The P. indica plants used in experiments in this study were collected and verified as described in a previous study (20). The roots of P. indica were prepared as a dried powder and deposited at the National Sun Yat-sen University (Kaohsiung, Taiwan, R.O.C.) and the H-PIRE was extracted using the method described by Kao et al (21). Briefly, the P. indica extract was prepared by immersing the root powder in 95% (v/v) ethanol at room temperature overnight, filtering with 0.45 µm filters, then partitioning the water and ethyl acetate (v/v, 1:1). The ethyl acetate fraction was further partitioned with a 1:1 (v/v) mixture of 75% ethanol and hexane. The hexane layer was collected and concentrated to obtain powdered H-PIRE. The powdered H-PIRE was then dissolved in dimethyl sulfoxide (DMSO) to desired concentrations in subsequent experiments.

Total phenolic compound content was determined with Folin-Ciocalteu reagent and presented as gallic acid equivalents in mg/g of extract. Total flavonoid content was determined by aluminum chloride colorimetric method and expressed as catechin equivalents in mg/g of extract (22). Condensed tannin (proanthocyanidin) content was analyzed by vanillin assay method as described by Butler et al (23) and presented as catechin equivalents in mg/g of extract. All assays were performed in triplicate.

Primary GBM cells were enzymatically dissociated from fresh glioblastoma samples as described by Pavon et al (24), according to a protocol approved by the Kaohsiung Medical University Institutional Review Board (Kaohsiung, Taiwan, R.O.C.). The U87 and primary GBM cell lines were cultured in Dulbecco's modified Eagle's medium with 10% fetal bovine serum (all from HyClone; GE Healthcare Life Sciences, Logan, UT, USA), 200 mM l-glutamine (Invitrogen Life Technologies; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 1% penicillin (10,000 U/ml)/streptomycin (10,000 U/ml) (HyClone; GE Healthcare Life Sciences Hyclone), 1X non-essential amino acids and 1 mM sodium pyruvate (both from Invitrogen Life Technologies; Thermo Fisher Scientific, Inc.). The cells were cultured in 5% CO₂ at 37°C in a humidified incubator. Institutional board approval was obtained before experimental use of primary GBM cells derived from patients.

Cell viability was determined using the WST-1 colorimetric assay method. The cells were cultured overnight in 96-well plates (2×10³ cells/well) and treated with 0–1,000 µg/ml H-PIRE or DMSO (negative control) for 24 to 72 h. The medium was then carefully removed and replaced with fresh medium containing the cell proliferation reagent WST-1 (Roche Diagnostics GmbH, Mannheim, Germany) and the cells were incubated for an additional 3 h at 37°C. Absorbance was measured at 450 nm with an ELISA plate reader (Micro Quant; BioTek Instruments, Inc., Winooski, VT, USA). The IC₅₀ values were determined as previously described (20) and presented as mean ± standard deviation (SD) (n=3).

Apoptosis was detected with an fluorescein isothiocyanate (FITC) Annexin V apoptosis detection kit I (BD Pharmingen, BD Biosciences, Franklin Lakes, NJ, USA). After 24 and 48 h of incubation with H-PIRE or DMSO, 1×10⁵ cells were collected and suspended in 100 µl staining buffer containing 5 µl of FITC Annexin V and 5 µl propidium iodide (PI) at room temperature in the dark. After 15 min, 400 µl 1X binding buffer was added and flow cytometric analysis was performed as previously described (20).

Cells were cultured in 60-mm dishes until 60% confluent and then treated with 300 µg/ml H-PIRE for 24 h. The cells were then stained with 1 mg/ml acridine orange (AO) for 15 min and washed with phosphate buffer saline (PBS) to detect acidic vesicular organelles (AVOs). A fluorescence microscope (Olympus IX70; Tokyo, Japan) with an excitation filter set to 490 nm and a 515 nm long-pass filter was used for cell analysis and digital imaging.

The cells were cultured in 60-mm dishes until 60% confluent and then treated with 300 µg/ml H-PIRE for 0 to 24 h. After incubation, the cells were washed with PBS and lysed in T-PER protein extraction reagent (Pierce). Cell lysates containing 20 µg total protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes. Antibodies against p-p38 (1:1,000, rabbit monoclonal, Cell Signaling Technology cat. no. 4511), p-JNK (1:1,000, rabbit monoclonal, Cell Signaling Technology, cat. no. 4668), LC3B (1:1,000, rabbit monoclonal, Cell Signaling Technology cat. no. 3868) and β-actin (1:10,000, mouse monoclonal, Sigma-Aldrich,) were used for immunoblotting. The immunoblots were visualized using enhanced chemiluminescence detection reagent (Amersham; GE Healthcare, Chicago, IL, USA). A LAS-3000 CCD system was used for imaging and a Multi Gauge V2.02 both from (Fujifilm Holdings Corporation, Tokyo, Japan) was used for analysis.

