Accumulation of vascular smooth muscle cells (VSMC) is a key event in the formation and development of lesions of atherosclerosis. The excessive accumulation of VSMC is due to a combination of directed migration from the media into the intima of the artery accompanied by proliferation and possibly decreased apoptosis. The proliferation and migration of VSMC are involved in vascular remodelling and intimal lesion formation, and these can be induced by various growth factors, such as platelet-derived growth factor (PDGF) or basic fibroblast growth factor(Reference Ross1–Reference Koyama, Hart and Clowes4). PDGF is expressed by platelets, smooth muscle cells, endothelial cells and macrophages, and PDGF-BB among the PDGF isomers is a potent inducer of VSMC proliferation and migration compared with PDGF-AA and PDGF-AB(Reference Raines2–Reference Koyama, Hart and Clowes4).
Although the role of PDGF on vascular biology is important, the mechanisms related to PDGF-stimulated smooth muscle cell proliferation and migration have not been fully elucidated. Several studies have reported that PDGF interferes with actin filament assembly and causes marked up-regulation of PDGF receptor (PDGFR) expression after injury to the vessel wall(Reference Sirois, Simons and Edelman5). In an animal model, infusion of PDGF-BB promoted intimal thickening and promoted VSMC migration from the media to intima(Reference Jawien, Bowen-Pope and Lindner6), and the blocking of PDGFR decreased in smooth muscle cell growth and neointimal formation in response to injury(Reference Banai7). Some studies also showed that activation of mitogen-activated protein kinase (MAPK) (extracellular signal-regulated kinases; ERK1/2) is implicated in the PDGF-mediated DNA synthesis and proliferation process(Reference Bornfeldt, Raines and Nakano8), and the ability of the PDGF to activate phospholipase C (PLC)γ1 and phosphoinositide 3-kinase (PI3K)/Akt correlated with its ability to induce migration of smooth muscle cells rather than DNA synthesis and proliferation(Reference Bornfeldt, Raines and Graves9). Furthermore, PDGFR transports its signal into the intracellular space by the generation of reactive oxygen species (ROS). For instance, H2O2 was related to stimulation of the phosphorylation of ERK1/2 in PDGF-stimulated VSMC proliferation(Reference Sundaresan, Yu and Ferrans10).
Isoflavone-rich soya foods, soya protein isolate and isoflavone extracts have been reported to improve cognitive performance(Reference Lephart, West and Weber11, Reference Lund, West and Tian12), menopausal symptoms(Reference Nagata, Shimizu and Takami13, Reference Nagata, Takatsuka and Kawakami14), bone health(Reference Anderson and Garner15), hypertension(Reference Lichtenstein16) as well as prevent lipid peroxidation(Reference Wiseman, O'Reilly and Adlercreutz17, Reference Mortensen, Kulling and Schwartz18), and breast(Reference Zava and Duwe19) and prostate cancers(Reference Davis, Singh and Bhuiyan20, Reference Onozawa, Fukada and Ohthani21). In addition, several studies demonstrated that soya protein with isoflavones exhibited atheroprotective effects by increasing HDL-cholesterol, reducing vascular inflammation and platelet aggregation, and inhibiting human aortic smooth muscle cell (HASMC) proliferation(Reference Dubey, Gillespie and Imthurn22). Isoflavones scavenge free radicals that are involved in platelet stimulation, suggesting novel anti-aggregatory mechanism of isoflavones(Reference Pignatelli, Pulcinelli and Lenti23). Furthermore, soya-isoflavone supplementation reduced NF-κB activation induced by TNF-α in human lymphocytes(Reference Davis, Kucuk and Djuric24). Genistein, in particular, activates endothelial NO synthase (eNOS) by stimulation of cAMP production, leading to the phosphorylation of eNOS. These results indicate that isoflavones have specific pathways by which these oestrogen-like components induce vasodilatory effects(Reference Satake and Shibata25). Treatment of human umbilical vein endothelial cells (HUVEC) with genistein resulted in a dose-dependent inhibition of TNF-α-induced expression of adhesion molecules such as E-selectin, vascular cell adhesion molecule-1 and inter-cellular adhesion molecule-1(Reference Weber, Negrescu and Erl26). Genistein and daidzein not only inhibited NO production via suppression of inducible NO synthase in activated macrophages but also reduced monocyte chemotactic protein (MCP)-1 secretion(Reference Gottstein, Ewins and Eccleston27).
