N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

Ava i l ab l e on l i ne a t www.sc i enced i r ec t . com

www.n r j ou rna l . com

Grape seed proanthocyanidins inhibit angiogenesis via thedownregulation of both vascular endothelial growth factor andangiopoietin signaling

Shuangsheng Huanga,⁎, 1, Ninggang Yangb, 1, Yuanyuan Liu c, Lamei Hud, Jin Zhaoa,Jing Gaoa, Yongquan Lia, Caili Li a, Xiaosu Zhanga, Tao Huanga

a Medical College of Northwest University for Nationalities, Lanzhou 730030, Chinab Department of Urology, Lanzhou First People's Hospital, Lanzhou, 730030, Chinac Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou 730000, Chinad Institute of Pharmacology, School of Basic Medical Science, Lanzhou University, Lanzhou 730000, China

A R T I C L E I N F O

Abbreviations: Ang, angiopoietin; CAM, choGSPs, grape seed proanthocyanidins; HMEC-matrix metalloproteinase; SRB, sulforhodamdomains 2; VEGF, vascular endothelial growt⁎ Corresponding author. Medical College of N

E-mail address: huangshsh02@qq.com (S.1 The first two authors contributed equally

0271-5317/$ – see front matter © 2012 Elsevidoi:10.1016/j.nutres.2012.05.012

A B S T R A C T

Article history:Received 2 February 2012Revised 27 May 2012Accepted 29 May 2012

Vascular endothelial growth factor (VEGF)/VEGF receptor 2 and angiopoietin 1/tyrosinekinase with immunoglobulin and epidermal growth factor homology domains 2 signalingpathways regulate different, but complementary, aspects of blood vessel growth intumors. Simultaneous inhibition of both pathways not only exhibits additiveantiangiogenic effects but also overcomes the resistance to anti-VEGF therapy. Grapeseed proanthocyanidins (GSPs) are widely consumed dietary supplements withantiangiogenic activity. However, the molecular mechanisms underlying theirantiangiogenic action have not been fully understood. We hypothesized that GSPsmodulate multiple signaling pathways to exhibit antiangiogenic effects. In the presentstudy, we aimed to test this hypothesis by examining the effects of GSPs on humanmicrovascular endothelial cell–1 and chick chorioallantoic membrane. Our results showedthat GSPs inhibited the migration, matrix metalloproteinase–2 and –9 secretion, and tubeformation of human microvascular endothelial cell–1 in vitro in a dose-dependentmanner. In addition, chick chorioallantoic membrane angiogenesis assay showed thatGSPs inhibited neovascularization in a dose-dependent manner. Furthermore, wedemonstrated that GSPs inhibited the phosphorylation of VEGF receptor 2 and tyrosinekinase with immunoglobulin and epidermal growth factor homology domains 2 as well asdownstream signaling component extracellular signal–regulated kinase 1/2. In summary,these data suggest that GSPs inhibit both VEGF and angiopoietin 1 signaling to executethe antiangiogenic effects and indicate that GSPs could be developed as a pharmacologicallysafe chemopreventive agent against cancer.

© 2012 Elsevier Inc. All rights reserved.

Keywords:Grape seed proanthocyanidinsAngiogenesisVEGFR2Tie2HMEC-1 endothelial cellsChick chorioallantoic membrane

rioallantoic membrane; ERK, extracellular signal–regulated kinase; FBS, fetal bovine serum;1, human microvascular endothelial cell-1; MAPK, mitogen-activated protein kinase; MMP,ine B; Tie2, tyrosine kinase with immunoglobulin and epidermal growth factor homologyh factor; VEGFR2, vascular endothelial growth factor receptor 2.orthwest University for Nationalities, Lanzhou, 730030, China. Tel.: +86 931 2928013.Huang).to the work.

er Inc. All rights reserved.

