Background
Gliomas are the most common, highly proliferative and invasive primary intracranial tumors [
1]. Currently, temozolomide (TMZ) is the first-line chemotherapeutic drug for glioma treatment with a definite curative effect. However, the efficacy of TMZ is frequently limited by the durability of the chemoresistance response. Therefore, research that facilitates the development of innovative drugs is urgently needed.
Celastrol is a predominantly active natural product extracted from the root bark of TCM
Tripterygium wilfordii and it has various biological properties, such as anti-tumor, immunosuppression and weight loss activities [
2‐
5]. In 2007, celastrol, along with artemisinin, triptolide, capsaicin, and curcumin, was reported to be a molecule that was most likely to be developed into a modern drug [
6]. Previous research has shown that celastrol exhibits potential cytotoxicity in multiple tumor cells. Xu et al. reported that celastrol could inhibit the growth of ovarian cancer cells by inducing apoptosis via increased intracellular ROS accumulation in vitro and in vivo [
7]. In non-small-cell lung cancer, celastrol inhibited cell proliferation and induced apoptosis through the degradation of the cancerous inhibitor of protein phosphatase 2A [
8]. As a potent low-molecular-weight inhibitor, celastrol inhibited the proliferation of AML cells in vitro and prolonged the survival of mice in an in vivo model of AML [
9]. Studies have shown that celastrol can inhibit the growth of glioma cells, although the detailed mechanism remains to be investigated [
10,
11]. In addition, celastrol has shown neuroprotective effects in various disease models (such as Parkinson’s Disease, Alzheimer’s Disease, and Amyotrophic Lateral Sclerosis), which means that celastrol can cross the blood–brain barrier [
12,
13], which may be an advantage of celastrol in the treatment of intracranial tumors.
Aberrant changes in the cell cycle commonly occur in tumor cells, and many cytotoxic agents act on cell cycle checkpoints [
14]. The G2/M check point arrest is an effective mechanism adopted by many cytotoxic agents. The cyclinB1/cdc2 complex, which plays a key role in controlling the progression of the cell cycle by regulating the phosphorylation status of various proteins, is regulated by a series of proteins, including p21, Cdc25C, and Chk2 [
15‐
17].
Studies have found that apoptosis and autophagy are two main pathways for death of tumor cells. Apoptosis is a common pattern of cell death observed with chemotherapies against all types of cancers [
18]. Apoptosis is usually accompanied by typical morphological changes, including cell membrane blebbing, cell shrinkage, nuclear condensation and fragmentation, and apoptotic body formation. Autophagy, which is also known as autophagic cell death, is an evolutionarily conserved intracellular self-digestive process that maintains cellular homeostasis via lysosome-dependent machinery [
19]. Beth Levine et al. demonstrated that autophagy played an extremely important role in tumor suppression [
20]. Moreover, autophagy is widely recognized as a mechanism for tumor cell survival by enhancing stress tolerance and providing an alternative pathway for cancer cells to provide substantial nutrient and energy requirements [
21]. Recent studies have demonstrated that a large number of antitumor drugs known to induce apoptosis also activated autophagy [
22]. Therefore, further research is needed to be focused on the possible mechanism underlying celastrol-induced apoptosis or autophagy in glioma cells and determine the role of these processes and their relationship.
Reactive oxygen species (ROS) are the main molecules produced under conditions of oxidative stress, and they have long been considered to be important factors in tumorigenesis and tumor development and recurrence [
23]. ROS include oxygen anions, superoxide (O
2−), hydroxyl radicals and peroxides such as hydrogen peroxide (H
2O
2). In glioma cells, treatment with H
2O
2 simultaneously activated autophagy and apoptosis, which induced the membrane potential and the release of cytochrome c [
24]. The generation of O
2− caused mitochondrial damage, selective degradation of mitochondria via autophagosomes and cell death of malignant glioma cells [
25]. ROS can activate various signaling pathways, such as members of the MAPK family including p38, JNK and ERK1/2 [
26,
27]. Activation of the JNK and p38 MAPK signaling pathways may be related to apoptosis and multiple pathophysiological processes during stress [
23]. As a classic signaling pathway, the AKT/mTOR pathway has also been reported to mediate antitumor drug-induced apoptosis and autophagy [
28].