Glyceollins, which are synthesised from daidzein in soyabeans cultured with Aspergillus sojae, have been shown to have cancer-preventive properties(Reference Payton-Stewart, Schoene and Kim28, Reference Salvo29) and represent another group of phytoalexins whose biosynthesis is increased in soyabean in response to various stress signals such as fungal infection(Reference Brent and Hollomon30, Reference Costa, Fernandes and Santos31). Flavonoids are a group of widely distributed plant polyphenols having antioxidant activity. As glyceollins have similar structure to other soyabean isoflavonoids (Fig. 1), it is presumed that they share common physiological properties. Research in our laboratory showed a marked oestrogenic activity of glyceollins in vitro on oestrogen receptor signalling(Reference Kim, Suh and Kim32). Moreover, our recent preliminary studies show that glyceollins have anti-inflammatory and antioxidant activity.
Fig. 1 Molecular structures of glyceollins I (a), II (b) and III (c).
In the present study, the anti-atherosclerotic effects of glyceollins derived from soyabean elicited with fungi were evaluated and found that these compounds can inhibit the PDGF-stimulated proliferation and migration of HASMC.
Experimental methods
Preparation of glyceollins
A mixture of glyceollins I, II and III was isolated according to the method described previously(Reference Kim, Suh and Kim32). Briefly, glyceollin-rich extract was prepared by inoculation of soyabean seed with A. sojae in a dark chamber at 26°C for 3 d. After homogenisation in 80 % aqueous ethanol and incubation, the filtered extract was freeze-dried. The ethyl acetate soluble fraction of the crude extract was subjected to silica gel column chromatography. One fraction was subjected to TLC and HPLC to confirm existence of glyceollins, and then the fraction was evaporated under reduced pressure.
Cell culture
HASMC were obtained from Cascade Biologics, Inc. (Portland, OR, USA). For routine maintenance, cells were grown in Dulbecco's modified Eagle's medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10 % heat-inactivated fetal bovine serum at 37°C in an atmosphere of 5 % CO2–95 % air under saturating humidity and passaged every week (1:4 split ratio) by trypsinisation with 0·25 % trypsin–0·02 % EDTA sodium salt solution (Thermo Fisher Scientific Inc., Waltham, MA, USA).
Cell proliferation assay
The experimental procedures were performed according to previously described methods with some modifications. Briefly, HASMC were trypsinised and plated into ninety-six-well plates with the DMEM at an initial concentration of 1 × 104 cells per well and incubated in an atmosphere consisting of 5 % CO2 in air at 37°C for 24 h. Then the medium was removed and replaced by the medium containing a range of concentrations of test compounds in serum-free DMEM. The bioassay was terminated on day 1 by removing the medium from the wells, and adding 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) solution (0·5 mg/ml in phenol red-free culture medium). After 2 h, the solution was removed and 200 μl dimethylsulfoxide (DMSO) were added. After 10 min, the absorbance was determined at 540 nm by a microplate reader (Tecan, Grödig, Austria). The cell proliferation of each group was calculated as the relative absorbance of treatment group compared with control.
Synthesis assay
Thymidine incorporation experiments were carried out according to a previously described method(Reference Cha, Shi and Yonezawa33). The HASMC were cultured until they reached about 40 % confluence. The culture medium was changed with serum-free medium and incubated for 24 h. The serum-deprived cells were then pre-incubated with glyceollins for 24 h and stimulated with/without PDGF or with/without glyceollins for 20 h. After the incubation, 1 μCi/ml (37 000 Bq) of [3H]thymidine was added to the cultures and incubated for 4 h at 37°C. The cells were harvested using a Universal Harvester (Perkin Elmer, Waltham, MA, USA), and transferred to a GF/C filter (Perkin Elmer). The filter was dried and the level of radioactivity was determined using a Microplated Scintillation and Luminescence Counter-Topcount NXT (Perkin Elmer). The values were calculated from absolute counts to a percentage of the control to allow comparison between the experimental groups.
