Bardoxolone methyl (CDDO-Me or RTA402) induces cell cycle arrest, apoptosis and autophagy via PI3K/Akt/mTOR and p38 MAPK/Erk1/2 signaling pathways in K562 cells

Abstract

Chronic myeloid leukemia (CML) treatment remains a challenge due to drug resistance and severe side effect, rendering the need on the development of novel therapeutics. CDDO-Me (Bardoxolone methyl), a potent Nrf2 activator and NF-κB inhibitor, is a promising candidate for cancer treatment including leukemia. However, the underlying mechanism for CDDO-Me in CML treatment is unclear. This study aimed to evaluate the molecular interactome of CDDO-Me in K562 cells using the quantitative proteomics approach stable-isotope labeling by amino acids in cell culture (SILAC) and explore the underlying mechanisms using cell-based functional assays. A total of 1,555 proteins responded to CDDO-Me exposure, including FANCI, SRPK2, XPO5, HP1BP3, NELFCD, Na⁺,K⁺-ATPase 1, etc. in K562 cells. A total of 246 signaling pathways and 25 networks regulating cell survival and death, cellular function and maintenance, energy production, protein synthesis, response to oxidative stress, and nucleic acid metabolism were involved. Our verification experiments confirmed that CDDO-Me down-regulated Na⁺,K⁺-ATPase α1 in K562 cells, and significantly arrested cells in G₂/M and S phases, accompanied by remarkable alterations in the expression of key cell cycle regulators. CDDO-Me caused mitochondria-, death receptor-dependent and ER stress-mediated apoptosis in K562 cells, also induced autophagy with the suppression of PI3K/Akt/mTOR signaling pathway. p38 MAPK/Erk1/2 signaling pathways contributed to both apoptosis- and autophagy-inducing effects of CDDO-Me in K562 cells. Taken together, these data demonstrate that CDDO-Me is a potential anti-cancer agent that targets cell cycle, apoptosis, and autophagy in the treatment of CML.

Introduction

Chronic myeloid leukemia (CML) is one of a group of diseases called chronic myeloproliferative disorders that also include chronic granulocytic leukemia, myelofibrosis/osteomyelosclerosis, polycythemia vera, and idiopathic thrombocythemia [1,2]. These diseases have overlapping clinical and molecular features, characterized by the unrestrained expansion of pluripotent hematopoietic stem cells. With a low incidence of 0.6 to 2 cases per 100,000 adults, CML accounts for ~15% of all newly diagnosed cases of leukemia in adults [3,4].The estimated number of new cases of CML is 8950 (representing 0.5% of all new cancer cases and 13.8% of all new leukemia cases)and there would be 1080 deaths due to CML in the United States (US) in 2017 (https://seer.cancer.gov/statfacts/html/cmyl.html). More than 95% of CML patients have a distinctive and characteristic cytogenetic abnormality, the Philadelphia chromosome (Ph⁺) arising from the translocation t(9;22)(q34;q11.2) which involves the ABL1 gene in chromosome 9 and the BCR gene in chromosome 22, resulting in a fused BCR-ABL gene encoding the constitutively active BCR-ABL of p210 or sometimes p185 that is necessary and sufficient for initiating CML [5-8]. The BCR-ABL transcript is continuously active with no dependence on other cellular signaling proteins. In turn, BCR-ABL activates a cascade of critical proteins controlling the cell cycle and accelerates cell division and proliferation. BCR-ABL also inhibits DNA repair, resulting in genomic instability and making the cell more susceptible to developing further genetic abnormalities [5-7]. With more understanding of the nature of BCR-ABL as the pathologic basis of CML and its action as an overactive tyrosine kinase, targeted biological therapies that specifically inhibit the activity of BCR-ABL have been developed in the past 20 years [9-12]. These tyrosine kinase inhibitors (TKIs) can induce complete remissions in CML and change the clinical course of CML. The first of these TKIs was imatinib mesylate (trade names: Gleevec and Glivec), which was approved by the US Food and Drug Administration (FDA) in 2001, and has been considered the standard of care for more than a decade. Imatinib inhibited the progression of 65-75% of CML patients, but approximately 20-30% patients developed resistance and/or intolerance to imatinib [13]. To overcome drug resistance and to increase clinical response, second generation TKIs targeting BCR-ABL and other oncogenic tyrosine kinases have been developed. The first, dasatinib, a more potent inhibitor of BCR-ABL, was approved in 2007 by the US FDA to treat CML patients who were either resistant to or intolerant of imatinib. Nilotinib and dasatinib were then approved by the FDA for first-line therapy of Ph⁺ CML in 2010. Both dasatinib and nilotinib are highly effective in newly diagnosed CML patients as well as those who fail imatinib. In 2012, radotinib was approved in South Korea only for use in CML patients resistant to or intolerant of imatinib. Another second generation TKI, bosutinib, received FDA approval in 2012 for the treatment of adult patients with Ph⁺ CML with resistance, or intolerance to prior therapy [14]. Second generation TKIs have been demonstrated to induce better and faster clinical responses compared to imatinib and are highly effective in patients resistant to and/or intolerant to imatinib and are extremely active against all the resistant BCR-ABL1 mutations, with the exception of T3151 [14]. However, no survival advantage has been seen in CML patients [11,13]. Ponatinib is a third generation TKI, which causes response in both early and advanced phases of CML and those bearing any resistant mutations, specifically T315I [15]. The successful implementation of above TKIs for the treatment of CML remains a flagship for molecularly targeted therapy in cancer. However, some patients still did not respond to these TKIs due to primary or secondary resistance to such therapy and some patients developed severe adverse effects [12,16]. Although mutations in the BCR-ABL gene have proven to be the most prominent mechanism of resistance to TKIs, other mechanisms dependent on BCR-ABL activity or supporting oncogenic properties of the leukemic cells independent of BCR-ABL signaling have been documented [17]. Clearly, there is a strong need to develop more efficacious and safer drugs for CML therapy when all TKI fail for the treatment.

