- Email: [email protected]
Raddeanin A, a triterpenoid saponin isolated from Anemone raddeana, suppresses the angiogenesis and growth of human colorectal tumor by inhibiting VEGFR2 signaling
Phytomedicine 22 (2014) 103–110
Contents lists available at ScienceDirect
Phytomedicine journal homepage: www.elsevier.com/locate/phymed
Raddeanin A, a triterpenoid saponin isolated from Anemone raddeana, suppresses the angiogenesis and growth of human colorectal tumor by inhibiting VEGFR2 signaling Ying-Yun Guan a,1, Hai-Jun Liu a,1, Xin Luan a, Jian-Rong Xu a, Qin Lu a, Ya-Rong Liu a, Yun-Ge Gao a, Mei Zhao b,∗, Hong-Zhuan Chen a,∗, Chao Fang a,∗ a Hongqiao International Institute of Medicine, Shanghai Tongren Hospital and Department of Pharmacology, Institute of Medical Sciences, Shanghai Jiao Tong University School of Medicine (SJTU-SM), Shanghai 200025, China b Department of Pharmacy, Shanghai Institute of Health Sciences and Health School Attached to SJTU-SM, 279 Zhouzhu Road, Shanghai 201318, China
a r t i c l e
i n f o
Article history: Received 10 July 2014 Revised 27 August 2014 Accepted 15 November 2014 Keywords: Anemone raddeana Raddeanin A Tumor angiogenesis VEGFR2 Colorectal tumor Zebrafish
a b s t r a c t Raddeanin A (RA) is an active triterpenoid saponin from a traditional Chinese medicinal herb, Anemone raddeana Regel. It was previously reported that RA possessed attractive antitumor activity through inhibiting proliferation and inducing apoptosis of multiple cancer cells. However, whether RA can inhibit angiogenesis, an essential step in cancer development, remains unknown. In this study, we found that RA could significantly inhibit human umbilical vein endothelial cell (HUVEC) proliferation, motility, migration, and tube formation. RA also dramatically reduced angiogenesis in chick embryo chorioallantoic membrane (CAM), restrained the trunk angiogenesis in zebrafish, and suppressed angiogenesis and growth of human HCT-15 colorectal cancer xenograft in mice. Western blot assay showed that RA suppressed VEGF-induced phosphorylation of VEGFR2 and its downstream protein kinases including PLCγ 1, JAK2, FAK, Src, and Akt. Molecular docking simulation indicated that RA formed hydrogen bonds and hydrophobic interactions within the ATP binding pocket of VEGFR2 kinase domain. Our study firstly provides the evidence that RA has high antiangiogenic potency and explores its molecular basis, demonstrating that RA is a potential agent or lead candidate for antiangiogenic cancer therapy. © 2014 Elsevier GmbH. All rights reserved.
Introduction Raddeanin A (RA) (Fig. 1a) is an oleanane-type triterpenoid saponin extracted from the root of Anemone raddeana Regel, a traditional Chinese medicinal herb (Luan et al. 2013). Previous study showed that RA exerted antitumor activity both in vitro and in vivo. RA inhibited proliferation and induced apoptosis of various human gastric cancer cell lines BGC-823, SGC-7901, MKN-28 and human nonsmall cell lung cancer H460 cells (Gao et al. 2010; Xue et al. 2013). Furthermore, RA exhibited antitumor efficacy in S180, H22 and U14 tumor xenograft in mice (Wang et al. 2008). However, it remains unknown whether RA can suppress angiogenesis, a crucial step in tumor development.
∗
Corresponding author. Tel.: +86 21 64674721. E-mail addresses: [email protected] (M. Zhao), [email protected] (H.-Z. Chen), [email protected] (C. Fang). 1 These authors contributed equally.
http://dx.doi.org/10.1016/j.phymed.2014.11.008 0944-7113/© 2014 Elsevier GmbH. All rights reserved.
Angiogenesis, which involves multiple cells and soluble factors for the formation of new blood vessels from the preexisting ones, plays a pivotal role in the process of tumor growth and metastasis (Grothey and Galanis 2009). Blocking angiogenesis is a validated effective therapeutic approach against cancer, and several antiangiogenic agents (Avastin, Sutent, Nexavar, Votrient, Inlyta, Zaltrap, Stivarga, etc.) (Meadows and Hurwitz 2012; Mullard 2013) have been successfully translated into cancer clinic. Besides the approved monoclonal antibody and small-molecule tyrosine kinase inhibitors (TKIs), natural products from medical herb are currently attracting a growing amount of researchers to excavate their antiangiogenic activity (Song et al. 2012; Wang et al. 2013; Xu et al. 2013; Li et al. 2014). The present study reveals the antiangiogenic potency of RA using human umbilical vein endothelial cell (HUVEC) (a classical in vitro cell model mimicking tumor vascular endothelial cells), chick chorioallantoic membrane (CAM) model, and transgenic zebrafish angiogenesis model. The tumor antiangiogenic efficacy of RA was evaluated in the subcutaneous HCT-15 xenograft mice model. Moreover, the antiangiogenic molecular mechanism of RA was explored by western blot and molecular docking assay.
104
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 1. RA more effectively inhibited the viability of HUVEC relative to HCT-15 cells, and it also significantly suppressed HUVEC motility, migration and tube formation. (a) Chemical structure of RA. (b) RA dose-dependently inhibited the viability of HUVEC, but not HCT-15 cells. Values are expressed as mean ± SD, n = 4. ∗∗∗ p < 0.001 as compared with HCT-15. (c) RA inhibited HUVEC motility. HUVEC treated with different concentration of RA were allowed to migrate on the blue fluorescent beads coated wells for 20 h, then the motion track area of HUVEC was analyzed using Thermo Scientific ArrayScan XTI High Content Analysis Reader. Bar, 200 μm. (d) RA inhibited HUVEC migration in wound healing assay. Dotted lines indicated the boundary of initial scraping. Bar, 500 μm. (e) RA suppresses HUVEC vertical migration in transwell assay. HUVEC migration after 8 h treatment of RA was assayed using Transwell Boyden chamber. The migrated cells were visualized by crystal violet and quantified using Image-Pro Plus 6.0 software. (f) RA inhibited tube formation of HUVEC. HUVEC were placed in the matrigel coated 96-well plate. After 10 h, the tubular structures were photographed. Representative photographs and quantitative analysis are shown in each panel. Values are expressed as mean ± SD, n = 4. ∗ p < 0.05, ∗∗∗ p < 0.001 as compared with control.
Materials and methods Materials, cell lines and animals RA was purchased from Pure-one Bio Technology Company (Shanghai, China). Recombinant human vascular endothelial growth factor (VEGF165 ) was obtained from ProSpec-Tany Technogene Ltd. (Ness Ziona, Israel). Antibodies for western blotting were purchased from Cell Signaling Technology (Danvers, MA). Primary human umbilical vascular endothelial cells (HUVEC) were obtained from Lifeline Cell Technology and cultured in completed endothelial cell medium (Lifeline Cell Technology, Frederick, MD). Human colorectal tumor cell line HCT-15 was obtained from the American Type Culture Collection (Manassas, VA) and cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal
bovine serum (Invitrogen, Carlsbad, CA) and antibiotics (100 mg/ml of streptomycin and 100 U/ml of penicillin). Both HUVEC and HCT15 were cultured at 37 °C in a humidified atmosphere containing 5% CO2 . Zebrafishes (fli1a:EGFP transgenic line) were raised and maintained under standard conditions (Westerfield 1995). Embryos were staged according to the previous protocol (Kimmel et al. 1995). BALB/c nude mice were provided by Shanghai Laboratory Animal Center (Chinese Academy of Sciences, Shanghai, China) and housed in an environmentally controlled quarters (20–25 °C, relative humidity 55–65%, 12 h light/12 h dark cycle) for 5 days before experiment. The food and water were available all the time. The animal experiment designed in this study was approved by the ethical committee of Shanghai Jiao Tong University School of Medicine.
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
105
Cell viability assay
Chick embryo CAM assay
Cell viability was determined by Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan). HUVEC or HCT-15 (5 × 103 cells/well) were seeded in 96-well culture plates and incubated overnight. RA of various concentrations (0.1–10 μM) was added into the wells for 48 h incubation. Then, CCK-8 solution (10 μl) was added and the cells were incubated for additional 2 h. Absorbance was measured at 450 nm using a microplate reader. The percentage of cell viability was calculated against control.
The chick embryonic eggs were incubated in 38 ◦ C with the relative humidity at 65–70%. Six days later, a 2 cm diameter window was opened and the shell membrane was removed to expose the chorioallantoic membrane (CAM). Six millimeter diameter Whatman filter disk as drug carrier that absorbed RA with different concentrations (0.1–10 μM) was put on the CAM. Vehicle (Saline) alone was the control group. Then the window was sealed with sterile parafilm and the eggs were incubated for another 48 h. The CAM was observed and photographed with digital camera, and the neovascularization was quantified by Image-Pro Plus 6.0 software.
