Celastrol induces apoptosis of gastric cancer cells by miR-146a inhibition of NF-κB activity

The inhibitory effects of celastrol on the growth of gastric cancer cells and normal gastric epithelial cells were evaluated using MTT assays. Celastrol treatment inhibited the growth of BGC-823, SGC-7901 and MGC-803 cells in a dose- and time-dependent manner (Figure 1A, 1B and 1C). And the IC 50 value at 72 h after treatment was 0.989 μM (BGC-823 cells), 0.798 μM (SGC-7901 cells) and 1.98 μM (MGC-803 cells). Normal gastric epithelial cells showed more resistance to the cytotoxicity effect of celastrol. And the IC 50 value at 72 h was 8.16 μM for GES-1 cells (Figure 1D).

Next, we wanted to determine whether the NF-κB pathway was inhibited by celastrol in gastric cancer cells. Therefore, phosphorylation of IκB was investigated as a potential response to treatment with celastrol. As shown by western blot analysis in Figure 1E, celastrol inhibited phosphorylation of IκB in a dose-dependent manner in BGC-823 cells, SGC-7901 cells and MGC-803 cells.

It is well known that the phosphorylation and degradation of IκB can result in the release of NF-κB complex followed by its rapid translocation into the nucleus [17]. To investigate the nuclear entry of NF-κB complex, the p65 content was measured by western blot in BGC-823, SGC-7901 and MGC-803 cells. Treatment with celastrol could inhibit p65 entry into the nucleus (Figure 1E). The inhibition of NF-κB signaling pathway by celastrol was also determined by the observation of the transcriptional activity of NF-κB. As shown in Figure 1F, the treatment of BGC-823, SGC-7901 and MGC-803 cell with celastrol strongly inhibited the NF-κB transcriptional activity in a dose-dependent manner.

To further identify the mechanism for celastrol inhibition of BGC-823, SGC-7901 and MGC-803 cells growth, we performed real-time PCR to detect miR-146a expression after treated with 2 μM of celastrol for 12 h. As shown in Figure 2, celastrol treatment significantly increased the expression of miR-146a in these cells (P < 0.01). However, celastrol can not significantly induce miR-146a expression in GES-1 cells.

To determine whether miR-146a can regulate signaling pathway in gastric cancer cells, we investigated the modulation of phosphorylation of IκB in cells transfected miR-146a mimic by western blot. The result of real-time PCR revealed that miR-146a mimic can significantly increase the expression of miR-146a in BGC-823, SGC-7901 and MGC-803 cells (P < 0.01) (Figure 3A), suggesting that miR-146a mimic is efficiently introduced into the cells and acts to up-regulate miR-146a expression. Furthermore, transfection of miR-146a mimic decreased phosphorylation of IκB and nuclear P65 as shown in Figure 3B.

We also measured the transcriptional activity of NF-κB using Luciferase reporter assay. Figure 3C showed that up-regulation of miR-146a expression decreased NF-κB activity, which was similar to Bay.

In order to evaluate the role of miR-146a in the effect of celastrol on inhibition of BGC-823, SGC-7901 and MGC-803 cells apoptosis, we treated cells with celastrol after transfected with miR-146a inhibitor. As shown in Figure 4A, miR-146a inhibitor can significantly decrease the expression of miR-146a in BGC-823, SGC-7901 and MGC-803 cells (P < 0.01). Interestingly, we found that celastrol treatment inhibited transcriptional activity of NF-κB, which can be restored by inhibition of miR-146a expression (Figure 4B). In addition, the number of cells that underwent apoptosis significantly increased after treated with 2 μM celastrol. Moreover, the celastrol induced apoptosis was attenuated by miR-146a inhibitor in BGC-823, SGC-7901 and MGC-803 cells (Figure 4C). Furthermore, BGC-823, SGC-7901 and MGC-803 cells growth was increased in celastrol-treated cells after transfected miR-146a inhibitor (Figure 4D).

Dulbecco’s modified Eagle’s medium (DMEM), sodium pyruvate were from Gibco-BRL (Rockville, MD, USA). Fetal bovine serum (FBS) was from PAA Laboratories (GmbH, Linz, Austria). Celastrol was purchased from Alexis Biochemicals (San Diego, CA). Bay 11–7082 (Bay) and dimethylthiazoly-2, 5-diphenyltetrazolium bromide (MTT) were obtained from Sigma Aldrich. Mouse anti-phospho-IκB (Ser32/36) antibody was obtained from Cell Signaling Technology (Beverly, MA, USA). Mouse anti-β-actin and goat anti-Lamin B antibodies were purchased from Santa Cruz Biotechnologies (Santa Cruz, CA, USA). Annexin V-EGFP ⁄ PI Apoptosis Detection Kit was from KeyGen (Nanjing, China). The Detergent Compatible (DC) Protein Assay kit was purchased from Bio-Rad Laboratories (Hercules, CA, USA). The miRNeasy Mini kit, the miScript Reverse Transcription kit and the miScript SYBR Green PCR kit were purchased from Qiagen (Hilden, Germany).