Oleanolic acid is naturally occurring triterpenoids that have been used in traditional medicine for centuries, showing antioxidant, antibacterial, antifungal, anticancer, and antiinflammatory activities [18]. To further improve their pharmacological efficacy, a series of novel derivatives have been synthesized, such as 2-cyano-3,12-dioxooleana-1,9(11)-dien-28-oic acid (CDDO), CDDO-imidazolide (CDDO-Im), the methyl amide of CDDO (CDDO-Ma), and CDDO methyl ester (CDDO-Me, also named as bardoxolone methyl, RTA402, TP-155 and NSC713200) (Figure 1A) [19]. These synthetic triterpenoids are potent inhibitors of the de novo synthesis of inflammatory enzymes such as inducible nitric oxide synthase (iNOS) and inducible cyclooxygenase 2 (COX-2) [20]. CDDO-Me is a promising candidate for prevention and treatment of cancer, which protects cells from oxidative stress at nanomolar concentrations, whereas exhibits cytotoxicity against various cancer cells at micromolar concentrations [21,22]. CDDO-Me is more potent than CDDO in anticancer and cancer-preventive activities and in the activation of Kelch-like erythroid cell-derived protein with CNC homology-associated protein 1/nuclear factor (erythroid-derived 2)-like 2/antioxidant response element (Keap1/Nrf2/ARE) pathway [23,24], which is involved in cytoprotection in the presence of excessive electrophiles or oxidative stress. Binding of CDDO-Me to Keap1 disrupts its critical cysteine residues, leading to the release of Nrf2, which hinders its ubiquitination and finally leads to stabilization and nuclear translocation of NF-κB. In the nucleus, Nrf2 activates the transcription of phase 2 response genes, leading to a coordinated antioxidant and anti-inflammatory response [24]. As a potent Nrf2 activator and NF-κB inhibitor, the therapeutic effects of CDDO-Me has been tested in Phase III for chronic kidney disease [25]. The antitumor effect of CDDO-Me has been demonstrated in different cancers by inhibition of proliferation and induction of apoptosis [26]. Moreover, preclinical studies have shown that CDDO-Me induced tumor regression in xenografted-mouse models [27-29]. It was evaluated in a few Phase I clinical trials for advanced solid tumor or lymphoid malignancy and showed good tolerance [30,31]. Notably, Samudio et al. [32] reported that CDDO-Me induced cytotoxicity in imatinib-resistant CML cells. However, the underlying mechanisms of the anticancer effects of CDDO-Me in the treatment of CML are not fully understood.

Figure 1.

Cytotoxicity of CDDO-Me on K562 cells and the regulating effect on Na⁺,K⁺-ATPase α1 expression. A. The chemical structure of CDDO-Me. B. Viability of K562 cells as examined by the MTT assay. C. Mass spectrum of Na⁺,K⁺-ATPase α1 quantified and identified by the SILAC-based proteomics assay. D. Representative blots and bar graphs showing the level of Na⁺,K⁺-ATPase α1, when cells were treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h or 0.5 μM for 72 h, then the protein samples were subject to Western blot assay. Data are expression as mean ± SD of three independent experiments. ***P<0.005 by one-way ANOVA.

Mass spectrometry-based proteomics is increasingly applied in a quantitative formatto investigate changes in protein abundances in biological samples, often based on labeling of samples with stable isotopes that are introduced chemically or metabolically. Stable-isotope labeling by amino acids in cell culture (SILAC) is a powerful and increasingly popular approach for quantitative proteomics studies [33-36]. In the SILAC method, two cell populations are cultured in the presence of heavy or light amino acids (typically lysine and/or arginine), one of them is subject to a perturbation (e.g. drug exposure), and then both are combined, processed, and analyzed. Incorporation of the “heavy” amino acid occurs through cell growth, protein synthesis, and turnover. SILAC allows “light” and “heavy” proteomes to be distinguished by mass spectrometry while avoiding any chemical derivatization and associated purification. SILAC can be applied to systemically assess global protein profile, evaluate the target network of drugs, estimate drug toxicity, and find new biomarkers for the diagnosis and treatment of cancers [35,37,38]. In this study, we evaluated the SILAC-based proteomic response of human CML K562 cells to CDDO-Me exposure and examined its effects on cell proliferation, cell cycle distribution, apoptosis, and autophagy in K562 cells.

Materials and methods

Chemicals and reagents

CDDO-Me (purity >98%) and JC-1 mitochondrial membrane potential assay kit were obtained from Cayman Chemical Inc. (Ann Arbor, MI, USA). MK-2206 was purchased from Selleckchem Inc. (Houston, TX, USA). SB202190, Alexa Fluor 488-conjugated secondary antibodies, 6-diamidino-2-phenylindole (DAPI) and Dulbecco’s modified Eagle’s medium (DMEM)/F12 (1:1) were bought from Invitrogen Inc. (Carlsbad, CA, USA). ¹³C₆ ¹⁵N₄-L-arginine, L-arginine, ¹³C₆-L-lysine, L-lysine, DMEM/F12 for the SILAC study, 4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES), ethylenediaminetetraacetic acid (EDTA), ribonuclease (RNase A), propidium iodide (PI), dimethyl sulfoxide (DMSO), fetal bovine serum (FBS), dialyzed FBS, Dulbecco’s phosphate buffered saline (PBS), and 2-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were sourced from Sigma-Aldrich Inc. (St Louis, MO, USA). The annexin V: phycoerythrin (PE) apoptosis detection kit was obtained from BD Biosciences Inc. (San Jose, CA, USA). The Cyto-ID^® Autophagy detection kit was bought from Enzo Life Sciences Inc.(Farmingdale, NY, USA). The caspase-3 colorimetric assay kit was purchased from R&D Systems Inc. (Minneapolis, MN, USA). The FASP^TM protein digestion kit was obtained from Protein Discovery Inc. (Knoxville, TN, USA). The Pierce^TM bicinchoninic acid (BCA) protein assay kit, radioimmunoprecipitation assay buffer (RIPA), skim milk and Western blotting substrate were obtained from Thermo Fisher Scientific Inc. (Hudson, NH, USA). The polyvinylidene difluoride (PVDF) membrane was purchased from Bio-Rad Inc. (Hercules, CA, USA). U0126 and primary antibody against human β-actin were obtained from Santa Cruz Biotechnology Inc. (Dallas, TX, USA). The rest of antibodies for signaling proteins related to cell cycle, apoptosis, and autophagy were all sourced from Cell Signaling Technology Inc. (Beverly, MA, USA).

Cell line and cell culture

The human chronic myeloid leukemia K562 cell line was obtained from American Type Culture Collection (Manassas, VA, USA) and cells were cultured in DMEM/F12 medium supplemented with 10% heat-inactivated FBS at 37°C in a 5% CO₂/95% air humidified incubator. CDDO-Me was dissolved in DMSO as stock solution of 50 mM. The stock solution was freshly diluted with culture medium at a final concentration of 0.05% DMSO (v/v). The control cells were treated with 0.05% DMSO only.

Cell viability

The effect of CDDO-Me on cell viability of K562 cells was examined using the MTT assay. Briefly, K562 cells were seeded in 96-well culture plates at a density of 9 × 10³ cells/well overnight, then treated with CDDO-Me at concentrations ranging from 0.05 to 10 μM for 24 or 48 h. Ten mL of MTT solution (5 mg/mL) was added into each well for another 4 h incubation. Then the solution was aspirated and 100 mL of DMSO was added into each well. After shaking for 10 min, the absorbance of the plate was measured at wavelengths of 560 nm (MTT formazan) and 670 nm (background) using a Synergy™ H4 Hybrid microplate reader (BioTek, Winooski, VT, USA). The half maximal inhibitory concentration (IC₅₀) value was calculated using the relative viability over CDDO-Me concentration curve by GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, CA, USA).