531N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

1. Introduction

Angiogenesis is involved in many physiological and patho-logical conditions, such as wound healing, embryonic devel-opment, atherosclerosis, arthritis, and tumor. During tumordevelopment, newly generated vessels not only provideoxygen and nutrients for tumor growth but also promotetumor metastasis [1]. The process of angiogenesis is regulatedinmalignant tissues by the balance of pro- and antiangiogenicfactors. Vascular endothelial growth factor (VEGF) is the mostimportant proangiogenic factor [2]. Vascular endothelialgrowth factor signaling involved in angiogenesis is mainlymediated by VEGF receptor 2 (VEGFR2) expressed on thesurface of endothelial cells. The binding of VEGF to VEGFR2facilitates autophosphorylation of VEGFR2 and then activatesdiverse intracellular signaling molecules, including phosphoi-nositide 3-kinase/AKT kinase, phospholipase Cγ, proteinkinase C, and mitogen-activated protein kinase (MAPK)/extracellular signal–related kinase (ERK), which ultimatelyleads to endothelial cell activation, proliferation, migration,and survival [3,4]. Many antiangiogenic agents targeting VEGFor VEGFR2 have been approved for the treatment ofmetastaticcolorectal cancer, renal cell cancer, and hepatocellular carci-noma [5]. However, tumor could circumvent the anti-VEGFtherapy via the activation and/or upregulation of otherproangiogenic signaling pathways, including fibroblastgrowth factor, ephrin, and angiopoietin (Ang) families thatdrive angiogenesis and tumor progression [6]. In addition,interference with VEGF-mediated signaling events is onlyeffective in preventing the early growth of neovessels. Maturevessels covered with pericytes from more established tumorsare largely resistant to these inhibitors [7].

Angiopoietin 1 (Ang1) is another important proangiogenicfactor. Angiopoietin 1 exerts its biological effects by binding totyrosine kinasewith Ig and epidermal growth factor homologydomains 2 (Tie2) on endothelial cells [8]. It is generallyaccepted that VEGF/VEGFR2 signaling is essential for drawingendothelial cells from preexisting blood vessels and stimulat-ing their growth, whereas Ang1/Tie2 signaling is important forsustaining the interaction between endothelial and muralcells and stabilizing the vasculature [9]. Accumulating evi-dence has suggested that blocking both VEGFR2 and Tie2signaling pathways could overcome the resistance to anti-VEGF therapy [10–12].

Grapes are one of the most widely consumed fruits in theworld and are rich in polyphenols of which about 60% to 70% isfound in grape seeds as dimers, trimers, and other oligomersof flavan-3-ols and known commonly as proanthocyanidins.Grape seed proanthocyanidins (GSPs) have demonstratedchemopreventive and/or chemotherapeutic effects in variouscancer cell cultures and animal models [13,14]. Recent studiesfurther showed that GSPs exhibited antiangiogenic effects inhuman prostate and breast cancer via the inhibition of VEGFsignaling [15,16]. However, GSPs could regulate multiplesignaling pathways such as nuclear factor–κB, MAPK, andphosphoinositide 3-kinase/Akt in tumor cells [17]. Therefore,we hypothesized that GSPs alter multiple signaling pathwaysthat contribute to its antiangiogenic actions. Thus, in thepresent investigation, we tested our hypothesis by examining

the effects of GSPs on human microvascular endothelial cells(HMEC-1) and chick chorioallantoic membrane (CAM) asmodels for understanding potential signaling mechanismsinfluenced by these compounds.

2. Methods and materials

2.1. Materials and reagents

Grape seed proanthocyanidins, consisting of at least 95%proanthocyanidins, 1.8% proanthocyanidins B2, and 60%oligomers, were purchased from Jianfeng Company (Tianjin,China). Recombinant human VEGF and Ang1 were purchasedfrom Peprotech (Rocky Hill, New Jersey). Growth factor–reduced Matrigel was purchased from Becton Dickinson(Bedford, Massachusetts). MCDB131 medium, epithelialgrowth factor, hydrocortisone, sulforhodamine B (SRB), andgelatin from porcine skin were purchased from Sigma (StLouis, Missouri). Fetal bovine serum (FBS) was purchased fromLanzhou HyClone (Lanzhou, China). Millicell cell cultureinserts were purchased from Millipore (Bedford, Massachu-setts). Primary antibodies to VEGFR2, Tie2, phospho-ERK1/2,and actin were purchased from Santa Cruz Biotechnology(Santa Cruz, California). Primary antibodies to phospho-VEGFR2 (Tyr1175), phospho-Tie2 (Tyr992), and ERK1/2 werepurchased from Cell Signaling (Beverly, Massachusetts).