Endothelial cell motility assay
Zebrafish angiogenesis study HUVEC motility assay was performed using Cellomics Cell Motility Kit (Thermo Scientific, Rockford, IL), which could quantify cell motility by measuring the size of tracks generated by migrating cells. Briefly, collagen-I coated 96-well plates were prepared, and then the blue fluorescent microsphere solution was added to each well for 1 h incubation in the dark. After washing plate five times, HUVEC (500 cells/well) suspended in 100 μl completed medium containing various concentrations of RA (0.1–10 μM) were added to the plate and incubated for 20 h at 37 °C. Then, 200 μl of 5.5% warmed fixation solution were added to each well and incubated in fume hood at room temperature for 1 h. The plate was then washed three times and added with permeabilization buffer and rhodamine-phalloidin staining solution successively. Then, the cell motility was assayed on the Thermo Scientific ArrayScan XTI High Content Analysis Reader (Rhodamine Conjugates, Ex: 542 nm, Em: 565 nm; Blue Fluorescent Beads, Ex: 365 nm, Em: 415 nm).
Endothelial cell wound healing assay HUVEC were seeded in 12-well plate and allowed to grow to confluence. After scratched with pipette tips, HUVEC were treated with different concentrations of RA (0.1–10 μM) for 24 h. After another 48 h, microphotographs were taken with EVOS microscope (Life Technologies, Grand Island, NY) for quantifying endothelial cell migration.
Endothelial cell transwell migration assay HUVEC migration assay was performed in 24-well Transwell Boyden chambers with polycarbonate filter of 8 mm pore size (Corning, Tewksbury, MA). Briefly, the bottom chambers were filled with 600 μl completed endothelial cell medium supplemented with 20 ng/ml VEGF165 . HUVEC (2×104 cells/well) suspended in 100 μl completed medium plus various concentrations of RA (0.1–10 μM) were seeded in the top chambers. Cells were allowed to migrate for 8 h. The nonmigrated cells on the top surface of the membrane were removed with cotton swab, and the migrated cells were fixed with cold 1% glutaraldehyde and stained with 0.1% crystal violet. Images were taken using an inverted microscope (Carl Zeiss), and migrated cells were quantified by Image-Pro Plus 6.0 software (Media Cybernetics, Bethesda, MD).
To evaluate blood vessels formation in zebrafish embryos, 18 hpf embryos were distributed into 12-well plates (30 embryos per well) for a treatment period of 30 h. RA was diluted in 0.1% dimethyl sulfoxide (DMSO) at a concentration of 0.4 μM. The positive control for this assay was 5 μM PTK787, a VEGFR antagonist (Chan et al. 2002; Bayliss et al. 2006), and the negative control was 0.1% DMSO. After treatment, embryos were anesthetized with 0.016% MS-222 (tricaine methanesulfonate, Sigma– Aldrich, St. Louis, MO) and the number of complete intersegmental vessels (ISVs), i.e. the number of ISVs that connect the dorsal aorta (DA) to the dorsal longitudinal anastomotic vessel (DLAV) was counted. Embryos and larvae were analyzed with Nikon SMZ 1500 Fluorescence microscope and subsequently photographed with digital cameras. Quantitative image analyses were performed using image based morphometric analysis (NIS-Elements D3.1, Japan). A subset of images was adjusted for levels, brightness, contrast, hue and saturation with Adobe Photoshop 7.0 software (Adobe, San Jose, California) to optimally visualize the expression patterns. The inhibition rate was calculated by following formula:
ISVamount of experimental group inhibition rate % = 1 − × 100 ISVamount of vehicle control
Anticancer therapy of RA in subcutaneous HCT-15 xenograft in mice HCT-15 cells were s.c. injected into the 5-week-old female BALB/c nude mice (1 × 106 cells/mouse) to establish xenograft tumor model. After tumors grew to about 100 mm3 , mice were randomly divided into two groups (6 mice per group) and then treated i.p. with or without RA (5 mg/kg) once every 2 days. The tumor volume and body weight were recorded every other day. The tumor sizes were measured using calipers and were calculated using the formula: volume (mm3 ) = (length × width2 )/2. To further investigate the antiangiogenic effect of RA in vivo, on 22 day, mice were sacrificed and the resected tumor tissues were prepared for paraffin sections and performed histological assay. The tumor vessels were stained using rabbit anti-mouse CD31 antibody (1:200, Abcam, Hong Kong); the tumor cell apoptosis was identified using ApopTag Peroxidase In Situ Apoptosis Detection Kit (Merck Millipore, Billerica, MA); and H&E staining was performed for tumor necrosis statistical analysis. All the slices were photographed by Leica DFC 320 photomicroscope. The slices were analyzed for tumor necrosis area, microvessel density (MVD), and TUNEL-positive cells using Image-Pro Plus 6.0 software.
Endothelial cell tube formation assay Western blot assay Matrigel (BD Biosciences, San Jose, CA) was pipetted into prechilled 96-well plates (50 μl per well) and polymerized for at least 30 min. HUVEC (1 × 104 cells/well) suspended in 100 μl completed medium plus various concentrations of RA (0.1–10 μM) were placed onto the layer of Matrigel. Cells were allowed to form tubes for 10 h and then photographed using an EVOS microscope, the tube length were quantified by Image-Pro Plus 6.0 software.
To determine the effect of RA on VEGF-dependent angiogenesis signaling cascade, western blot was performed. Briefly, HUVEC were seeded in 6-well plates (1×105 cells/well) and incubated overnight. After starvation in serum-free medium for 6 h, HUVEC were pretreated with or without various concentrations of RA for 30 min followed by stimulation with 100 ng/ml of VEGF for 4 min. The whole
106
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 2. RA blocked angiogenesis in CAM. (a) Representative photographs of CAM treated with different concentrations of RA. (b) Statistical analysis of inhibitory effect of RA. Values are expressed as mean ± SD, n = 5. ∗∗∗ p < 0.001 as compared with control.
cell extracts were sampled with RIPA Lysis Buffer supplemented with PMSF (Beyotime, Shanghai, China) and PhosSTOP Phosphatase Inhibitor Cocktail (Roche, Rotkreuz, Switzerland). Protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Rockford, IL) and equalized before loading. Forty micrograms of membrane protein from each sample was applied to 8–10% SDSpolyacrylamide gel and probed with specific antibodies followed by exposure to a horseradish peroxidase-conjugated goat anti-rabbit antibody (Cell Signaling Technology, Danvers, MA). Molecular docking By using MOE 2011.10 (Molecular Operating Environment), computational docking was performed to explore molecular interactions between VEGFR2 and RA. X-ray crystal structures (PDB ID: 3VHE) of VEGFR2 kinase domain and its ligand were obtained from Protein Data Bank (http://www.rcsb.org). Water molecules and other heteroatom were manually removed out from the protein structures, and 3D hydrogen coordinates were given through protonate 3D module under MMFF94x. The 3D structure of RA was generated through energy minimization in MOE. With Site Finder module, the potential binding sites were searched for VEGFR2 kinase domain. The sites containing key residues, such as Leu840, Glu885, Leu1035 and Asp1046, responsible for other VEGFR2 kinase inhibitor were kept as the target sites. Then, RA was docked into these sites, and the molecular interactions were analyzed and visualized by Ligand Interaction module and PyMOL. Statistical analysis All data are presented as mean ± SD. Statistical analysis and graphical representation of the data were performed using GraphPad Prism 6.0 (GraphPad Software, San Diego, CA). Differences between groups were examined using Student’s t-test or ANOVA with Bonferroni’s multiple comparison tests. Differences were considered significant if p value was less than 0.05. Results RA inhibited HUVEC proliferation at concentrations not affecting HCT-15 RA can dose-dependently inhibit HUVEC proliferation as shown in decreased cell viability (Fig. 1b). The proliferation of HUVEC was almost completely prohibited at RA concentration of 10 μM. In contrast, HCT-15 cells were not affected at all tested concentrations. These
results suggest that RA can selectively kill tumor endothelial cells relative to tumor cells. RA suppressed HUVEC motility, migration, and tube formation Angiogenesis requires directed cell motility for efficient migration (Vitorino and Meyer 2008), thus we used Cellomics Cell Motility Kit to quantify the motion track area of HUVEC treated with RA. It was shown that RA dose-dependently restrained HUVEC movement and RA at the concentration higher than 3 μM completely inhibited the motion of HUVEC (Fig. 1c). Two migration assay methods were used to evaluate the influence of RA on EC migration. It was shown that RA significantly inhibited HUVEC horizontal migration to the wound area in the wound healing assay (Fig. 1d). In the transwell assay, even at the lowest tested concentration (0.1 μM), RA could block nearly 40% EC migration (Fig. 1e). RA also significantly inhibited HUVEC tube formation at the concentration above 0.3 μM (Fig. 1f). It is noted that RA could suppress HUVEC motility, migration, and tube formation in non-toxic concentrations (Fig. 1c–f). RA restrained the angiogenesis in chick embryo CAM To further evaluate the antiangiogenic effect of RA on vascular development in vivo, we applied the chick embryo CAM model to examine the inhibitory effect of RA on angiogenesis. After 48 h, the formation of new blood vessels was significantly blocked in RA-treated CAM comparing to that in control, and the inhibitory effect of RA was dose-dependent (Fig. 2). RA inhibited angiogenic vessel growth in zebrafish The potential antiangiogenic effect of RA was further translated and verified using a zebrafish model. In the vehicle control, complete ISVs formation was clearly visible in the embryo (Fig. 3A, D, H). In contrast, average less than 10 ISVs were observed, corresponding to nearly 68% ISVs inhibition, in the embryos treated with 0.4 μM RA (Fig. 3B, E, I, K, L). The positive control at the concentration of 12.5-fold higher than RA resulted in 100% inhibition of ISVs (Fig. 3C, F, J, K). RA inhibited HCT-15 xenograft tumor growth through antiangiogenic effect To investigate the effect of RA on tumor growth of human colorectal cancer, nude mice bearing subcutaneous HCT-15 xenograft were
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
107
Fig. 3. RA suppressed the trunk angiogenesis in zebrafish. (A–J) Representative bright field and fluorescent images of zebrafish embryos at 49 h post-fertilization (hpf) treated with 0.1% dimethyl sulfoxide (DMSO) (control), 0.4 μM RA or 5 μM PTK787 (positive control) for 30 h. (D–J) Compared with control, embryos treatment with RA presented an increased number of incomplete ISVs and only occasional sprouts (asterisk) of dorsal aorta could be observed. The boxed regions are shown at higher magnification in the right panels. Black arrows in B and C indicate the pericardial edema after RA or PTK787 treatment. DLAV, dorsal longitudinal anastomotic vessels; ISV, intersegmental vessels; DA, dorsal aorta. (K) Statistical analysis of number of ISVs. (L) Inhibition rate of ISVs after RA and PTK787 treatment. ∗∗∗ p < 0.001 as compared with control.
