This article is mentioned in:

Introduction

Current cancer medications are costly and often cause serious side effects. The US National Cancer Institute began investigating antitumor plant extracts in the 1960s, and the premise that natural compounds obtained from therapeutic plants could produce anticancer medications has henceforth been of great research interest. Traditional Chinese medicines (TCMs) using dried plants or plant extracts have provided low cost diet and pharmaceutical therapies for thousands of years and experimental and clinical studies have proven that >400 plant species used in TCMs as anticancer herbal medications are significantly effective in the prevention or treatment of various cancers (1–4). However, much work remains to be done to determine the effectiveness of the individual compounds present in the TCMs.

Paeonia suffruticosa, or Paeoniaceae, is a widely utilized Chinese medicinal plant within the Paeonia genus. This genus comprises ~35 species that are classified into three groups: Oneapia, Paeonia, and Moutan (5). The Cortex Moutan (root cortex) of Paeonia has been recorded by China's Pharmacopoeia as a significant source of herbal medicine (6). Extracts of Paeonia have been shown to possess cytotoxic, antitumor, anti-inflammatory and anti-oxidative activities (5). Previous photochemical research on Paeonia identified >260 bioactive compounds, including phenols, monoterpenoidglucosides, paeonols, flavonoids, tannins, steroids, triterpenoids and stilbenes (7). A more recent study showed that the seeds of Paeonia contain considerable quantities of stilbenes compared to the other compounds (7,8).

Stilbenes are a class of polyphenols widely found in plants that contain a 1.2-diphenylethylene nucleus in their structure (9). Stilbenes have aroused great interest due to their antitumor, anti-steroidal, anti-mutagenic, anti-oxidative, anti-malarial, and anti-inflammatory bioactivities (10–16). One well-known example of the stilbenes is resveratrol, and its antitumor activity has been intensively studied. Several in vivo and in vitro studies have shown that resveratrol inhibits the growth of cancer cells and effects various molecular targets associated with cancer progression such as the Wnt signaling pathway, nuclear factor-kappa B (NF-κB), and the MAPK/ERK pathway in different types of cancer (17,18).

Previously, two novel stilbenes, cis- and trans-suffruticosol D, were extracted from the seeds of Paeonia (5). The two chemicals have similar structures as the mass fragmentation pattern of trans-suffruticosol D was very similar to cis-suffruticosol D, with cis-suffruticosol D varying only from trans-suffruticosol D in its olefinic hydrogen signal (Fig. 1). In this study, we investigated the antitumor activities of cis- and trans-suffruticosol D and examined how these two chemicals act against cancer cells in vitro.

Figure 1 Chemical structures of cis- and trans-suffruticosol D.

Materials and methods

Plant material and compound isolation

The seeds of P. suffruticosa were collected in Tongling, Anhui, China, and identified in September 2012. A voucher specimen (2012001) has been deposited in the Seed Resource Bank at the Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College. cis- and trans-suffruticosol D were extracted and isolated from the dried seeds of P. suffruticosa as described previously (5). Compounds were re-suspended in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO, USA) at a concentration of 10 mM and stored at 4°C.

Cell culture

Four human cancer cell lines including A549 (lung carcinoma), BT20 (estrogen receptor-negative human breast adenocarcinoma), MCF-7 (estrogen receptor-positive human breast adenocarcinoma) and U2OS (human osteosarcoma) were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). An A549 cell line that stably expresses green fluorescent protein (GFP) was purchased from Cell BioLabs Inc. (San Diego, CA, USA). A549, A549-GFP and BT20 cells were cultured in RPMI-1640 media (Sigma-Aldrich), MCF-7 cells were cultured in DMEM medium (ATCC), and U20S cells were cultured in McCoy's 5A medium (ATCC). As a control, HPL1A cells (human peripheral lung epithelial cells) were obtained from Nagoya University and cultured in DMEM/F12 medium (Sigma-Aldrich); HMEC cells (primary human mammary breast epithelial cells) were purchased from ATCC and cultured in McCoy's 5A medium. HPL1A cells (primary human peripheral lung epithelial cells) were obtained from Nagoya University, Japan, and cultured in DMEM/F-12K medium (Sigma-Aldrich). All medium contained 10% FBS (Sigma-Aldrich) and 1% streptomycin and penicillin (Sigma-Aldrich). These cells were incubated in a humid environment with 5% CO2 at 37°C.