Quantitative proteomics

A SILAC-based approach was used to identify the molecular targets of CDDO-Me in K562 cells as previously described [39,40]. In brief, K562 cells were cultured in DMEM/F12 medium (for SILAC) with (heavy) or without (light) stable isotope labeled amino acids (¹³C₆ L-lysine and ¹³C₆ ¹⁵N₄ L-arginine) and 10% dialyzed FBS. After treatment with 0.5 μM CDDO-Me for 24 h, cellular proteins were collected for the subsequent digestion and desalting. Five mL of the peptide mixtures in 0.1% formic acid were subject to hybrid linear ion trap-Orbitrap (LTQ Orbitrap XL, Thermo Scientific Inc., Hudson, NH, USA) for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. The peptide SILAC ratio was calculated using MaxQuant version 1.2.0.13 (Max Planck Institute of Biochemistry, Munich, Germany). And the proteins were identified using Scaffold 4.3.2. The pathway was analyzed using ingenuity pathway analysis (IPA, www.ingenuity.com) from QIAGEN Inc. (Redwood City, CA, USA).

Determination of cell cycle distribution

The effect of CDDO-Me on the cell cycle distribution of K562 cells was determined by flow cytometry as previously described [41]. After treatment with CDDO-Me at concentrations of 0.25, 0.5, and 1 μM for 24 h, K562 cells were harvested, washed by PBS and fixed in 70% ethanol at -20°C overnight. Then, the cells were resuspended in 1 mL PBS containing 1 mg/mL RNase A and 50 µg/mL PI in dark at room temperature for 30 min. A total number of 1 × 10⁴ cells were subject to cell cycle analysis using a flow cytometer with CellQuest^TM software (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA).

Determination of cellular apoptosis

The effect of CDDO-Me on the apoptosis of K562 cells was evaluated using the annexin V:PE apoptosis detection kit as previously described [21,41]. In short, the cells were collected after CDDO-Me treatment at different concentrations over 24 h, or evaluated for different time intervals, and resuspended and incubated in 100 μL 1 × binding buffer containing 5 mL annexin V:PE and 5 μL 7-amino-actinomycin D (7-AAD) in the dark at room temperature for 15 min. The number of apoptotic cells was analyzed by flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) within 1 h.

Determination of caspase 3 activity

Caspase 3 activity was determined using the caspase 3 colorimetric assay kit following the manufacturer’s instructions. Briefly, K562 cells were harvested and lysed on ice for 1 h. Cell lysates were placed in 96-well plates and then 100 mL reaction buffer (containing DTT) was added. The plates were incubated at 37°C for 1 h and the caspase activity was determined using a Synergy^TM H4 Hybrid microplatereader (BioTek Inc.) at 380 nm (excitation wavelength) and 440 nm (emission wavelength).

Immunofluorescence

For the immunofluorescence assay, cells were fixed with fresh 4% formaldehyde in PBS for 10 min at room temperature, and subsequently penetrated with 0.25% Triton X-100 for 5 min and blocked with 5% BSA for 30 min. The samples were incubated with primary antibodies (1:500 dilution) at 4°C overnight. Then the cells were incubated for 1 h with the Alexa Fluor 488 goat anti-rabbit secondary antibodies (1:500 dilution) conjugated to FITC and stained with DAPI. Finally, the specimens were analyzed with a TCS SP2 laser scanning confocal microscope (Leica, Wetzlar, Germany).

Mitochondrial membrane potential (Δψ_m) changes in apoptosis

The mitochondrial membrane potential changes in apoptosis of K562 cells induced by CDDO-Me treatment was assayed using JC-1 following the manufacturer’s instructions. JC-1 exists either as a green fluorescent monomer at depolarized membrane potential with low Δψ_m or as a red fluorescent J-aggregate at hyperpolarized membrane potential with high Δψ_m. Briefly, K562 cells were cultured in 6-well plate. After CDDO-Me treatment, cells were loaded with culture medium containing 10 µmol/L JC-1 for 20 min at 37°C. The fluorescence was analyzed using a TCS SP2 laser scanning confocal microscope. Healthy cells with mainly JC-1 J-aggregates can be detected with fluorescence settings designed to detect rhodamine (excitation/emission = 540/570 nm). Apoptotic cells with mainly JC-1 monomers can be detected with settings designed to detect FITC (excitation/emission = 488/535 nm).

Determination of cellular autophagy

The effect of CDDO-Me on the autophagy of K562 cells was detected by flow cytometry as previously described [21,41]. In brief, cells were collected after CDDO-Me treatment at different conditions and resuspended in 250 mL of assay buffer containing 5% FBS. Following with the addition of 250 mL of the diluted Cyto-ID^® Green stain solution, cells were incubated at room temperature in the dark for 20 min, then cells were collected and washed with 1 × assay buffer. The percentage of autophagy cells was analyzed using the green (FL1) channel of a flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA).

Western blotting assay

The expression levels of targeted proteins were determined using Western blotting assay. Protein samples were collected in the RIPA buffer (50 mmol HEPES at pH 7.5, 150 mmol NaCl, 10% glycerol, 1.5 mmol MgCl₂, 1% Triton-X 100, 1 mmol EDTA at pH 8.0, 10 mmol sodium pyrophosphate, 10 mmol sodium fluoride) containing the protease inhibitor and phosphatase inhibitor cocktails, and centrifuged at 3,000 × g for 10 min at 4°C. Nuclear and cytoplasmic protein were separated using NE-PER^® Nuclear and Cytoplasmic Extraction Reagents as previously described [21,41]. Protein concentrations were determined using the BCA assay and 20 µg samples were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer and electrophoresed on 7-12% SDS-PAGE mini-gel after thermal denaturation at 95°C for 5 min. Proteins were transferred onto PVDF membrane at 400 mA for 2 h at 4°C. Membranes were probed with indicated primary antibody overnight at 4°C and then blotted with respective secondary anti-mouse or anti-rabbit antibody. Visualization was performed using Bio-Rad ChemiDoc™ XRS system with an enhanced chemiluminescence kit. The blots were analyzed using Image Lab 3.0 (Bio-Rad) and the protein level was normalized to the matching densitometric value of β-actin or histone H3.

Statistical analysis

Data are presented as the mean ± standard deviation (SD). The comparisons of multiple groups were tested by one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison procedure. P<0.05 was considered to be statistically significant. The experiments were performed at least three times independently.

