Stewartia koreana extract stimulates proliferation and migration of human endothelial cells and induces neovasculization in vivo

Copyright © 2009 John Wiley & Sons, Ltd. Phytother. Res. 24: 20–25 (2010)DOI: 10.1002/ptr

20 T. H. LEE ET AL.

Copyright © 2009 John Wiley & Sons, Ltd.

PHYTOTHERAPY RESEARCHPhytother. Res. 24: 20–25 (2010)Published online 22 June 2009 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/ptr.2851

Stewartia koreana Extract StimulatesProliferation and Migration of HumanEndothelial Cells and Induces NeovasculizationIn Vivo

Tae Hoon Lee1, Guy Wilhem Lee1, Chan Woo Kim1, Myun-Ho Bang2, Nam-In Baek1,Sung-Hoon Kim3, Dae Kyun Chung1,2 and Jiyoung Kim1*1Graduate School of Biotechnology and Institute of Life Sciences and Resources, Kyung Hee University, Yongin 446-701, Korea2Skin Biotechnology Center, Kyung Hee University, Yongin 446-701, Korea3Cancer Preventive Material Development Research Center and Institute, College of Oriental Medicine, Kyung Hee University,Seoul, 131-701, Korea

Angiogenesis, the growth of new blood vessels from preexisting vasculature, plays an important role in physi-ological and pathological processes such as embryonic development and wound healing. This study investigatedthe effects of methanol extracts of Stewartia koreana leaves (SKE) on angiogenesis. Stewartia koreanasignificantly promoted the proliferation and migration of human umbilical vein endothelial cells in a dose-dependent manner. The SKE induced endothelial cell proliferation in the range of 50 µµµµµg/mL without cytotoxicity.Treatment of HUVECs resulted in the activation of the mitogen-activated protein kinases that was correlatedwith endothelial cell proliferation and migration. SKE also stimulated angiogenesis in a chick chorioallantoicmembrane assay, demonstrating promotion of new blood vessel formation in vivo. Local administrationof SKE onto skin punched wounds resulted in increased von Willebrand Factor antigen, indicating that itstimulated neovasculization in the wound region. The results suggest that Stewartia koreana extracts maypotentially be useful for the development of agents to accelerate vascular wound healing or to promote thegrowth of collateral blood vessels in ischemic tissues. Copyright © 2009 John Wiley & Sons, Ltd.

Keywords: angiogenesis; migration; proliferation: endothelial cell; plant extract; Stewartia koreana.

Received 18 November 2008Revised 9 March 2009

Accepted 9 March 2009

INTRODUCTION

A variety of herbs and plants have traditionally beenused in folk medicine in Asian countries for the treat-ment of inflammatory diseases and wounds (Phan et al.,2000; Biswas and Mukherjee, 2003; Nayak and Pinto,2006; Kimura et al., 2007; Kumara et al., 2007). It has beenreported that extracts of Stewartia koreana exhibit vari-ous biological activities, such as antioxidant activity,HIV-1 protease inhibitory activities and inhibition ofNO production in LPS-stimulated macrophage cells(Min et al., 1999; Han et al., 2003; Kim et al., 2003a,2003b; Choi et al., 2005). Furthermore, a recent reportshowed that Stewartia koreana extracts inhibited LPS-induced transactivation of NF-κB, which resulted inthe suppression of nitric oxide production in an LPS-stimulated murine macrophage cell line, RAW 264.7(Lee et al., 2007).

Angiogenesis plays an important role in physiologicaland pathological processes such as embryonic develop-ment, wound healing, chronic inflammation and tumorgrowth (Pandya et al., 2006). Angiogenesis is regulated

