neuropilin-like membrane protein discoidin, CUB and LCCL domain-containing protein 2, leading to increased E3 ubiquitin ligase activity and subsequent activation of AKT, and accelerating epidermal growth factor receptor-driven tumorigenesis (7). As a result of the close association between TRAF6 and AKT, the present study speculated that AKT can be indirectly targeted by developing small molecules that can bind to TRAF6 and disrupt its interaction with AKT, and that this may be a promising approach in cancer therapeutics. To determine this, the present study performed a computational docking study to screen small molecules, which can bind to TRAF6 in regions critical for its interaction with AKT (residues 50–159, RING and zinc finger domains). The present study found that quinine, a member of well-known Cinchona alkaloids, has a high affinity for TRAF6 and may indeed induce apoptosis, as well as reduce cell viability. The underlying mechanism by which quinine inhibited AKT activation was also examined using a range of ex vivo cellular assays.
Materials and methods
Materials
A549 human lung adenocarcinoma and HeLa cervical cancer cell lines were kindly provided by Dr. Yaozhou Zhang (Tianjin International Joint Academy of Biomedicine, Tianjin, China). RPMI-1640 medium and fetal bovine serum (FBS) were obtained from Tianjin HOPE Biotechnology Co., Ltd. (Tianjin, China). Rabbit polyclonal antibodies against BCL-2 (cat. no. 2876), BCL-2-associated X protein (BAX; cat. no. 2772), total AKT (cat. no. 4691), phosphorylated (p-)AKT (Thr308; cat. no. 4056s) and p-AKT (Ser473; cat. no. 4051s) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Rabbit polyclonal antibody against β-actin (cat. no. PR0255) was purchased from Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd. (Beijing, China). Horseradish peroxidase (HRP)-conjugated goat poly-clonal antibody against rabbit and mouse (cat. no. M21003) was obtained from Shanghai Abmart Biotechnology Co., Ltd. (Shanghai, China). Radioimmunoprecipitation assay (RIPA) buffer and protease inhibitor cocktail were purchased from Beijing Kangwei Biotechnology Co., Ltd. (Beijing, China). Phosphatase inhibitor cocktail and lipopolysaccharide (LPS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Quinine was purchased from Tianjin Xiensi Biochemical Technology Co., Ltd. (Tianjin, China). Cis-platinum was purchased from Jiangsu Hansoh Pharmaceutical Co., Ltd. (Tianjin, China). Protein A sepharose beads were purchased from Beijing Kangwei Biotechnology Co., Ltd. (Pierce, Rockford, IL, USA). All chemicals and solvents were of analytical grade.
Virtual screening
A three-dimensional (3D) structure library containing 1,792 compounds was established for the virtual screen. Structures of these compounds were generated using a two-dimensional (2D)/3D editor-sketcher and were minimized to a local energy minimum using the CHARMm-like force field implemented within with Catalyst 4.11 software (8). The 3D structure of TRAF6 (residues 50–159, RING and zinc finger domains) was retrieved from the Protein Data Bank as the docking receptor (http://www.rcsb.org/pdb; accession no. 3HCT). Both the TRAF6 crystal structure and candidate ligands were prepared using AutoDock Tools v.1.5.2 software (http://autodock.scripps.edu).
For virtual screening, flexible ligands and rigid receptor docking calculations were performed with the AutoDock4.10 software package using the following parameters: The Lamarckian genetic algorithm; a population size of 300; energy evaluations of 25,000,000; search runs of 100. The docking area was defined as a dimension of 60×60×60 points with grid spacing of 0.375 Å. The grid box was centered on the binding site of the ubiquitin-conjugating E2 enzyme, Ubc13. The results were ranked, according to the binding free energy. The conformations from the docking experiments were analyzed using Chimera software (9), which also identified the hydrophobic interactions between the receptor and the ligands.
Cell culture
A549 and HeLa cells were cultured in RPMI-1640 medium, supplemented with 10% FBS at 37°C in a humidified atmosphere of 5% CO2.