Results

Proteomic response to CDDO-Me treatment in K562 cells

SILAC-based proteomics assay was firstly performed to examine proteomic responses to CDDO-Me treatment in K562 cells. A total number of 1,555 protein molecules was identified in response to CDDO-Me treatment, of which 657 proteins expression level were up-regulated and 898 ones were down-regulated (Table S1). These proteins include FANCI, SRPK2, XPO5, HP1BP3, NELFCD, TH1L, HMGA1, ZC3HC1, PCK2, SOD1, GSR, etc. Interestingly, we observed a reduction of Na⁺,K⁺-ATPase α1 expression.

Subsequently, the identified proteins were subject to the IPA analysis. The results showed that 246 signaling pathways (Table S2), and 25 networks of signaling pathways and cellular functions (Table S3 and Table 1) responded to CDDO-Me in K562 cells. The signaling pathways involved included G₁ and G₂ checkpoint regulation pathways, mTOR signaling pathway, PI3K/Akt signaling pathway, Erk/MAPK signaling pathway, Nrf2-mediated oxidative stress response pathway, unfolded protein response (UPR) pathway, mitochondrial dysfunction signaling pathway, and apoptosis signaling pathway. The networks involved have important roles in pathophysiological functions and the development of cancer, diabetes, Alzheimer’s disease, and chronic inflammatory diseases (Table 1). In aggregate, the IPA results have demonstrated that CDDO-Me modulates various molecular proteins and signaling pathways, including cell cycle, response to oxidative stress, apoptosis, and autophagy, eventually, leading to cell proliferation inhibition and deathin K562 cells. To validate the proteomic results, we next investigated the effects of CDDO-Me on cell cycle distribution, apoptosis, and autophagy and the role of key signaling pathways.

Table 1.

Top 5 network functions regulated by CDDO-Me in K562 cells

ID	Associated network functions	Score
1	RNA post-transcriptional modification, protein synthesis, & cancer	55
2	Protein synthesis, gene expression, & RNA post-transcriptional modification	49
3	RNA post-transcriptional modification, cell morphology, & cellular assembly and organization	44
4	Connective Tissue disorders, developmental disorder, & hereditary disorder	44
5	Gene expression, protein synthesis, & amino acid metabolism	43

CDDO-Me inhibits the proliferation of K562 cells

We evaluated the effect of CDDO-Me on the viability of K562 cells using the MTT assay. The cell viability was markedly decreased when exposed to CDDO-Me at concentrations from 0.05 to 10 μM (Figure 1B). The IC₅₀ values were 2.15 and 1.58 μM for 24 and 48 h incubation with CDDO-Me, respectively. The results show that CDDO-Me significantly inhibits the proliferation of K562 cells.

CDDO-Me suppresses Na⁺,K⁺-ATPase α1 expression in K562 cells

Compelling evidence shows that Na⁺,K⁺-ATPase has a role in cancer development and is a potential target for cancer therapy [42]. As our proteomic data revealed the reduction of Na⁺,K⁺-ATPase α1 expression in response to CDDO-Me treatment (Figure 1C), we verified this effect using Western blotting assay. Consistently, CDDO-Me concentration- and time-dependently decreased Na⁺,K⁺-ATPase α1 expression level in K562 cells (P<0.005, Figure 1D). These results suggest that CDDO-Me may target Na⁺,K⁺-ATPase α1 to exhibit the cancer cell killing effect.

CDDO-Me induces K562 cell cycle arrest

Since the IPA pathway analysis results showed that CDDO-Me had effects on G₂/M and G₁/S checkpoint regulation (Table S2 and Figure S1), we examined the cell cycle distribution of K562 cells when treated with CDDO-Me at different concentrations. The data showed that CDDO-Me hampered cell cycle progression by arresting cells at G₂/M and S phases (Figure 2A and 2B). When treated with CDDO-Me at 1 μM for 24 h, the percentage of K562 cells arrested at G₂ and S phases ascended from 4.7% to 18.7% (P<0.01) and 59.7% to 69.3% (P<0.05), respectively, with concomitant decrease in G₁ phase from 35.3% to 11.3% (P<0.001). In addition, CDDO-Me treatment markedly increased the number of sub-G₁ cells in a dose-dependent manner, which reflected the proportion of apoptotic cells in K562 cells.

Figure 2.

CDDO-Me induces cell cycle arrest in G₂/M and S phases and modulates key cell cycle regulators in K562 cells. The cells were incubated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. (A) The flow cytometric histograms showing the cell cycle arresting effect of CDDO-Me in K562 cells. (B) Bar graphs showing a concentration-dependent cell cycle arresting effect of CDDO-Me in K562 cells. (C) Representative blots and (D) bar graphs showing the expression of cyclin B1, CDK1/CDC2, CDK2, p27 Kip1, p21 Waf1/Cip1, p-p53 (Ser15), p53, p-H2A and β-actin. Data are expression as mean ± SD of three independent experiments. *P<0.05, **P<0.01, ***P<0.005, and ****P<0.001 by one-way ANOVA.

Next, we further tested the effect of CDDO-Me on the expression levels of several key regulators in cell cycle checkpoints. As shown in Figure 2C and 2D, compared with the control cells, there was a significant decrease in the expression of cyclin B1, CDK1/CDC2 and CDK2 when treated with 1 μM CDDO-Me for 24 h. In contrast, the level of p27 Kip1 was elevated 2.1- and 4.2-fold when cells were treated with 0.5 and 1 μM CDDO-Me for 24 h, respectively (P<0.01 or 0.005, Figure 2C and 2D), and p21 Waf1/Cip1 level was up-regulated 6.7-fold when cells were exposed to 1 μM CDDO-Me (P<0.005, Figure 2C and 2D). Moreover, there was a concentration-dependent increase in p-p53 (Ser15) and p53 levels, resulted in a 1.7-, 2.3-, and 2.6-fold increase in the ratio of p-p53 (Ser15)/p53 (P<0.05, Figure 2C and 2D) with 0.25, 0.5, and 1 μM CDDO-Me for 24 h, respectively. Meanwhile, we detected level of p-H₂A, a DNA damage marker associated with G₂/M phase, was elevated with treatment of CDDO-Me in K562 cells (Figure 2). In aggregate, the results indicate that CDDO-Me alters the cell cycle distribution and induces G₂ and S phase arrest with DNA damage, contributing to its anticancer effect in K562 cells.