by a number of growth factors, including basic fibroblastgrowth factor (bFGF) and vascular endothelial cell growthfactor (VEGF) which promote proliferation, migrationand other angiogenic activities of endothelial cells (Tsuboiand Rifkin, 1990; Bates and Jones, 2003; Scharpfeneckeret al., 2007). It has been suggested that proangiogenicgrowth factors would be effective as therapeutic agentsfor the treatment of wound healing (Schultz et al., 1991;Roesel and Nanney, 1995), ischemic cardiovascular dis-eases and bone fracture healing (Tsuboi and Rifkin,1990; Scharpfenecker et al., 2007). Therefore, identifica-tion of agents with pro-angiogenic activity that are costeffective and non-toxic would be very useful for thedevelopment of a potential alternative agent for thetreatment of wound healing and ischemic diseases.During a search for angiogenic substances produced bynatural products, it was observed that extracts fromStewartia koreana stimulated the migration of humanendothelial cells. In this study, it was demonstrated thatextracts of Stewartia koreana leaves (SKE) stimulatedangiogenic properties in vitro and in vivo. Further, itwas demonstrated the enhancement of neovasculizationin punched skin wound of mouse by SKE.

MATERIALS AND METHODS

Materials and cell culture. The methanol extract ofStewartia koreana was prepared as reported previously

* Correspondence to: Jiyoung Kim, Graduate School of Biotechnologyand Institute of Life Sciences and Resources, Kyung Hee University,Yongin 446-701, Korea.E-mail: [email protected]/grant sponsor: Korea Healthcare technology R&D Project,MIHWAF, Republic of Korea; contract/grant number: MIHWAF-A080339.

STEWARTIA KOREANA INDUCES ANGIOGENESIS IN VITRO AND IN VIVO 21


(Lee et al., 2007). Briefly, the leaves of Stewartia koreanawere dried and refluxed twice with 80% methanol. Theextract was then filtered and centrifuged at 800 × gfor 10 min. Methanol in the extract was removed underreduced pressure by rotary evaporation and the waterresidue was removed by lyophilization. The percent yieldwas calculated on the basis of the dried plant materialweight. The dried powder was dissolved in DMSO inaliquots and diluted to the desired concentrations.

Human umbilical vein endothelial cells (HUVECs)were maintained in M199 containing 10% fetal bovineserum (Invitrogen, Carlsbad, CA, USA), 3 ng/mL bFGF,5 units/mL heparin, 100 unit/ml of penicillin and 100 µg/mL of streptomycin in a humidified incubator of 5%CO2 at 37 °C.

Analysis of ααααα-amyrin in SKE by UPLC. Samples werefiltered through a 0.2 µm microspin PVDF filter. Theapparatus was a UPLC system equipped with a UVdetector (Waters, Milford, MA, USA). The constituentsof the SKE were separated on a Waters Acquity UPLC™

BEH C8 column (1.7 µm, 2.1 × 100 mm). The mobile phasewas eluted by using a gradient system containing 0.1%acetic acid in water and 0.1% acetic acid in acetonitrile.The flow rate was 100 µL/min and the UV detector wasset at an absorbance of 210 nm.

Cell viability and proliferation assay. Cell viability wastested by the 3-[4,5-dimetnythiazol-2-yl]-2,5-diphenyl-thetazolium bromide (MTT)-based colorimetric assay asdescribed previously (Pokharel et al., 2007). Proliferationassay was performed by using 5-bromo-2′-deoxy-uridine(BrdU) labeling and detection kit (Roche, Indianapolis,IN, USA). In brief, HUVECs were seeded in each wellof gelatin 96-well plates for 24 h at 37 °C. After removingmedium, the cells were starved for 6 h and then growthmedium and SK extract were added at various con-centrations for 24 h. Vascular endothelial specific growthfactor (VEGF) was obtained from R&D Systems(Minneapolis, MN, USA) and used for positive con-trol. The BrdU-integrated DNA was quantitated by therelative luminescence unit (RLU) of each well usinga Wallace Victor2 1420 multilabel counter as describedpreviously (Kim et al., 2006).

Western blot analysis. HUVEC cells were treated withextracts of Stewartia koreana for 20 min and lysedin RIPA buffer containing proteinase inhibitors. Theprotein concentration was quantified with a proteinassay kit. Proteins (20 µg/lane) were resolved with SDS-polyacrylamide gel electrophoresis, and western blotanalysis was performed as described previously (Leeet al., 2007).