treated with 5 mg/kg RA every other day for total 11 i.p. injections. It showed that RA significantly inhibited HCT-15 tumor growth and decreased the tumor size. The tumor volume in RA-treated group was 765.3 ± 156.6 mm3 , much smaller than that of control group (1919.0 ± 670.6 mm3 ) (Fig. 4a, c). The average tumor weight of control group was 2.2 ± 0.5 g, whereas that of RA-treated group was only 1.2 ± 0.3 g (Fig. 4b). The mice weight loss after RA treatment was not significant, suggesting that RA did not cause obvious toxicity to the mice at curative dose (Fig. 4d). Immunohistochemical and pathological examinations showed that RA treatment significantly decreased intratumoral MVD, elevated TUNEL-positive cells, and led to increased tumor necrotic area compared with control group (Fig. 4e–g). These results indicated that the antitumor effect of RA was related to its antiangiogenic activity. RA down-regulated the activation of VEGFR2 and its downstream proteins To investigate the underlying molecular mechanism of RAmediated antiangiogenesis, we performed western blot assay to elucidate whether RA could inhibit VEGFR2 phosphorylation, and prohibit the activation of its downstream signaling pathway, which closely related to the survival, migration and proliferation of EC. RA strongly inhibited VEGF-induced VEGFR2 phosphorylation in a dose-dependent manner (Fig. 5a). Furthermore, the activation of downstream signaling of VEGFR2, including PLCγ 1, JAK2, FAK, Src, and Akt, was also decreased when treated with different concentration of RA (Fig. 5a). RA docked into the ATP binding pocket of VEGFR2 kinase domain As RA down-regulated the phosphorylation of VEGFR2 and its downstream signaling molecules, it was hypothesized that RA may interact with VEGFR2 kinase domain. Then, molecular docking simulation was performed to investigate the possible binding pattern between RA and VEGFR2 kinase domain. It was shown that RA was well docked into the ATP binding pocket of VEGFR2 kinase domain (Fig. 5b). The pentacyclic triterpene moiety of RA was trapped in the
hydrophobic pocket, which was composed of Leu840, Leu1035, and Phe1047 (Fig. 5b, c). There were also hydrogen bonds between the saccharide moiety of RA and the key residues of ATP binding pocket, including Lys868, Glu885, and Asp1046 (Fig. 5b, c). In addition, RA also moderately interacted with other amino acid residues, including Gly922, Val848, Val916, Val899, Ala881, Leu1049 and Arg1027 through the hydrophobic interaction and hydrogen bonds (Fig. 5c).
Discussion Antiangiogenic intervention mediated by natural products is a promising research area that provides an effective anticancer strategy. Many natural compounds have been proved to possess antiangiogenic activity in vitro and in vivo. Previous studies showed that RA exerted antitumor activity on several kinds of cancer in vitro and in vivo by inducing apoptosis and inhibiting invasion, migration and adhesion of tumor cells (Wang et al. 2008; Gao et al. 2010; Xue et al. 2013). However, there has been no report regarding the antiangiogenic activity of RA so far. In this study, we firstly found that RA possessed remarkable antiangiogenic effect both in vitro and in vivo. RA interrupted a series of orchestrated processes of angiogenesis, including EC proliferation, motility, migration and capillary-like tube formation in a dose-dependent manner (Fig. 1b–f), and inhibited in vivo chick CAM angiogenesis (Fig. 2). Transgenic zebrafish was also adopted to further evaluate the antiangiogenic activity of RA. The quantification of the number of complete intersegmental vessels (ISVs) is one of the reliable parameter to evaluate angiogenesis in vivo (Vasil et al. 2009; Murphy et al. 2010). RA could inhibit the ISVs formation of zebrafish in a non-lethal dose. As shown in Fig. 3, 0.4 μM RA effectively disrupted the completion of ISVs and only occasional sprouts of dorsal aorta were observed. Both PTK787 (positive control) and RA significantly blocked blood vessel formation, which may be responsible for the observed pericardial edema (Chan et al. 2002). It is also noted that the morphology of zebrafishes treated with RA deformed and curved to some extent. Prompted by the robust antiangiogenic activity of RA, we applied
108
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 4. RA (5 mg/kg, i.p. once every 2 days) inhibited tumor growth through antiangiogenic activity in a subcutaneous HCT-15 tumor model of mice. (a) Representative photographs of the tumor bearing mice and the resected tumor in the control and RA-treated group. (b) The tumor weight at the end of the study. (c) Mice tumor volume. (d) Mice body weight. Representative staining sections and quantitative analysis of CD31 for microvessel density (MVD) (e), TUNEL for apoptotic cells (f), and H&E for necrosis area (g) of tumor treated with control and RA. Bar, 100 μm in (e) and (f), and bar, 1 mm in (g). RA could significantly reduce the tumor volume and tumor weight in HCT-15 tumor bearing mice without obvious toxicity (no distinct weight loss observed). The MVD was significant decreased and the percentage of TUNEL positive cells and necrosis area were pronouncedly increased after RA treatment. Values are expressed as mean ± SD, n = 6. ∗∗ p < 0.01, ∗∗∗ p < 0.001 as compared with control.
HCT-15 human colorectal tumor xenograft in mice to investigate whether RA could exert antitumor effect through its high antiangiogenic potency. RA significantly suppressed tumor growth and tumor weight, induced tumor cell apoptosis, and increased tumor necrosis without obvious adverse effects such as body weight loss (Fig. 4a, b, c, d, f, g). The antitumor efficacy of RA was found to be closely related to the antiangiogenic activity, which was shown in remarkably decreased MVD in tumor treated by RA (Fig. 4e). The phosphorylation of VEGFR2 is critical for angiogenesis, which mediates several effects of EC function, including EC survival, proliferation, and migration. Western blot assay showed that RA could decrease the VEGF induced VEGFR2 phosphorylation and its downstream signals, including PLCγ 1, JAK2, FAK, Src, and Akt, in a dosedependent manner (Fig. 5a). RA of 3 μM almost totally inhibited the phosphorylation of VEGFR2, while the total protein of VEGFR2 remained unaffected, suggesting that RA can be a potent VEGFR2 inhibitor. The down-regulation of PLCγ 1 is related to the inhibition of EC proliferation (Sase et al. 2009). The inactivation of FAK and Src can suppress EC migration. JAK2 can mediate the phosphorylation of STAT3, a potent transcription activator in tumor angiogenesis (Zhang et al. 2011). Akt plays a key role in several cellular
functions including cell survival, proliferation, migration, and protein synthesis. The suppression of Akt and JAK2 phosphorylation can affect EC function in various aspects. Taken together, these results indicated that RA modulated VEGF-mediated angiogenesis by blocking the phosphorylation of VEGFR2 and its multiple downstream protein kinases as summarized in Fig. 5d. It is noted that some other triterpenoid natural products and their derivatives also have antiangiogenic activity by influencing various signaling pathway, including VEGF/VEGFR2 pathway (Pang et al. 2010b; Mu et al. 2012; Kim and Kim 2014), bFGF/FGFR1 pathway (Kim and Kim 2014), mTOR/S6K pathway (Pang et al. 2010b), etc. Moreover, many other saponins, such as deltonin (Tong et al. 2011), terrestrosin D (Wei et al. 2014) and ASC (Zeng et al. 2014), also possessed antitumor and antiangiogenic activity. The phosphorylation of VEGFR2 is an ATP consuming process. Molecular docking test indicated that RA could stably locate at the ATP-binding pocket of VEGFR2 kinase domain (Fig. 5b). Six amino acids (Leu840, Leu1035, Phe1047, Lys868, Glu885 and Asp1046) at the ATP binding pocket were essential for the stable conformation of VEGFR2-RA complex (Fig. 5b, c). Some other amino acid residues of ATP binding pocket also have moderate interaction with RA (Fig. 5c). Such bioinformatics of the binding pattern of RA and VEGFR2 can help
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
109
Fig. 5. RA inhibited the kinase activity of VEGFR2 and its downstream signaling molecules. (a) RA suppressed the activation of VEGFR2 and its downstream signaling kinases induced by VEGF in HUVEC. HUVEC were pretreated with or without various concentrations of RA for 30 min followed by the stimulation with 100 ng/ml of VEGF for 4 min. The activation of VEGFR2 and its downstream cascade including PLCγ 1, JAK2, FAK, Src, and Akt were analyzed by western blot and probed with specific antibodies. (b) The predicted binding interactions between RA and VEGFR2. RA was well docked into the ATP binding pocket of VEGFR2 kinase domain. Side chains of crucial residues in the binding site are labeled and shown as purple sticks. (c) Two-dimensional interaction map of RA and ATP binding pocket of VEGFR2 kinase domain was analyzed and visualized by Ligand Interaction module. (d) Signaling pathway diagram of RA-mediated antiangiogenesis.