In the present study, we aimed to investigate the antitumor effects and possible mechanisms underlying the impact of celastrol on glioma cells both in vitro and in vivo. We elucidated that celastrol induced G2/M-phase arrest, apoptosis, and autophagy in glioma cells by modulating the ROS/JNK and Akt/mTOR signaling pathways. In addition, we found that autophagy caused by celastrol played a role in promoting cell survival. Celastrol-induced apoptosis and autophagy inhibited each other.
Methods
Cells and cell culture
The human glioma cell lines U251 and U87 and rat glioma cell line C6 were purchased from the Cell Resource Center (IBMS, CAMS/PUMC, Beijing, China). The U251 cells were cultured in MEM (Corning, NY, USA), and the U87 cells were cultured in MEM-NEAA supplemented with 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. The C6 cells were cultured in F10 with 15% horse serum (HyClone, Logan, UT, USA) containing 2.5% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. The cells were maintained at 37 °C in a humidified incubator with 5% CO2. Astrocytes were prepared from primary cell cultures of neocortical tissues from one-day-old SD rats. Briefly, following the removal of the meninges, the cortices were mechanically dissociated and digested with 0.25 mg/mL trypsin, triturated in DMEM by pipetting, and filtered to remove large debris. The isolated cells were suspended in DMEM containing 10% FBS and plated on flasks coated with Poly-D-lysine. After one week, mixed glial cells were shaken for 16 h at 240 rpm. Detached cells, which consisted of microglia and oligodendrocytes, were removed. Attached cells were maintained as astrocyte-enriched cultures. After 3–5 days of additional culture, the cells were plated into appropriate plates for the experiments.
Reagents and antibodies
Celastrol (purity of > 99%) was purchased from Pharmacodia Co., Ltd. (Beijing, China). N-Acetyl-L-cysteine (NAC), 3-methyladenine(3-MA), z-VAD-fmk (z-VAD), SP600125 (SP), MK2206 (MK) and rapamycin (Ra) were purchased from Selleckchem (Houston, TX, USA). Chloroquine diphosphate salt (CQ) and TMZ were purchased from Sigma (St. Louis, MO, USA). Antibodies against caspase-3 (cat. no. 9664), SQSTM1/p62 (cat. no. 5114), cyclin B1 (cat. no. 4138) and phospho-cdc2 (cat. no. 4539) were obtained from Cell Signaling Technology (Beverly, MA, USA). Antibodies against caspase-8 (cat. no. ab 25,901), caspase-9 (cat. no. ab32539), cleaved PARP (cat. no. ab32064), Beclin-1(cat. no. ab207612), Akt (cat. no. ab32505), phospho-Akt (cat. no. ab192623), JNK (cat. no. ab179461), phospho-JNK (cat. no. ab124956), mTOR (cat. no. ab2732), phospho-mTOR (cat. no. ab109268), p38 (cat. no. ab170099), phospho-p38 (cat. no. ab195049), Chk2 (cat. no. ab47433), phospho-Chk2 (cat. no. ab59408), Cdc25C (cat. no. ab226958), phospho-Cdc25C (cat. no. ab47322) and p21 (cat. no. ab109199) were purchased from Abcam (Cambridge, MA, UK). Antibody against LC3B (cat. no. 6008–1-Ig) was purchased from Sigma. Antibody against β-actin (cat. no. 6008–1-Ig) was purchased from Proteintech (Chicago, IL, USA). The plasmid ptfLC3 was purchased from Addgene Repository (
http://www.addgene.org).