Cell proliferation assay

The resazurin reduction reagent AlamarBlue (Invitrogen, Frederick, MD, USA) was used to evaluate the cytotoxicity of the compounds. Cells were plated at a density of 5×103 cells per well in 96-well microplates with 100 μl culture medium, and were allowed to attach for 16 h prior to treatment. Next, all the medium was replaced with medium containing the cis- or trans-suffruticosol D compounds at seven different concentrations: 320, 100, 32, 10, 3.2, 1.0 and 0.32 μM. 1% DMSO was used as vehicle control. The cells were placed in an incubator for 48 h at 37°C. Cells that were treated with medium containing vehicle only serve as negative control. Subsequently, AlamarBlue solution was added to the medium and the cells were incubated in the CO2 incubator for 1 h. The fluorescent intensity change of the dye was measured at Ex 555 nm and Em 590 nm using a plate reader (Molecular Devices, Sunnyvale, CA, USA). The cytotoxicity was examined by determining by IC50, the dose that inhibited 50% of cell growth, using GraphPad Prism software (GraphPad Software, La Jolla, CA, USA). For the N-acetylL-cysteine (NAC) attenuation assay, the cells were treated with different concentrations of cis- or trans-suffruticosol D followed by incubation with or without 10 mM NAC (Sigma-Aldrich) for 48 h, and then cell viability was assessed using the AlamarBlue assay.

Apoptosis assay

The FlowCellect Annexin Red kit (EMD Millipore, Billerica, MA, USA) was used to determine the apoptosis rate in A549 cells according to the manufacturer's instructions. Briefly, A549 cells were plated in 96-well plates. After a 24-h treatment with cis- or trans-suffruticosol D at concentrations of 100, 32 and 10 μM, the floating and attached cells were collected for analysis. The cells were centrifuged at 700 × g for 7 min and were resuspended in 100 μl assay buffer (EMD Millipore). Afterwards, the cells were stained with Annexin V for 15 min and 7-amino-actinomycin D (7-AAD) for 5 min, and examined with a Guava EasyCyte Flow Cytometer (EMD Millipore). Data were analyzed using Guava InCyte software.

Apoptosis antibody array

The Human Apoptosis Antibody Array kit (RayBiotech, Inc., Norcross, GA, USA) was used to evaluate apoptotic protein expression according to the manufacturer's instructions. A549 cells were plated at 8,000 cells/well intensity in a 96-well plate and then treated with cis- or trans-suffruticosol D at a concentration of 50 μM for 6 h. The cells were lysed in lysis buffer with protease inhibitors. The cell lysates were concentrated using a protein concentration column (EMD Millipore) to a total protein concentration of 2 mg/ml. The samples were then diluted 10-fold with assay buffer and incubated with an array membrane for 2 h at room temperature, and washed with washing buffer for five times. Subsequently, the cocktail of biotin-conjugated antibody mix was added to the membrane and incubated overnight at 4°C. The samples were then incubated with HRP-conjugated streptavidin for 2 h at room temperature and chemiluminescence substrate was used to detect the signal. Image Studio software (LI-COR Biotechnology, Lincoln, NE, USA) was used to quantify the intensity of each array dot and then normalized to the internal control.