Scratch wound closure assay. Scratch wound closureassay was performed as described previously (Saitohet al., 2006). Confluent HUVEC monolayers in 6-wellplates (Corning Costar, Cambridge, MA) were scratchedand wounded using a universal sterile 200 µL pipette tipand rinsed with PBS. Each cell was treated with vari-ous concentrations of SKE for 8 h. VEGF were addedas a positive control (R&D Systems). The cells werestained with Diff-Quick and the width of the woundwas measured using an Olympus digital camera. Eachcondition was studied in triplicate wells, and eachexperiment was performed three times.

Cell migration assay. HUVECs migration assay was con-ducted in a 48-well chemotaxis chamber (Neuro Probe Inc.,Cabin John, MD, USA) as described previously (Hwanget al., 2004). PVP-free polycarbonate membrane 12 µm pores(Neuro Probe Inc.) were coated with 0.1% gelatin (Sigma,St Louis, MO, USA). The bottom chamber was loadedwith 30 000 cells, and polycarbonate membrane was laidover the cells. The chemotaxis chamber was then invertedand incubated at 37 °C for 2 h. The upper wells were thenloaded with RPMI 1640 containing 0.1% BSA and variousconcentrations of SK extract. VEGF was added to severalwells as a positive control. The chamber was re-incubatedat 37 °C for 2 h, and the filters were fixed and stainedusing Diff-Quick (Baxter Healthcare Corp., McGraw Park,IL, USA). Each condition was studied in triplicate wells,and each experiment was performed at least three times.

Tube formation assay on matrigel. UnpolymerizedMatrigel (Collaborative Biomedical Products, MA, USA)was added to 24-well plates, with a total volume of 300 µLin each well, and allowed to polymerize overnight at37 °C. Various concentrations of SKE, 10 ng/mL of VEGFwere plated onto the layer of Matrigel at a density of1 × 105 cells/well in control medium. After 8 h, the cellswere photographed, and the extent of tube formationwas analysed using the Scion image program.

Chick chorioallantoic membrane assay. The modifiedchorioallantoic membrane (CAM) assay was carried outas described previously (Hwang et al., 2004). Briefly,10 µL of test samples in type I collagen were applied ontoThermonox disks and polymerized at room temperature.The discs were loaded onto the CAM of 10-day-oldembryos. After 72 h incubation at 37 °C, the area aroundthe loaded disk was photographed with an Olympusdigital camera and the number of newly formed vesselswas counted by two observers in a double-blind manner.

Immunohistochemical analysis. The immunohistoche-mical assay was performed, 10 µm sections were excisedfrom embedded tissues and mounted on to poly-L-lysinecoated slides (overnight at 37 °C). Tissues were thendeparaffinized and rehydrated. After washing in dH2O(5 min, 2 times) and PBST, the sections were then in-cubated with 10 mM citric buffer (pH 7.0) overnight at70 °C. After incubation with 10% goat serum in TBST for1 h at room temperature, the sections were incubated withrabbit anti-vWF (von Willebrand Factor) polyclonal anti-body (1:500, Chemicon) overnight at 4 °C. After beingrinsed with TBST, the tissues were stained with goatanti-rabbit IgG-FITC (1:200, Santa Cruz) and mountedwith buffered glycerol. Wound areas were observed byusing a confocal microscope (Carl Zeiss, Germany).

Statistical analysis. Unless otherwise stated, all experimentswere performed with triplicate samples and repeated atleast three times. The data are presented as mean ± SD andstatistical comparisons between groups were performedusing one-way ANOVA followed by Student’s t-test.

RESULTS

UPLC analysis of methanol extract of Stewartia koreana(SKE) was performed to obtain profiles of the extracts.