Cell cycle analysis
Cell cycle progression analysis was measured as described previously(Reference Erl34). HASMC were seeded in six-well plates at 2 × 105 cells/well and incubated until the cell density reached 80 %. Then the medium was replaced with serum-free medium and incubated for 24 h. The cells were then pre-incubated with glyceollins for 24 h and stimulated with/without PDGF or with/without glyceollins for 24 h. The HASMC were harvested, re-suspended in 70 % ethanol, and incubated at − 20°C for 4 h. The fixed cells were washed twice with PBS, and incubated at 37°C for 60 min using a solution containing DNAse-free RNAse (200 μg/ml). After incubation, the cells were stained with propidium iodide (50 μg/ml) at 4°C for 30 min and subjected to a flow cytometric analysis (Becton Dickinson, Franklin Lakes, NJ, USA).
Migration assay
Cell migration was measured using a modified Boyden transwell chamber (twenty-four-well plate) assay as described in our previous study(Reference Cha, Shi and Yonezawa33). Starved cells were trypsinised and cell suspension in basal medium (500 μl, 5 × 104 cells/well) seeded in the upper chamber. Then, 760 μl of serum free medium with or without experimental compounds were added to the lower chamber. After incubation for 24 h, the cells were labelled by fluorescence dye calcein-acetoxymethyl ester (AM) (4 μg/ml). The cells migrated to the lower chamber were measured using a fluorescence microplate reader at 485 nm excitation and 530 nm emission wavelengths. Each treatment was repeated in three independent transwells.
Wound healing assay
When the HASMC had grown to 80 % confluence in the eight-well plate, a scratch was made with a sterile cell scraper as described in a previous study(Reference Mu, Ohashi and Yang35). Cells were washed with PBS twice, were incubated for 18 h with or without glyceollins (3 μg/ml) in the presence or absence of PDGF (20 ng/ml) in serum-free medium, stained with haematoxylin and eosin. The plates were photographed and the wound width was calculated by a microruler.
Dichlorofluorescein assay
Oxidative stress was quantified in cells by the dichlorofluorescein (DCF) assay according to the method described by Wang & Joseph(Reference Wang and Joseph36), with slight modifications. Briefly, HASMC were seeded into black-bottom ninety-six-well plates, and cultured for 24 h. After attachment, plates were washed with PBS and the cells were incubated with H2O2 (1000 μm) for 2 h before treatment with increasing concentrations of each generation of experimental compounds prepared in serum-free media for 24 h. The stimulated cells were washed with PBS and incubated for 30 min with dichlorofluoresceindiacetate, dissolved in DMSO (final concentration 50 μm). Fluorescence was measured at 0 and 40 min using excitation and emission wavelengths of 485 and 535 nm, in a Tecan microplate reader (Tecan). All dye and oxidant incubations, washing steps and fluorimetric determination were performed in a dark room. The intensity of fluorescence was calculated as ((F40 min − F0 min)/F0 min) × 100 as described. Results are expressed as relative intensity of fluorescence (as % of H2O2 control).
Dihydroethidium assay
The production of superoxide anion in HASMC was measured using the oxidation-sensitive dye dihydriethidium(Reference Li, Miller and Brown37). HASMC were stained by incubation for 30 min at 37°C with 10 μm-dihydriethidium (Sigma). The cells were washed with PBS twice, and photographed using a Nikon Eclipse 80i fluorescence microscope and a Nikon Digital Sight DS-Fi1c camera (Nikon, Tokyo, Japan).
Western blotting
HASMC were homogenised in pre-cooled radio-immunoprecipitation assay (RIPA) buffer (pH 7·4) containing 2-amino-2-hydroxymethyl-propane-1,3-diol (Tris)-HCl (50 mm), NaCl (150 mm), EDTA (1 mm), Triton-X (1 %), sodium deoxycholate (0·5 %) and SDS (0·1 %). The homogenates were cleared by centrifugation at 9000 rpm for 5 min at 4°C, and the supernatant fractions were denatured in sample buffer for 5 min at 95°C. Proteins were separated by electrophoresis on a 10 % SDS-polyacrylamide gel for 1·5 h at 70–150 V and transferred onto nitrocellulose membranes (Amersham Biosciences, Freiburg, Germany) for 2 h at 250 mA. Membranes were incubated with antibodies to phosphorylated PDGFR-β, PDGFR-β, phosphorylated Akt, Akt, phosphorylated PLCγ, PLCγ, phosphorylated MAPKinases, MAPKinases, cyclin-dependent kinase (CDK)1, CDK4, cyclin D1, cyclin E, p27kip1, p21cip1, p53 and β-actin at dilutions of 1:1000 overnight at 4°C. The bands were detected using a chemiluminescence kit (Pierce, Chester, Cheshire, UK). Densitometry analysis was performed with Lab Image software (Scion Corp., Walkerville, MD, USA).
