U87 glioma cell lines were purchased from American Type Culture Collections (Manassas, VA, USA). These cell lines were cultured in Dulbecco’s modified Eagle’s medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum and 25 U/ml penicillin/streptomycin. Anti-phospho-p53, and phospho-Akt polyclonal antibody were purchased from Cell Signaling Inc. (Danvers, MA, USA). Antibodies for p21, p27, β-actin were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Secondary horseradish peroxidase-conjugated anti-rabbit and anti-mouse antibodies were from Zymed Laboratories (San Francisco, CA, USA); chemiluminescent detection systems were from Pierce (Rockford, IL, USA); and Lipodin-Pro protein transfection reagent was from Abbiotec (San Diego, CA, USA). Small interfering RNA duplexes against Akt, p53, and p21 were purchased from Dharmacon Co. Transfection reagent (Oligofectamine^™ 2000) was purchased from Invitrogen (Eugene, OR, USA). N-acetyl-l-cysteine and doxorubicin were obtained from Calbiochem (San Diego, CA, USA), and 2′, 7′-dichlorofluorescein diacetate and MitoSOX was purchased from Invitrogen (Carlsbad, CA, USA). 20(S)-ginsenoside Rg3 were obtained from International laboratory (USA). All other biochemical reagents were from Sigma or Invitrogen.

Cellular senescence of cells was confirmed by a senescence-associated β-galactosidase activity assay as described by Dimri et al (13). After being grown in a semi-confluent state, senescence-associated β-galactosidase, pH 6.0, activity was examined. Cells were washed with phosphate-buffered saline (PBS) and fixed with 2% paraformaldehyde containing 0.2% glutaraldehyde in PBS for 5 min at room temperature (RT). After washing with PBS, cells were incubated with β-galactosidase reagent [1 mg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-gal), 40 mm citric acid/sodium phosphate buffer, pH 6.0, 5 mm potassium ferrocyanide/potassium ferricyanide, 150 mm NaCl, 2 mm MgCl₂] at 37°C.

Akt, p16, p21, p27 and p53 band were determined by western blotting. Briefly, cells were lysed with lysis buffer [50 mM Tris pH7.4, 150 mM NaCl, 1 mM EDTA pH 8.0, 1 mM protease inhibitor (Roche), 1 mM PMSF, 1 mM NaF, 1 mM sodium orthovanadate], and protein contents were determined using Bradford reagent. Equal amounts of protein (40 μg) were then separated by SDS-PAGE and transferred to nitrocellulose membranes (Schleicher & Schuell BioScience Inc). After blocking with TBS containing Tween-20 in the presence of 2.5% nonfat dry milk, the membranes were incubated with the primary antibodies at 4°C for 16 h. Secondary antibodies were added for 1 h at RT. The antibody-antigen complexes were detected using the ECL detection system (Pierce).

The amount of phosphatidylserine (PS) on cell surfaces (a measure of apoptosis) was determined using an Annexin V-FITC apoptosis detection kit (Abcam, USA), according to the manufacturer’s instructions. Briefly, following H₂O₂ or staurosporine treatment, cells were harvested and washed twice with PBS. The cells were then resuspended in 0.5 ml of cold 1X binding buffer (10 mM Hepes, pH 7.4, 150 mM NaCl, 2.5 mM CaCl₂, 1 mM MgCl₂, 4% BSA), and 2.5 μl of Annexin V-FITC (fluorescein isothiocyanate) was added. Following incubation for 15 min, stained cells were analyzed by flow cytometry (Becton-Dickinson FACSorter) using CELLQuest software. Error bars represent standard deviations of the means.

U87 cells were plated in a 100 mm dish, transfected with 0.5 nmol of siRNA and Oligofectamine reagent in serum-free medium and incubated for 4 h at 37°C in a CO₂ incubator. Following incubation, the cells were supplemented with growth medium containing 10% fetal bovine seerum. Cells were harvested after 72 h.

Cells were stained with 50 μM of DCF-DA for 30 min and then harvested. For quantitation of mitochondrial ROS, the cells were incubated with 0.2 μM MitoSOX for 30 min at 37°C, washed with PBS, trypsinized, collected in PBS, and analyzed on a FACSCalibur. To examine the effect of N-acetylcysteine (NAC), cells were treated with 20 mM of NAC for 2 days.

Cells were plated in 24-well tissue culture plates and allowed to attach overnight. MTT was added to each well to a final concentration of 200 μg/ml, and cells were incubated for 4 h. After removing the medium completely, the formazan product was solubilized with dimethylsulfoxide. Optical densities (OD) were measured at 490 nm. Each experiment was performed three times. Error bars represent standard deviations of the means.

Data are presented as the means ± S.D. and p-values were calculated using a Student’s t-test. A value of p<0.05 was considered to indicate a statistically significant difference. All data presented are the representative of at least three separate experiments.

