Celastrol impairs tumor growth by modulating the CIP2A-GSK3β-MCL-1 axis in gastric cancer cells

Cell culture

The Gastric cancer cell lines AGS, BCG-823, HGC-27 and KATO-III were purchased from the American Tissue Culture Collection. The AGS and KATO-III cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Hyclone) and BCG-823 and HGC-27 were cultured in RPMI 1640 (Hyclone) containing 10% fetal bovine serum (FBS, Hyclone), 100 units/ml penicillin and 100 mg/ml streptomycin at 37°C in 5% CO₂ in a humidified incubator.

Reagents and antibodies

Celastrol and MG-132 were obtained from Calbiochem (Cat #474790; San Diego, CA, USA). PS-341 was obtained from Millennium Pharmaceuticals (Cambridge, MA, USA). Cycloheximide (CHX) was purchased from Beyotime (Jiangsu, China). These reagents were dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich, MO, USA). The primary antibodies were used: anti-CIP2A (HL1925; Santa Cruz Biotechnology, Santa Cruz, CA, USA); anti-PP2A, anti-MCL-1 and anti-Actin (ab137849, ab32087, ab8226; Abcam, Cambridge, MA, UK); anti-GSK3β, anti-BCL-2 and anti-BCL-x_L (#9336, #15071, #2764; Cell Signaling Technology, Danvers, MA, USA). Enhanced chemiluminescence kit was obtained from Yeasen Technology (36222ES60; Shanghai, China).

RNA extraction, reverse transcription-PCR and RT-PCR analysis

Total cellular RNAs were extracted using a Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. 1 μg total RNA were reverse synthesized into cDNAs using random or oligo-dT primers and MMLV reverse transcriptase (AT101; Transgen, Beijing, China). qRT-PCR analysis was subsequently performed with ABI7700 detector (Applied Biosystem, Waltham MA, USA) using the SYBR Premix ExTaq system (RR390Q; Takara, Japan). Sequence of primers used was as follows:

The mRNA was normalized to β-actin and calculated using delta method from threshold cycle numbers.

siRNA and plasmid transfection

The chemical modified siRNAs targeting CIP2A and control were designed and purchased from Shanghai GenePharma. The sequences for CIP2A and control were used as follows: 5′-CUGUGGUUGUGUUUGCACUTT-3′ (CIP2A) and 5′-UUCUCCGAACGUGUCACGUTT-3′ (NC). siRNA transfection into AGS and BCG-823 cells at 100 nM was performed using Lipofectamine 3000 (L3000001; Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocols.

The pcDNA3.1-FLAG-CIP2A expression plasmid was constructed and transiently transfected with 2 μg plasmid into AGS and BCG-823 cells using Lipofectamine 3000.

Western blotting

Cell pellets were extracted with RIPA buffer on ice for 30 min. Protein extracts were normalized and loaded on 10–15% SDS-PAGE and electrophoresed. Gels were transferred onto nitrocellulose membranes followed by blocking with 5% non-fat milk in Tris-buffered saline. The membranes were rinsed and probed with the specific primary and corresponding secondary antibodies. Each Band was visualized by using Luminescent Image Analyzer LSA 4000 (GE, Fairfield, CO, USA).

Apoptosis analysis

After celastrol or siRNA treatment, the cells were harvested by trypsinization and then subjected to Annexin V-FITC/PI apoptosis detection kit (C1062S; Beyotime, Shanghai, China). Cells were rinsed twice with ice-cold PBS and incubated with FITC-conjugated annexin V for 30 min and then stained with PI added immediately prior to analysis. Apoptosis was determined by flow cytometer (FACSCalibur, by flow cytometer BD Biosciences, San Diego, CA, USA).

Cell proliferation assay

The anti-proliferative effect of celastrol on AGS and HCG-823 cells was determined by CCK-8 assay according to manufacturer’s protocols. Briefly, cells were plated into 96-well plates with a density of 5 × 10³ cells/well, treated indicated time points later with different doses of celastrol. The fresh mixture containing 10% CCK-8 reagent was added to the cells to replace spent medium and incubated at 37°C for 1–4 h, and absorbance was measured at 492 nm using Microplate spectrophotometer.

