The seeds of P. suffruticosa (1.2 kg) were collected in Tongling, Anhui province, China, and identified in 2012. The sample was authenticated by Professor Peigen Xiao and Dr Chunnian He from the Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences. A voucher specimen (2012001) was deposited in the Seed Resource Bank of the Institute of Medicinal Plant Development and Peking Union Medical College, Beijing, China.

The ten oligostilbenes, trans-resveratrol, cis-ε-viniferin, trans-ε-viniferin, suffruticosol A, suffruticosol B, suffruticosol C, cis-suffruticosol D, trans-suffruticosol D, cis-gnetin H and trans-gnetin H, were simultaneously purified from the dried seeds of P. suffruticosa as described previously (16). Their structures were characterized by ultraviolet (UV), infrared (IR), mass and nuclear magnetic resonance (NMR) spectroscopy, and the purities of all compounds were determined to be >95% (16). The compounds were resuspended in dimethyl sulfoxide (DMSO) (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) to yield a concentration of 10 mM, and stored at 4°C.

Six human cancer cell lines were used in this study, including lung carcinoma (A549), breast carcinoma (BT20, MCF-7 and MDA-MB-231), osteosarcoma (U2OS) and cervix adenocarcinoma (HeLa). All cell lines were purchased from American Type Culture Collection (Manassas, VA, USA), except for the HPL1A cell line, which was obtained from Nagoya University, Japan. A549, BT20 and HeLa cells were grown in RPMI-1640 medium (Sigma-Aldrich; Merck Millipore). MCF-7 and MDA-MB-231 cells were grown in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma-Aldrich; Merck Millipore) supplemented with 2 mM L-glutamine (Sigma-Aldrich; Merck Millipore). U2OS cells were grown in McCoy's 5A medium (ATCC). HPL1A cells were grown in DMEM/F-12 medium (Sigma-Aldrich; Merck Millipore). All mediums were supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Inc., Waltham, MA, USA), 1% penicillin and streptomycin, and incubated in a humidified atmosphere containing 5% CO₂ at 37°C.

Anti-proliferation activity was determined using a fluorescent staining assay. Cells were seeded into a 96-well tissue culture plate at a density of 4,000 cells/well and treated with the one of the ten compounds at final concentrations of 100, 50, 25, 12.5, 6.25 or 3.13 µM for 48 h. Cells treated with vehicle (1% DMSO) only was also included as an experimental control. Subsequently, 10% AlamarBlue dye (Invitrogen; Thermo Fisher Scientific, Inc.) was added to the medium and the cells were incubated at 37°C in the CO₂ incubator for 1 h. The fluorescence intensity was read in a SpectraMax M5 microplate reader (Molecular Devices, LLC, Sunnyvale, CA, USA) at excitation/emission (Ex/Em) wavelengths of 550/590 nm. The results are expressed as a percentage relative to the untreated control, and the IC₅₀ values were calculated using non-linear regression analysis with GraphPad Prism software 6.0 (GraphPad Software, Inc., La Jolla, CA, USA).

MDB-MA-231 cells were seeded into a 96-well plate at a density of 8,000 cells/well, and treated with the test compound at 50 µM for 24 h. Apoptosis was assessed using a live-cell assay by fluorescein isothiocyanate (FITC)-Annexin V and propidium iodide (PI) staining, as described previously (20,21). Briefly, 10 µl of 10X binding buffer containing 1 µl Hoechst 33342, 5 µl FITC-Annexin V and 5 µl mg/ml PI (BD Biosciences, Franklin Lakes, NJ, USA) was added to cell culture medium following treatment with test compounds. Staurosporine (Sigma-Aldrich; Merck Millipore) at a concentration of 1 µM served as a positive control and cells treated with vehicle only served as a negative control. Removal of cell culture medium from wells was avoided because necrotic and poorly attached cells would be detached and removed during this process. Immediately following incubation at room temperature in the dark for 10 min, cells were imaged using an ArrayScan VTI High-Content Screening (HCS) reader (Thermo Fisher Scientific, Inc.) and the fluorescence intensity of each channel (Hoechst 33342, FITC-Annexin V and PI) was analyzed using HCS Studio software 6.5.0.