Statistical analysis
All values are expressed as mean values and standard deviations. For statistical significance, unpaired Student's t tests were used. A significant difference was indicated at P < 0·05.
Results
Effects of glyceollins on human aortic smooth muscle cell proliferation and platelet-derived growth factor-stimulated cycle progression
When HASMC were stimulated by PDGF-BB (20 ng/ml), the number of cells increased by approximately 34·5 % compared with the non-stimulated group. Treatment with glyceollins (0·3–3 μg/ml) for 24 h in PDGF-stimulated HASMC resulted in a significant decrease in cell number without any cytotoxicity. The ratio of inhibition was 13·6 (sd 2·3), 16·7 (sd 4·1) and 30·7 (sd 4·6) % at 0·3, 1·5 and 3 μg/ml, respectively (Fig. 2(a)). There was no significant change in cell morphology in cultures treated with PDGF plus glyceollins (Fig. 2(c)).
Fig. 2 Effects of glyceollins on platelet-derived growth factor (PDGF)-induced human aortic smooth muscle cell (HASMC) proliferation. HASMC were treated with or without glyceollins (3 μg/ml) in the presence or absence of PDGF (20 ng/ml) after pre-incubation with or without glyceollins for 24 h. After incubation for 20–24 h, the cells were processed for the 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay (a) or [3H]thymidine incorporation (b) as described in the Materials and methods section. Values are means of three separate experiments, with standard deviations represented by vertical bars. * Mean value was significantly different from that without PDGF (P < 0·05). † Mean value was significantly different from that in the absence of glyceollins (P < 0·05). Morphology of cells cultured under different conditions was observed by microscopy (c): (i) control; (ii) glyceollins (3 μg/ml); (iii) PDGF; (iv) PDGF+glyceollins (0·3 μg/ml); (v) PDGF+glyceollins (1·5 μg/ml); (vi) PDGF+glyceollins (3 μg/ml). Magnification 40 × .
The effect of glyceollins on DNA synthesis was assayed by [3H]thymidine incorporation. As shown in Fig. 2(b), PDGF at 20 ng/ml led to a significant increase in [3H]thymidine incorporation that corresponded to approximately 4·2-fold of non-stimulated cells. Glyceollins effectively inhibited the PDGF-induced [3H]thymidine incorporation into DNA in a dose-dependent manner. The inhibition percentage of glyceollins was about 27·3 (sd 6·0), 52·6 (sd 4·7) and 41·0 (sd 2·7) % at 0·3, 1·5 and 3 μg/ml, respectively.
The DNA content was analysed by propidium iodide staining to investigate the effect of glyceollins on cell cycle progression in HASMC. PDGF-BB (20 ng/ml) induced an increase in the S phase from 2·3 % to 18·1 % compared with the control. Glyceollins dose-dependently decreased accumulation at the S phase to 11·3 % at 0·3 μg/ml, 8·2 % at 1·5 μg/ml and 5·6 % at 3 μg/ml in PDGF-treated cells in a dose-dependent fashion. These results indicate that glyceollins were effective in arresting PDGF-stimulated cell cycle progression in HASMC.
Effects of glyceollins on the expression of cell cycle-related proteins
Cell cycle progression is tightly regulated through a complex network of positive and negative cell cycle-regulatory molecules such as CDK and cyclins. To investigate whether inhibition of proliferation by glyceollins involves the regulation of cell cycle-related proteins, we assessed the expression of G1-checkpoint proteins in glyceollin-treated cell cycle arrest. As shown in Fig. 3(c), glyceollins significantly decreased the level of CDK2 and cyclin E, but did not change the levels of CDK4 and cyclin D. Therefore, these findings suggest that glyceollins affect the transition of HASMC from the G1 to the S phase, together with its inhibitory effects on HASMC proliferation. The effect of glyceollins on the expression level of CDK inhibitor p27kip1 and p53 was further clarified. The protein levels of p27kip1 and p53 were reduced at 24 h after stimulation by PDGF, while they were significantly enhanced in HASMC treated with glyceollins in a dose-dependent manner. The expression of p21cip1 protein was not influenced by PDGF or glyceollins.