us better understand the antiangiogenic effect of RA. It appears that the interactions between the sapogenin (but not the saccharide chains of RA) to VEGFR2 contribute more to the stable conformation VEGFR2RA complex. Consistent to our speculation, some non-saponin triterpenoid compounds such pristimerin (Mu et al. 2012) and celastrol (Pang et al. 2010a) also possess potent antiangiogenic activity by inhibiting VEGFR2 activation. However, the saccharide moiety of RA may help improve the hydrophility of the compound and facilitate in vivo distribution. In summary, our findings demonstrated for the first time that RA possessed potent antiangiogenic activity both in vitro and in vivo as a VEGFR2 inhibitor, and it may be a potential drug candidate or lead compound for antiangiogenic cancer therapy. Acknowledgments This work was supported by National Basic Research Program of China (No. 2010CB529806), National Natural Science Foundation of China (No. 81272569), Shanghai Pujiang Program (No. 12PJD023), Innovation Program of Shanghai Municipal Education Commission (Nos. 12ZZ200, 13ZZ087), and Shanghai Municipal Science and Technology Commission (No. 14JC1491900).
References Bayliss, P.E., Bellavance, K.L., Whitehead, G.G., Abrams, J.M., Aegerter, S., Robbins, H.S., Cowan, D.B., Keating, M.T., O’Reilly, T., Wood, J.M., Roberts, T.M., Chan, J., 2006. Chemical modulation of receptor signaling inhibits regenerative angiogenesis in adult zebrafish. Nat. Chem. Biol. 2, 265–273. Chan, J., Bayliss, P.E., Wood, J.M., Roberts, T.M., 2002. Dissection of angiogenic signaling in zebrafish using a chemical genetic approach. Cancer Cell 1, 257–267. Gao, Y., An, J., Zhu, Q., 2010. Inhibitory effect of Raddeanin A in human non-small cell lung cancer H460 cells. Chin. J. Appl. Environ. Biol. 16, 637–641. Grothey, A., Galanis, E., 2009. Targeting angiogenesis: progress with anti-VEGF treatment with large molecules. Nat. Rev. Clin. Oncol. 6, 507–518. Kim, H.J., Kim, J.K., 2014. Antiangiogenic effects of cucurbitacin-I. Arch. Pharm. Res. doi:10.1007/s12272-014-0386-5. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310. Li, K.K., Liu, C.L., Tam, J.C., Kwok, H.F., Lau, C.P., Leung, P.C., Ko, C.H., Ye, C.X., 2014. In vitro and in vivo mechanistic study of a novel proanthocyanidin, GC-(4→8)GCG from cocoa tea (Camellia ptilophylla) in antiangiogenesis. J. Nutr. Biochem. 25, 319–328. Luan, X., Guan, Y.Y., Wang, C., Zhao, M., Lu, Q., Tang, Y.B., Liu, Y.R., Yu, D.H., Wang, X.L., Qi, H., Fang, C., Chen, H.Z., 2013. Determination of Raddeanin A in rat plasma by liquid chromatography–tandem mass spectrometry: application to a pharmacokinetic study. J. Chromatogr. B 923-924, 43–47. Meadows, K.L., Hurwitz, H.I., 2012. Anti-VEGF Therapies in the Clinic. Cold Spring Harb. Perspect. Med. 2, a006577. Mu, X., Shi, W., Sun, L., Li, H., Jiang, Z., Zhang, L., 2012. Pristimerin, a triterpenoid, inhibits tumor angiogenesis by targeting VEGFR2 activation. Molecules 17, 6854–6868. Mullard, A., 2013. 2012 FDA drug approvals. Nat. Rev. Drug Discov. 12, 87–90.
110
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Murphy, E.A., Shields, D.J., Stoletov, K., Dneprovskaia, E., McElroy, M., Greenberg, J.I., Lindquist, J., Acevedo, L.M., Anand, S., Majeti, B.K., Tsigelny, I., Saldanha, A., Walsh, B., Hoffman, R.M., Bouvet, M., Klemke, R.L., Vogt, P.K., Arnold, L., Wrasidlo, W., Cheresh, D.A., 2010. Disruption of angiogenesis and tumor growth with an orally active drug that stabilizes the inactive state of PDGFRβ /B-RAF. Proc. Natl. Acad. Sci., 107, pp. 4299–4304. Pang, X., Yi, Z., Zhang, J., Lu, B., Sung, B., Qu, W., Aggarwal, B.B., Liu, M., 2010. Celastrol suppresses angiogenesis-mediated tumor growth through inhibition of AKT/mammalian target of rapamycin pathway. Cancer Res. 70, 1951–1959. Pang, X., Zhang, L., Wu, Y., Lin, L., Li, J., Qu, W., Safe, S., Liu, M., 2010. Methyl 2-cyano-3,11-dioxo-18-olean-1,12-dien-30-oate (CDODA-Me), a derivative of glycyrrhetinic acid, functions as a potent angiogenesis inhibitor. J. Pharmacol. Exp. Ther. 335, 172–179. Sase, H., Watabe, T., Kawasaki, K., Miyazono, K., Miyazawa, K., 2009. VEGFR2-PLCγ 1 axis is essential for endothelial specification of VEGFR2+ vascular progenitor cells. J. Cell Sci. 122, 3303–3311. Song, Y., Dai, F., Zhai, D., Dong, Y., Zhang, J., Lu, B., Luo, J., Liu, M., Yi, Z., 2012. Usnic acid inhibits breast tumor angiogenesis and growth by suppressing VEGFR2-mediated AKT and ERK1/2 signaling pathways. Angiogenesis 15, 421–432. Tong, Q.Y., Qing, Y., Shu, D., He, Y., Zhao, Y.L., Li, Y., Wang, Z.L., Zhang, S.Y., Xing, Z.H., Xu, C., Wei, Y.Q., Huang, W., Wu, X.H., 2011. Deltonin, a steroidal saponin, inhibits colon cancer cell growth in vitro and tumor growth in vivo via induction of apoptosis and antiangiogenesis. Cell. Physiol. Biochem. 27, 233–242. Vasil, M.L., Stonehouse, M.J., Vasil, A.I., Wadsworth, S.J., Goldfine, H., Bolcome 3rd, R.E., Chan, J., 2009. A complex extracellular sphingomyelinase of Pseudomonas aeruginosa inhibits angiogenesis by selective cytotoxicity to endothelial cells. PLOS Pathogens 5, e1000420.
Vitorino, P., Meyer, T., 2008. Modular control of endothelial sheet migration. Genes Dev. 22, 3268–3281. Wang, M.K., Ding, L.S., Wu, F.E., 2008. Antitumor effects of raddeanin A on S180, H22 and U14 cell xenografts in mice. Chin. J. Cancer 27, 910–913. Wang, F.L., Sun, J.Y., Wang, Y., Mu, Y.L., Liang, Y.J., Chong, Z.Z., Qin, S.H., Yao, Q.Q., 2013. Oldhamianoside II, a new triterpenoid saponin, prevents tumor growth via inducing cell apoptosis and inhibiting angiogenesis. Oncol. Res. 20, 369–376. Wei, S., Fukuhara, H., Chen, G., Kawada, C., Kurabayashi, A., Furihata, M., Inoue, K., Shuin, T., 2014. Terrestrosin D, a steroidal saponin from Tribulus terrestris L., inhibits growth and angiogenesis of human prostate cancer in vitro and in vivo. Pathobiology 81, 123–132. Westerfield, M., 1995. The Zebrafish Book: A Guide for the Laboratory Use of Zebrafish, 3rd ed. University of Oregon Press, Eugene, USA. Xu, H.Y., Pan, Y.M., Chen, Z.W., Lin, Y., Wang, L.H., Chen, Y.H., Jie, T.T., Lu, Y.Y., Liu, J.C., 2013. 12-Deoxyphorbol 13-palmitate inhibit VEGF-induced angiogenesis via suppression of VEGFR-2-signaling pathway. J. Ethnopharmacol. 146, 724–733. Xue, G., Zou, X., Zhou, J.Y., Sun, W., Wu, J., Xu, J.L., Wang, R.P., 2013. Raddeanin A induces human gastric cancer cells apoptosis and inhibits their invasion in vitro. Biochem. Biophys. Res. Commun. 439, 196–202. Zeng, K., Song, F., Li, N., Dong, X., Jiang, Y., Tu, P., 2014. ASC, a bioactive steroidal saponin from Ophitopogin japonicas, inhibits angiogenesis through interruption of Src tyrosine kinase-dependent matrix metalloproteinase pathway. Basic Clin. Pharmacol. Toxicol. doi:10.1111/bcpt.12305. Zhang, X., Song, Y., Wu, Y., Dong, Y., Lai, L., Zhang, J., Lu, B., Dai, F., He, L., Liu, M., Yi, Z., 2011. Indirubin inhibits tumor growth by antitumor angiogenesis via blocking VEGFR2-mediated JAK/STAT3 signaling in endothelial cell. Int. J. Cancer 129, 2502–2511.
