S. chinensis was collected between June and July 2013 from a local site in Jianguo, China, and the plant material was confirmed by Professor JW Chen (College of Pharmaceutical Science, Nanjing University of Chinese Medicine, Nanjing, China). The aerial parts of S. chinensis were washed thoroughly with tap water, air dried and then sectioned into small pieces. Methanol (95%) was used for the hot extraction, which was performed after 4 h using a Soxhlet extraction apparatus (BSXT-02; Shanghai Bilon Instrument Co., Ltd. Shanghai, China). In this method, the finely ground crude drug is placed in a porous bag made of strong filter paper, which is placed in chamber E of the Soxhlet apparatus The extract was concentrated under reduced pressure in a rotary evaporator at 45°C, and was maintained at in a refrigerator at 4°C prior to use.

The LC-MS equipment consisted of a chromatographic system (LC-MS Infinity; Agilent Technologies, Inc., Santa Clara, CA, USA) coupled with an Agilent 1100 Series LC system (Agilent Technologies, Inc.), which was equipped with a binary solvent delivery system, auto-sampler, column temperature controller, photo diode array detector and Finnigan LCQ Deca XP Plus ion trap mass spectrometer (Thermo Finnigan; Thermo Fisher Scientific, Waltham, MA, USA) via an ESI interface. MS spectra were obtained using positive and negative modes; nebulizer gas, 45 Psi; capillary voltage, 4,000 V. The operating parameters for MS were as follows: Collision gas, ultrahigh-purity helium (He); nebulizing gas, high purity nitrogen (N₂); ion spray voltage, −5.5 kV; sheath gas (N₂) at a flow rate of 70 arbitrary units; auxiliary gas (N₂) at a flow rate of 30 arbitrary units; capillary temperature, 360°C; capillary voltage, −15 V; and tube lens offset voltage, −30 V. Full scan data acquisition was performed between 80 and 1,800 m/z in MS scan mode.

HPLC analysis was performed on an Agilent 1260 Infinity series (Agilent Technologies, Inc.) using a Chromolith RP-18e column (4.6 mm ID, 60 mm length). The mobile phase consisted of (A) 0.5% aqueous acetic acid and (B) methanol. Mobile phase gradient: 0–10 min, linear gradient between 10 and 20% of B; 10–15 min, isocratic conditions at 25% of B; 15–20 min, linear gradient between 25 and 40% of B; 20–40 min, linear gradient between 40 and 50% of B; 40–50 min; linear gradient between 50 and 100% of B. Flow rate: 1.5 ml/min.

RPMI-1640 growth medium was purchased from Hangzhou Sijiqing Biological Products Co., Ltd. (Hangzhou, China). Fetal calf serum (Gibco Life Technologies, Carlsbad, CA, USA), trypsin, penicillin, MTT, streptomycin, dimethyl sulfoxide and phosphate-buffered saline (PBS) were used in the present study (all purchased from Sigma-Aldrich, St. Louis, MO, USA). The MTT kit was obtained from Roche Diagnostics (Indianapolis, IN, USA). Camptothecin was used as a positive control for the mitochondrial membrane loss and was purchased from Sigma-Aldrich. An Annexin V-Fluorescein Isothiocyanate (FITC)-Propidium Iodide (PI) Apoptosis Detection kit was purchased from Sigma-Aldrich. All other chemicals and solvents used were of the highest purity grade. Cell culture plasticware was purchased from Falcon^® (Corning Life Sciences, Tewkesbury, MA, USA).

MCF-7 human breast cancer cells, A549 human lung cancer cells, HCT-116 human colon cancer cells, COLO-205 human colon cancer cells, MiapaCa-2 human pancreatic cancer cells, and the normal cell line, NIH-3T3 mouse embryonic fibroblasts, were obtained from the Shanghai Institute of Cell Resource Center of Life Science (Shanghai, China). All the cell lines were cultured in a humidified atmosphere of 5% CO₂ at 37°C in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum, 100 IU/ml penicillin and 100 µg/ml streptomycin.

The inhibition of cell proliferation following treatment with the extract was measured using an MTT assay. Briefly, the MCF-7, A-549, HCT-116, COLO-205, MiapaCa-2 and NIH-3T3 cells were plated in separate 96-well culture plates (1×10⁵ cells/well). Following 24 h incubation at 37°C, the cells were treated with the polyphenol-rich extract (10, 20, 40, 60, 80 and 100 µg/ml; eight wells per concentration) for 12, 24 or 48 h. MTT solution (5 mg/ml) was subsequently added to each well. Following incubation for 4 h, the formazan precipitate was dissolved with 100 µl dimethyl sulfoxide, and the absorbance was measured using an ELISA reader (SpectraMax Plus 384 microplate reader; Molecular Devices, LLC, Sunnyvale, CA, USA) at a wavelength of 570 nm. The cell viability ratio was calculated using the following formula: Inhibitory ratio (%) = [(ODcontrol − ODtreated) / (ODcontrol)] × 100%; OD, optical density. Cytotoxicity was expressed as the half maximal inhibitory concentration.