Fig. 3 (a) Effect of glyceollins on cell cycle progression in human aortic smooth muscle cells (HASMC) stimulated by platelet-derived growth factor (PDGF). HASMC were treated as described in Fig. 2(a). (□), G2/M; (
), S; (
), G0/G1. (b) Representative DNA histograms of propidium iodide fluorescence in cells as assessed by flow cytometry. The profiles are representative examples of experiments measured in the presence of PDGF, together with different doses of glyceollins (0·3, 1·5, 3 μg/ml). DNA histograms for untreated control and for cells treated with PDGF or glyceollins alone are also shown for reference. (c) For evaluation of the expression of protein related to cell cycle progression, whole cell extracts (20 μg) were subjected to Western blot analysis for phosphorylation of cyclin-dependent kinase (CDK)4, cyclin D1, CDK2, cyclin E, p21, p27, p53 or β-actin. Upper histogram: (□), CDK4:β-actin; (
), cyclin D1:β-actin; (), CDK2:β-actin; (), cyclin E:β-actin. Lower histogram: (□), p27:β-actin; (), p21:β-actin; (), p53:β-actin.
Effects of glyceollins on signalling pathway related to platelet-derived growth factor
PDGF-induced expression of proteins such as PDGFR-β, PLCγ1 and Akt was detected to investigate the inhibitory mechanism of glyceollins on the signal transduction pathway of PDGF for HASMC proliferation and migration. HASMC were stimulated with PDGF-BB (20 ng/ml) for 15 min, which caused obvious phosphorylation of PDGFR-β, PLCγ1, Akt and ERK1/2, and the increased phosphorylation was suppressed by glyceollins of various concentrations, suggesting that the anti-proliferative effect of glyceollins was implicated in early signal transduction by PDGF (Fig. 4(a)).
Fig. 4 Effects of glyceollins on the platelet-derived growth factor (PDGF) signalling pathway. Western blot of phosphorylation of PDGF receptor (PDGFR)-β, phospholipase (PLC)γ1 and Akt (a) and activation of mitogen-activated protein kinase (b) was performed when human aortic smooth muscle cells were treated with glyceollins (3 μg/ml) with or without PDGF-BB (20 ng/ml) for 15 min after pre-incubation with glyceollins for 24 h. (a): (□), Phosphorylated PDGFR-β:PDGFR-β; (), phosphorylated PLCγ1:PLCγ1; (), phosphorylated Akt:Akt. (b): (□), Phosphorylated P38:P38; (), phosphorylated c-Jun N-terminal protein kinase (JNK):JNK; (), phosphorylated extracellular signal-regulated kinase (ERK):ERK.
Since PDGF is capable of inducing the activation of MAPK for VSMC proliferation(Reference Bornfeldt, Raines and Graves9), we investigated the effects of glyceollins on the PDGF signalling pathway by measuring the phosphorylation levels of MAPK including p38, c-Jun N-terminal protein kinase (JNK) and ERK1/2 with Western blotting (Fig. 4(b)) in HASMC. As shown in Fig. 4(b), PDGF substantially increased ERK1/2 phosphorylation, but not p38 and JNK phosphorylation, at 15 min after stimulation, whereas cells incubated with glyceollins at various concentrations (0·3, 1·5 and 3 μg/ml) diminished the ERK1/2 phosphorylation. Unlike ERK1/2, JNK and p38 phosphorylation were not affected by glyceollins, indicating that glyceollins influence ERK1/2 signalling pathways on PDGF-stimulated HASMC.
Role of reactive oxygen species on glyceollin-mediated inhibition of intracellular signalling transduction by platelet-derived growth factor
Various ROS are associated with PDGFR-mediated signalling events(Reference Sundaresan, Yu and Ferrans10, Reference Wedgwood and Black38) and can affect the activation of MAPK(Reference Parinandi, Kleinberg and Usatyuk39). To confirm the role of ROS in glyceollin-mediated inhibition of the PDGF signalling pathway, dichlorofluorescein and dihydroethidium (DHE) assays were carried out. The results revealed that PDGF-BB (20 ng/ml) induced ROS formation up to 2·1-fold compared with DMSO-treated cells as control, whereas treatment with glyceollins (0·3, 1·5 and 3 μg/ml) markedly inhibited the reaction to 10·5 (sd 1·2), 12·3 (sd 1·2) and 25·1 (sd 1·3) %, respectively (Fig. 5(a)). Furthermore, the dihydriethidium assay (Fig. 5(b)) showed that glyceollins also remarkably inhibited superoxide anion production stimulated by PDGF. These results indicate that glyceollins negatively regulate PDGF-increased ROS generation in VSMC.