22 T. H. LEE ET AL.

tionally, the level of endothelial cell migration inducedby 50 µg/mL of SKE was similar to the levels observedfor the positive control, VEGF, at 10 ng/mL. The woundclosure assay also similarly demonstrated that SKEenhances cell migration in vitro. When a confluentmonolayer was scratched, migration of cells into thewounded area was significantly increased in the pres-ence of 100 µg/mL of SKE compared with migration inmedia alone, as observed 8 h after wounding endothelialcells (Fig. 2C). The number of the migrated cells on the

Figure 1. UPLC profiles and structure of α-amyrin. α-Amyrinwas used as a reference substance and the chemical structureof α-amyrin is shown (A). The UPLC fraction appearing at6.364 min (B) was standard α-amyrin, and the α-amyrin in SKE(C) was detected at 6.364 min. * detection time of α-amyrin.

Figure 2. SKE stimulates proliferation, migration and tube-likeformation in HUVECs (A) HUVECs were plated in 96-well plates;the total cells in each well were then detected using a BrdUincorporation assay, as described in Materials and Methods.* p < 0.05 versus media alone. Data are shown as the meanvalue ± SD of over three independent experiments conductedon triplicate samples. (B) Quantification of migrated endothelialcells in the presence of different concentrations of SKE (�) wascompared with that observed in the presence of either mediumalone or VEGF (�). * p < 0.01 versus medium alone. Data areshown as the mean value ± SD of over three independentexperiments conducted on triplicate samples. (C) HUVECs weretreated with various concentrations of SKE for 8 h. SKE sig-nificantly accelerates the wound closure of human endothelialcells. Representative photographs of the medium two differentconcentrations of SKE, and cell specific positive controls.

α-Amyrin was chosen as a reference substance in thepresent study because it was one of the abundant sub-stances in SKE as deduced by TLC analysis (data notshown) and by UPLC analysis (Fig. 1) and also becauseα-amyrin belongs to the triterpenoid family, which areknown to promote wound healing processes (Maquartet al., 1999). α-Amyrin was isolated from SKE and thecontent of α-amyrin in SKE was estimated to be about5.4% of the total extract. Based on the standardcalibration curve, the R2 value of amyrin was 0.99, whichshowed a highly linear relationship between the stand-ard amyrin peak and the sample peak. As shown inFig. 1, α-amyrin in SKE was identified by matching theretention time against reference α-amyrin.

The effects of various concentrations of SKE onthe proliferation of HUVECs were investigated usinga BrdU incorporation assay. Additionally, the studyinvestigated the effects of SKE on the viability of cellsusing a colorimetric MTT assay. SKE stimulated theproliferation of HUVECs in a dose-dependent mannerup to 100 µg/mL (Fig. 2A). Proliferation was increasedby approximately 2-fold over cells grown in mediumalone when HUVECs were treated with 100 µg/mL ofSKE. Furthermore, cell viability was greater than 80%up to a concentration of 100 µg/mL (data not shown),indicating that SKE is not cytotoxic to human endothelialcells in the ranges of concentrations tested in this study.

The effects of SKE on endothelial cell migrationwere next measured using a Boyden chamber assay andwound closure assay. The results showed that SKEinduced endothelial cell migration in a dose-dependentmanner (Fig. 2B). After 4 h, cell migration in responseto SKE was enhanced by two- to three-fold (p < 0.01)over migration in the presence of media alone. Addi-



signaling pathways for proliferation and migrationof endothelial cells. Thus, the effects of SKE on thephosphorylation of Akt and MAP kinases were inves-tigated. As shown in Fig. 3C, treatment of HUVECswith SKE stimulated phosphorylation of ERK and Aktkinases in a dose-dependent manner. Strong activationof phosphorylation was observed in HUVECs treatedwith SKE at concentrations of 50 µg/mL and 100 µg/mL,which were stronger than that of the positive control,VEGF. These results demonstrate that SKE enhancesproliferation and migration of endothelial cells via theactivation of ERK and Akt pathways.

Angiogenic activity of SKE in vivo was confirmed usinga modified chorioallantoic membrane (CAM) assay.SKE significantly induced neovascularization from pre-existing blood vessels. The presence of 0.1, 1, 10 and100 µg of SKE per egg caused an approximate 4-foldincrease in the number of newly formed blood vesselscompared with that of the negative control (Fig. 4Aand B).