Cell viability assay
The effect of quinine on HeLa and A549 cell viability and proliferation was determined using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Briefly, 200 µl HeLa or A549 cell suspensions were seeded into 96-well plates at densities of 4 and 3×103 cells/well, respectively. The cells were allowed to adhere overnight and were subsequently treated with LPS (20 µg/ml) in the absence or presence of quinine at various concentrations (1, 5, 10, 25, 50, 100, 250 and 500 µM). Cis-platinum (16.7 µM) was used as a positive control. Cell viability was assessed by adding 20 µl of 5 mg/ml MTT to each well and incubating for 4–6 h. The medium was replaced with 150 µl dimethyl sulfoxide, which was stirred to dissolve the formazan crystals. The absorbance (A) was measured at 490 nm using an ST-360 microplate reader (Shanghai Kehua Bio-engineering Co., Ltd., Shanghai, China). The survival rate of each group of cells was calculated relative to the vehicle-treated control cells (designated as 100%). Cell inhibition ratio was calculated using the following formula: Cell inhibition ration = (Acontrol − Atreated / Acontrol) × 100.
Detection of apoptosis by flow cytometry
Apoptosis was assessed by flow cytometry using an Apoptosis Detection kit (Becton-Dickinson, Franklin Lakes, NJ, USA), according to the manufacturer's protocol. Briefly, HeLa cells were seeded into a 6 cm plate at a density of 6×105 cells/well and allowed to attach overnight. The medium was replaced with fresh medium containing various concentrations of quinine (100, 150 or 200 µM) for 48 h at 37°C. The cells were collected and resuspended in 1X binding buffer at a concentration of 1×106 cells/ml, and 100 µl of the cell suspension was transferred to a 5 ml culture tube, to which 5 µl annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI) were added (from the Apoptosis Detection kit). The mixture was incubated in the dark for 15 min at room temperature (25°C). The apoptotic index was immediately determined with a FACSCalibur instrument (BD Biosciences, Franklin Lakes, NJ, USA). Cell Quest Pro software v5.1 (BD Biosciences) was used to analyze flow cytometry data.
Western blot analysis
Western blotting was performed according to a standard protocol. Briefly, HeLa and A549 cells at 60–70% confluence were incubated with 150 µM quinine. The cells were rinsed with ice-cold phosphate-buffered saline and resuspended in RIPA buffer, supplemented with phenylmethylsulfonyl fluoride and phosphatase inhibitors. Following an incubation on ice for 30 min, the lysates were centrifuged at 12,000 × g for 10 min and the supernatant was collected. Protein concentrations were determined using a bicinchoninic acid protein assay kit (CWBIO, Beijing, China) and 40 µg protein was resolved by 3% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The proteins were transferred onto a polyvinylidene difluoride membrane (EMD Millipore, Bedford, MA, USA), which was blocked with a solution of 5% non-fat dry milk or 5% bovine serum albumin in Tris-buffered saline (10 mM Tris-HCl and 150 mM NaCl) with 0.05% Tween-20. After 2 h, the membranes were washed and incubated with primary antibodies against β-actin (1:1,500), AKT (1:1,000), p-AKT (1:1,000), BCL-2 (1:1,000), BAX (1:1,000), and TRAF6 (1:1,000) overnight at 4°C, followed by washing and incubation with HRP-conjugated goat anti-rabbit or anti-mouse immunoglobulin G (1:10,000) for 1 h at 25°C. Antibody binding was detected using an enhanced chemiluminescence kit (CWBIO), according to the manufacturer's protocol. Image J software v.1.48u (National Institutes of Health, Bethesda, MD, USA) was used to quantify the intensity of protein bands.
Statistical analysis
The data are presented as the mean ± standard deviation of multiple independent experiments. One-way analysis of variance was used to compare group means, followed by Dunnett's t-test, using StataSE12 software (Stata Corp., College Station, TX, USA). P<0.05 was considered to indicate a statistically significant difference.
Results
Selection of quinine using computational virtual screening
The 1,792 compounds were docked into the binding pocket of TRAF6 using AutoDock 4.10 software, and the top 50 compounds were selected with the lowest estimated inhibition constant. Subsequently, binding conformations for each of these selected compounds were exported and evaluated manually, according to the estimated binding free energy and hydrophobic interactions. Quinine (ΔG = −6.8 kcal/mol) was selected based on these calculations and purchased for further experimental testing.