Fig. 5 Effects of glyceollins on platelet-derived growth factor (PDGF)-stimulated reactive oxygen species production. (a) Relative fluorescence of human aortic smooth muscle cells (HASMC) stained with dichlorofluoresceindiacetate. Cells were treated with glyceollins in the presence or absence of PDGF (20 ng/ml). Values are means of six separate experiments, with standard deviations represented by vertical bars. (b) Fluorescence micrographs of HASMC stained with dihydriethidium. (i) Control; (ii) glyceollins (3 μg/ml); (iii) PDGF; (iv) PDGF+glyceollins (0·3 μg/ml). Magnification × 100.
Effects of glyceollins on platelet-derived growth factor-stimulated human aortic smooth muscle cell migration
In order to investigate the effect of glyceollins on HASMC migration, the cells were treated for 24 h with different concentrations of glyceollins in the presence or absence of PDGF in a Boyden chamber. The cells that migrated to the lower chamber were stained with calcein-AM, and were photographed with a fluorescence microscope. As shown in Fig. 6, treatment with glyceollins alone had no significant effect on HASMC migration compared with the DMSO-treated cells. However, PDGF significantly increased migration up to about 291·7 (sd 8) %, while glyceollins effectively inhibited the PDGF-stimulated cell migration at 3 μg/ml.
Fig. 6 Effect of glyceollins on platelet-derived growth factor (PDGF)-induced human aortic smooth muscle cell (HASMC) migration. HASMC were seeded onto the transwell plate with or without glyceollins (3 μg/ml) in the presence or absence of PDGF-BB (20 ng/ml) for 24 h. Migrated HASMC were analysed with fluorescence calcein-acetoxymethyl ester (AM) staining. (a)–(d) Representative photomicrographs of the migrated cells to the lower chambers after calcein-AM staining. Magnification × 20. (a) Control; (b) glyceollins (3 μg/ml); (c) PDGF; (d) PDGF+glyceollins (0·3 μg/ml). (e) Percentage of migrated cells relative to the control. Values are means of three separate experiments, with standard deviations represented by vertical bars. * Mean value was significantly different from that without PDGF and glyceollins (P < 0·05). † Mean value was significantly different from that with PDGF in the absence of glyceollins (P < 0·05).
To confirm the effect of glyceollins on the cell migration assay, we carried out a wound healing assay, and observed the cellular morphology and migration behaviour after treatment with glyceollins. As shown in Fig. 7, glyceollins alone did not induce any change in the migration compared with control, whereas treatment with PDGF substantially promoted cell migration and cellular morphological change during the wound-healing process at the same time. However, glyceollins significantly reduced the PDGF-induced cell migration up to same level as the non-treated group.
Fig. 7 Effect of glyceollins on wound healing. Confluent human aortic smooth muscle cells (HASMC) were wounded by a scratch injury line made with a sterile cell scraper. After washing, the cells were treated with or without glyceollins (3 μg/ml) in the presence or absence of platelet-derived growth factor (PDGF) (20 ng/ml). After incubation for 16 h, photographs were taken (a–d). Dotted black lines delimit the initially wounded regions. Magnification × 20. (a) Control; (b) glyceollins (3 μg/ml); (c) PDGF; (d) PDGF+glyceollins (0·3 μg/ml). (e) Percentage of migrated area relative to control. Values are means of three separate experiments, with standard deviations represented by vertical bars. * Mean value was significantly different from that without PDGF and glyceollins (P < 0·05). † Mean value was significantly different from that with PDGF in the absence of glyceollins (P < 0·05).
Effect of glyceollins on rearrangement of actin filaments by platelet-derived growth factor
Since it is known that the active rearrangement of actin filaments is related to VSMC migration, we attempted to determine the effects of glyceollins in cells stimulated by PDGF-BB (Fig. 8). Treatment with glyceollins alone did not result in any change in actin microfilaments of HASMC. On the contrary, incubation with glyceollins (3 μg/ml) prevented the PDGF-induced loss of microfilaments, and the appearance of actin filaments looked similar in cells treated with either DMSO or glyceollins only.