Data were expressed as the mean ± SD of three independent experiments. Statistical significance was determined by performing independent Student's t-test with SPSS 12.0 software (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Table I shows that the preliminary phytochemical analysis of H-PIRE revealed a high content of phenols, flavonoids and proanthocyanidins, which are biologically active phytochemical compounds found in many medicinal herbal extracts. Secondary metabolites such as polyphenols and flavonoids interact with proteins and are useful for inhibiting multidrug resistance (25), suppressing tumor angiogenesis and preventing inflammation (26). Thus, the phytochemical composition of H-PIRE was consistent with the reported medicinal properties of P. indica.

The cytotoxic effects of H-PIRE on GBM cells were examined by assays of WST-1 cell proliferation. Two GBM cell lines were treated with varying concentrations of H-PIRE (range, 0–1,000 µg/ml) at 24 and 48 h. Fig. 1 shows the experimental results. In both cell lines, H-PIRE reduced cell viability in a dose-dependent manner. At 24 and 48 h, the IC₅₀ values (mean ± SD, µg/ml) for H-PIRE were 422.6±38.3 and 353.3±18.7 in U87 cells and 416.4±35.5 and 334.1±29.6 in GBM cells, respectively. That is, the IC₅₀ values indicated that H-PIRE suppressed proliferation in both human GBM cell lines.

The effect of H-PIRE on the cell cycle was investigated by flow cytometry. The U87 and GBM cells were incubated with 300 µg/ml H-PIRE for 24 h. The incubated cells were then collected, fixed and stained with PI for flow cytometric analysis of DNA content. Fig. 2 shows that the H-PIRE treatment significantly decreased the cell populations in the S and G2/M phases. However, the sub-G1 population showed no increase, which indicated that the treatment did not induce apoptosis. Thus, H-PIRE suppressed progression of the human GBM cell cycle through the G0/G1 phase.

Apoptotic effects of H-PIRE were investigated by flow cytometry of unfixed cells stained with Annexin V and PI. Fig. 3 shows that, at 24 and 48 h after treatment with 300 µg/ml H-PIRE, the U87 and GBM cells showed significantly increased percentages of dead cells in quadrant 3 but no increase in early apoptosis in quadrant 2. These experimental results suggest that cell death induced by H-PIRE in glioblastoma cells is probably not associated with apoptosis.

Formation of AVOs, which is a morphological characteristic of cells undergoing autophagy, can be detected by fluorescence microscopy of living cells stained with AO, which emits red fluorescence at low pH. The staining results in Fig. 4 show that AVO expression was increased in U87 and GBM cells at 24 h after treatment with 300 µg/ml H-PIRE. These results clearly indicate that H-PIRE induces autophagy in human glioblastoma cells.

The mechanism of cell death induced by H-PIRE was investigated by examining the expression of autophagy-related proteins in U87 and GBM cells. The cells were treated with 300 µg/ml of H-PIRE for 0 to 24 h. Western blot analysis was then used to examine the expression of microtubule-associated protein light chain 3 (LC3)-I, LC3-II, phosphorylated-p38 and phosphorylated-JNK. Fig. 5A shows that, at 1 h after the H-PIRE treatment, LC3-II was significantly upregulated relative to LC3-I, which significantly increased autophagosome formation (27). In both cell lines, expression of phosphorylated-p38 protein was significantly increased at 1 h after H-PIRE treatment and peaked at 3 h after treatment. Phosphorylated-p38 protein expression remained at the peak level for >12 h in U87 cells and for 24 h in GBM cells before dropping back to the base level. Expression of phosphorylated-JNK protein was substantially increased at 1 h after H-PIRE treatment and then gradually decreased to the base level in the two cell lines. The results indicated that, in both cell lines, the stress induced by H-PIRE activated the p38 and JNK signaling pathways, which are known to play key roles in the induction of autophagy (27).

Fig. 5B shows that the LC3-II/LC3-I ratio increased in a time-dependent manner in both cell lines. The ratio peaked at 0.30 at 1 h and started to decline after 12 h in U87 cells. By contrast, the ratio increased steadily and peaked at 0.73 at 24 h in GBM cells. In addition, H-PIRE treatment resulted in different FITC-AV and PI staining pattern at 48 h (Fig. 3) and cell morphology at 24 h (Fig. 4) in U87 and GBM cells. These data may reflect the intrinsic difference in the manner the two types of cells responded to autophagic stress. The overall results indicate that H-PIRE activates p38 and JNK and induces autophagy in human GBM cells.

Rapamycin and LY294002 (a pan-PI3K inhibitor) reportedly exert synergistic effects on the induction of autophagy and reduction of cell survival in malignant glioma cell lines, including U87 (28). Specifically, rapamycin induces autophagy by deactivating mTOR (29). To determine whether inhibition of the PI3K pathway and H-PIRE treatment synergistically affect cell viability, U87 cells were treated with H-PIRE (0–1,000 µg/ml) with or without 10 µM LY274002. Fig. 6 shows the cell viability assay results, which indicated that the addition of LY294002 enhanced the reduction of cell viability induced by H-PIRE treatment. The addition of LY294002 reduced IC₅₀ of H-PIRE from 353.3±18.7 to 298.6±20.3 µg/ml. These results indicate that the combined treatment of H-PIRE and LY294002 have additive effects on the reduction of cell survival in human GBM cells.