Oxidative stress assay

The Hitkit oxidative stress kit (Thermo Scientific, Waltham, MA, USA) was used to determine the generation of reactive oxygen species (ROS) according to the manufacturer's instructions. Briefly, A549 cells were treated with cis- or trans-suffruticosol D for 24 h, fixed with warm 37% formaldehyde and stained with Hoechst and dihydroethidium (DHE) dye for 30 min at 37°C with 5% CO2. Doxorubicin (DOX) at 1 μM concentration was used as a positive control and cells treated with vehicle only were used as negative control. ROS generation in the nuclei was indicated by the production of the fluorescent ethidium, and assessed by measuring the nuclear fluorescent intensity using an ArrayScan VTI High-content screening (HCS) reader (Thermo Scientific). Images were acquired and data was analyzed by vHCS Scan software.

Cell motility assay

A 96-well collagen plate (Corning, Corning, NY, USA) was coated with blue fluorescent beads (Life Technologies, Eugene, OR, USA) as follows. The beads were centrifuged for 1 min at 14,000 g and washed twice with PBS, then 75 μl beads were added to each well of the 96-well collagen plate and incubated for 1 h at 37°C. The cells were seeded on the lawn of fluorescent beads and the sizes of the tracks generated by migrating cells were measured. After the plate was washed 5 times with PBS, A549-GFP cells were seeded at 500 cells/well in the coated plate and incubated for 1 h at 37°C. Subsequently the cells were treated with different concentrations of cis- or trans-suffruticosol D in medium containing 10% FBS for 18 h. Cells treated with serum-free medium serve as the negative control and cells treated with medium containing 10% FBS serve as the positive control. Cell tracks were imaged using an Arrayscan VTI HCS reader (Thermo Scientific) and the data were analyzed by vHCS Scan software. The mean of the full track area per cell for the test compound and the controls was calculated.

Multi-parameter cytotoxicity assay

HCS analysis was used to measure nuclear morphology, cell membrane permeability, and mitochondrial membrane potential changes, the three parameters associated with cytotoxicity. A549 cells were treated with different concentrations of cis- or trans-suffruticosol D for 24 h. The cells were then fixed and stained with a warm solution containing Hoechst dye, membrane permeability dye, and mitochondrial membrane potential dye (Thermo Scientific). Cells were imaged using an Arrayscan VTI HCS reader (Thermo Scientific). Data on nuclear size, cell permeability, and mitochondria membrane potential were collected and analyzed using vHCS Scan software.

Western blot analysis

A549 cells were treated with 50 μM of cis- or trans-suffruticosol D for 3 h then incubated with 10 ng/ml of TNF-α for 30 min. Cells treated with the NF-κB inhibitor Bay11-7082 (10 μM) (Sigma-Aldrich) were used as a positive control, and cells treated with vehicle only were used as a negative control. After treatment, the cells were lysed using M-PER mammalian protein extraction reagent (Thermo Scientific) containing proteinase and phosphatase inhibitors (Sigma-Aldrich) and centrifuged at 13,000 rpm for 5 min at 4°C. A Pierce BCA protein assay kit (Thermo Scientific) was used to determined protein concentrations. Proteins were separated on a 4–20% Tris-glycine gel (Thermo Scientific), and electrophoretically transferred to a PVDF membrane. The following primary antibodies were used: phosphorylated-NF-κB p65, NF-κB p65 (Cell Signaling Technology, Danver, MA, USA) and actin (Santa Cruz Biotechnology, Dallas, TX, USA). The membrane was incubated with the primary antibodies at a 1:1,000 concentration at 4°C overnight. After washing with 1X PBS 5 times, the membrane was incubated for 2 h at room temperature with HRP linked anti-rabbit IgG secondary antibodies. Membranes were developed with chemiluminescent substrates (Thermo Scientific) and scanned with a chemiDoc MP imaging system (Bio-Rad, Hercules, CA, USA).