2.2. Cell culture

Human microvascular endothelial cells were cultured inMCDB131 medium supplemented with 1.18 mg/mL NaHCO3,20% inactivated FBS, 10 ng/mL epithelial growth factor, and1 μg/mL hydrocortisone in a humidified atmosphere of 5% CO2

at 37°C.

2.3. Cell viability assay

To examine the effect of GSPs on the viability of HMEC-1 cells,we performed cell viability assay using SRB method asdescribed previously [18]. Briefly, exponentially growing cellswere seeded in 96-well plates with the final volume 100 μL perwell. After 24 hours, cells were treated with various concen-trations of GSPs for indicated time. The cultures were thenfixed at 4°C for 1 hour with ice-cold 50% trichloroacetic acid togive a final concentration of 10%. Fixed cells were rinsed 5times with deionized water and stained for 10 minutes with0.4% SRB dissolved in 0.1% acetic acid. The wells were thenwashed 5 times with 0.1% acetic acid and left to dry overnight.The absorbed SRB was dissolved in 150 μL 1% Tris base (pH10.5). The absorbance of extracted SRB was measured on amicroplate reader at 515 nm.

2.4. Cell migration assay

To examine the effect of GSPs on the migration ability ofHMEC-1 cells, we performed cell migration assay as describedpreviously [19]. Briefly, 2 × 105 cells were suspended in 0.4 mLserum-free MCDB131 medium supplemented with differentconcentrations of GSPs or vehicle and added to upper

0

20

40

60

80

100

120

Cel

l via

bilit

y (%

of

Con

trol

)

24 h 72 h

***

**

***

Control 25 50 100 200GSP (ug/ml)

Fig. 1 – Effect of GSPs on the cell viability of HMEC-1 cells.After treatment with GSPs for indicated time, the viabilitywas measured with SRB assay. Results were representativeof 3 independent experiments and are expressed asmeans ±SE of 6 cultures with the vehicle control as 100%. **P < .01,***P < .001 vs vehicle control.

532 N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

compartment of cell culture inserts. The lower compartmentcontained 0.6 mL MCDB131 medium supplemented with 20%FBS. After incubation for 24 hours at 37°C, the nonmigratedcells on the upper face of the membrane were removed with acotton swab. The migrated cells on the bottom surface of themembrane were fixed with methanol and stained with 0.1%crystal violet. Images from 5 randomly selected microscopicfields were obtained under light microscopy. The number ofmigrated cells was counted with Image-plus software. Eachsample was repeated 3 times.

2.5. Gelatin zymography

To examine the effect of GSPs on the enzymatic activities ofmatrix metalloproteinase (MMP)–2 and MMP-9 in HMEC-1cells, we performed gelatin zymography as described previ-ously [19]. Briefly, subconfluent HMEC-1 cells were incubatedfor 24 hours in the absence or presence of different concen-trations of GSPs in serum-free MCDB131 medium. Theconditioned medium was collected. After incubation withsample buffer for 0.5 hours at 37°C, the sample was separatedby electrophoresis on a 7.5% sodium dodecyl sulfate poly-acrylamide gel containing 1% gelatin under nonreducedconditions. The gel was washed twice with washing buffer(50 mmol/L Tris-HCl [pH 7.5], 100 mmol/L NaCl, and 2.5%Triton X-100) for 1 hour and then incubated in incubationbuffer (50 mmol/L Tris-HCl [pH 7.5], 150 mmol/L NaCl, and 10mmol/L CaCl2) at 37°C for 36 hours. Next, the gel was stainedwith 0.25% Coomassie Brilliant Blue R250 and then destainedwith 25% methanol and 7.5% acetic acid. Gelatinase activitywas detected as clear bands against a dark blue background.