Fig. 8 Effect of glyceollins on platelet-derived growth factor (PDGF)-stimulated cytoskeletal reorganisation. The cells grown on coverslips were treated with or without glyceollins (3 μg/ml) in the presence or absence of PDGF-BB (20 ng/ml) for 24 h. The cells were fixed, the actin filaments were stained with rhodamine phalloidine, and photographed using a fluorescent microscope. (a) Control; (b) glyceollins (3 μg/ml); (c) PDGF; (d) PDGF+glyceollins (0·3 μg/ml). Magnification × 200.
Discussion
Many studies have shown anti-atherosclerotic effects by the components of various natural products such as polyphenols(Reference Arts and Hollman40–Reference Heber42) and flavonoids with antioxidant activity(Reference Kim, Zhang and Kim43–Reference Miller and Ruiz-Larrea45). Oestrogen has been demonstrated to have anti-atherosclerotic effects(Reference Jiang, Liu and Li46). The beneficial effects of soya isoflavones, a group of natural phyto-oestrogens, on atherosclerosis have been known to be associated with oestrogenic properties. Several studies on the oestrogenic soya isoflavone genistein showed its inhibitory effects on not only the immune reaction(Reference Sosroseno, Bird and Seymour47), but also the proliferation and migration of VSMC(Reference Pan, Ikeda and Takebe48). As the signalling pathway associated with VSMC, genistein inhibited smooth muscle cell proliferation through the increase in p27kip1 in vivo (Reference Yu, Lee and Lim49).
Glyceollins have been known to be de novo synthesised in raw soyabean exposed to some biotic elicitors such as fungi and have also been reported to have oestrogenic activity(Reference Kim, Suh and Kim32). Our preliminary studies showed that glyceollins have anti-inflammatory effects on the lipopolysaccharide-induced immune response of RAW264·7 cells. Furthermore, our previous study demonstrated that the glyceollins have strong antioxidant potential in radical-scavenging activity, and inhibitory activity against singlet oxygen and lipid peroxidation. The radical-scavenging activity of glyceollins as assessed by the 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) assay has been shown to be higher than daidzein but similar to genistein(Reference Kim, Suh and Kim50). Considering that 84 % of the mixture is glyceollin I, it is possible to speculate that glyceollin I could be the major compound for the antioxidant activity of the glyceollin mixture. Several studies also supported our presumption that glyceollin isomer I had stronger physiological activity than isomers II and III(Reference Zimmermann, Tilghman and Boué51, Reference Payton-Stewart, Khupse and Boué52).
In the present study, we investigated the effects of glyceollins on PDGF-induced proliferation and migration of HASMC, and evaluated the mechanism of action of glyceollins. The present results demonstrated that glyceollins potently inhibited the PDGF-stimulated proliferation and DNA synthesis without cytotoxicity. The cell cycle is controlled by changes in the expression of cell cycle-regulatory proteins such as phosphorylated retinoblastoma protein (pRB), CDK, cyclins, CDK inhibitor proteins p21cip1 and p27kip1, and tumour suppressors including p53(Reference Braun-Dullaeus, Mann and Sedding53–Reference Sherr and Roberts55). CDK2 and CDK4 are key mediators during the G0/G1 to S phase progression of the cell cycle and form complexes with cyclin E and cyclin D1, respectively. These complexes phosphorylate pRB and then release transcription factors that promote DNA synthesis(Reference Sherr54–Reference Martin, Odajima and Hunt57). The inhibitory effects of glyceollins on PDGF-induced proliferation might be due to the G0/G1-S phase arrest, which was, in turn, mediated by reduced expression of cyclin E and CDK2, and increased p27kip1 and p53 expressions. On the other hand, glyceollins did not have any effect on the expression of cyclin D1, CDK4 and p21cip1 in PDGF-stimulated HASMC. Although numerous CDK are involved in the activation of pRB, the suppression of CDK2 might be enough to inhibit the phosphorylation of pRB(Reference Akiyama, Ohuchi and Sumida58, Reference Connell-Crowley, Elledge and Harper59). In fact, the previous study suggested that inhibition of CDK4 might not be necessary for cell cycle progression arrest(Reference Brooks, Gray and Joly60, Reference Sasaguri, Ishida and Kosaka61). Therefore, inhibition of CDK2/cyclin E expression may be sufficient for cell cycle arrest. This result indicates that glyceollins have influence on signal transduction events in the G0/G1-S interphase rather than those involved in the S-G2/M phase.