NF-κB nuclear translocation assay

The Multiplexed NF-κB activation HCS kit (Thermo Scientific) was used to assess NF-κB nuclear translocation. A549 cells were pre-treated with different concentrations of cis- or trans-suffruticosol D for 4 h, then 10 ng/ml of TNF-α (Sigma-Aldrich) was added to the cells for an additional 30 min. After treatment, cells were fixed and permeabilized prior to detection. NF-κB distribution was detected by adding NF-κB p65 primary antibodies and then staining with a secondary antibody conjugated with DyLight 549 and Hoechst dye (Thermo Scientific). Cells treated with medium containing only the vehicle were used as negative control, and cells treated with 25 ng/ml TNF-α were used as a positive control. Cells were imaged using an Arrayscan VTI HCS reader. Data on the mean difference of NF-κB fluorescent intensity between the nuclear and cytoplasmic areas were collected and analyzed by vHCS Scan software.

Results

Cytotoxicity of cis- and trans-suffruticosol D in lung, breast, and bone cancer cells

After 48-h treatment, both cis- and trans-suffruticosol D showed significant cytotoxic effects against A549 (lung), BT20 (breast), MCF-7 (breast) and U2OS (osteosarcoma) cancer cell lines. IC50 values for cis- and trans-suffruticosol D against these cancer cells ranged from 9.93 to 46.79 μM as shown in Table I. Interestingly, we observed that trans-suffruticosol D had lower IC50 values (9.93–15.84 μM) than cis-suffruticosol D (13.42–46.79 μM) in all four cancer cell lines. In addition, both cis- and trans-suffruticosol D showed notably weaker cytotoxicity against normal breast epithelial cells HMEC (IC50 values of 146.3 and 269.5 μM, respectively) and normal lung epithelial cells HPL1A (IC50 values of 78.3 and 177.5 μM, respectively) (Table I).

Table I IC50 values of cis- and trans-suffruticosol D in selected cancer and normal cell lines.

Table I

IC50 values of cis- and trans-suffruticosol D in selected cancer and normal cell lines.

IC50 (μM)	trans-SD	cis-SD
A549	11.9±1.2	17.1±1.0
BT20	9.9±3.8	13.4±2.5
MCF-7	15.8±1.6	46.8±3.3
U2OS	11.3±2.3	24.6±4.4
HPL1A	78.3±6.1	177.5±9.3
HMEC	146.3±2.7	269.5±2.2

[i] Cells were treated with various concentrations of cis- or trans-suffruticosol D for 48 h, and the viability of cells was evaluated with the AlamarBlue dye. Data are expressed as mean ± SD, and experiments were performed in triplicates.

cis- and trans-suffruticosol D induce apoptosis in A549 lung cancer cells

To find out whether these cytotoxic properties were due to apoptosis, we conducted an apoptosis assay using A549 cells treated with cis- or trans-suffruticosol D. Following a 24-h treatment, both compounds showed significant apoptosis induction at a wide range of concentrations compared with the non-treated cells (*P<0.05, **P<0.01 or ***P<0.001) and the apoptotic effects were concentration-dependent (Fig. 2A–D). trans-suffruticosol D induced 30.1, 39.8 and 41.9% of A549 cells into apoptosis at concentrations of 10, 32 and 100 μM, respectively. cis-suffruticosol D induced 22.2, 27.1 and 45.3% of A549 cells into apoptosis at concentrations of 10, 32 and 100 μM, respectively.

Figure 2 Induction of apoptosis by cis- and trans-suffruticosol D in A549 cells. After 24-h treatment by cis- or trans-suffruticosol D, A549 cells were stained with Annexin V/7-AAD, and the percentage of apoptotic cells was assessed by flow cytometry. Cells treated with the vehicle only served as a negative control. (A–C) Annexin V/7-AAD double staining of A549 cells treated with various concentrations of cis- or trans-suffruticosol D. (D) Percentage of apoptotic cells induced by cis- or trans-suffruticosol D. (E and F) Effect of cis- or trans-suffruticosol D on key regulatory proteins of apoptosis. The error bars indicate the standard deviation from three experiments. *P≤0.05; **P≤0.01; ***P≤0.001.