2.6. Tube formation assay

To examine the effect of GSPs on the tube formation of HMEC-1 cells, we performed tube formation assay as describedpreviously [20]. Briefly, Matrigel was distributed in 96-wellplates with 60 μL per well and allowed to solidify at 37°C for atleast 1 hour. TheHMEC-1 cells were seeded at a density of 7.5 ×104 per well in 100-μL culture medium containing differentconcentrations of GSPs or vehicle control. Plates were thenincubated at 37°C and 5% CO2 for 8 hours until forming anintact network in the control group. Images were capturedunder inverted microscope (Olympus, IX41, Tokyo, Japan).

2.7. Chick CAM angiogenesis assay

To examine the in vivo antiangiogenic effect of GSPs, weperformed chick CAM assay as described previously [21].Briefly, fertilized white Leghorn chicken eggs were incubatedunder conditions of constant humidity at 37°C. On day 8 ofincubation, a small hole was punched over the air sac todetach the CAM from the eggshell; and then a square windowwas opened on the broad side of the egg to expose CAM. Sterilecellulose disks adsorbed with 40-μL solution containingvarious concentrations of GSPs or vehicle were placed ontoan avascular area of CAM. The eggswere returned to incubatorand incubated for 3 days. During this period, GSPs were addedevery day. Three days later, the neovascular zones werephotographed and calculated.

2.8. Western blot analysis

To determine the effects of GSPs on VEGF- and Ang1-dependent signaling cascades, HMEC-1 cells cultured inserum-free MCDB131 medium were pretreated with or with-out various concentrations of GSPs for 4 hours and thenstimulated with 50 ng/mL VEGF for 10 minutes or 200 ng/mLAng1 for 30minutes. The whole-cell extracts were prepared inlysis buffer supplemented with proteinase inhibitors (50mmol/L NaF, 0.2 mmol/L Na3VO4, 1 mmol/L phenylmethyl-sulfonyl fluoride, 2 μg/mL aprotinin, and 2 μg/mL leupeptin).Equal amounts of protein from each sample were resolved by6% to 10% sodium dodecyl sulfate polyacrylamide gels,transferred onto the polyvinylidene fluoride membrane, andprobed with specific antibodies, followed by exposure to ahorseradish peroxidase–conjugated secondary antibody. Im-munoreactivity was visualized by exposure to x-ray film usingenhanced chemiluminescence detection.

2.9. Statistical analyses

Results were expressed as means ± standard error (SE).Statistical analyses were performed by Student t test using theSPSS version 12 statistical analysis package (SPSS Inc, Chicago,Illinois). A P value less than .05 was considered significant.

3. Results and discussion

3.1. Effect of GSPs on the cell viability of HMEC-1 cells

Grape seed proanthocyanidin treatment for 72 hours signif-icantly inhibited cell viability of HMEC-1 cells in a dose-dependent manner, with an IC50 value of 99.32 μg/mL.However, treatment with GSPs (25-200 μg/mL) for 24 hoursdid not change cell viability of HMEC-1 cells (Fig. 1). Therefore,

0

20

40

60

80

100

120

Mig

rate

d ce

lls (

% o

f C

ontr

ol)

***

***

***

Control 25 50 100 200

A

Control GSP (200 ug/ml)

Control 25 50 100 200

MMP-9

MMP-2

B

C

GSP (ug/ml)

GSP (ug/ml)

533N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

all the following experiments were performed with GSPtreatment not exceeding 24 hours.