Proliferation of VSMC plays an important role in the progression of atherosclerosis and restenosis after vascular injury and PDGF, a peptide growth factor, triggers signals for the proliferation of VSMC through the autophosphorylation of PDGFR and the interaction with cytoplasmic proteins including PLCγ, PI3K/Akt, the signal transducers and activators of transcription (STAT) family, and ROS production(Reference Heldin and Westermark62–Reference Sundaresan, Yu and Ferrans64). In agreement with previous studies on inhibitory effects of an antioxidant derivative, a PDGF-Rβ antagonist or protein kinase inhibitor through inactivation of proteins mentioned above(Reference Heldin and Westermark62–Reference Choi, Lee and Kim65), the present study demonstrated that glyceollins reduced the activation of PDGFR-β, ERK1/2, Akt and PLCγ1 in PDGF-stimulated HASMC. In addition, PDGF increases the intracellular concentration of ROS in VSMC. These effects were abolished by a ROS scavenger, resulting in a decrease in VSMC proliferation(Reference Wedgwood and Black38, Reference Sundaresan, Yu and Ferrans64). H2O2 derived from superoxide anions can form hydroxyl radicals, and which, in turn, leads to phosphorylation of ERK1/2 and PLCγ1 in VSMC(Reference Brown and Griendling63–Reference Choi, Lee and Kim65). It was shown that an antioxidant derivative abolished this reaction of H2O2(Reference Hsieh, Liu and Hsu66). The present results show that glyceollins obviously inhibit the PDGF-induced production of superoxide anions as well as other ROS in HASMC, and diminish the PDGF phosphorylation of PDGFR-β, PLCγ1, Akt and ERK. Particularly, unlike the activation of Akt by PDGF, the total Akt protein was decreased. According to a previous study, PDGF gave rise to regulated proteolytic down-regulation of Akt by PI3K and proteasome activities, resulting in controlling the activity and function of Akt in VSMC(Reference Adachi, Katsumura and Fujii67). In agreement with the results from proliferation, cell cycle progression and PDGF-signal transduction mechanisms, glyceollins also blocked the PDGF-treated HASMC migration.
The purity of glyceollins isolated from a methanolic extract of A. sojae-inoculated sprouting soyabeans was determined to be over 95 % by HPLC-mass analysis and the relative ratio of glyceollin isomers I, II and III was 12:1·3:1, respectively. Therefore, the biological activity observed with crude glyceollins in the study is most likely to be attributable to the glyceollins themselves.
The concentrations of glyceollin mixture used in the present study were 0·3, 1·5 and 3 μg/ml which are equivalent to 1, 5 and 10 μm (glyceollin I), unless specially indicated. The concentration of isoflavones used in most studies ranged from 16 to 500 μm, although some studies reported physiological activity of genistein or daidzein at doses as low as 0·1–20 μm(Reference Chen, Garner and Anderson68, Reference Chacko, Chandler and Mundhekar69). The concentration of glyceollins used in the study is physiologically relevant because it could be readily achievable by a typical dietary intake of isoflavones in adults, and glyceollins may be better absorbed in the gastrointestinal tract due to their relatively high hydrophobicity(Reference Arai, Uehara and Sato70). The present results indicate that glyceollins attenuate the growth of VASMC at 0·3 μg/ml (about 1 μm-glyceollin I) under the PDGF stimulation while glyceollins at 3 μg/ml (about 10 μm-glyceollin I) completely abrogated the effect of PDGF.
Conclusion
The present results indicate that glyceollins inhibit PDGF-induced HASMC proliferation through G0/G1-S phase arrest by decreasing the expression of cyclin E and CDK2, and increasing the expression of p27kip1 and p53. Inactivation of PDGFR-β, PLCγ1, Akt, ERK, and reduction of ROS are presumed to be the possible mechanisms of the inhibitory effects of glyceollins. In addition to anti-proliferative effects, glyceollins have an anti-migrative effect on PDGF-stimulated HASMC. This property of glyceollins may play an important role for the beneficial effects on CVD and angioplasty.
Acknowledgements
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (no. 2010-0009928 and 2010-0027204).
H. J. K. and B.-Y. C. contributed equally as corresponding author to the study.
J. S. K. and J.-T. W. planned the study. H. J. K. and B.-Y. C. performed the proliferation, Western blotting, migration, wound healing, and cytoskeletal reorganisation assay. B. C. and J. S. L. performed the ROS and flow cytometric analysis. H. J. K., B.-Y C. and J.-S. K. wrote the manuscript.
The authors have declared no conflict of interest.