Cell viability assay
Inhibition of glioma cell proliferation by celastrol was detected by a CCK8 assay (Dojindo, Kumamoto, Japan). Briefly, the cells were passaged in 96-well plates with a density of 2.5 × 103 cells/well. After 24 h, the cells were treated with gradient concentrations of celastrol (0.3, 1, 3 and 10 μM) for various durations (12, 24, 36 and 48 h). After incubation, the cells were treated with a mixture of CCK8 and MEM and incubated for 0.5–1 h. The absorbance at 450 nm was determined by a microplate reader (Molecular Devices, Sunnyvale, CA, USA).
Cells were cultured in 6-well plates with a density of 100–400 cells/well. After 24 h, the cells were treated with various concentrations of celastrol (0.03, 0.1, 0.3 and 1 μM) for approximately 10 days or treated with celastrol for 24 h and left untreated for approximately 10 days to allow for the generation of colonies. Then, the medium was discarded, and the cells were washed with PBS 3 times. After being fixed with 4% paraformaldehyde, the colonies were stained with 0.1% crystal violet for 15 mins. The clones with more than 50 cells were counted. The images were captured by a digital camera.
Cell cycle analysis by flow cytometry
The role of celastrol in cell cycle distribution was monitored by flow cytometry with PI/RNase staining buffer (BD Biosciences, San Jose, CA, USA). Briefly, cells were passaged in 6-well plates at a density of 3 × 105 cells/well. After 24 h, the cells were exposed to various concentrations of celastrol (0, 0.3, 1, 3, and 10 μM) for 24 h. Then, the cells were harvested, fixed with 75% ethanol at -20 °C overnight, and stained with PI/RNase staining buffer for 15 mins. The cell cycle analyses were performed with a NovoCyte instrument (ACEA Biosciences, San Diego, CA, USA.), and the data were analyzed by using NovoExpress software (ACEA).
Morphological changes due to apoptosis
Characteristic morphological changes associated with apoptosis were assessed by fluorescence microscopy using Hoechst 33342 staining. Briefly, cells were cultured at a density of 3 × 104/ml in 96-well plates and then treated with celastrol for 24 h. Then, the cells were fixed with 4% paraformaldehyde for 15 mins and stained with Hoechst 33342 solution for 30 mins at room temperature in the dark. Nuclear fragmentation and chromatin condensation were observed with a fluorescence microscope (Olympus, Tokyo, Japan) after the cells were washed 3 times with PBS.
Flow cytometric analysis of apoptosis
Apoptosis was assessed by a FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, San Diego, CA, USA) according to the provided protocol. The samples were analyzed by a flow cytometer (LSRFortessa SORP, BD, San Jose, CA, USA).
Measurement of mitochondrial membrane potential
Mitochondrial membrane potential (MMP) was measured with a MMP Assay Kit with JC-1 (Beyotime, Jiangsu, China) according to the manufacturer’s instructions. Briefly, cells were seeded in six-well plates at a density of 2.5 × 105/ml and then treated with celastrol (0–10 μM) for 24 h. The cells were then harvested and resuspended in 500 μl of MEM. Then, 500 μl of JC-1 staining solution was added, and the cells were incubated for 20 mins at 37 °C in a CO2 incubator. The results were analyzed by flow cytometry, and mitochondrial depolarization was evaluated by measuring the decrease in the red/green fluorescence intensity ratio.
mRFP-EGFP-LC3 puncta assay
U251 cells were transiently transfected with ptfLC3 (mRFP-EGFP-LC3) plasmid to examine the formation of fluorescent puncta of autophagosomes. Cells were cultured in 24-well plates and transfected with 0.8 μg/well mRFP-EGFP-LC3 plasmid using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After transfection, the cells were treated with or without 1.5 μM celastrol for 24 h and then incubated with DAPI for 15 mins. Image acquisition was performed using a Leica confocal laser scanning microscope.
Measurement of intracellular ROS generation
Intracellular ROS production was detected using the ROS Assay Kit (Beyotime). Cells were plated in six-well plates at a density of 2.5 × 105/ml and treated with celastrol in the absence or presence of NAC (a ROS scavenger) and SP (a JNK inhibitor). The cells were then incubated with 10 μM DCFH-DA at 37 °C for 30 mins. Then, then cells were washed with serum-free MEM, and the ROS levels were determined by fluorescence microscopy (Leica, Wetzlar, Germany) and flow cytometry (BD Biosciences, San Jose, CA, USA).