Human gastric adenocarcinoma cell line BGC-823 cells and SGC-7901 cells, and gastric mucinous adenocarcinoma cell line MGC-803 were purchased from the Shanghai Institute of Cytobiology in Chinese Academy of Sciences. Human normal gastric epithelial cell line GES-1 was obtained from the Institute of Oncology in Beijing University. The cells were grown in DMEM medium supplemented with 10% FBS, 100 units/ml penicillin and 100 μg/mL streptomycin, and routinely passaged at 3-day intervals. All cell lines were maintained at 37°C in a humidified incubator containing 5% CO₂ and treated with celastrol (dissolved in DMSO) in complete DMEM medium. To obtain reliable results, the final concentration of DMSO in the culture medium was kept less than 0.1%.

Cell viability was determined using MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assays. Briefly, the cells were seeded in 96-well dishes at 1 × 10⁴ cells per well, and treated with different concentrations of celastrol. Then each well was supplemented with 10 μL MTT and incubated for 4 h at 37°C. The medium was then removed, and 150 μL DMSO (Sigma Aldrich) was added to solubilize the MTT formazan. The optical density was read at 490 nm. The inhibition rate was calculated as follows: inhibition rate = [(mean control absorbance - mean experimental absorbance) ⁄ mean control absorbance] × 100 (%). To evaluate the effect of celastrol on cellular growth, the concentration that caused 50% growth inhibition (IC 50) was calculated by the modified Karbers method according to the formula: IC50 = log^− 1[X_k − i (∑P  − 0.5)], in which X_k represents the celastrol of the highest drug concentration; i is that of ratio of adjacent concentration; and ∑P is the sum of the percentage of growth inhibition at various concentrations [21].

This assay is based on the translocation of phosphatidylserine from the inner leaflet of the plasma membrane to the cell surface in early apoptotic cells [22]. Briefly, cells were resuspended in a binding buffer. Next, annexin V-EGFP and PI were added and the solution was incubated at room temperature for 15 min in the dark, followed by assay on FACScan (Becton Dickin-son). The percentage of apoptosis was computed using Cell-Quest software (Becton Dickinson).

miR-146a was knocked down or overexpressed by transfection with miRNA inhibitor or miRNA mimic using siPort Neo-FX (Ambion) according to the manufacturer’s recommendations. miR-146a mimics (5′-UGAGAACUGAAUUCCAUGGGUU-3′), miR-146a inhibitor (5′-AACCCAUGGAAUUCAGUUCUCA-3′) and negative mimics/inhibitors (NC, 5′-UUGUACUACACAAAAGUACUG-3′) were synthesized by RIBOBIO (Ribobio Co. Ltd, Guangzhou, China). All of the oligonucleotides were transfected at a final concentration of 100 nM. BGC-823, SGC-7901 and MGC-803 cells were transfected with miR-146a inhibitor or mimic using siPort Neo-FX (Ambion) according to the manufacturer’s recommendations.

Approximately 5 × 10⁶ cells were treated without (control) or with celastrol for 12 h. miRNAs were isolated and purified using Trizol reagent (Invitrogen, USA), according to manufacturer’s protocol. The miR-146a level was quantified by real-time PCR using TransStartTM SYBR Green qPCR Supermix (TransGen Biotech, Beijing, China), and with U6 small nuclear RNA as an internal normalized reference. For miR-146a, the primers were as follows: forward, 5′- GCCGATTGGAGTGGTAAAC-3′ and reverse, 5′-AAGACCCCTCGTTGCAGT-3′. For U6, the primers were as follows: forward, 5′-CTCGCTTCGGCAGCACA-3′ and reverse, 5′-AACGCTTCACGAATTTGCGT-3′.

The reporter plasmid, pNF-κB-luc, containing the κB-enhancer consensus sequences [(TGGGGACTTTCCGC) × 5] and NF-κB-dependent firefly luciferase gene was purchased from Stratagene (La Jolla, CA, USA). BGC-823, SGC-7901 and MGC-803 cells were transiently transfected with two plasmids (pNF-κB-luc plasmid and β-galactosidase) using the LipofectAMINE Plus regent. Cells were passaged on 24-wells plate the day prior to transfection to achieve 80%-85% confluence on the following day. Twelve hours after transfection, cells were incubated for an additional 24 h in medium containing different concentration of celastrol and harvested for luciferase reporter assays. Luciferase activity was measured with a luminometer (TD-20/20; Turner Designs, CA, USA) using the Luciferase Assay System. β-Galactosidase activity was detected to normalise any variations in the transfection efficiency.

Total protein was extracted using a lysis buffer containing 50 mmol/l Tris–HCl, pH 7.4; 1% NP-40; 150 mmol/l NaCl; 1 mmol/l EDTA; 1 mmol/l phenylmethylsulfonyl fluoride; and complete proteinase inhibitor mixture (one tablet per 10 ml; Roche Molecular Biochemicals, Indianapolis, IN, USA). The cytosol and nuclear extracts were prepared by NE-PER nuclear and cytoplasmic extraction reagents (Pierce Biotechnology, Rockford, IL, USA) according to the manufacturer’s instructions. Protein concentration in the cell lysate was quantified using the DC protein assay kit (Bio-Rad). Then, Western blot analysis was performed. β-actin and Lamin B were used as internal controls for the total and nuclear extracts, respectively.

Statistical analysis was performed with statistical analysis software SPSS 13.0 software. Statistical analyses were performed using either an analysis of variance (ANOVA) or Student’s t-test. Data were expressed as mean ± standard deviation. P < 0.05 was considered to be significant.