To investigate the growth-regulatory activity of 20(S)-Rg3 in glioma cells, we treated glioma cells with various concentrations of 20(S)-Rg3. Exposure of U87 cells to 20(S)-Rg3 for 3 days resulted in a dose-dependent inhibition of cell proliferation (Fig. 1A and B). Concentrations of ≥10 μM 20(S)-Rg3 suppress the growth of U87 cells (Fig. 1B) whereas concentrations <10 μM exhibit little effect (data not shown). Next we examined the involvement of apoptosis in 20(S)-Rg3-induced growth arrest. Direct detection of apoptosis with Annexin V staining showed that apoptotic cells were not significantly increased in a sublethal dose of 20(S)-Rg3 (10 μM, 20 μM) compared with a high dose of 20(S)-Rg3 (50 μM, 100 μM) (Fig. 1C). We further examined the effects of chronic exposure of sub-apoptotic 20(S)-Rg3 on cell proliferation. Direct cell count assay indicated that the growth of U87 cells was gradually inhibited by chronic exposure of a sublethal dose of 20(S)-Rg3, which became more obvious on day 9 (Fig. 1D). 20(S)-Rg3 (20 μM) completely suppressed cell proliferation for at least nine days, while inducing only modest levels of cell death (data not shown). To test the possibility that cell growth arrest in response to 20(S)-Rg3 was caused by the induction of cellular senescence, we treated glioma cells with various sublethal concentrations of 20(S)-Rg3 and examined the expression of senescence-associated-β-galactosidase (SA-β-gal), a commonly accepted marker of senescence. As shown in Fig. 1E, ∼67% of U87 cells after chronic 20(S)-Rg3 treatment at a 20 μM concentration were stained positively compared with only ∼5% cells with positive staining in the DMSO control group (Fig. 1E). Cells that were positive for SA-β-gal showed an enlarged and flattened morphology that was consistent with cellular senescence (Fig. 1E). These findings indicated that 20(S)-Rg3 could induce senescence in addition to apoptosis. An immunoblot analysis showed that the level of p53 and cyclin-dependent kinase inhibitor p21CIP were markedly increased in 20(S)-Rg3-treated cells (Fig. 1F). However, there was no increase in the expression of p27Kip1 and only a moderate elevation of p16INK4 in 20(S)-Rg3-treated U87 cells (Fig. 1F).

As reactive oxygen species (ROS) are well-known inducers of senescence, we evaluated whether treatment of glioma cells with 20(S)-Rg3 increases oxidative stress. ROS levels, as assessed by dichlorofluorescein (DCF), were increased after 20(S)-Rg3 treatment of U87 glioma cells (Fig. 2A). As mitochondrial ROS are the major cellular source of ROS, we also examined the fluorescence of MitoSOX Red as a mitochondrial superoxide indicator. Fluorescence intensity of MitoSOX Red was significantly increased in U87 cells, consistent with increased intracellular ROS levels (Fig. 2B). To address the role of ROS in senescence induced by 20(S)-Rg3, we treated U87 cells with the ROS scavenger N-acetyl-l-cysteine before 20(S)-Rg3 exposure. NAC treatment reduced percentages of SA-β-gal-positive cells (Fig. 2C). These data suggest that elevated ROS levels contribute to senescence induction by 20(S)-Rg3.

It has been reported that PI3K/Akt signaling cascades are frequently deregulated in glioma and Akt activation is involved in glioma cell senescence (14,15). To examine whether Akt activity is involved in 20(S)-Rg3-induced senescent glioma cells, we measured phospho-Akt levels. U87 cells treated with 20 μM 20(S)-Rg3 were characterized by increases in phospho-Akt (Fig. 3A). To address whether the increase in levels of phospho-Akt is important for the induction of senescence in 20(S)-Rg3-treated human glioma cells, we employed si-RNA against Akt (Fig. 3B). As shown in Fig. 3C, depletion of Akt significantly decreased the SA-β-gal activity compared with control siRNA-infected cells. As ROS generation has been shown to be involved in the Akt in glioma cells (16), we next examined whether the increased ROS levels caused by 20(S)-Rg3 occurred as a result of activation of Akt or was a secondary effect of this drug. Transfection of Akt siRNA reduced ROS generation by approximately 32% (Fig. 3D). These data strongly suggest that AKT may play an important role in regulating cellular ROS status and senescence in glioma cells.

The p53/p21 signaling pathway is a major senescence-triggering pathway in response to various stresses. As shown in Fig. 1F, the treatment of U87 cells with 20(S)-Rg3 significantly increased the expression levels of p53 and p21. Therefore, we examined whether 20(S)-Rg3-mediated senescence requires signaling of the p53/p21 pathway in U87 cells. We transfected U87 cells with p53 and/or p21 si-RNA and looked for senescent phenotypes. As shown in Fig. 4B, p53 siRNA transfected cells showed a significant decrease in SA-β-Gal activity. Cells transfected with p21 siRNA or co-transfected p53 siRNA showed similar results with p53 depleted cells (Fig. 4B). These data suggest that the p53/p21 plays an essential role in the maintenance of senescence triggered by 20(S)-Rg3.