Fluorescence microscopy was performed to evaluate morphological alterations in the MiapaCa-2 cancer cells, following treatment with the extract. The cells (1×10⁶ cells/ml) were seeded in 6-well plates and treated with the extract (20, 40, 60 and 80 µg/ml concentrations). Following 24 h incubation at 37°C, the cells were centrifuged at 112 × g for 5 min at 4°C. The resuspended pellet was then dissolved in PBS. The air-dried smears were then fixed in methanol at −20°C, stained with 4′,6-diamidino-2-phenylindole (DAPI; 2 µg/ml) and incubated at 37°C for 20 min. The culture plates were subsequently observed under an inverted light microscope (Eclipse Ti-E; Nikon Corporation, Tokyo, Japan) for morphological analysis.

To establish and confirm that the cells were undergoing apoptosis, an annexin V binding assay was performed using flow cytometry. Briefly, the MiapaCa-2 pancreatic cancer cells (2×10⁶ cells/ml) were treated with the polyphenol-rich extract at 20, 40, 60 and 80 µg/ml for 24 h at 37°C. Subsequently, the treated and untreated cells were harvested by trypsinization. The harvested cells were then incubated with annexin V-FITC (80 ng/ml) and PI (50 µg/ml), at room temperature in the dark for 20 min, and analyzed using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA). A minimum of 2×10⁴ cells were measured in each sample.

For the detection of DNA fragmentation, the cells were lysed in a solution containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA and 0.5% Triton X-100) at room temperature for 30 min. The lysates were then vortexed and cleared by centrifugation at 15,000 rpm for 15 min. DNA was extracted from the supernatant using a 20:20:10 (v/v/v) equal volume of neutral phenol:chloroform:isoamyl alcohol. The DNA was then separated by electrophoresis on 1.0% agarose gels containing 0.1 µg/ml ethidium bromide (Sigma-Aldrich), and DNA fragmentation was detected under UV illumination.

MiapaCa-2 cells (5×10⁶) were seeded into 60 mm dishes and subjected to various concentrations (20, 40, 60 and 80 µg/ml) of polyphenol-rich extract for 48 h at 37°C. The floating and adherent cells were collected by trypsinization and washed twice with PBS. The remaining cells were then incubated in 70% ethanol at −20°C overnight, treated with 10 µg/ml RNase A, and stained with 2.0 µg/ml PI. The stained cells were subsequently analyzed using flow cytometry at a wavelength of 488 nm (FACSCalibur; BD Biosciences), equipped with CellQuest 3.3 software.

Mitochondrial membrane potential (ΛΨm) was measured using 1 mM Rhodamine-123 (Rh-123) dye. Rhodamine fluorescence can be used as a measure of membrane polarization in live cell assays within mitochondria. Briefly, 5×10⁵ MiapaCa-2 cells were treated with different concentrations (20, 40, 60 and 80 µg/ml) of the polyphenol-rich extract for 48 h at 37°C. Subsequently, ΛΨm was measured using flow cytometry (FACSCalibur; BD Biosciences). Rh-123 (2 mM) was added 1.5 h prior to termination of the experiment. The cells were then collected, washed in PBS and incubated with PI (10 µg/ml) for 15 min at room temperature. The reduction in fluorescence intensity, due to loss of ΛΨm, was analyzed using flow cytometry. The mean fluorescence intensity was detected using the FL1 channel of the FACSCalibur.

All the data were analyzed using a one-way analysis of variance, followed by Dunnett's test for pair wise comparison with GraphPad Prism 4.0 (GraphPad Software, Inc., La Jolla, CA, USA). Values are presented as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

The polyphenol-rich extract was evaluated for antiproliferative activity using an MTT assay. The MCF-7 human breast cancer cells, A549 human lung cancer cells, HCT-116 human colon cancer cells, COLO-205 human colon cancer cells, MiapaCa-2 human pancreatic cancer cells and NIH-3T3 mouse embryonic fibroblast normal cell line were treated with the extract for 24 h (Fig. 1). The extract exhibited potent dose-dependent cytotoxic activity against the different cancer cell lines. The COLO-205 and MCF-7 cancer cells were the most susceptible to treatment with the extract, and exhibited increased growth inhibition. The HCT-116 and MiapaCa-2 cells exhibited higher levels of growth inhibition only following treatment with higher concentrations of the extract. In order to examine the toxic effects of the extract on normal cells, the cytotoxic effects of the extract were assessed against the NIH-3T3 mouse embryonic fibroblast cell line. The extract demonstrated reduced cytotoxicity towards the normal cell line, compared with the cancer cell lines, suggesting that its effects are specific to cancer cells. The effect of the polyphenol-rich extract on the growth of MiapaCa-2 pancreatic cancer cells was evaluated using an MTT assay at three different time intervals (12, 24 and 48 h). The cytotoxic effect of the extract on the cells was dose- and time-dependent. At increased time intervals, high levels of growth inhibition were observed (Fig. 2).