Western blotting analysis
The glioma cells or tissues were lysed with cool RIPA lysis buffer. Proteins were extracted following the standard protocol and the concentration was measured using a BCA protein assay kit. Equal amounts of protein (30 μg) were separated by 10–15% SDS-PAGE and electrotransferred to PVDF membranes. The membranes were blocked with 5% skim milk for 1 h and then incubated overnight at 4 °C with the following antibodies: β-action (1:50000), LC3B (1:10000), SQSTM1/p62 (1:3000), (Beclin1 (1:4000), Akt (1:10000), phospho-Akt (p-Akt, 1:5000), p38 (1:5000), phospho-p38 (p-p38, 1:1500), JNK (1:3000), phospho-JNK (p-JNK, 1:3000), mTOR (1:5000), phospho-mTOR (p-mTOR, 1:30000), caspase-8 (1:3000), caspase-9 (1:3000), cleaved PARP (1:3000), cleaved caspase-3 (1:3000), Chk2 (1:1000), phospho-Chk2 (p-Chk2,1:500), Cdc25C (1:1000), phospho-Cdc25C (p-Cdc25C, 1:1000), cyclin B1 (1:1500), phospho-cdc2 (p-cdc2, 1:2000), p21 (1:1500). Membranes were then incubated with HRP-conjugated goat anti-mouse or anti-rabbit IgG antibodies (1:5000) for 1 h at 37 °C, detected by an enhanced chemiluminescent detection system and quantified using ImageJ.
Human glioma xenograft experiment
All animal procedures were performed according to the guidelines of the Animal Experiments and Experimental Animal Welfare Committee of Capital Medical University (Approval number: AEEI-2017-119). Healthy male BALB/c-nu mice (6-8 weeks old, 18-20 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were kept in standard cages in a room with a controlled environment in terms of temperature (25 ± 2 °C), humidity (40-50%) and light (12 h light/dark cycle), with commercial standard solid rodent chow and water provided ad libitum in the Experimental Animal Center of Capital Medical University. Approximately 3.5 × 105 U251 cells (> 95% viability) in a volume of 5 μl were stereotactically injected into the right striatum (2 mm lateral, 1 mm anterior to the bregma, and 3 mm deep) by using a small animal stereotactic frame (RWD Life Science, Shenzhen, China).
TUNEL assay
The apoptotic response of tumor tissues was identified using a TUNEL assay with an In Situ Cell Death Detection Kit, Fluorescein (Roche Diagnostics, Mannheim, Germany), according to the manufacturer’s instructions.
Histopathology and immunohistochemistry
Formalin-fixed tissue samples were embedded in paraffin, and 4-μm sections were cut. Primary tumor, heart, liver, spleen, lung, and kidney sections were stained with hematoxylin and eosin (H&E) for routine histological examinations and morphometric analysis. Brain slices were immunostained with Ki67 (1:400), phospho-JNK (1:100), and cleaved caspase-3 (1:100). Images were captured using a Leica microscope.
Statistical analysis
Quantitative data were presented as the mean ± SD from at least three independent experiments. Fisher’s exact test, one-way ANOVA and the Mann-Whitney U-test were used to compare differences between groups. All data were analyzed using SPSS version 21.0 software (IBM Corporation, Chicago, USA). Statistical significance was indicated as P < 0.05.
Discussion
Glioma is a lethal human malignant tumor with the characteristics of high incidence, high recurrence rate, high mortality and limited treatment process [
31]. Celastrol has attracted widespread concern, especially for its potent anti-tumor activity, including glioma [
32‐
34]. Previous studies have shown that celastrol inhibited the growth of rat glioma cells and human glioma cells via the suppression of VEGFR expression and induction of apoptosis and cell cycle arrest [
11,
35]. However, whether autophagy is involved in the inhibitory effect of celastrol on glioma cells is unclear and the mechanism of apoptosis has not been clarified. In the present study, our results further demonstrated that celastrol could inhibit the proliferation of glioma cells and cause G2/M phase arrest and apoptosis. More importantly, we found that celastrol simultaneously trigged apoptosis and autophagy by activating ROS/JNK signaling and suppressing the AKT/mTOR signaling pathway.