Caspase-3/7 activity was determined using the SensoLyte Homogeneous AMC Caspase-3/7 Assay kit (AnaSpec, Fremont, CA, USA), according to the manufacturer's instructions. Briefly, MDA-MB-231 cells (8,000 cells/well) were seeded into a 96-well plate and incubated at 37°C overnight. The cells were treated with test compounds at 50 µM at 37°C for 24 h. Staurosporine at a concentration of 1 µM served as a positive control and cells treated with vehicle only served as a negative control. After 24 h, 50 µl caspase-3/7 substrate solution was added to each well, and the plate was incubated in the dark at room temperature for 1 h. Fluorescence intensity was measured using a SpectraMax M5 microplate reader at Ex/Em wavelengths of 350/440 nm. Caspase-3/7 activities were calculated by subtracting the fluorescence levels of wells containing medium only.

Prior to performing the in vitro migration and invasion assay, the cytotoxicity of the test compound was determined after a 16-h treatment. Only the concentration that had <20% toxicity on MDA-MB-231 cells was used for the migration and invasion assay. The in vitro migration assay was performed as described previously (22), with slight modifications. Millicell^® Cell Culture Inserts (8-µm pore size; EMD Millipore, Billerica, MA, USA) were placed into a 24-well cell culture plate. MDA-MB-231 cells (2×10⁵ cells/ml) in 500 µl serum-free medium were seeded into the upper chamber. The lower chamber was filled with 500 µl culture medium supplemented with 10% FBS. Each compound was added to the medium in both the upper and lower chambers to the desired concentration (10 µM). Doxycycline (Sigma-Aldrich; Merck Millipore) at a concentration of 100 µM served as a positive control and cells treated with vehicle only served as a negative control. The cells were incubated for 16 h to allow the cells to migrate through the filter pores. Subsequently, cells on the upper surface of the filter membrane were removed using a sterile cotton swab. Cells that had migrated to the lower surface were stained with 10% AlamarBlue dye at 37°C for 1 h and read in a SpectraMax M5 microplate reader at Ex/Em wavelengths of 550/590 nm.

The in vitro invasion assay followed the same procedure as the in vitro migration assay except that the filter inserts were pre-coated with Matrigel matrix. Briefly, 100 µl Matrigel matrix coating (Corning Incorporated, Corning, NY, USA) at 5 mg/ml was added to each insert and incubated at 37°C for 2 h to form a continuous thin layer.

All experiments were performed in triplicate, and each experiment was repeated 3 times. Data were presented as the mean ± standard deviation. Statistical analysis was performed using GraphPad Prism 6.0 (GraphPad Software, Inc.). Data were analyzed using one-way analysis of variance tests, where P<0.05 was considered to indicate a statistically significant difference.

The anti-proliferative activity of ten oligostilbenes was initially evaluated in six different human cancer cell lines (A549, MCF-7, BT20, MDA-MB-231, U2OS and HeLa). All oligostilbenes showed mild-to-potent in vitro cytotoxicity against these human cancer cells, and the oligomers of resveratrol showed superior antitumor activities compared with the resveratrol monomer (Table I). Concentration-dependent anti-proliferation effects were observed for all oligostilbenes in most of the tested cancer cell lines after 48 h of treatment. Cis- and trans-gnetin H showed the most potent anti-proliferation activities, with IC₅₀ values ranging from 0.9 to 10.0 µM against the six cancer cell lines, representing a >20-fold increase in potency compared with resveratrol (data not shown). Generally, trans-ε-viniferin, trans-suffruticosol D and trans-gnetin H were more potent than their respective cis-forms, cis-ε-viniferin, cis-suffruticosol D and cis-gnetin H. Additionally, in most of the cancer cell lines, with the exception of A549, the trimers of resveratrol, suffruticosol A-C, cis- and trans-suffruticosol D and cis- and trans-gnetin H were more potent than the dimers of resveratrol cis- and trans-ε-viniferin.