3.2. GSPs inhibit cell migration of HMEC-1 cells

Cell migration is necessary for endothelial cells to formblood vessels during angiogenesis. After stimulating HMEC-1 cells with 20% serum for 24 hours, a large number ofcells migrated to the lower side of the filter. Grape seedproanthocyanidins significantly inhibited serum-inducedmigration of HMEC-1 cells in a dose-dependent manner.The inhibition of HMEC-1 cell migration by GSPs at 25, 50,100, and 200 μg/mL was 12.7%, 26.3%, 51.5%, and 87.9%,respectively (Fig. 2A).

3.3. GSPs inhibit the secretion of MMP-2 and MMP-9 byHMEC-1 cells

The secretion of MMPs is crucial for extracellular matrixdegradation, which is required for angiogenesis. After HMEC-1cells were treated with various concentrations of GSPs for 24hours in serum-free medium, the conditioned medium wascollected and assayed for MMP activity. Gelatin zymographyshowed that MMP-2 and MMP-9 activities in the conditionedmediumwere markedly reduced by GSPs in a dose-dependentmanner (Fig. 2B).

3.4. GSPs suppress the tube formation of HMEC-1 cells

Maturation of migrated endothelial cells into a tubelikestructure is a critical step for the formation of functionalvessels. As shown in Fig. 2C, HMEC-1 cells seeded on thegrowth factor–reduced Matrigel formed elongated androbust tubelike structures within 8 hours. However, GSPssignificantly suppressed tube formation of HMEC-1 cells ina dose-dependent manner. In particular, when treatedwith 200 μg/mL GSPs, most of the cells appeared asunorganized cell aggregates; and the tube formation wascompletely disrupted.

Control GSP (25 ug/ml)

GSP (50 ug/ml) GSP (100 ug/ml)

GSP (200 ug/ml)

Fig. 2 – Grape seed proanthocyanidins inhibit the cellmigration, MMP-2 and -9 secretion, and tube formation ofHMEC-1 cells. A, Themigration of HMEC-1 cells was assessedby cell migration assay as described in “Methods andmaterials.” Representative photomicrographs were shown.Values were representative of 3 independent experimentsand are expressed as means ± SE of 5 fields with the vehiclecontrol as 100%. *** P < .001 vs vehicle control. B, Thesecretion of MMP-2 and -9 by HMEC-1 cells was detected bygelatin zymography. The clear zone of gelatin digestionindicated the presence of MMP-2 and -9. The photographswere the representative of 3 independent experiments. C,The HMEC-1 cells were seeded on the Matrigel-coated96-well plates at the density of 7.5 × 104 per well in thepresence of different concentrations of GSPs or vehiclecontrol. After 8 hours, photographs of endothelial tubeformation were taken. Results were representative of 3independent experiments.

534 N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

3.5. GSPs inhibit angiogenesis in vivo

To further confirm the antiangiogenic effect of GSPs, weexamined the antiangiogenic activity of GSPs in vivo. Theresults showed that CAM in the control group showed well-developed zones of neovascularization around the sterilecellulose disks after 3 days of incubation. In contrast, CAMneovascularization, especially the small vessels, was signifi-cantly suppressed by GSPs in a dose-dependent manner(Fig. 3). However, GSP treatment did not affect the preexistingvessels. These results indicate that GSPs dramatically inhib-ited angiogenesis in vivo and that the antiangiogenic effectwas not due to their cytotoxic effect.

3.6. GSPs inhibit VEGF-induced phosphorylation ofVEGFR2 and ERK1/2 in HMEC-1 cells

Vascular endothelial growth factor/VEGFR2 signaling is a keypathway that regulates endothelial cell proliferation, migra-tion, and differentiation [22]. As shown in Fig. 4A and C,stimulation of HMEC-1 cells with exogenous VEGF (50 ng/mL)for 10 minutes significantly enhanced the phosphorylation ofVEGFR2. Pretreatment with GSPs significantly blocked VEGF-induced phosphorylation of VEGFR2 without affecting totalVEGFR2 expression level. To identify the downstream signal-ing pathway targeted by GSP treatment, we examined theexpression and phosphorylation of ERK1/2, one of the key