Recent studies have indicated that the heterogeneity of tumors is a major factor in determining cancer progression and the drug resistance response [
36‐
38]. Tumor heterogeneity occurs not only between different individuals, but also between the same individuals because of spatial and temporal heterogeneity [
39,
40]. Our results showed that celastrol exhibited different cytotoxicity in U251, U87-MG and C6 cells. U251 and C6 cells treated with celastrol for 24 h had a similar IC
50 whereas those treated with celastrol for 48 h had significantly different IC
50 values. In addition, U251 and U87 malignant glioma cell lines are derived from different individuals who suffered from a pleomorphic glioma. C6 glioma cell line is derived from N-nitrosomethylurea-induced rat glioma. Studies have shown that the U87 and U251 cell lines show differences in protein expression profiles [
41]. These differences may also account for the different biological phenotypes such as proliferation, migration and invasion of glioma cell lines.
The G2/M cell cycle checkpoint plays a key role in normal cell proliferation. When DNA is damaged, Chk2 is firstly phosphorylated, and finally activation of Chk2 is achieved by autophosphorylation and transphosphorylation. Activation of Chk2 results in the phosphorylation of Cdc25, which leads to the inhibition of Cdc25. Cdc25 normally activates the cdc2/cyclin B1complex, which is a specific regulator of the G2/M phase [
15,
42,
43]. Previous studies have reported celastrol induced cell cycle arrest at the G2/M phase in C6 cells [
35]. According to the flow cytometry analysis, celastrol enhanced the proportion of glioma cells in the G2/M phase in all three glioma cell lines. However, in the U87-MG cells, celastrol also induced S phase arrest in the cell cycle. Similar conclusions have not been reported before. A further western blot analysis showed that celastrol upregulated the expression of the G2/M-phase-related proteins Chk2, p-Chk2, p-cdc2 and p-Cdc25C and downregulated the expression of Cdc25C. Surprisingly, celastrol increased the expression of cyclin B1 which was consistent with previous research [
34]. Moreover, we found that celastrol also increased the protein level of p21, which plays a crucial role in blocking the activation of Cdk1/cyclin B1 in a p53-dependent or p53-independent manner [
44]. Whether p53 participates in the inhibition of glioma cell proliferation induced by celastrol needs to be further tested.
Recent studies have extensively shown that two main apoptotic pathways participate in the regulation of apoptosis: extrinsic pathway and intrinsic pathway [
45]. In the extrinsic apoptotic pathway, caspase-8 is first activated to cause subsequent downstream cascade reactions. The other pathway is the mitochondria-mediated intrinsic pathway, in which caspase-9 and additional caspase molecules such as caspase-3, -6 and -7 are activated [
46]. Mitochondrial dysfunction has been shown to participate in the induction of apoptosis [
47]. The present study found that a sharp decrease in MMP occurred after celastrol treatment. In normal cells, each caspase is present in an inactive state until it is cleaved after apoptotic signaling events [
48]. Further western blotting analysis showed that treatment of celastrol led to the activation of caspase-3, caspase-8, caspase-9 and cleaved-PARP. Immunohistochemical and western blotting analyses confirmed that celastrol enhanced cleaved caspase-3 levels in vivo, and the TUNEL assay demonstrated a distinct increase in apoptosis in tumor tissues following celastrol treatment. In summary, celastrol can induce cell apoptosis in both the extrinsic and intrinsic pathways in glioma cells.