Contents lists available at ScienceDirect
Phytomedicine journal homepage: www.elsevier.com/locate/phymed
Raddeanin A, a triterpenoid saponin isolated from Anemone raddeana, suppresses the angiogenesis and growth of human colorectal tumor by inhibiting VEGFR2 signaling Ying-Yun Guan a,1, Hai-Jun Liu a,1, Xin Luan a, Jian-Rong Xu a, Qin Lu a, Ya-Rong Liu a, Yun-Ge Gao a, Mei Zhao b,∗, Hong-Zhuan Chen a,∗, Chao Fang a,∗ a Hongqiao International Institute of Medicine, Shanghai Tongren Hospital and Department of Pharmacology, Institute of Medical Sciences, Shanghai Jiao Tong University School of Medicine (SJTU-SM), Shanghai 200025, China b Department of Pharmacy, Shanghai Institute of Health Sciences and Health School Attached to SJTU-SM, 279 Zhouzhu Road, Shanghai 201318, China
a r t i c l e
i n f o
Article history: Received 10 July 2014 Revised 27 August 2014 Accepted 15 November 2014 Keywords: Anemone raddeana Raddeanin A Tumor angiogenesis VEGFR2 Colorectal tumor Zebrafish
a b s t r a c t Raddeanin A (RA) is an active triterpenoid saponin from a traditional Chinese medicinal herb, Anemone raddeana Regel. It was previously reported that RA possessed attractive antitumor activity through inhibiting proliferation and inducing apoptosis of multiple cancer cells. However, whether RA can inhibit angiogenesis, an essential step in cancer development, remains unknown. In this study, we found that RA could significantly inhibit human umbilical vein endothelial cell (HUVEC) proliferation, motility, migration, and tube formation. RA also dramatically reduced angiogenesis in chick embryo chorioallantoic membrane (CAM), restrained the trunk angiogenesis in zebrafish, and suppressed angiogenesis and growth of human HCT-15 colorectal cancer xenograft in mice. Western blot assay showed that RA suppressed VEGF-induced phosphorylation of VEGFR2 and its downstream protein kinases including PLCγ 1, JAK2, FAK, Src, and Akt. Molecular docking simulation indicated that RA formed hydrogen bonds and hydrophobic interactions within the ATP binding pocket of VEGFR2 kinase domain. Our study firstly provides the evidence that RA has high antiangiogenic potency and explores its molecular basis, demonstrating that RA is a potential agent or lead candidate for antiangiogenic cancer therapy. © 2014 Elsevier GmbH. All rights reserved.
Introduction Raddeanin A (RA) (Fig. 1a) is an oleanane-type triterpenoid saponin extracted from the root of Anemone raddeana Regel, a traditional Chinese medicinal herb (Luan et al. 2013). Previous study showed that RA exerted antitumor activity both in vitro and in vivo. RA inhibited proliferation and induced apoptosis of various human gastric cancer cell lines BGC-823, SGC-7901, MKN-28 and human nonsmall cell lung cancer H460 cells (Gao et al. 2010; Xue et al. 2013). Furthermore, RA exhibited antitumor efficacy in S180, H22 and U14 tumor xenograft in mice (Wang et al. 2008). However, it remains unknown whether RA can suppress angiogenesis, a crucial step in tumor development.
∗
Corresponding author. Tel.: +86 21 64674721. E-mail addresses: [email protected] (M. Zhao), [email protected] (H.-Z. Chen), [email protected] (C. Fang). 1 These authors contributed equally.
http://dx.doi.org/10.1016/j.phymed.2014.11.008 0944-7113/© 2014 Elsevier GmbH. All rights reserved.
Angiogenesis, which involves multiple cells and soluble factors for the formation of new blood vessels from the preexisting ones, plays a pivotal role in the process of tumor growth and metastasis (Grothey and Galanis 2009). Blocking angiogenesis is a validated effective therapeutic approach against cancer, and several antiangiogenic agents (Avastin, Sutent, Nexavar, Votrient, Inlyta, Zaltrap, Stivarga, etc.) (Meadows and Hurwitz 2012; Mullard 2013) have been successfully translated into cancer clinic. Besides the approved monoclonal antibody and small-molecule tyrosine kinase inhibitors (TKIs), natural products from medical herb are currently attracting a growing amount of researchers to excavate their antiangiogenic activity (Song et al. 2012; Wang et al. 2013; Xu et al. 2013; Li et al. 2014). The present study reveals the antiangiogenic potency of RA using human umbilical vein endothelial cell (HUVEC) (a classical in vitro cell model mimicking tumor vascular endothelial cells), chick chorioallantoic membrane (CAM) model, and transgenic zebrafish angiogenesis model. The tumor antiangiogenic efficacy of RA was evaluated in the subcutaneous HCT-15 xenograft mice model. Moreover, the antiangiogenic molecular mechanism of RA was explored by western blot and molecular docking assay.
104
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 1. RA more effectively inhibited the viability of HUVEC relative to HCT-15 cells, and it also significantly suppressed HUVEC motility, migration and tube formation. (a) Chemical structure of RA. (b) RA dose-dependently inhibited the viability of HUVEC, but not HCT-15 cells. Values are expressed as mean ± SD, n = 4. ∗∗∗ p < 0.001 as compared with HCT-15. (c) RA inhibited HUVEC motility. HUVEC treated with different concentration of RA were allowed to migrate on the blue fluorescent beads coated wells for 20 h, then the motion track area of HUVEC was analyzed using Thermo Scientific ArrayScan XTI High Content Analysis Reader. Bar, 200 μm. (d) RA inhibited HUVEC migration in wound healing assay. Dotted lines indicated the boundary of initial scraping. Bar, 500 μm. (e) RA suppresses HUVEC vertical migration in transwell assay. HUVEC migration after 8 h treatment of RA was assayed using Transwell Boyden chamber. The migrated cells were visualized by crystal violet and quantified using Image-Pro Plus 6.0 software. (f) RA inhibited tube formation of HUVEC. HUVEC were placed in the matrigel coated 96-well plate. After 10 h, the tubular structures were photographed. Representative photographs and quantitative analysis are shown in each panel. Values are expressed as mean ± SD, n = 4. ∗ p < 0.05, ∗∗∗ p < 0.001 as compared with control.
Materials and methods Materials, cell lines and animals RA was purchased from Pure-one Bio Technology Company (Shanghai, China). Recombinant human vascular endothelial growth factor (VEGF165 ) was obtained from ProSpec-Tany Technogene Ltd. (Ness Ziona, Israel). Antibodies for western blotting were purchased from Cell Signaling Technology (Danvers, MA). Primary human umbilical vascular endothelial cells (HUVEC) were obtained from Lifeline Cell Technology and cultured in completed endothelial cell medium (Lifeline Cell Technology, Frederick, MD). Human colorectal tumor cell line HCT-15 was obtained from the American Type Culture Collection (Manassas, VA) and cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal
bovine serum (Invitrogen, Carlsbad, CA) and antibiotics (100 mg/ml of streptomycin and 100 U/ml of penicillin). Both HUVEC and HCT15 were cultured at 37 °C in a humidified atmosphere containing 5% CO2 . Zebrafishes (fli1a:EGFP transgenic line) were raised and maintained under standard conditions (Westerfield 1995). Embryos were staged according to the previous protocol (Kimmel et al. 1995). BALB/c nude mice were provided by Shanghai Laboratory Animal Center (Chinese Academy of Sciences, Shanghai, China) and housed in an environmentally controlled quarters (20–25 °C, relative humidity 55–65%, 12 h light/12 h dark cycle) for 5 days before experiment. The food and water were available all the time. The animal experiment designed in this study was approved by the ethical committee of Shanghai Jiao Tong University School of Medicine.
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
105
Cell viability assay
Chick embryo CAM assay
Cell viability was determined by Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan). HUVEC or HCT-15 (5 × 103 cells/well) were seeded in 96-well culture plates and incubated overnight. RA of various concentrations (0.1–10 μM) was added into the wells for 48 h incubation. Then, CCK-8 solution (10 μl) was added and the cells were incubated for additional 2 h. Absorbance was measured at 450 nm using a microplate reader. The percentage of cell viability was calculated against control.
The chick embryonic eggs were incubated in 38 ◦ C with the relative humidity at 65–70%. Six days later, a 2 cm diameter window was opened and the shell membrane was removed to expose the chorioallantoic membrane (CAM). Six millimeter diameter Whatman filter disk as drug carrier that absorbed RA with different concentrations (0.1–10 μM) was put on the CAM. Vehicle (Saline) alone was the control group. Then the window was sealed with sterile parafilm and the eggs were incubated for another 48 h. The CAM was observed and photographed with digital camera, and the neovascularization was quantified by Image-Pro Plus 6.0 software.
Endothelial cell motility assay
Zebrafish angiogenesis study HUVEC motility assay was performed using Cellomics Cell Motility Kit (Thermo Scientific, Rockford, IL), which could quantify cell motility by measuring the size of tracks generated by migrating cells. Briefly, collagen-I coated 96-well plates were prepared, and then the blue fluorescent microsphere solution was added to each well for 1 h incubation in the dark. After washing plate five times, HUVEC (500 cells/well) suspended in 100 μl completed medium containing various concentrations of RA (0.1–10 μM) were added to the plate and incubated for 20 h at 37 °C. Then, 200 μl of 5.5% warmed fixation solution were added to each well and incubated in fume hood at room temperature for 1 h. The plate was then washed three times and added with permeabilization buffer and rhodamine-phalloidin staining solution successively. Then, the cell motility was assayed on the Thermo Scientific ArrayScan XTI High Content Analysis Reader (Rhodamine Conjugates, Ex: 542 nm, Em: 565 nm; Blue Fluorescent Beads, Ex: 365 nm, Em: 415 nm).