Control GSP (5 u

BA

0

20

40

60

80

100

120

Vas

cula

r de

nsit

y (%

of

Con

trol

)D

Control

Fig. 3 – Grape seed proanthocyanidins inhibit in vivo angiogenesand then cellulose disks saturated with different concentrations opatterns of angiogenesis on the CAMwere photographed. VehicleVascular densities around cellulose diskwere quantified. Valuesvehicle control as 100%. ***P < .001 vs vehicle control.

signaling pathway components supporting endothelial cellsurvival, migration, and differentiation. We found that GSPsinhibited the phosphorylation of ERK1/2 in HMEC-1 cells, butdid not affect total ERK1/2 level. Our results are consistentwith previous reports that grape seed extracts inhibitedangiogenesis via the suppression of VEGFR/MAPK signalingpathway [16].

3.7. GSPs inhibit Ang1-induced phosphorylation of Tie2and ERK1/2 in HMEC-1 cells

Activation of Tie2 by Ang1 has been linked to the promotion ofendothelial cell survival, and the induction of migration andsprouting [23]. Therefore, we examined the activation of Tie2in HMEC-1 cells. Following the stimulation with Ang1 (200 ng/mL) for 30 minutes, Tie2 was strongly tyrosine phosphorylat-ed in HMEC-1 cells. However, preincubation of HMEC-1 cellswith various concentrations of GSPs led to dramatic suppres-sion of Ang1-stimulated Tie2 tyrosine phosphorylation.Furthermore, we found that Ang1 markedly increased thephosphorylation of ERK-1/2 in HMEC-1 cells, but that pre-treatment with GSPs significantly inhibited Ang1-stimulatedERK-1/2 activation (Fig. 4B and D).

Recent studies have shown that proanthocyanidins ex-hibit a range of beneficial effects such as reducing hepaticlipid accumulation [24], inhibiting autoimmunity [25], andproviding cardioprotection [26]. In the present study, we

g/egg) GSP (20 ug/egg)

C

***

***

5 20 GSP (ug/egg)

is in CAM. Fertilized chicken eggs were incubated for 7 days,f GSPs were placed on the CAMs every day. Three days later,control (A), 5 μg GSPs per egg (B), or 20 μg GSPs per egg (C). D,

were expressed asmeans ± SE of 4 to 5 separate eggswith the

VEGF

GSP (ug/ml) 0 0 50 100 200

A

P-VEGFR2

VEGFR2

P-ERK1/2

ERK1/2

Actin

Ang1

GSP (ug/ml) 0 0 50 100 200

B

P-Tie2

Tie2

P-ERK1/2

ERK1/2

Actin

0

0.2

0.4

0.6

0.8

1

1.2

p-V

EG

FR

2/V

EG

FR

2 ra

tio

GSP (ug/ml) 0 0 50 100 200 VEGF

**

*

C

0

0.2

0.4

0.6

0.8

1

1.2

p-T

ie/T

ie r

atio

GSP (ug/ml) 0 0 50 100 200 Ang1

***

***

*

*

D

Fig. 4 – Grape seed proanthocyanidins inhibit VEGF-induced VEGFR2 phosphorylation, Ang1-induced Tie 2 phosphorylation,and the activation of downstream effector ERK1/2. The HMEC-1 cells were pretreated with various concentrations of GSPs for 4hours and stimulated with 50 ng/mL VEGF for 10 minutes (A) or 200 ng/mL Ang1 for 30 minutes (B). Western blotting analysiswas performed using anti-p-VEGFR2, VEGFR2, anti-p-Tie2, Tie2, p-ERK, and ERK antibodies. Actin served as loading control.Densitometric analysis (values are means ± SE) revealed that GSPs reduced the ratios of phosphorylated VEGFR2/VEGFR2 (C)and phosphorylated Tie/Tie (D).The significance was assessed with a Student t test. *P < .05, **P < .01, and ***P < .001 comparedVEGF or Ang1 alone.