As a modulator of pathogenesis, autophagy has been widely studied as a promising, novel therapeutic target in diverse diseases [
49]. In the present study, we found that celastrol triggered autophagy and elevated the expression of Beclin-1 and LC3B. Interestingly, the level of p62 also increased, which is related to the degradation of autophagy. When we used CQ (an autophagy inhibitor) to limit the degradation of the autophagosome, the expression of LC3B did not significantly change compared with that of the celastrol treatment alone. Moreover, the higher LC3B levels with celastrol + CQ than with CQ alone may indicate that celastrol increases the synthesis of autophagy-related membranes and partially blocks autophagic flux [
50]. Meanwhile, the confocal results indicated that the number of yellow autophagic vesicles increased. Therefore, celastrol may partly inhibit the function of lysosomes. In addition, p62 has a critical role in oxidative stress response pathways [
51]. The increased expression of p62 may also be associated with the generation of ROS. We further revealed that the autophagy inhibitors 3-MA and CQ moderately reinforced the inhibitory effect of celastrol on cell viability, indicating that celastrol-triggered autophagy may promote survival, which is consistent with the basic function of autophagy.
Autophagy and apoptosis can collectively trigger cell death through synergy, complementary cooperation, or alternative mechanisms [
52]. In our study, suppression of autophagy following 3-MA and CQ treatment enhanced the level of apoptosis induced by celastrol. Moreover, when apoptosis was blocked by z-VAD, the expression of autophagy-related proteins was moderately increased. Therefore, the relationship of apoptosis and autophagy triggered by celastrol in glioma cells may be interdependent and interactive, which needs to be further confirmed. The pro-survival effect of autophagy caused by celastrol also indicated that autophagy and apoptosis caused by celastrol were antagonistic.
In glioma cells, anticancer agents generally induce an increase in ROS levels triggering the cell death [
53]. However, tumor cells can survive under high levels of ROS by the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway, a transcription gene regulator [
54]. Moreover, ROS-dependent ERK activation has been shown to contribute to the invasion/migration of U87 glioma cells [
55]. In the present study, celastrol induced a significant increase in ROS generation. The ROS inhibitor NAC markedly repressed the proliferation inhibition, cell cycle arrest, apoptosis and autophagy triggered by celastrol. All these results indicated that celastrol induced the generation of ROS, which contributed to cell death.
Usually, ROS is the upstream signal molecule of JNK, which can phosphorylate JNK and induce sustained JNK activation [
56‐
58]. In addition, the activation of JNK promotes ROS production. A positive feedback effect or antagonistic effect may occur between ROS and JNK [
34,
59]. Our study showed that NAC could slightly attenuate the phosphorylation of JNK while ROS generation was generally increased after pretreatment with SP in the cells treated with celastrol. An antagonistic effect may exist between ROS and JNK. Celastrol also upregulated the expression level of p38, another member of the MAPK family. Similar conclusions have been reported before [
60]. The AKT/mTOR signaling pathway is one of the main growth regulatory pathways in both normal and cancer cells, and it can negatively regulate autophagy [
61,
62]. Our study showed that celastrol inhibited Akt and mTOR phosphorylation and deactivated them. MK could slightly attenuate the phosphorylation of mTOR, while obvious changes in Akt phosphorylation were not observed after pretreatment with Ra in the cells treated with celastrol. Further, the present study suggested that the inhibition of ROS/JNK significantly inhibited autophagy and apoptosis, while the inhibition of Akt/mTOR had the opposite effect. Taken together, the above results suggested that celastrol triggered autophagy and apoptosis, which may be mediated through the activation of ROS/JNK signaling and inhibition of the Akt/mTOR signaling pathway.
Zhou and Huang et al. demonstrated that celastrol inhibited the growth of glioma in vitro and in vivo using murine ectopic xenografts models of glioma [
10,
11]. Our in vivo data demonstrated that after celastrol treatment at doses of 2 and 4 mg/kg, significant antitumor effects were observed in the murine orthotopic transplantation model of glioma. Furthermore, the western blotting and immunohistochemical analyses showed significant increases in the level of cleaved caspase-3, LC3B and JNK phosphorylation but decreases in the level of Ki67, Akt and mTOR phosphorylation.