CDDO-Me induces apoptosis of K562 cells via both extrinsic and intrinsic pathways

After observation of a clear increase in the number of sub-G₁ cells, we next examined the effect of CDDO-Me on apoptosis of K562 cells. CDDO-Me concentration- and time-dependently induced apoptosis of K562 cells (Figure 3). Incubation of cells with 0.25, 0.5, and 1 μM CDDO-Me for 24 h, the percentage of apoptotic cells (early plus late apoptosis) up-regulated from 5.3% to 6.3%, 10.5%, and 49.1%, respectively (P<0.005, Figure 3A and 3C). When cells were exposed to 0.5 μM CDDO-Me for 6, 12, 24, 48, and 72 h, the percentage of apoptotic cells elevated from 4.3% to 9.1%, 11.5%, 12.9%, 21.9% and 32.5%, respectively (P<0.01 or 0.005, Figure 3B and 3C).

Figure 3.

CDDO-Me induces apoptosis in K562 cells. A. Representative flow cytometric plots of apoptotic K562 cells when treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. B. Representative flow cytometric plots of apoptotic K562 cells when treated with CDDO-Me at 0.5 μM over 72 h. C. Bar graphs showing the percentage of apoptotic cells when treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h, or 0.5 μM over 72 h. Data are expression as mean ± SD of three independent experiments. **P<0.01, and ***P<0.005 by one-way ANOVA.

Following this, we further investigated the underlying mechanisms for the pro-apoptotic effect of CDDO-Me in K562 cells. Caspase 3 is a critical executioner of apoptosis, which can be cleaved and activated in apoptosis. Figure 4A and 4B showed that CDDO-Me remarkably induced a dose- and time-dependent elevation in casepase3 activity. Besides, activation of caspase 3 was further confirmed by Western blotting and immunofluorescence assays (Figure 4C-G). PARP is one of the main cleavage targets of caspase 3, and the cleavage of PARP facilitates cellular disassembly and serves as a marker of cells undergoing apoptosis [43]. Our study also showed that the level of cleaved PARP increased significantly (Figure 4C-E) after cells were treated with CDDO-Me.

Figure 4.

CDDO-Me induces caspases and PARP cleavage in K562 cells. (A) Activation of caspase 3 in K562 cells induced by CDDO-Me. Cells were treated with or without CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. (B) The time course of activation of caspase 3 in K562 cells induced by CDDO-Me. Cells were treated with 0.5 μM CDDO-Me for various periods of time. (C) Representative blots of caspases and cleaved PARP, and K562 cells were incubated with CDDO-Me at different concentrations for 24 h. (D) Representative blots of caspases and cleaved PARP, and K562 cells were incubated with 0.5 μM CDDO-Me over 72 h. (E) Bar graphs showing the expression level of the proteins above mentioned. (F) Representative Immunofluorescence staining the expression levels of cleaved caspase 3. (G) Bar graph showing quantitation results from (F). Data are expression as mean ± SD of three independent experiments. *P<0.05, **P<0.01, and ***P<0.005 by one-way ANOVA.

Extrinsic death receptor pathway and the intrinsic mitochondrial-mediated pathway are two main routes leading to apoptosis with the involvement of different caspases [43]. In this study, exposure of K562 cells to CDDO-Me led to an activation of caspases 8 and 9 (Figure 4C-E) in a dose- and time-dependent manner, indicating that both intrinsic and extrinsic pathways are involved in CDDO-Me-induced apoptosis in K562 cells.

CDDO-Me induces mitochondrial dysfunction of K562 cells involving the Bcl-2 family

Mitochondria plays a key role in the regulation of apoptosis, which can integrate the apoptotic signals originating from both extrinsic and intrinsic apoptosis pathways [44]. As our proteomics data indicated that mitochondrial dysfunction was a critical signaling pathway responding to CDDO-Me exposure (Table S2 and Figure S2), we detected mitochondrial membrane potential changes using JC-1 as a molecular probe in K562 cells. As shown in Figure 5A, cells exhibited red fluorescence in the control group, whereas CDDO-Me exposure increased the portion of K562 cells with green fluorescence exclusively, indicating loss of mitochondrial membrane potential. Ratios of JC-1 aggregates/monomeric was reduced by 31%, 67%, and 88%, when cells were exposed to CDDO-Me at 0.25, 0.5, and 1 μM, respectively (P<0.05, Figure 5B). These results demonstrate that CDDO-Me dose-dependently induces a significant reduction or loss of mitochondrial membrane potential due to membrane disruption.

Figure 5.

CDDO-Me induces mitochondrial dysfunction by regulating the expression of Bcl-2 family proteins in K562 cells. (A) Representative fluorescence microscopy images of K562 cells treated with 0.25, 0.5, and 1 μM CDDO-Me for 24 h, and stained with the JC-1 dye. (B) Bar graph showing quantitation results from (A). (C) Representative blots of Bax, Bak, Bcl-2, Bcl-xL, Mcl-1, and cytochrome C, when K562 cells were exposed to CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. (D) Representative blots of caspases and cleaved PARP, and K562 cells were incubated with 0.5 μM CDDO-Me over 72 h. (E) Bar graphs showing the expression level of the proteins above mentioned. Data are expression as mean ± SD of three independent experiments. *P<0.05, **P<0.01, and ***P<0.005 by one-way ANOVA.

The disruption of the mitochondrial membrane function results in the release of the cytochrome c, which is coupled to the activation of caspase 9, and the mitochondrial permeability is regulated by the Bcl-2 family [45]. As such, we next evaluated the level of cytochrome c and selected pro-survival and pro-apoptotic Bcl-2 family proteins by Western blotting assay. As shown in Figure 5C and 5E, the cytosolic level of cytochrome c was increased 1.6- and 2.1-fold in response to the treatment with 0.5 and 1 μM CDDO-Me, respectively (P<0.005). Also, CDDO-Me significantly decreased the Bcl-2 level, while only slightly increasing Bax level. As a consequence, the ratio of Bcl-2/Bax was decreased (Figure 5C and 5E). Compared with the control, there was a 57%, 76%, and 80% reduction in Bcl-xl level when treated with 0.25, 0.5, and 1 μM CDDO-Me, and a 61% and 79% decline in Mcl-1 level when incubated with 0.5 and 1 μM CDDO-Me, respectively (P<0.001, Figure 5C and 5E). Conversely, the expression level of puma was raised 3-fold by incubation with 1 μM CDDO-Me (P<0.005, Figure 5C and 5E), and the expression of Bak was slightly affected.

Following this, we tested the expression of cytochrome c and selected Bcl-2 family proteins over 72 h. In comparison with the control cells, there was a marked increase in the cytosolic level of cytochrome c, and a remarked decrease in the ratio of Bcl-2/Bax, when cells were exposed to CDDO-Me at 0.5 μM for 12, 24, 48, and 72 h (Figure 5D and 5E). The level of anti-apoptosis proteins, both Bcl-xl and Mcl-1, were down-regulated. Interestingly, CDDO-Me treatment led to a significant up-regulation in the level of Bak over 72 h. Collectively, these results reveal that CDDO-Me dose- and time-dependently induces mitochondrial dysfunction in K562 cells through Bcl-2 family, leading to apoptosis.