535N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

provide several lines of evidence to show that GSPs exhibitantiangiogenic effects both in vitro and in vivo and that thisis attributed to the inhibition of both VEGF and Ang1signaling. Vascular endothelial growth factor/VEGFR2 andAng1/Tie2 signaling pathways are known to regulate differ-ent but complementary aspects of blood vessel growth intumors. Vascular endothelial growth factor/VEGFR2 signalingplays important role in new blood vessels formation,whereas Ang1/Tie2 signaling contributes to new bloodvessels' maturation and stabilization. Simultaneous inhibi-tion of both pathways not only exhibits additive inhibitoryeffects on angiogenesis but also overcomes the resistance toanti-VEGF therapy. Our new findings provide support for ourhypothesis that GSPs inhibit multiple angiogenic pathwayssuch as VEGF/VEGFR2 and Ang1/Tie2 signaling to execute theantiangiogenic actions.

However, it is important to note that GSPs are a complexmixture of various polyphenol compounds including procya-

nidins B1 to B5, procyanidin C1, and procyanidin B5-3′-gallate.Further studies will be needed to determine whether theinhibition of VEGFR2 and Tie2 phosphorylation is due to onekind of active component alone or the combinative effects ofdifferent components. In addition, the effective concentrationof GSPs in this study is relatively higher than those in previousreports. This may be related to the fact that the methods forpreparing GSPs are different, resulting in the variation of theconcentration of active components in GSPs.

In summary, our results showed that GSPs inhibited themigration, MMP-2 and -9 secretion, and tube formation ofHMEC-1 cells in vitro in a dose-dependent manner. Inaddition, chick CAM assay showed that GSPs inhibitedneovascularization in a dose-dependent manner. Further-more, we demonstrated that GSPs inhibited the phosphory-lation of VEGFR2 and Tie2 as well as downstream signalingcomponent ERK1/2. The inhibition of both VEGF and Ang1signaling by GSPs may explain the strong antiangiogenic

536 N U T R I T I O N R E S E A R C H 3 2 ( 2 0 1 2 ) 5 3 0 – 5 3 6

effects of GSPs, which could inhibit both sprouting angiogen-esis and the maturation of blood vessels. These data indicatethat GSPs could be developed as a pharmacologically safechemopreventive agent against cancer.

Acknowledgment

This work was supported by the grant from National NaturalScience Foundation of China (no. 30700142 to ShuangshengHuang) and Scientific Research Innovation Teamof NorthwestUniversity for Nationalities (no. XBMUCXTD-2011-1).

The authors declare no conflicts of interest.

R E F E R E N C E S

[1] Bergers G, Benjamin LE. Tumorigenesis and the angiogenicswitch. Nat Rev Cancer 2003;3:401–10.

[2] Kerbel RS. Tumor angiogenesis. N Engl J Med 2008;358:2039–49.

[3] Holmes K, Roberts OL, Thomas AM, Cross MJ. Vascularendothelial growth factor receptor–2: structure, function,intracellular signaling and therapeutic inhibition. Cell Signal2007;19:2003–12.

[4] Olsson AK, Dimberg A, Kreuger J, Claesson-Welsh L. VEGFreceptor signaling-in control of vascular function. Nat RevMol Cell Biol 2006;7:359–71.

[5] Cook KM, FiggWD. Angiogenesis inhibitors: current strategiesand future prospects. CA Cancer J Clin 2010;60:222–43.

[6] Loges S, Schmidt T, Carmeliet P. Mechanisms of resistanceto anti-angiogenic therapy and development ofthird-generation anti-angiogenic drug candidates. GenesCancer 2010;1:12–25.

[7] Bergers G, Song S, Meyer-Morse N, Bergsland E, Hanahan D.Benefits of targeting both pericytes and endothelial cells inthe tumor vasculature with kinase inhibitors. J Clin Invest2003;111:1287–95.

[8] Saharinen P, Eklund L, Pulkki K, Bono P, Alitalo K. VEGF andangiopoietin signaling in tumor angiogenesis andmetastasis.Trends Mol Med 2011;17:347–62.