Next, we performed an apoptotic protein array analysis to investigate the effect of cis- and trans-suffruticosol D on apoptotic proteins. Two proteins from the inhibitor of apoptosis proteins family (IAPs), X-linked inhibitor of apoptosis protein (XIAP) and survivin, as well as the heat shock proteins Hsp60 and Hsp70, showed significant downregulation after treatment by cis- and trans-suffruticosol D (Fig. 2E). Whereas, death receptor 6 (DR6), also known as tumor necrosis factor receptor superfamily member 21 (TNFRSF21), the cyclin-dependent kinase inhibitor 1B (p27), and the BH3 interacting-domain death agonist (BID), were upregulated by both cis- and trans-suffruticosol D (Fig. 2F).

cis- and trans-suffruticosol D induce ROS generation in A549 lung cancer cells

We examined the cellular ROS levels in A549 cells to determine whether cis- and trans-suffruticosol D induced oxidative stress. As shown in Fig. 3A, both cis- and trans-suffruticosol D converted non-fluorescent DHE to fluorescent ethidium, which binds to DNA, suggesting they induced ROS generation in A549 cells. Quantitative data showed both compounds significantly induced ROS generation in a concentration-dependent manner (**P<0.01, ***P<0.001 or ****P<0.0001). After treatment for 24 h, trans-suffruticosol increased the ROS levels by 32.8, 34.6 and 87.2% at concentrations of 10, 32 and 100 μM, respectively, while cis-suffruticosol increased the ROS levels by 32.8, 55.6 and 73.1% at concentrations of 10, 32, and 100 μM, respectively, in A549 cells (Fig. 3B). To further investigate whether the cytotoxicity induced by cis- and trans-suffruticosol D was associated with ROS levels, we co-treated A549 cells with the anti-oxidant N-acetyl-L-cysteine (NAC) and different concentrations of cis- or trans-suffruticosol D for 48 h. We observed that 10 mM NAC attenuated the cell death induced by cis- or trans-suffruticosol D in A549 cells at all of the concentrations that were tested (Fig. 3C).

Figure 3 Induction of oxidative stress by cis- and trans-suffruticosol D in A549 cells. A549 cells were treated with various concentrations of cis- or trans-suffruticosol D for 24 h, then stained with Hoechst and DHE dye. Cells treated with doxorubicin served as a positive control, and cells treated with vehicle only served as a negative control. The ROS levels were measured by the fluorescent intensity of DHE that was converted to ethidium bromide. (A) Fluorescent cell images by the HCS reader. Scale bar, 100 μm. (B) ROS levels in A549 cells treated with various concentrations of cis- or trans-suffruticosol D. (C) The anti-oxidant NAC attenuated the cell death of A549 cells induced by cis- or trans-suffruticosol D. The error bars indicate the standard deviation from three experiments. *P≤0.05; **P≤0.01; ***P≤0.001.

cis- and trans-suffruticosol D inhibit the motility of A549 lung cancer cells

To test if cis- and trans-suffruticosol D affected cancer cell motility, we measured the size of the tracks generated by migrating cells after treatment, which is proportional to the magnitude of cell movement. As shown in Fig. 4A, A549 cells treated with cis- or trans-suffruticosol D in serum-containing medium showed less motility activity evidenced by a smaller track area per cell than the untreated cells. Both cis- and trans-suffruticosol D significantly inhibited cell movement at all the concentrations that were tested in A549 cells (***P<0.001 or ****P<0.0001) (Fig. 4B). trans-suffruticosol D decreased the A549 cell motility by 40.7, 40.7 and 54.9% at concentrations of 10, 32 and 100 μM, respectively, while cis-suffruticosol D decreased the A549 cell motility by 42.3%, 42.0 and 50.4% at concentrations of 10, 32 and 100 μM, respectively.