In order to establish whether the polyphenol-rich extract of S.chinensis induced apoptosis in MiapaCa-2 cells, the cells were treated with various concentrations of the extract (0, 20, 40,60 and 80 µg/ml) for 48 h. Subsequently, the representative morphological features of apoptosis were examined under an inverted light fluorescence microscope, using DAPI as a staining agent. As shown in Fig. 3, compared with the untreated viable cells, treatment with the extract resulted in the appearance of cell contraction and membrane blebbing, both of which are distinguishing features of apoptosis. When treated with a higher concentration of the extract (100 µg/ml), the majority of the cancer cells had shrunk considerably, and no cells exhibiting normal morphological features were detected.

The translocation of phosphatidylserine to the exterior surfaces of the plasma membrane is a distinguishing feature of early apoptosis, which can be identified and detected by annexin V-FITC binding. If cell death occurs, fragmented and damaged DNA becomes permeable for binding with PI (34). Following staining of cells with annexin V in combination with PI, this reagent enters the cells only when the plasma cell membrane has deteriorated. In the present study, flow cytometry revealed that, in the extract-treated cells, a higher number of annexin V-positive cells were detected, compared with the untreated control cells (Fig. 4A and B). The percentage of apoptotic cells was low following treatment with lower concentrations of the extract. However, at higher extract concentrations (60 and 80 µg/ml), the total number of apoptotic cells increased considerably. This assay provided a quantitative estimation of the rate of apoptotic cell death following drug exposure.

Apoptosis and cell cycle dysfunction are closely associated biochemical processes, and any disturbance in cell cycle progression may lead to apoptotic cell death (35). In order to determine the mechanism underlying the growth inhibitory effect of the extract on MiapaCa-2 cancer cells, flow cytometric analysis was performed to detect whether the extract induced cell cycle arrest. Treatment with different concentrations of the extract for 48 h induced G₀/G₁-phase growth arrest in the MiapaCa-2 cells. As shown in Fig. 5, following treatment of the MiapaCa-2 cells with different concentrations of the extract (20, 40, 60 and 80 µg/ml), considerable G₀/G₁ cell cycle growth arrest was observed. The apoptotic cells were observed as shrunken cells with degraded chromatin, increased side scatter and decreased forward scatter properties. The increase in the sub-G₁ cell population (hypodiploid DNA content) may be due to DNA fragmentation, which eventually results in apoptotic cell death. Inhibition of cell cycle progression may be one of the molecular events associated with the cytotoxic activities of the extract. Following treatment of the cells with low concentrations of the extract, no significant differences were observed in the levels of apoptosis; however, following treatment with extract concentrations of 60 and 80 µg/ml, the percentage of cells undergoing apoptosis (G₀/G₁ arrest) increased significantly.

Depolarization of the mitochondrial membrane and subsequent seepage of the outer membrane is a key step in the intrinsic apoptotic pathway. This is usually followed by the release of cytochrome c and pro-apoptotic molecules (36). The present study used the fluorescent probe, Rh-123, to detect the ΛΨm in living cells. Treatment with the extract induced a substantial reduction in the number of cells with intact membrane potential, and increased the number of cells with a low ΛΨm after 48 h (Fig. 6). Loss of ΛΨm is a crucial event in the mitochondrial apoptotic pathway. The loss of ΛΨm was found to exhibit dose-dependence, and the number of cells with reduced ΛΨm increased with increasing concentrations of the extract.

A DNA fragmentation assay also revealed that treatment with the extract resulted in DNA laddering, which is indicative of apoptosis (Fig. 7A). DNA fragmentation in the polypheno-rich extract-treated cells was confirmed using agarose gel electrophoresis, which detected the presence of DNA laddering, a marker of apoptosis, in the extract-treated MiapaCa-2 cells. By contrast, the untreated control cells demonstrated no evidence of DNA laddering. As shown in Fig. 7B, the number of cells exhibiting degraded DNA (sub-G₁ DNA content) increased in a dose-dependent manner.

In the present study, analysis of the S. chinensis methanol extract was performed using LC-ESI-MS as well as HPLC techniques. The extract was run under positive and negative ESI-MS conditions, and demonstrated several major and minor peaks. The total ion MS chromatogram, and the structure of the identified molecules are shown in Figs. 8 and 9, respectively. Fragmentation of the major peaks was used for identification of the compounds. Identification of the chemical compounds was also achieved by comparing the molecular ion peaks and MS fragmentation patterns with those described in the literature. The six chemical constituents identified in the extract were as follows: Protocatechuic acid, salvianolic acid B, salvianolic acid D, xeractinol, kaempferol and apigenin. All these compounds belong to the polyphenol class of natural products.