It is notable that all the tested oligostilbenes inhibited the proliferation of three representative subtypes of breast carcinoma cells, including Basal A phenotype BT20 cells [estrogen receptor (ER)⁻ progesterone (PR)⁻ human epidermal growth factor receptor 2 (HER2)⁻], Luminal A phenotype MCF-7 cells (ER⁺PR⁺HER2⁻) and Basal B phenotype MDA-MB-231 cells (ER⁻PR⁻HER2⁻). Generally they showed higher potency against BT20 cells than MCF-7 and MDA-MB-231 cells. Since Basal B cells are much more invasive than Basal A and luminal cells (23), Basal B phenotype MDA-MB-231 cells were selected for use in the following experiments to evaluate the anti-metastasis activity of the compounds.

In an examination of the cancerous cells using an inverted microscope, it was observed that A549, BT20, MCF-7, MDA-MB-231, U2OS and HeLa cells treated with oligostilbenes at various concentrations (3.13–100 µM) for 24 and 48 h showed the characteristic morphological changes of apoptosis, including cell shrinkage and apoptotic bodies (data not shown). Therefore, a multiplex apoptosis assay of MDA-MB-231 cells was conducted using HCS to confirm the occurrence of apoptosis. Hoechst 33342 was used to stain the nucleus, and FITC-Annexin V and PI were used as indicators of apoptosis. To compare the apoptotic effect, the same dose of 50 µM was used for all oligostilbenes and the cells were treated for 24 h. As shown in Fig. 2A, cells treated with vehicle only exhibited blue fluorescence due to Hoechst staining, cells undergoing early apoptosis exhibited green fluorescence due to Hoechst and FITC-Annexin V staining, and cells undergoing late apoptosis exhibited green and red fluorescence due to FITC-Annexin V and PI staining, respectively. In cells treated with 50 µM oligostilbenes, both FITC-Annexin V and PI staining exhibited a significant increase compared with the negative control (P<0.01 or P<0.001; Fig. 2B). These results suggest that the oligostilbenes inhibit cancer cell growth by inducing apoptosis in the cancer cells.

To further examine the activities of caspase-3/7, which are the major effectors activated by the formation of the apoptosome in apoptotic cells (24), a fluorescent assay of MDA-MB-231 cells was performed. Subsequent to a 24-h treatment with oligostilbenes, caspase-3/7 levels were significantly increased in cancer cells. All oligostilbenes at 50 µM concentration had induced caspase-3/7 activity levels by at least 3-folds (P<0.001; Fig. 2C). Among the oligostilbenes, trans-gnetin H showed the most significant effect. The caspase-3/7 levels in trans-gnetin H-treated cells were 46-fold higher compared with the negative control. This result was consistent with the multiplex apoptosis assay where trans-gnetin H also showed the highest Annexin V and PI staining.

Subsequently, the effects of oligostilbenes at a low-toxicity dosage on the migration and invasion of MDA-MB-231 breast cancer cells were investigated. Since all oligostilbenes displayed <20% cytotoxicity on the MDA-MB-231 cells at the dose of 10 µM following 18 h of treatment (Fig. 3A), we selected this concentration for the subsequent migration and invasion assays. All oligostilbenes inhibited the migration and invasion of MDA-MB-231 cells in vitro. In the migration assay, all oligostilbenes significantly affected the number of cells that migrated through the pores of the filter insert; the inhibition rate ranged from 24.4 to 88.9% (P<0.001; Fig. 3B). In the invasion assay, only those cells that passed through the Corning Matrigel Matrix layer and the filter membrane were detected, and the inhibition rate by oligostilbenes ranged from 22.0 to 54.6% (P<0.001; Fig. 3C).