CDDO-Me triggers endoplasmic reticulum (ER) stress involving UPR signaling in K562 cells

Na⁺,K⁺-ATPase inhibitors have gained increasing interest for its anticancer potential and Na⁺,K⁺-ATPase inhibitor activates UPR signaling, which may provide a further explanation for the anticancer effect [42,46,47]. As shown by the IPA pathway analysis (Table S2 and Figure S3) on the alteration of the UPR signaling response to CDDO-Me exposure in K562 cells, we examined the expression level of the UPR proteins. In comparison with the control group, CDDO-Me significantly increased the level of Bip and the ratio of p-PERK/PERK and p-IRE1α/IRE1α, indicating a triggered ER stress (Figure 6A, 6B). It is known that excessive and prolonged ER stress triggers apoptosis [48]. Notably, the expression of CHOP is dramatically up-regulated with increasing concentrations of CDDO-Me for 24 h (Figure 6). Taken together, CDDO-Me triggers ER stress involving UPR signals, contributing to CDDO-Me-elicited apoptosis.

Figure 6.

CDDO-Me induces UPR in K562 cells. K562 cells were treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. A. Representative blots of Bip, p-PERK, PERK, p-IRE1α, IRE1α, CHOP, and β-actin. B. Bar graphs showing the ratio of p-PERK/PERK and p-IRE1α/IRE1α, the expression of Bip and CHOP. Data are expression as mean ± SD of three independent experiments. *P<0.05 and ***P<0.005 by one-way ANOVA.

CDDO-Me induces autophagy of K562 cells via inhibition of the PI3K/Akt/mTOR signaling pathway

To further study the anticancer mechanisms of CDDO-Me on K562 cells, we investigated the effect of CDDO-Me on autophagy of K562 cells. As shown in Figure 7A and 7B, there was a 3.9-fold increase in the autophagic cells when treated with 1 μM CDDO-Me for 24 h (P<0.005), without significant change at lower concentrations of CDDO-Me. As two important markers of vesicle expansion and formation during autophagic process, LC3 and beclin 1 were determined. As expected, CDDO-Me treatment increased LC3-II level, while decreasing LC3-I level. Correspondingly, the ratio of LC3-II/LC3-I was tripled at 1 μM CDDO-Me (P<0.01, Figure 7C and 7D). Also, the expression level of beclin 1 was elevated 3.0- and 4.2-fold when cells were incubated with CDDO-Me at 0.5 and 1 μM for 24 h, respectively (P<0.01 or 0.005, Figure 7C and 7D). These results indicate that CDDO-Me exerts a promoting effect on autophagy of K562 cells.

Figure 7.

CDDO-Me induces autophagy of K562 cells via PI3K/Akt/mTOR signaling pathway. K562 cells were treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h. A. Representative flow cytometric plots of autophagic K562 cells. B. Bar graphs showing the percentage of autophagy cells. C. Representative blots of phosphorylated PI3K, Akt, mTOR, and the expression of PI3K, Akt, mTOR, beclin 1, LC3-I, and LC3-II. D. Bar graphs showing the ratio of p-mTOR/mTOR, p-Akt/Akt, LC3-II/LC3-I, and the level of p-PI3K and beclin 1. E. K562 cells were pretreated with 10 μM MK2206, and then incubated in the presence or absence of 0.5 μM CDDO-Me for 24 h. The treated cells were analyzed by flow cytometry. F. Bar graphs showing the percentage of autophagy cells. Data are expression as mean ± SD of three independent experiments. *P<0.05, **P<0.01, and ***P<0.005 by one-way ANOVA.

We further explored the possible mechanisms for the autophagy-inducing effect of CDDO-Me in K562 cells. The IPA canonical pathway analysis showed that mTOR signaling as well as upstream PI3K/Akt signaling were critical for the effect of CDDO-Me on K562 cells (Table S2, Table S3 and Figure S4). Thus, we examined the level of proteins in PI3K/Akt/mTOR pathways using Western blotting assay. After exposure of the cells to 1 μM CDDO-Me for 24 h, the level of p-PI3K (Tyr458) dropped by 30% (P<0.05, Figure 7C and 7D). Similarly, the ratio of p-Akt/Akt was decreased by 30% and 51% when treated with CDDO-Me at 0.5 and 1 μM for 24 h, respectively (P<0.01, Figure 7C and 7D). Additionally, CDDO-Me significantly down-regulated the phosphorylation level of mTOR at Ser2448, but only slightly affected the level of total mTOR, resulting in 62%, 30% and 21% decrease in the ratio of p-mTOR/mTOR when incubated with CDDO-Me at 0.25, 0.5, and 1 μM, respectively (P<0.005, Figure 7C and 7D). These results show that suppression of PI3K/Akt/mTOR pathway contributes to CDDO-Me-induced autophagy in K562 cells.

In order to further testify the role of PI3K/Akt/mTOR pathway in CDDP-Me-induced autophagy in K562 cells, we subsequently employed 10 μM MK-2206 (anAkt inhibitor and a blocker of autophagosome formation) to examine the autophagy of K562 cells. As shown in Figure 7D and 7E, co-incubation of CDDO-Me with MK-2206 remarkably enhanced the autophagy-inducing effect of CDDO-Me in K562 cells, with the percentage of autophagic cells being elevated from 8.1% to 58.2% (P<0.001). It indicates that PI3K/Akt/mTOR play an important role in CDDO-Me-induced autophagy in K562 cells.

p38 MAPK and Erk1/2 play roles in CDDO-Me-induced apoptosis and autophagy in K562 cells

p38 MAPK and Erk1/2 play a vital role in the regulation of cell death and cell growth. Therefore, we determined p38 MAPK and Erk1/2 signals in response to CDDO-Me treatment.We observed that CDDO-Me suppressed p38 MAPK signaling but enhanced Erk signaling in K562 cells, evidenced by the reduction in the ratio of p-p38 MAPK/p38 MAPK and increase in p-Erk1/2/Erk1/2 (Figure 8A and 8B). Following the observation on the regulatory effect of CDDO-Me on p38 MAPK and Erk1/2 signaling pathway, we explored the roles of p38 MAPK and Erk1/2 in the cancer cell killing effect of CDDO-Me in K562 cells, and examined apoptosis and autophagy by flow cytometry simultaneously. As shown in Figure 8C and 8D, incubation with SB202190 (a selective p38 MAPK inhibitor and autophagy inducer) alone for 24 h led to a 2.0-fold elevation (P<0.01) in the percentage of autophagic cells compared to the control cells. In comparison with cells incubated with CDDO-Me alone, co-incubation with SB202190 significantly enhanced the CDDO-Me-induced apoptosis (2.4-fold, P<0.01, Figure 8C and 8D) and autophagy (3.8-fold, P<0.001, Figure 8C and 8D). On the other hand, pre-treatment with U0126 (an Erk1/2 inhibitor) resulted in a 53.3% decline (P<0.01, Figure 8C and 8D) in CDDO-Me-induced apoptosis, while only exhibiting a marginal effect on autophagy in K562 cells.