Figure 4 Cell motility changes induced by cis- and trans-suffruticosol D in A549 cells. A549-GFP cells were seeded in a 96-well plate with a monolayer of fluorescent beads. After treatment with cis- or trans-suffruticosol D for 18 h, individual cell movement was evaluated by measuring the fluorescent track area. Cells treated with serum-free medium served as a negative control and cells treated with medium containing 10% serum served as a positive control. (A) Fluorescent track area showing the movement of the cells. Scale bar, 200 μm. (B) Measurement of the cell track areas of cells treated with various concentrations of cis- or trans-suffruticosol D. The error bars indicate the standard deviation from three experiments. *P≤0.05; **P≤0.01; ***P≤0.001.

cis- and trans-suffruticosol D decreased mitochondrial membrane potential in A549 cells

To determine the cytotoxic effect of cis- and trans-suffruticosol D in human lung cancer cells, we measured three cell health parameters, nuclear morphology, cell membrane permeability and mitochondrial membrane potential changes, using an HCS reader. As shown in Fig. 5, in the mitochondrial potential channel, untreated A549 cells exhibited bright fluorescent intensity, indicating intact mitochondrial membranes. In comparison, in cells treated with cis- or trans-suffruticosol D the fluorescent intensity of the dye was significantly decreased at all tested concentrations, indicating that cis- and trans-suffruticosol D induced a significant decrease of the mitochondrial membrane potential in A549 cells (***P<0.001). We also observed nuclei shrinkage and increased cell membrane permeability in cells treated with a high-concentration (100 μM) of trans-suffruticosol D (*P<0.05 or **P<0.01). However, no significant change was detected in nuclear size and cell membrane permeability in cells treated with cis-suffruticosol D.

Figure 5 Multi-parameter cytotoxicity induced by cis- and trans-suffruticosol D in A549 cells. A549 cells were treated with various concentrations of cis- or trans-suffruticosol D for 24 h, then stained with three dyes simultaneously (Hoechst, cell permeability dye and mitochondrial membrane potential dye). Cells treated with vehicle only served as a negative control and cells treated with 10 μM valinomycin served as a positive control. (A) Fluorescent cell images by HCS reader. Scale bar, 100 μm. (B–D) Evaluation of nuclear size, cell permeability and mitochondrial membrane potential of cells treated with cis- or trans-suffruticosol D. The error bars indicate the standard deviation from three experiments. *P≤0.05; **P≤0.01; ***P≤0.001.

cis- and trans-suffruticosol D inhibit TNF-α-induced NF-κB activation

We performed western blot analysis to examine the effects of cis- and trans-suffruticosol D on the expression of NF-κB in A549 cells. As shown in Fig. 6A, upon TNF-α stimulation, overexpression of phosphorylated NF-κB p65 was detected, and the overexpression was significantly inhibited by cis- and trans-suffruticosol D. In trans-suffruticosol D-treated cells, the expression of phosphorylated NF-κB p65 was almost completely blocked, and in cis-suffruticosol D-treated cells, the expression of phosphorylated NF-κB p65 was blocked as effectively as the blockage caused by the Bay11-7082 inhibitor control.

Figure 6 Inhibition of NF-κB translocation by cis- and trans-suffruticosol D in A549 cells. A549 cells were treated with various concentrations of cis- or trans-suffruticosol D for 4 h and then stimulated with 25 ng/ml TNF-α for 30 min. Cells treated with TNF-α alone or the vehicle only served as controls. The NF-κB translocation index was measured by the fluorescent intensity difference between the nucleus and cytoplasm. (A) Western blot analysis of the expression of phosphorylated-NF-κB p65 and total NF-κB p65. (B) Fluorescent cell images by the HCS reader. Scale bar, 100 μm. (C) Evaluation of the NF-κB translocation index in A549 cells. The error bars indicate the standard deviation from three experiments. *P≤0.05; **P≤0.01; ***P≤0.001.