[9] Augustin HG, Koh GY, Thurston G, Alitalo K. Control ofvascular morphogenesis and homeostasis through theangiopoietin-Tie system. Nat RevMol Cell Biol 2009;10:165–77.

[10] Jendreyko N, Popkov M, Rader C, Barbas III CF. Phenotypicknockout of VEGF-R2 and Tie-2 with an intradiabody reducestumor growth and angiogenesis in vivo. Proc Natl Acad Sci U SA 2005;102:8293–8.

[11] Erber R, Thurnher A, Katsen AD, Groth G, Kerger H, HammesHP, et al. Combined inhibition of VEGF and PDGF signaling

enforces tumor vessel regression by interfering withpericyte-mediated endothelial cell survival mechanisms.FASEB J 2004;18:338–40.

[12] Huang J, Bae JO, Tsai JP, Kadenhe-Chiweshe A, Papa J, Lee A,et al. Angiopoietin-1/Tie-2 activation contributes to vascularsurvival and tumor growth during VEGF blockade. Int J Oncol2009;34:79–87.

[13] Nandakumar V, Singh T, Katiyar SK. Multi-targeted preven-tion and therapy of cancer by proanthocyanidins. Cancer Lett2008;269:378–87.

[14] Kaur M, Agarwal C, Agarwal R. Anticancer and cancerchemopreventive potential of grape seed extract and othergrape-based products. J Nutr 2009;139:1806S–12S.

[15] Singh RP, Tyagim AK, Dhanalakshmim S, Agarwalm R,Agarwalm C. Grape seed extract inhibits advanced humanprostate tumor growth and angiogenesis and up-regulatesinsulin-like growth factor binding protein-3. Int J Cancer2004;108:733–40.

[16] Wen W, Lu J, Zhang K, Chen S. Grape seed extract inhibitsangiogenesis via suppression of the vascular endothelialgrowth factor receptor signaling pathway. Cancer Prev Res(Phila) 2008;1:554–61.

[17] Fraga CG, Oteiza PI. Dietary flavonoids: role of (−)-epicatechinand related procyanidins in cell signaling. Free Radic Biol Med2011;51:813–23.

[18] Vichai V, Kirtikara K. Sulforhodamine B colorimetric assay forcytotoxicity screening. Nat Protoc 2006;1:1112–6.

[19] Lee HS, Na MH, Kim WK. alpha-Lipoic acid reduces matrixmetalloproteinase activity in MDA-MB-231 human breastcancer cells. Nutr Res 2010;30:403–9.

[20] Zhang C, Yang F, Zhang XW, Wang SC, Li MH, Lin LP, et al.Grateloupia longifolia polysaccharide inhibits angiogenesisby downregulating tissue factor expression in HMEC-1endothelial cells. Br J Pharmacol 2006;148:741–51.

[21] Nguyen M, Shing Y, Folkman J. Quantitation of angiogenesisand antiangiogenesis in the chick embryo chorioallantoicmembrane. Microvasc Res 1994;47:31–40.

[22] Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and itsreceptors. Nat Med 2003;9:669–76.

[23] Brindle NP, Saharinen P, Alitalo K. Signaling and functions ofangiopoietin-1 in vascular protection. Circ Res 2006;98:1014–23.

[24] Hwang YP, Choi JH, Han EH, Kim HG, Wee JH, Jung KO, et al.Purple sweet potato anthocyanins attenuate hepatic lipidaccumulation through activating adenosine monopho-sphate–activated protein kinase in human HepG2 cells andobese mice. Nutr Res 2011;31:896–906.

[25] Hushmendy S, Jayakumar L, Hahn AB, Bhoiwala D, BhoiwalaDL, Crawford DR. Select phytochemicals suppress humanT-lymphocytes and mouse splenocytes suggesting their usein autoimmunity and transplantation. Nutr Res 2009;29:568–78.

[26] Leifert WR, Abeywardena MY. Cardioprotective actions ofgrape polyphenols. Nutr Res 2008;28:729–37.