Figure 8.

p38 MAPK and Erk1/2 play roles in CDDO-Me-induced apoptosis and autophagy in K562 cells. (A) K562 cells were treated with CDDO-Me at 0.25, 0.5, and 1 μM for 24 h or at 0.5 μM over 72 h. Representative blots of phosphorylated p38 MAPK, Erk1/2, p38 MAPK, andp-Erk1/2. (B) Bar graphs showing the ratio of p-p38/p38 and p-Erk1/2/Erk1/2. (C) Representative flow cytometric plots. (D) Bar graphs showing apoptotic and autophagic K562 cells. Cells were pretreated with 10 μM SB202190 or 10 μM U0126, then incubated with CDDO-Me for 24 h. (E) Representative fluorescence microscopic images. (F) Bar graphs showing K562 cells pretreated with 10 μM SB202190, then incubated with CDDO-Me for 24 h, and stained with JC-1. (G) Representative blots and (H) bar graphs of the expression of cleaved caspase 3, LC-3, Na⁺,K⁺-ATPase α1, CHOP, p-Akt, and p-Erk1/2. Data are expression as the mean ± SD of three independent experiments. *P<0.05, **P<0.01, ***P<0.005, and ****P<0.001 by one-way ANOVA.

We further investigate the mechanism for SB202190-enhanced CDDO-Me-induced apoptotic and autophagic effects, the mitochondrial membrane potential change and related protein expression levels were examined. In comparison with the cells treated with CDDO-Me alone, co-incubation CDDO-Me with SB202190 significantly increased cleaved caspase 3 (2.2-fold), although the alteration in the ratios of JC-1 aggregates/monomeric was not significant (35.7% reduction, P>0.05, Figure 8E-H). These changes suggest that SB202190 enhanced CDDO-Me-induced apoptosis via mitochondrial-dependent pathway. Furthermore, in comparison with the cells treated with CDDO-Me only, there was a 3.2-fold increase in CHOP level and a 22.4% decrease in Na⁺,K⁺-ATPase level when co-treatment with SB202190, indicating that p38 MAPK also influenced Na⁺,K⁺-ATPase expression and UPR signaling pathway (Figure 8G and 8H). In agreement with the flow cytometric results stated above, exposure of K562 cells to CDDO-Me plus SB202190 remarkably elevated the ratio of LC3-II/LC3-I (Figure 8G and 8H), compared with the control cells receiving CDDO-Me only. Moreover, the level of p-Akt was reduced by 78.8%, whereas the level of p-Erk1/2 was elevated 10.8-fold, in comparison with the cells exposed to CDDO-Me alone (Figure 8G and 8H), suggesting that SB202190 enhanced the effect of CDDO-Me-induced autophagy involving the PI3K/Akt/mTOR pathway. Interestingly, our results also show that CDDO-Me-induced Erk1/2 phosphorylation was increased 10.8-fold by SB202190 (P<0.001, Figure 8G and 8H), which indicates that p38 MAPK exerts an inhibitory effect on CDDO-Me-stimulated Erk1/2 activation. Taken together, there are interactions between CDDO-Me-induced apoptosis and autophagy, involving p38 MAPK/Erk1/2 signaling pathway.

Discussion

Up to now, treatment of CML is still a challenge because of poor response/drug resistance and intolerance in a substantial proportion of patients, thus there is an urgent need to develop new drugs and identify new therapeutic targets for better clinical outcomes. The SILAC-based proteomics approach is a high-throughput quantitative analytical method, which can comprehensively evaluate the effect of a given compound and recognize its potential molecular targets and related signaling pathways at cellular levels and in vivo [33-36]. In order to find the possible molecular targets and mechanisms for the anticancer effects, our earlier studies have employed this technique to disclose the molecular interactome of 5,6-dimethylxanthenone 4-acetic acid (DMXAA, a tumor vascular disrupter) and alisertib (an Aurora kinase A inhibitor) in different cancer cell lines [39,40,49,50]. CDDO-Me is a multi-targeting molecule exerting the potent anticancer effect in the treatment of various types of cancer in preclinical and clinical studies [21,26]. No studies have reported its proteomic responses in CML cells. In the present study, we evaluated the proteomic responses to CDDO-Me treatment in K562 cells. The results have shown that the responding functional proteins and signaling pathways were mainly involved in cell survival and death, cellular function and maintenance, energy and nutrition metabolism. We have verified that CDDO-Me suppressed Na⁺,K⁺-ATPase expression, arrested K562 cells in G₂ and S phases, induced marked apoptosis, promoted autophagy, and triggered ER stress.

Na⁺,K⁺-ATPase is a transmembrane protein complex serving as a central energy-consuming pump to maintain ionic and osmotic balance in cells [51]. It also serves as plasma membrane receptors bound by a family of cardiotonic steroids and signal transducers that can provide a feedback loop between Na⁺,K⁺-ATPase and the mitochondria [52]. Na⁺,K⁺-ATPaseis composed of 4 α isoforms (α1, α2, α3 and α4) and 3 β isoforms (β1, β2 and β3); and α1 or α3 isoforms are often overexpressed in cancer whereas the β1 isoform acts as a tumor suppressor [42,46,47]. Na⁺,K⁺-ATPase is a modulator of apoptosis and autophagy in tumor cells [47]. Here, we first reported the down-regulation of Na⁺,K⁺-ATPase α1 by the treatment of CDDO-Me in human leukemia cells, suggesting that it is a potential novel target protein of CDDO-Me. The mechanism for the anticancer effect of CDDO-Me via targeting Na⁺,K⁺-ATPase in human leukemia cells deserves further investigations.

It is known that the eukaryotes cell cycle is regulated by cyclins, CDKs and the CDK inhibitors (CKIs). The CDK1-cyclin B1 complexes promote transition and mitosis in G₂/M phase, and the CDK2-cyclin A complexes predominate in Sphase [53,54]. The CKIs, including p21Waf1/Cip1 and p27Kip1, inhibit the CDK-cyclin activities and prevent cell cycle progression. In this study, we observed a remarkable decrease in cyclin B1, CDK1/CDC2, and CDK2 expression, meanwhile an increase in p21Waf1/Cip1 and p27Kip1 expression, which might explain the G₂ and S phase arrest by CDDO-Me in K562 cells. We also observed CDDO-Me induced an elevation in the level of p53 and p-p53 at Ser15. As a direct p21Waf1/Cip1 upstream target, p53 can lead to either cell cycle arrest and DNA repair or apoptosis [55]. Phosphorylation at Ser15 impairs the ability of MDM2 to bind p53, promoting both the accumulation and activation of p53 in response to DNA damage [56]. Taken together, the proteomic and verification data reveal that CDDO-Me exerts a cell cycle arresting effect via regulation of key functional proteins of cell cycle in K562 cells.