Endothelial cell wound healing assay HUVEC were seeded in 12-well plate and allowed to grow to confluence. After scratched with pipette tips, HUVEC were treated with different concentrations of RA (0.1–10 μM) for 24 h. After another 48 h, microphotographs were taken with EVOS microscope (Life Technologies, Grand Island, NY) for quantifying endothelial cell migration.
Endothelial cell transwell migration assay HUVEC migration assay was performed in 24-well Transwell Boyden chambers with polycarbonate filter of 8 mm pore size (Corning, Tewksbury, MA). Briefly, the bottom chambers were filled with 600 μl completed endothelial cell medium supplemented with 20 ng/ml VEGF165 . HUVEC (2×104 cells/well) suspended in 100 μl completed medium plus various concentrations of RA (0.1–10 μM) were seeded in the top chambers. Cells were allowed to migrate for 8 h. The nonmigrated cells on the top surface of the membrane were removed with cotton swab, and the migrated cells were fixed with cold 1% glutaraldehyde and stained with 0.1% crystal violet. Images were taken using an inverted microscope (Carl Zeiss), and migrated cells were quantified by Image-Pro Plus 6.0 software (Media Cybernetics, Bethesda, MD).
To evaluate blood vessels formation in zebrafish embryos, 18 hpf embryos were distributed into 12-well plates (30 embryos per well) for a treatment period of 30 h. RA was diluted in 0.1% dimethyl sulfoxide (DMSO) at a concentration of 0.4 μM. The positive control for this assay was 5 μM PTK787, a VEGFR antagonist (Chan et al. 2002; Bayliss et al. 2006), and the negative control was 0.1% DMSO. After treatment, embryos were anesthetized with 0.016% MS-222 (tricaine methanesulfonate, Sigma– Aldrich, St. Louis, MO) and the number of complete intersegmental vessels (ISVs), i.e. the number of ISVs that connect the dorsal aorta (DA) to the dorsal longitudinal anastomotic vessel (DLAV) was counted. Embryos and larvae were analyzed with Nikon SMZ 1500 Fluorescence microscope and subsequently photographed with digital cameras. Quantitative image analyses were performed using image based morphometric analysis (NIS-Elements D3.1, Japan). A subset of images was adjusted for levels, brightness, contrast, hue and saturation with Adobe Photoshop 7.0 software (Adobe, San Jose, California) to optimally visualize the expression patterns. The inhibition rate was calculated by following formula:
ISVamount of experimental group inhibition rate % = 1 − × 100 ISVamount of vehicle control
Anticancer therapy of RA in subcutaneous HCT-15 xenograft in mice HCT-15 cells were s.c. injected into the 5-week-old female BALB/c nude mice (1 × 106 cells/mouse) to establish xenograft tumor model. After tumors grew to about 100 mm3 , mice were randomly divided into two groups (6 mice per group) and then treated i.p. with or without RA (5 mg/kg) once every 2 days. The tumor volume and body weight were recorded every other day. The tumor sizes were measured using calipers and were calculated using the formula: volume (mm3 ) = (length × width2 )/2. To further investigate the antiangiogenic effect of RA in vivo, on 22 day, mice were sacrificed and the resected tumor tissues were prepared for paraffin sections and performed histological assay. The tumor vessels were stained using rabbit anti-mouse CD31 antibody (1:200, Abcam, Hong Kong); the tumor cell apoptosis was identified using ApopTag Peroxidase In Situ Apoptosis Detection Kit (Merck Millipore, Billerica, MA); and H&E staining was performed for tumor necrosis statistical analysis. All the slices were photographed by Leica DFC 320 photomicroscope. The slices were analyzed for tumor necrosis area, microvessel density (MVD), and TUNEL-positive cells using Image-Pro Plus 6.0 software.
Endothelial cell tube formation assay Western blot assay Matrigel (BD Biosciences, San Jose, CA) was pipetted into prechilled 96-well plates (50 μl per well) and polymerized for at least 30 min. HUVEC (1 × 104 cells/well) suspended in 100 μl completed medium plus various concentrations of RA (0.1–10 μM) were placed onto the layer of Matrigel. Cells were allowed to form tubes for 10 h and then photographed using an EVOS microscope, the tube length were quantified by Image-Pro Plus 6.0 software.
To determine the effect of RA on VEGF-dependent angiogenesis signaling cascade, western blot was performed. Briefly, HUVEC were seeded in 6-well plates (1×105 cells/well) and incubated overnight. After starvation in serum-free medium for 6 h, HUVEC were pretreated with or without various concentrations of RA for 30 min followed by stimulation with 100 ng/ml of VEGF for 4 min. The whole
106
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 2. RA blocked angiogenesis in CAM. (a) Representative photographs of CAM treated with different concentrations of RA. (b) Statistical analysis of inhibitory effect of RA. Values are expressed as mean ± SD, n = 5. ∗∗∗ p < 0.001 as compared with control.
cell extracts were sampled with RIPA Lysis Buffer supplemented with PMSF (Beyotime, Shanghai, China) and PhosSTOP Phosphatase Inhibitor Cocktail (Roche, Rotkreuz, Switzerland). Protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Rockford, IL) and equalized before loading. Forty micrograms of membrane protein from each sample was applied to 8–10% SDSpolyacrylamide gel and probed with specific antibodies followed by exposure to a horseradish peroxidase-conjugated goat anti-rabbit antibody (Cell Signaling Technology, Danvers, MA). Molecular docking By using MOE 2011.10 (Molecular Operating Environment), computational docking was performed to explore molecular interactions between VEGFR2 and RA. X-ray crystal structures (PDB ID: 3VHE) of VEGFR2 kinase domain and its ligand were obtained from Protein Data Bank (http://www.rcsb.org). Water molecules and other heteroatom were manually removed out from the protein structures, and 3D hydrogen coordinates were given through protonate 3D module under MMFF94x. The 3D structure of RA was generated through energy minimization in MOE. With Site Finder module, the potential binding sites were searched for VEGFR2 kinase domain. The sites containing key residues, such as Leu840, Glu885, Leu1035 and Asp1046, responsible for other VEGFR2 kinase inhibitor were kept as the target sites. Then, RA was docked into these sites, and the molecular interactions were analyzed and visualized by Ligand Interaction module and PyMOL. Statistical analysis All data are presented as mean ± SD. Statistical analysis and graphical representation of the data were performed using GraphPad Prism 6.0 (GraphPad Software, San Diego, CA). Differences between groups were examined using Student’s t-test or ANOVA with Bonferroni’s multiple comparison tests. Differences were considered significant if p value was less than 0.05. Results RA inhibited HUVEC proliferation at concentrations not affecting HCT-15 RA can dose-dependently inhibit HUVEC proliferation as shown in decreased cell viability (Fig. 1b). The proliferation of HUVEC was almost completely prohibited at RA concentration of 10 μM. In contrast, HCT-15 cells were not affected at all tested concentrations. These
results suggest that RA can selectively kill tumor endothelial cells relative to tumor cells. RA suppressed HUVEC motility, migration, and tube formation Angiogenesis requires directed cell motility for efficient migration (Vitorino and Meyer 2008), thus we used Cellomics Cell Motility Kit to quantify the motion track area of HUVEC treated with RA. It was shown that RA dose-dependently restrained HUVEC movement and RA at the concentration higher than 3 μM completely inhibited the motion of HUVEC (Fig. 1c). Two migration assay methods were used to evaluate the influence of RA on EC migration. It was shown that RA significantly inhibited HUVEC horizontal migration to the wound area in the wound healing assay (Fig. 1d). In the transwell assay, even at the lowest tested concentration (0.1 μM), RA could block nearly 40% EC migration (Fig. 1e). RA also significantly inhibited HUVEC tube formation at the concentration above 0.3 μM (Fig. 1f). It is noted that RA could suppress HUVEC motility, migration, and tube formation in non-toxic concentrations (Fig. 1c–f). RA restrained the angiogenesis in chick embryo CAM To further evaluate the antiangiogenic effect of RA on vascular development in vivo, we applied the chick embryo CAM model to examine the inhibitory effect of RA on angiogenesis. After 48 h, the formation of new blood vessels was significantly blocked in RA-treated CAM comparing to that in control, and the inhibitory effect of RA was dose-dependent (Fig. 2). RA inhibited angiogenic vessel growth in zebrafish The potential antiangiogenic effect of RA was further translated and verified using a zebrafish model. In the vehicle control, complete ISVs formation was clearly visible in the embryo (Fig. 3A, D, H). In contrast, average less than 10 ISVs were observed, corresponding to nearly 68% ISVs inhibition, in the embryos treated with 0.4 μM RA (Fig. 3B, E, I, K, L). The positive control at the concentration of 12.5-fold higher than RA resulted in 100% inhibition of ISVs (Fig. 3C, F, J, K). RA inhibited HCT-15 xenograft tumor growth through antiangiogenic effect To investigate the effect of RA on tumor growth of human colorectal cancer, nude mice bearing subcutaneous HCT-15 xenograft were
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
107
Fig. 3. RA suppressed the trunk angiogenesis in zebrafish. (A–J) Representative bright field and fluorescent images of zebrafish embryos at 49 h post-fertilization (hpf) treated with 0.1% dimethyl sulfoxide (DMSO) (control), 0.4 μM RA or 5 μM PTK787 (positive control) for 30 h. (D–J) Compared with control, embryos treatment with RA presented an increased number of incomplete ISVs and only occasional sprouts (asterisk) of dorsal aorta could be observed. The boxed regions are shown at higher magnification in the right panels. Black arrows in B and C indicate the pericardial edema after RA or PTK787 treatment. DLAV, dorsal longitudinal anastomotic vessels; ISV, intersegmental vessels; DA, dorsal aorta. (K) Statistical analysis of number of ISVs. (L) Inhibition rate of ISVs after RA and PTK787 treatment. ∗∗∗ p < 0.001 as compared with control.