Statistical analysis

The quantitative data were presented as the mean ± SD unless noted otherwise. All statistical analyses were two-tailed and computed using GraphPad Prism 5 and SPSS 17.0 software. Student’s t-test or Mann-Whitney U tests were used to determined statistically significant differences between two or more groups, respectively. P values less than 0.05 indicated statistically significant in all cases.

CIP2A is rapidly downregulated in cells treated with celastrol

To investigate the effect of celastrol on CIP2A expression level, we treated AGS, BCG-823, HGC-27 and KATO-III cells with celastrol or not and evaluated CIP2A expression level by quantitative real-time PCR and western blot analysis. Results indicated treatment of AGS cells with celastrol at 5 μM for 6–18 h drastically lowered CIP2Aexpression at protein level and became undetectable at 18 h, but very slight change at mRNA level (Figure 1B). Similarly, expression level of CIP2A was reduced response to celastrol treatment at 1–5 μM for 6 h also reduced the (Figure 1B). Furthermore, these similar observations were confirmed in other gastric cancer cells (Figure 1C–1E). In summary, celastrol induced reduction of CIP2A at protein level in a dose- and time-dependent manner, but not at mRNA level.

Celastrol-induced CIP2A degradation is mediated by the ubiquitin–proteasome pathway

The underlying mechanism of CIP2A reduction is important for celastrol clinical application. CIP2A was downregulated at 2–6 h after celastrol treatment using protein synthesis inhibitor cycloheximide (50 μg/mL) to block de novo protein synthesis, indicating that CIP2A ablation was due to its proteolytic degradation. Of course, only cycloheximide could not reduce CIP2A expression level within 6 h (Figure 2A, 2B).

Figure 2. Celastrol-induced CIP2A degradation is mediated by the ubiquitin–proteasome pathway. (A, B) Effects of 50 ug/ml cycloheximide (CHX) alone or in combination with 5 μM celastrol on CIP2A expression were evaluated by western blotting in AGS and BGC-823 cells. (C, D) Western blotting analyzed CIP2A expression in AGS and BGC-823 cells preincubated with PS-341 (100 nM) and MG-132 (10 μM) for 2 h followed by treatment of celastrol (5 μM) for additional 6 h.

The ubiquitin-proteasome system is involved in targeted degradation of intracellular short lived proteins in mammalian cytosol and nucleus [21]. To explore and clarify the mechanism of CIP2A degradation, we first detected the effect of classic proteasome inhibitors PS-341 and MG-132 on the degradation of CIP2A. Interestingly, western blot results demonstrated only proteasome inhibitors stimulation could not influence CIP2A stability. However, the phenomenon of celastrol-triggered CIP2A degradation was blocked when these inhibitors were added to the medium with 5 μM celastol within 6 h (Figure 2C, 2D). These observations clearly indicated that ubiquitin-proteasome was the essential pathway of celastrol-triggered CIP2A degradation.

Celastrol suppresses proliferation and triggers apoptosis of gastric cancer cells

Our previous studies indicated anti-cancer effect of celastrol was associated with CIP2A reduction. We have known CIP2A was a direct target of celastrol. Next, our research focused on the anti-gastric cancer effect of this compound. Cell apoptosis induced by celastrol was detected by using the technique of annexin V-FITC/PI staining and flow cytometry analysis. AGS and BCG-823 cells were treated with celastrol at 1–5 μM for 24 h or 2.5 μM for indicated time points. The results demonstrated that celastrol induced cell apoptosis in a dose- and time-dependent manner. (Figure 3A, 3B) In order to evaluate another character of celastrol, CCK8 assay was applied to analyze cell proliferation. The results indicated the proliferation of AGS and BCG-823 cells was significantly decreased after celastrol administration (Figure 3C).