Next, we used HCS analysis to test whether cis- or trans-suffruticosol D could block NF-κB nuclear translocation in A549 cells. As shown in Fig. 6B, NF-κB fluorescent staining remained in the cytoplasmic area and no fluorescence was detected in the nuclear area in non-treated cells, however, in cells treated with TNF-α the NF-κB fluorescent staining was detected in the nuclear area, indicating that NF-κB was translocated from the cytoplasm to the nucleus. In A549 cells treated with cis- or trans-suffruticosol D, NF-κB fluorescent staining remained in the cytoplasm, suggesting that NF-κB translocation to the nucleus was blocked. Treatment with trans-suffruticosol D at all the tested concentrations, caused a significant inhibition of NF-κB activation (***P<0.001) (Fig. 6C). In contrast, treatment with cis-suffruticosol D only caused a significant inhibition of NF-κB at 100 μM (***P<0.001).

Discussion

Oligostilbenes have been widely considered to be valuable resources of antitumor agents. Previously, two novel oligostilbenes, cis- and trans-suffruticosol D, were extracted from the seeds of P. suffruticosa, but their antitumor activities were not determined. In this study, we found that both of these oligostilbenes exhibited remarkable anti-proliferation activities against several types of cancer cell lines, and their cytotoxicity effects and related mechanisms were investigated.

trans-suffruticosol D exhibited lower IC50 values (9.93–20.8 μM) than cis-suffruticosol D (13.42–46.79 μM) in all of the cancer cell lines that were tested, indicating that trans-suffruticosol D is more cytotoxic than its cis-isomer. Consistent with this conclusion, trans-suffruticosol D had stronger effects than cis-suffruticosol D on three cytotoxicity parameters, changes in nuclear size, cell membrane permeability and mitochondrial transmembrane potential. trans-suffruticosol D also showed higher inhibition activity of NF-κB activation than cis-suffruticosol D. These observations are consistent with a previous report, which showed that trans-resveratrol had stronger cytotoxicity than its cis-isomer (19). In addition, both chemicals showed selective cytotoxicity against cancer cell lines versus a normal cell line.

Cancer cells usually develop the ability to escape apoptosis (programmed cell death), which is a homeostatic mechanism to maintain cell populations in the body (20). Hence, targeting apoptotic induction has become an important strategy of anticancer therapies. It is commonly known that there are two apoptotic pathways, the extrinsic, or the death receptor pathway, and the intrinsic, or the mitochondrial pathway. Previous studies have shown that mitochondria play a critical role in apoptosis, especially in the intrinsic apoptosis pathways (21,22). Mitochondria are the main source of ROS inside the cell, and increases in ROS production can damage the mitochondrial membrane and subsequently lead to the release of pro-apoptotic proteins and cytochrome c, thus activating the apoptotic pathway (23–25). In this study, we found that cis- and trans-suffruticosol D induced apoptosis in A549 lung cancer cells after 24-h treatment in a concentration-dependent manner. Both oligostilbenes significantly decreased the mitochondrial membrane potential in lung cancer cells, suggesting they might induce the mitochondrial apoptosis pathway. Since both chemicals significantly increased cellular ROS levels in lung cancer cells and their cytotoxicity was associated with ROS levels as shown by the NAC attenuation assay, it can be speculated that the excessive ROS induced by cis- and trans-suffruticosol D act as an apoptosis mediator by damaging the mitochondrial membrane, causing the release of the mitochodria's contents, which eventually leads to apoptosis. In addition, cis- and trans-suffruticosol D affected the expression of several key regulators involved in apoptosis; XIAP, survivin, Hsp60 and Hsp70 were downregulated, while BID, DR6 and p27 were upregulated.