Fig. 1 demonstrated the binding mode of quinine: The oxygen atom of the hydroxyl group of quinine forms H-bonds to the OD1 atom of residue Asp-57; the quinoline ring is in π–π stacking interaction with Phe-60 and also in good hydrophobic interactions with the surrounding residues, Phe-60, Pro-63 and Leu-64.
Quinine suppresses the growth of A549 and HeLa cells
As a result of the molecular docking results, which suggested quinine binds to TRAF6 and affects its function, the present study therefore used HeLa and A549 cell lines that express high levels of endogenous TRAF6 (5,10) in the following experiments. As shown in Fig. 2A, quinine inhibited HeLa cell proliferation in a concentration- and time-dependent manner, with concentrations >5 µM markedly reducing cell viability. An increase in the concentration between 10 and 500 µM increased the inhibition rate from 12.80 to 82.53%, when treated for 72 h. Similarly, prolonged incubation with quinine for 96 h increased the rate of inhibition from 17.67 to 89.73% when the dose was increased between 10 and 500 µM. These effects were also reproduced in the A549 cells (Fig. 2B).
Quinine induces apoptosis in HeLa cells
The present study next investigated whether the decreased cell viability following quinine treatment was due to the occurrence of apoptosis. To confirm this hypothesis, annexin V-FITC and PI staining were used to discriminate between cells undergoing early apoptosis (annexin V+/PI−) and necrosis (PI+). The percentage of apoptotic cells increased in a dose-dependent manner upon quinine treatment (Fig. 3). The fraction of viable cells decreased from 91.28 down to 42.80% following treatment with 200 µM quinine. This was simultaneously accompanied with an increase (5.64 to 49.83%) in the fraction of early apoptotic cells in the population. Treatment with the positive control, Cis-platinum, reduced the number of viable cells to 64.99% and was accompanied by a concomitant increase in the fraction of early apoptotic cells to 29.94%. By contrast, the percentage of PI+ cells increased from 0.92 to 6.48% upon treatment with 200 µM quinine. These results indicated that quinine induces cell apoptosis, rather than necrosis.
Quinine-induced apoptosis activates BAX in HeLa and A549 cells
The above results showed that quinine application resulted in the inhibition of HeLa and A549 cell growth and proliferation, and induced early apoptosis. Since the BCL-2 family serves a critical role in the execution of programmed cell death, the present study assessed whether the apoptosis induced by quinine was associated with changes in the levels of either BCL-2 or BAX (Fig. 4). Indeed, the quinine treatment led to an increase in the levels of the pro-apoptotic factor, BAX, in a time-dependent manner, as well as a decrease in the expression of BCL-2. These results suggested that quinine exposure is responsible for inducing apoptosis, which underlies the reduced viability of HeLa and A549 cells.
Quinine treatment regulates the phosphorylation of AKT
To elucidate the underlying molecular mechanism in which quinine induces apoptosis, the phosphorylation status of AKT was analyzed in these two cell lines following LPS treatment. The present study decided to incorporate LPS here, since it is a robust inducer of AKT activity in HeLa and A549 cells (11,12). Stimulation of HeLa and A549 cells with LPS for 3 h resulted in an increase in p-AKT levels, and this effect was inhibited by quinine treatment (Fig. 5). In addition, HeLa and A549 cell proliferation was enhanced by treatment with LPS, which was abrogated in the presence of quinine (Fig. 6). These results indicated that quinine induced apoptosis via the activation of AKT.
Discussion
Cancer has been the top health threat for over a century and is second only to cardiovascular diseases in mortality. AKT is activated in >50% of human cancer cell lines. The AKT signaling pathway has been linked to a variety of cellular processes, including proliferation and migration, and it also serves a role in response to anticancer therapeutics (13,14). An important step in AKT activation is its phosphorylation, and this activation is subsequently enhanced by TRAF6-dependent ubiquitination. Yang et al (4) demonstrated that the formation of a stable complex between AKT and TRAF6 is responsible for the enhanced activation. Previous evidence suggested that TRAF6 is associated with tumors, since TRAF6 depletion in osteosarcoma cells was found to reduce NF-κB expression and the survival rate for cancer patients overexpressing TRAF6 is relatively low (5,10–17). Therefore, targeting TRAF6-mediated AKT activation may constitute an excellent therapeutic strategy for the treatment of cancer.