treated with 5 mg/kg RA every other day for total 11 i.p. injections. It showed that RA significantly inhibited HCT-15 tumor growth and decreased the tumor size. The tumor volume in RA-treated group was 765.3 ± 156.6 mm3 , much smaller than that of control group (1919.0 ± 670.6 mm3 ) (Fig. 4a, c). The average tumor weight of control group was 2.2 ± 0.5 g, whereas that of RA-treated group was only 1.2 ± 0.3 g (Fig. 4b). The mice weight loss after RA treatment was not significant, suggesting that RA did not cause obvious toxicity to the mice at curative dose (Fig. 4d). Immunohistochemical and pathological examinations showed that RA treatment significantly decreased intratumoral MVD, elevated TUNEL-positive cells, and led to increased tumor necrotic area compared with control group (Fig. 4e–g). These results indicated that the antitumor effect of RA was related to its antiangiogenic activity. RA down-regulated the activation of VEGFR2 and its downstream proteins To investigate the underlying molecular mechanism of RAmediated antiangiogenesis, we performed western blot assay to elucidate whether RA could inhibit VEGFR2 phosphorylation, and prohibit the activation of its downstream signaling pathway, which closely related to the survival, migration and proliferation of EC. RA strongly inhibited VEGF-induced VEGFR2 phosphorylation in a dose-dependent manner (Fig. 5a). Furthermore, the activation of downstream signaling of VEGFR2, including PLCγ 1, JAK2, FAK, Src, and Akt, was also decreased when treated with different concentration of RA (Fig. 5a). RA docked into the ATP binding pocket of VEGFR2 kinase domain As RA down-regulated the phosphorylation of VEGFR2 and its downstream signaling molecules, it was hypothesized that RA may interact with VEGFR2 kinase domain. Then, molecular docking simulation was performed to investigate the possible binding pattern between RA and VEGFR2 kinase domain. It was shown that RA was well docked into the ATP binding pocket of VEGFR2 kinase domain (Fig. 5b). The pentacyclic triterpene moiety of RA was trapped in the
hydrophobic pocket, which was composed of Leu840, Leu1035, and Phe1047 (Fig. 5b, c). There were also hydrogen bonds between the saccharide moiety of RA and the key residues of ATP binding pocket, including Lys868, Glu885, and Asp1046 (Fig. 5b, c). In addition, RA also moderately interacted with other amino acid residues, including Gly922, Val848, Val916, Val899, Ala881, Leu1049 and Arg1027 through the hydrophobic interaction and hydrogen bonds (Fig. 5c).
Discussion Antiangiogenic intervention mediated by natural products is a promising research area that provides an effective anticancer strategy. Many natural compounds have been proved to possess antiangiogenic activity in vitro and in vivo. Previous studies showed that RA exerted antitumor activity on several kinds of cancer in vitro and in vivo by inducing apoptosis and inhibiting invasion, migration and adhesion of tumor cells (Wang et al. 2008; Gao et al. 2010; Xue et al. 2013). However, there has been no report regarding the antiangiogenic activity of RA so far. In this study, we firstly found that RA possessed remarkable antiangiogenic effect both in vitro and in vivo. RA interrupted a series of orchestrated processes of angiogenesis, including EC proliferation, motility, migration and capillary-like tube formation in a dose-dependent manner (Fig. 1b–f), and inhibited in vivo chick CAM angiogenesis (Fig. 2). Transgenic zebrafish was also adopted to further evaluate the antiangiogenic activity of RA. The quantification of the number of complete intersegmental vessels (ISVs) is one of the reliable parameter to evaluate angiogenesis in vivo (Vasil et al. 2009; Murphy et al. 2010). RA could inhibit the ISVs formation of zebrafish in a non-lethal dose. As shown in Fig. 3, 0.4 μM RA effectively disrupted the completion of ISVs and only occasional sprouts of dorsal aorta were observed. Both PTK787 (positive control) and RA significantly blocked blood vessel formation, which may be responsible for the observed pericardial edema (Chan et al. 2002). It is also noted that the morphology of zebrafishes treated with RA deformed and curved to some extent. Prompted by the robust antiangiogenic activity of RA, we applied
108
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Fig. 4. RA (5 mg/kg, i.p. once every 2 days) inhibited tumor growth through antiangiogenic activity in a subcutaneous HCT-15 tumor model of mice. (a) Representative photographs of the tumor bearing mice and the resected tumor in the control and RA-treated group. (b) The tumor weight at the end of the study. (c) Mice tumor volume. (d) Mice body weight. Representative staining sections and quantitative analysis of CD31 for microvessel density (MVD) (e), TUNEL for apoptotic cells (f), and H&E for necrosis area (g) of tumor treated with control and RA. Bar, 100 μm in (e) and (f), and bar, 1 mm in (g). RA could significantly reduce the tumor volume and tumor weight in HCT-15 tumor bearing mice without obvious toxicity (no distinct weight loss observed). The MVD was significant decreased and the percentage of TUNEL positive cells and necrosis area were pronouncedly increased after RA treatment. Values are expressed as mean ± SD, n = 6. ∗∗ p < 0.01, ∗∗∗ p < 0.001 as compared with control.
HCT-15 human colorectal tumor xenograft in mice to investigate whether RA could exert antitumor effect through its high antiangiogenic potency. RA significantly suppressed tumor growth and tumor weight, induced tumor cell apoptosis, and increased tumor necrosis without obvious adverse effects such as body weight loss (Fig. 4a, b, c, d, f, g). The antitumor efficacy of RA was found to be closely related to the antiangiogenic activity, which was shown in remarkably decreased MVD in tumor treated by RA (Fig. 4e). The phosphorylation of VEGFR2 is critical for angiogenesis, which mediates several effects of EC function, including EC survival, proliferation, and migration. Western blot assay showed that RA could decrease the VEGF induced VEGFR2 phosphorylation and its downstream signals, including PLCγ 1, JAK2, FAK, Src, and Akt, in a dosedependent manner (Fig. 5a). RA of 3 μM almost totally inhibited the phosphorylation of VEGFR2, while the total protein of VEGFR2 remained unaffected, suggesting that RA can be a potent VEGFR2 inhibitor. The down-regulation of PLCγ 1 is related to the inhibition of EC proliferation (Sase et al. 2009). The inactivation of FAK and Src can suppress EC migration. JAK2 can mediate the phosphorylation of STAT3, a potent transcription activator in tumor angiogenesis (Zhang et al. 2011). Akt plays a key role in several cellular
functions including cell survival, proliferation, migration, and protein synthesis. The suppression of Akt and JAK2 phosphorylation can affect EC function in various aspects. Taken together, these results indicated that RA modulated VEGF-mediated angiogenesis by blocking the phosphorylation of VEGFR2 and its multiple downstream protein kinases as summarized in Fig. 5d. It is noted that some other triterpenoid natural products and their derivatives also have antiangiogenic activity by influencing various signaling pathway, including VEGF/VEGFR2 pathway (Pang et al. 2010b; Mu et al. 2012; Kim and Kim 2014), bFGF/FGFR1 pathway (Kim and Kim 2014), mTOR/S6K pathway (Pang et al. 2010b), etc. Moreover, many other saponins, such as deltonin (Tong et al. 2011), terrestrosin D (Wei et al. 2014) and ASC (Zeng et al. 2014), also possessed antitumor and antiangiogenic activity. The phosphorylation of VEGFR2 is an ATP consuming process. Molecular docking test indicated that RA could stably locate at the ATP-binding pocket of VEGFR2 kinase domain (Fig. 5b). Six amino acids (Leu840, Leu1035, Phe1047, Lys868, Glu885 and Asp1046) at the ATP binding pocket were essential for the stable conformation of VEGFR2-RA complex (Fig. 5b, c). Some other amino acid residues of ATP binding pocket also have moderate interaction with RA (Fig. 5c). Such bioinformatics of the binding pattern of RA and VEGFR2 can help
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
109
Fig. 5. RA inhibited the kinase activity of VEGFR2 and its downstream signaling molecules. (a) RA suppressed the activation of VEGFR2 and its downstream signaling kinases induced by VEGF in HUVEC. HUVEC were pretreated with or without various concentrations of RA for 30 min followed by the stimulation with 100 ng/ml of VEGF for 4 min. The activation of VEGFR2 and its downstream cascade including PLCγ 1, JAK2, FAK, Src, and Akt were analyzed by western blot and probed with specific antibodies. (b) The predicted binding interactions between RA and VEGFR2. RA was well docked into the ATP binding pocket of VEGFR2 kinase domain. Side chains of crucial residues in the binding site are labeled and shown as purple sticks. (c) Two-dimensional interaction map of RA and ATP binding pocket of VEGFR2 kinase domain was analyzed and visualized by Ligand Interaction module. (d) Signaling pathway diagram of RA-mediated antiangiogenesis.