XIAP and survivin are known apoptosis inhibitors that prevent apoptosis by inhibiting caspase-3, −7, and −9 (26–28). Downregulation of XIAP or survivin has been demonstrated to inhibit the progression of cancer and increase the sensitivity of cancer cells to chemo-reagents (29–32). Heat shock proteins Hsp60 and Hsp70 are chaperones that play essential roles in tumor cell survival and proliferation due to their ability to block both the intrinsic and extrinsic apoptosis pathways (33,34). BID is a pro-apoptotic member of the Bcl-2 protein family, and is a mediator of mitochondrial damage induced by caspase-8 (35). p27, the cyclin-dependent kinase inhibitor, controls the cell cycle progression at G1 by preventing the activation of cyclin E-Cdk2 or cyclin D1-Cdk4 complexes (36,37). DR6, also known as TNFRSF21, is a member of the death receptor family, which induces apoptosis in mammalian cells and its apoptotic function is inhibited by survivin (38). Downregulation of XIAP, survivin, Hsp60 and Hsp70, as well as upregulation of BID, DR6 and p27 by cis- and trans-suffruticosol D at least partially contribute to the apoptotic effect of cis- and trans-suffruticosol D.

Tumor cells have the ability to migrate to surrounding tissues and organs through reorganization of the actin cytoskeleton (39,40). Most of the fatality from tumors occurs when cells move from the initial organs where they originated (41). Therefore, control of cancer cell motility and migration is an essential issue in cancer treatment and represents a new opportunity for a potential tumor therapy (42). cis- and trans-suffruticosol D significantly inhibited the mobility of lung cancer cells after treatment for 18 h at all the concentrations that were tested. Therefore, both chemicals exhibit therapeutic potential as an inhibitor of cancer cell mobility.

The NF-κB pathway is known to control cell growth and survival, and the transcription factor NF-κB has been found to be permanently activated in various tumors (21). Activation of NF-κB in cancer cells is often associated with drug resistance as both radio- and chemo-therapies induce constitutive activation of the NF-κB pathway (43). Therefore, a compound's ability to block the NF-κB pathway is important for the efficacy of cancer therapy (44). In this study, we evaluated cis- and trans-suffruticosol D for their abilities to inhibit TNF-α induced NF-κB activation in lung cancer cells. After a 4-h treatment both chemicals significantly blocked NF-κB p65 phosphorylation as well as NF-κB p65 translocation from the nucleus to the cytoplasm, suggesting they might act as an inhibitor of the NF-κB pathway. Since NF-κB affects the transcription of a number of anti-apoptotic proteins, including cellular inhibitor of apoptosis proteins (cIAPs), XIAP, bcl-2, bcl-XL, and FADD-like IL-1β-converting enzyme-inhibitory protein (c-FLIP), blocking NF-κB nuclear translocation decreases the expression of anti-apoptotic proteins and subsequently promotes apoptosis. In addition, several studies have shown that an increase of ROS can block the NF-κB pathway by the inhibition of cytokines, such as TNF and IL-1 (45). Because cis- and trans-suffruticosol D increased ROS generation in lung cancer cells, the block in the NF-κB pathway may be associated with the inhibition of the inducer cytokines by excessive ROS.

In conclusion, this study provides evidence that cis- and trans-suffruticosol D have promising antitumor activities. Both compounds selectively inhibited the growth of various cancer cells, induced apoptosis in A549 lung cancer cells, as well as inhibited A549 cell movement. The induction of apoptosis may be associated with ROS generation and inhibition of the NF-κB pathway. Collectively, our results suggest a potential mechanism for the cytotoxicity of cis- and trans-suffruticosol D. As shown in Fig. 7, in A549 lung cancer cells, cis- and trans-suffruticosol D trigger oxidative stress, which in turn leads to mitochondrial damage, blocks NF-κB activation and ultimately triggers apoptosis. Our findings suggest that both cis- and trans-suffruticosol D have promising chemo-therapeutic potential for treating cancer.

Figure 7 Proposed cytotoxicity mechanism of cis- and trans-suffruticosol D.

Acknowledgements

We thank the Tennessee Center of Botanical Medicine Research (TCBMR) for providing the funding for this study. We also thank Dr Takashi Takahashi at Nagoya University for the HPL1A cell line.

References