The present study used computational methods to select small molecules, which can bind with TRAF6, and pursued ex vivo assays to evaluate their antitumor activities and to investigate possible mechanisms of action. Firstly, the computational docking data revealed that quinine may be an excellent ligand for TRAF6, thereby representing a potential candidate for modulating TRAF6-mediated AKT activation. Subsequently, the antitumor activity of quinine was assessed and it was revealed that this compound inhibited the growth of HeLa and A549 cells in a dose-dependent manner. Secondly, it is well-known that apoptosis is closely associated with the initiation, progression and metastasis of cancer, as well as the elimination of malignant or infected cells (18–21). Apoptosis provides a vital homeostatic mechanism for maintaining an appropriate number of cells in the body through a balance of cell division and cell death (22). Consequently, the present study evaluated the cytotoxic and apoptotic effects of quinine through appropriately selected cell lines and found that quinine caused a significant loss of cell viability in HeLa and A549 cells in a concentration- and time-dependent manner. Under these experimental conditions, the loss of cell viability following quinine treatment was predominantly due to the induction of apoptosis.
Next, to further investigate the underlying mechanism of quinine-induced apoptosis, western blotting was used to analyze the expression levels of BCL-2 and BAX in HeLa and A549 cells. Quinine was demonstrated to suppress the expression of BCL-2, whilst stimulating that of BAX in a time-dependent manner. These results supported that quinine induces apoptosis in HeLa and A549 cells via modulation of these key apoptosis-regulating proteins. It has been previously suggested that mitochondria serve a key role in cell apoptosis, since numerous BCL-2 family proteins interact with the mitochondrial outer membrane (23,24). The mutual antagonism that exists between subsets of BCL-2 family members, including BCL-2 and BAX, directly impacts the mitochondrial membrane permeability, which ultimately determines whether cytochrome c is released in order to activate caspases and induce apoptosis (25).
AKT can also directly or indirectly affect mitochondria-dependent apoptosis via the modulation of BCL-2 family proteins. Previous evidence demonstrates that phosphorylation of AKT can inactivate BAX (26,27). Others have found that activated AKT can phosphorylate BCL-2-associated death domain (BAD) at Ser136, and the association of p-BAD with 14-3-3 proteins can inhibit apoptosis by preventing p-BAD from interacting with the antiapoptotic BCL-2 family members, including BCL-2 or BCL-extra large (28). These collective observations along with the present results indicated that activated AKT can attenuate mitochondria-dependent apoptosis by inactivating BAD, BAX and other forkhead transcription factors, and that quinine can counter this process by inhibiting the activation of AKT.
Additionally, while serving a critical role in the regulation of cell survival and apoptosis in a variety of cell types (29–31), AKT activation is accomplished via the phosphorylation at Ser473 in the C-terminal hydrophobic region by phosphoinositide-dependent kinase (PDK)-2 and at Thr308 in the catalytic domain by PDK-1 (32). Through computational virtual docking, the present study quickly identified appropriate ligands, including quinine, which can bind effectively with TRAF6, and subsequently, detected the effect of quinine on the activity of AKT. These experimental results demonstrated that quinine can inhibit the phosphorylation of AKT at both Thr-308 and Ser-473, thereby suggesting that quinine-induced apoptosis is likely a result of inhibition of AKT phosphorylation.
In conclusion, the present study demonstrated that quinine can reduce cell viability and induce apoptosis in HeLa and A549 cells. These novel activities were associated with the inhibition of AKT activation by inhibiting its phosphorylation at Thr-308 and Ser-473, as well as AKT-induced suppression of BCL-2 and upregulation of BAX (Fig. 7). Further investigations are currently underway to determine whether quinine is a viable anticancer agent and whether TRAF6 can serve as a therapeutic target for developing novel anticancer drugs.