Figure 3. Celastrol triggers apoptosis of gastric cancer cells. (A, B) AGS and BCG-823 cells were treated with various concentrations of celastrol for 24 h or 5 μM celastrol for indicated times, and cell apoptosis was assessed by annxin V-FITC/PI staining and flow cytometry. (C) AGS and BCG-823 cells were treated with gradually increasing doses of celastrol for 24 h. The cell viability was measured by CCK8 assay. (D) AGS and BCG-823 cells were transfected with CIP2A-flag expression vector for 48 h, followed by celastrol (5 μM) for 6 h, cell apoptosis was assessed by annxin V-FITC/PI staining and flow cytometry.

To investigate CIP2A’s role on cell apoptosis induced by celastrol, protein overexpression technique was applied in the next experimental system. Firstly, we transfected the CIP2A-flag expression construct into AGS and BCG-823 cells for 48 h, followed with or without celastrol for another 6 h, and cell apoptosis was evaluated. As expected, the rate of cell apoptosis induced by celastrol was partly reversed by CIP2A overexpression (Figure 3D). Consistent to our previous study, CIP2A exerts its anti-proliferative and pro-apoptotic effects of celastrol on gastric cancer cells.

CIP2A mediates cytotoxicity of celastrol by triggering MCL-1 Degradation

Therefore, to further explore whether Myeloid Cell Leukemia 1 (MCL-1), as a downstream molecule of CIP2A signaling pathway, took part in regulating cell apoptosis, we knocked down CIP2A expression level by using RNA interference technology. Results demonstrated silencing of CIP2A specifically decreased MCL-1 expression, but not influenced Bcl-2 and Bcl-x_L expression (Figure 4). Our above data confirmed the specificity of CIP2A modulation in cell apoptosis pathway.

Figure 4. CIP2A mediates cytotoxicity of celastrol by triggering MCL-1 Degradation. (A, B) AGS and BCG-823 cells were transfected with siNC or siCIP2A for 48 h, the expression of CIP2A, BCL-2, BCL-x_L and MCL-1 were detected by real-time PCR (A) and western blotting (B) assays.

Dephosphorylation by GSK3β enhanced MCL-1 degradation

To identify the signaling pathway(s) mediating the observed apoptosis by celastrol, we firstly detected Protein phosphatase 2A (PP2A) expression in the AGS and BCG-823 cells treated with 5μM celastrol or transfected with siCIP2A. PP2A expression level was upregulated in response to celastrol stimulation and inverse correlation with CIP2A level (Figure 5A). Glycogen synthase kinase 3β (GSK3β) owns a Ser/Thr kinase activity and regulates protein synthesis, cell proliferation, differentiation, motility and apoptosis [22]. We therefore wondered to test whether PP2A regulated GSK3β phosphorylation to influence cell apoptosis. Interestingly, our results showed when GSK3β was silenced, MCL-1 was also downregulated (Figure 5B). Our above data provided the possibility that PP2A-pGSK3β-MCL1 Axis regulated celastrol-triggered apoptosis. To further dissection of this hypothesis, AGS and BCG-823 cells were treated with 5 μM Celastrol or CIP2A-specific knockdown followed by celastrol treatment or not. Results demonstrated PP2A was upregulated and inhibited phosphorylation of GSK3β, and further low level of GSK3β phosphorylation down-regulated MCL-1 expression in response to celastrol-induced degradation of CIP2A (Figure 5C). These results obtained a deeper insight that pro-apoptosis effects of celastrol were involved in GSK3β dephosphorylation activated by PP2A, finally leading to a reduction of pro-survival protein MCL-1.

Figure 5. Dephosphorylation by GSK3β enhanced MCL-1 reduction. (A) AGS and BCG-823 cells were transfected with siNC or siCIP2A for 48 h followed by celastrol (5 μM) for 6 h, lysed, and the lysates were probed with indicated antibodies. (B) AGS and BCG-823 cells were transfected with siNC or siGSK3β for 48 h, lysed, and the lysates was probed with anti-MCL-1 antibody. (C) AGS and BCG-823 cells were transfected with siNC or siCIP2A for 48 h, followed by celastrol (5 μM) for 6 h, lysed, the expression of PP2A, GSK-3β, pGSK-3β and MCL-1 were detected by Western blot assay.