us better understand the antiangiogenic effect of RA. It appears that the interactions between the sapogenin (but not the saccharide chains of RA) to VEGFR2 contribute more to the stable conformation VEGFR2RA complex. Consistent to our speculation, some non-saponin triterpenoid compounds such pristimerin (Mu et al. 2012) and celastrol (Pang et al. 2010a) also possess potent antiangiogenic activity by inhibiting VEGFR2 activation. However, the saccharide moiety of RA may help improve the hydrophility of the compound and facilitate in vivo distribution. In summary, our findings demonstrated for the first time that RA possessed potent antiangiogenic activity both in vitro and in vivo as a VEGFR2 inhibitor, and it may be a potential drug candidate or lead compound for antiangiogenic cancer therapy. Acknowledgments This work was supported by National Basic Research Program of China (No. 2010CB529806), National Natural Science Foundation of China (No. 81272569), Shanghai Pujiang Program (No. 12PJD023), Innovation Program of Shanghai Municipal Education Commission (Nos. 12ZZ200, 13ZZ087), and Shanghai Municipal Science and Technology Commission (No. 14JC1491900).
References Bayliss, P.E., Bellavance, K.L., Whitehead, G.G., Abrams, J.M., Aegerter, S., Robbins, H.S., Cowan, D.B., Keating, M.T., O’Reilly, T., Wood, J.M., Roberts, T.M., Chan, J., 2006. Chemical modulation of receptor signaling inhibits regenerative angiogenesis in adult zebrafish. Nat. Chem. Biol. 2, 265–273. Chan, J., Bayliss, P.E., Wood, J.M., Roberts, T.M., 2002. Dissection of angiogenic signaling in zebrafish using a chemical genetic approach. Cancer Cell 1, 257–267. Gao, Y., An, J., Zhu, Q., 2010. Inhibitory effect of Raddeanin A in human non-small cell lung cancer H460 cells. Chin. J. Appl. Environ. Biol. 16, 637–641. Grothey, A., Galanis, E., 2009. Targeting angiogenesis: progress with anti-VEGF treatment with large molecules. Nat. Rev. Clin. Oncol. 6, 507–518. Kim, H.J., Kim, J.K., 2014. Antiangiogenic effects of cucurbitacin-I. Arch. Pharm. Res. doi:10.1007/s12272-014-0386-5. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310. Li, K.K., Liu, C.L., Tam, J.C., Kwok, H.F., Lau, C.P., Leung, P.C., Ko, C.H., Ye, C.X., 2014. In vitro and in vivo mechanistic study of a novel proanthocyanidin, GC-(4→8)GCG from cocoa tea (Camellia ptilophylla) in antiangiogenesis. J. Nutr. Biochem. 25, 319–328. Luan, X., Guan, Y.Y., Wang, C., Zhao, M., Lu, Q., Tang, Y.B., Liu, Y.R., Yu, D.H., Wang, X.L., Qi, H., Fang, C., Chen, H.Z., 2013. Determination of Raddeanin A in rat plasma by liquid chromatography–tandem mass spectrometry: application to a pharmacokinetic study. J. Chromatogr. B 923-924, 43–47. Meadows, K.L., Hurwitz, H.I., 2012. Anti-VEGF Therapies in the Clinic. Cold Spring Harb. Perspect. Med. 2, a006577. Mu, X., Shi, W., Sun, L., Li, H., Jiang, Z., Zhang, L., 2012. Pristimerin, a triterpenoid, inhibits tumor angiogenesis by targeting VEGFR2 activation. Molecules 17, 6854–6868. Mullard, A., 2013. 2012 FDA drug approvals. Nat. Rev. Drug Discov. 12, 87–90.
110
Y.-Y. Guan et al. / Phytomedicine 22 (2014) 103–110
Murphy, E.A., Shields, D.J., Stoletov, K., Dneprovskaia, E., McElroy, M., Greenberg, J.I., Lindquist, J., Acevedo, L.M., Anand, S., Majeti, B.K., Tsigelny, I., Saldanha, A., Walsh, B., Hoffman, R.M., Bouvet, M., Klemke, R.L., Vogt, P.K., Arnold, L., Wrasidlo, W., Cheresh, D.A., 2010. Disruption of angiogenesis and tumor growth with an orally active drug that stabilizes the inactive state of PDGFRβ /B-RAF. Proc. Natl. Acad. Sci., 107, pp. 4299–4304. Pang, X., Yi, Z., Zhang, J., Lu, B., Sung, B., Qu, W., Aggarwal, B.B., Liu, M., 2010. Celastrol suppresses angiogenesis-mediated tumor growth through inhibition of AKT/mammalian target of rapamycin pathway. Cancer Res. 70, 1951–1959. Pang, X., Zhang, L., Wu, Y., Lin, L., Li, J., Qu, W., Safe, S., Liu, M., 2010. Methyl 2-cyano-3,11-dioxo-18-olean-1,12-dien-30-oate (CDODA-Me), a derivative of glycyrrhetinic acid, functions as a potent angiogenesis inhibitor. J. Pharmacol. Exp. Ther. 335, 172–179. Sase, H., Watabe, T., Kawasaki, K., Miyazono, K., Miyazawa, K., 2009. VEGFR2-PLCγ 1 axis is essential for endothelial specification of VEGFR2+ vascular progenitor cells. J. Cell Sci. 122, 3303–3311. Song, Y., Dai, F., Zhai, D., Dong, Y., Zhang, J., Lu, B., Luo, J., Liu, M., Yi, Z., 2012. Usnic acid inhibits breast tumor angiogenesis and growth by suppressing VEGFR2-mediated AKT and ERK1/2 signaling pathways. Angiogenesis 15, 421–432. Tong, Q.Y., Qing, Y., Shu, D., He, Y., Zhao, Y.L., Li, Y., Wang, Z.L., Zhang, S.Y., Xing, Z.H., Xu, C., Wei, Y.Q., Huang, W., Wu, X.H., 2011. Deltonin, a steroidal saponin, inhibits colon cancer cell growth in vitro and tumor growth in vivo via induction of apoptosis and antiangiogenesis. Cell. Physiol. Biochem. 27, 233–242. Vasil, M.L., Stonehouse, M.J., Vasil, A.I., Wadsworth, S.J., Goldfine, H., Bolcome 3rd, R.E., Chan, J., 2009. A complex extracellular sphingomyelinase of Pseudomonas aeruginosa inhibits angiogenesis by selective cytotoxicity to endothelial cells. PLOS Pathogens 5, e1000420.
Vitorino, P., Meyer, T., 2008. Modular control of endothelial sheet migration. Genes Dev. 22, 3268–3281. Wang, M.K., Ding, L.S., Wu, F.E., 2008. Antitumor effects of raddeanin A on S180, H22 and U14 cell xenografts in mice. Chin. J. Cancer 27, 910–913. Wang, F.L., Sun, J.Y., Wang, Y., Mu, Y.L., Liang, Y.J., Chong, Z.Z., Qin, S.H., Yao, Q.Q., 2013. Oldhamianoside II, a new triterpenoid saponin, prevents tumor growth via inducing cell apoptosis and inhibiting angiogenesis. Oncol. Res. 20, 369–376. Wei, S., Fukuhara, H., Chen, G., Kawada, C., Kurabayashi, A., Furihata, M., Inoue, K., Shuin, T., 2014. Terrestrosin D, a steroidal saponin from Tribulus terrestris L., inhibits growth and angiogenesis of human prostate cancer in vitro and in vivo. Pathobiology 81, 123–132. Westerfield, M., 1995. The Zebrafish Book: A Guide for the Laboratory Use of Zebrafish, 3rd ed. University of Oregon Press, Eugene, USA. Xu, H.Y., Pan, Y.M., Chen, Z.W., Lin, Y., Wang, L.H., Chen, Y.H., Jie, T.T., Lu, Y.Y., Liu, J.C., 2013. 12-Deoxyphorbol 13-palmitate inhibit VEGF-induced angiogenesis via suppression of VEGFR-2-signaling pathway. J. Ethnopharmacol. 146, 724–733. Xue, G., Zou, X., Zhou, J.Y., Sun, W., Wu, J., Xu, J.L., Wang, R.P., 2013. Raddeanin A induces human gastric cancer cells apoptosis and inhibits their invasion in vitro. Biochem. Biophys. Res. Commun. 439, 196–202. Zeng, K., Song, F., Li, N., Dong, X., Jiang, Y., Tu, P., 2014. ASC, a bioactive steroidal saponin from Ophitopogin japonicas, inhibits angiogenesis through interruption of Src tyrosine kinase-dependent matrix metalloproteinase pathway. Basic Clin. Pharmacol. Toxicol. doi:10.1111/bcpt.12305. Zhang, X., Song, Y., Wu, Y., Dong, Y., Lai, L., Zhang, J., Lu, B., Dai, F., He, L., Liu, M., Yi, Z., 2011. Indirubin inhibits tumor growth by antitumor angiogenesis via blocking VEGFR2-mediated JAK/STAT3 signaling in endothelial cell. Int. J. Cancer 129, 2502–2511.