Apoptosis is a process of programmed cell death necessary for cell growth, development and maintenance of homeostasis in metazoans associated with G₂/M arrest [43,57,58]. Caspases are a family of cysteine proteases and the central regulators in cell apoptosis. In agreement with our previous findings in esophageal squamous cancer cells [21], we observed a concentration- and time-dependent apoptosis induced by CDDO-Me in K562 cells. In this study, enhanced expression level of caspase 9 and cleaved caspase 8 reflected the activation of both intrinsic/mitochondrial-mediated and extrinsic/death receptor pathways, which in turn activated cleavage caspase 3 and PARP and ultimately induced apoptosis. There is a crosstalk between two apoptotic pathways through Bid, which transferred the apoptotic signal from the cell surface to mitochondria [59]. Our proteomics data also showed that CDDO-Me regulated mitochondrial function. Mitochondrial permeabilization is an important cellular event in apoptotic cell death and regulated by Bcl-2 family members, including pro-survival proteins Bcl-2, Bcl-xL,and Mcl-1; pro-apoptotic members Bax, Bak, and “BH3 only” proteins PUMA [60,61]. We verified the depolarization of mitochondrial membrane potential induced by CDDO-Me in K562 cells and the release of cytochrome c into cytosol and subsequent activation of caspase 9, which might be attributed to down-regulation of Bcl-2/Bax, Bcl-xL, Mcl-1, and up-regulation of Bak and PUMA. Taken together, our results suggest that CDDO-Me exhibits its apoptotic effects through intrinsic/mitochondrial-mediated as well as extrinsic apoptosis pathways.

In addition to the two major pathways, intrinsic mitochondria-mediated pathway and the extrinsic death receptor-induced pathway, apoptosis can be induced via ER stress [61-64]. When the cellular energy level, the redox state, and Ca²⁺ concentration are perturbed, the ER stress is initiated, triggering the UPR. Interestingly, Jeong et al. [65] reported that CDDO-Me increased intracellular Ca²⁺ concentration. In this study, we observed an up-regulated level of BIP and activated PERK and IREα, indicating that CDDO-Me activates UPR in K562 cells. As well known, BiP is the marker of ER stress, controlling the activation of transmembrane ER stress sensors (PERK, IRE1, and ATF6) through a binding-release mechanism [48,63,66]. Our results also displayed that the exposure of K562 cells to CDDO-Me increased the expression of CHOP. Although UPR is a protective factor of cell, if the stress cannot be resolved, it switches from pro-survival to pro-apoptosis [48]. CHOP is a downstream of PERK and IREα, once activated, it can trigger the expression of pro-apoptotic proteins, targeting the Bcl2 family, acting on the mitochondrial membrane to release cytochrome c and initiating the caspase cascade [67,68]. Similar with our results, Zou et al. [66] have reported that CDDO-Me triggered ER stress, leading to CHOP-mediated apoptosis in lung cancer cells. Our study suggests that UPR signaling is implicated in CDDO-Me-induced apoptosis, although more studied are needed to further validate the UPR-inducing effect in the treatment of CML by CDDM-Me.

Autophagy as type II programmed cell death, extremely affects diverse stages of occurrence and development of cancer with the contribution to overlapped signaling pathways of autophagy, apoptosis and carcinogenesis [64,69-71]. The PI3K/Akt/mTOR is a central pathway involved in autophagy. The phosphorylation of PI3K activated Akt, then mTOR can integrate upstream activating signals through PI3K/Akt pathway and become phosphorylated form, which negatively regulates autophagy by limiting the inhibitory effect on the ULK1 kinase complex in response to the deprivation of nutrient or stress [72]. CDDO-Me has been found to inhibit proliferation of cancer cells via PI3K/Akt/mTOR signaling pathway by blocking the activation of Akt and downstream targets, including mTOR [22], but the association with autophagy is unclear. In this study, a concentration-dependent autophagy induced by CDDO-Me was observed. In addition, MK2206 was employed and significantly increased the CDDO-Me-induced autophagic cell percentage. Therefore, our results suggest that CDDO-Me-induced autophagy of K562 cells by inhibition of PI3K/Akt/mTOR pathway.

p38 MAPK and Erk1/2, known as the members of MAPK family, play a critical role in the regulation of cell growth, differentiation, and control of cellular responses to cytokines and stress [73]. Erk1/2 and p38 MAPK have opposing effects on cancer cell death, however each of them was involved in stress-induced apoptosis. In the present study, we have observed that CDDO-Me treatment remarkably decreased the phosphorylation of p38 MAPK, contributing to the recent findings that Akt suppresses the activation of p38a by phosphorylation of ASK1 on Ser [74,75], but increases the phosphorylation of Erk1/2 in K562 cells in concentration- and time-dependent manners. Furthermore, SB202190 significantly enhanced the apoptosis and autophagy induced by CDDO-Me in K562 cells, whereas the reverse regulating effects of U0126 were observed. Interestingly, our results also showed that CDDO-Me-induced Erk1/2 phosphorylation was increased by SB202190, which might be ascribed to a crosstalk between p38 MAPK and Erk1/2 signaling pathways via PP2A. Our results indicate a crosstalk between the autophagic and apoptotic pathways, with a series of key molecules or pathways being synchronized and mediating the complex interplay, including mTOR and UPR signaling pathways, Akt, Erk1/2, and Na⁺,K⁺-ATPase α1. Herein, CDDO-Me can induce apoptosis and autophagy in a coordinated manner in K562 cells.

Conclusion

The SILAC proteomics and validating cellular studies demonstrate that CDDO-Me inhibits cellular proliferation, induces cell cycle arrest, triggers mitochondria-, death receptor-dependent, and ER stress-mediated apoptosis, and promotes autophagy by regulating numerous functional proteins and signaling pathways. In particular, the PI3K/Akt/mTOR signaling pathway is involved in autophagy and p38 MAPK/Erk1/2 signaling pathways contributes to apoptosis- and autophagy-inducing effects in K562 cells. Na⁺,K⁺-ATPase may be a novel target of CDDO-Me. Taken together, CDDO-Me may represent a promising anticancer agent for CML therapy, and further studies are warranted to uncover the biochemical mechanisms.

Disclosure of conflict of interest

None.