Angiogenesis is known to play an important role inthe cutaneous wound healing process. Immunohistoche-mical analysis of skin punched wound of mice using

wound scratch areas could be influenced by prolifera-tion of endothelial cells. However, SKE-enhanced pro-liferation might not have significantly affected theincrease in the migrated cells in wound closure areas,because HUVECs were treated with SKE only for8 h in the wound closure assays. The levels of woundclosure in response to SKE were similar to those ofthe positive control, the cells treated with 10 ng/mLVEGF. Wound closure of HUVEC cells treated withSKE was accelerated significantly when compared withthe negative control. Overall, the results suggest thatSKE may possess potent angiogenic activity.

The study next tested that the stimulatory potencyof SKE on the differentiation of endothelial cells intotube-like structures. Under the SKE-free condition, themean intensity of HUVEC tube-like structures formedwas 5684 ± 133.5; whereas in the presence of SKEsignificantly increased tube-like structures formationin a dose dependent manner was found. With SKE at100 µg/mL, the mean intensity of tube-like structuresformed was 28,657 ± 334.3 tubes/field, a about a 5-foldstimulation compared with the control (Fig. 3A and B).

The phosphorylation of Akt and MAP kinasesignaling pathways are known to be involved in the

Figure 3. Effects of SKE on tube-like formation and MAPKsphosphorylation in HUVECs. (A) HUVECs were treated with dif-ferent concentration of SKE, 10 ng/mL VEGF on Matrigel. Afterincubation for 8 h and fixation, the cells were observed underthe microscope (magnification, 100×) and photographed. Digitalimage of untreated HUVECs (control), VEGF and 100 µg/mL SKE.(B) Tubes were counted per field in at least four fields aftereach experiment, and results were expressed as the number oftubes formed. Data are expressed as the mean ± SD of threeindependent experiments. * p < 0.01 compared with control. (C)Activation of MAPKs by SKE was determined western blottingusing antibody against phosphorylated form of MAPKs and AKTin HUVECs. The membranes were stripped and reprobed withantibody against MAPKs.

Figure 4. SKE enhances blood formation in the chick CAM assayand wound areas. SKE or VEGF was loaded onto the CAMs of10 day old chick embryos. After 72 ± 4 h incubation, a fat emulsionwas injected under the CAMs for better visualization of the vessels.Disks and surrounding CAMs were photographed. (A) Repre-sentative photographs of the vehicle, VEGF and 100 µg/egg ofSKE. (B) Quantification of newly formed blood vessels. * p < 0.01versus vehicle. Twelve to fifteen eggs were used for each datapoint and mean value ± SD are shown. (C) 9 days after wound-ing, immunohistochemically stained skins were observedthrough a confocal microscope. Tissue showed an enormousfluorescence in wound areas, whereas fewer antibodies againstvWFs were seen in PBS and normal wound tissues.


24 T. H. LEE ET AL.

antibody against von Willebrand Factors (vWFs), amarker for endothelial cells, was performed. Skin sectionsof the wound region 9 days after operation were observedusing FITC-coupled secondary antibody against vWFthrough a confocal microscope. There are two adjacentpictures (left: without fluorescence, right: fluorescence)for each photograph. As shown in Fig. 4C, SKE at 100 µg/mice treated tissues and EGF treated tissues showed amarked increase in fluorescence in wound areas, whereasfewer antibodies against vWFs were seen in PBS treatedand normal tissues.

DISCUSSION

The process of wound healing involves several phases,including the formation of granulation tissue, re-epithelialization, neovascularization and tissue reorgani-zation (Broughton et al., 2006). Among these phases,neovascularization, which is also called angiogenesis,plays a critical role in the early phases of woundhealing. Agents that control angiogenesis are beneficialfor the development of therapeutics for treatment ofwound healing and tissue regeneration. In this paper,it was demonstrated that methanol extracts of Stewartiakoreana enhance angiogenesis in vitro and in vivo.

The in vitro experiments examined the effect of SKEon the proliferation of human endothelial cells andit was found that SKE stimulated human endothelialcells in a dose-dependent manner. Cell proliferation ofSKE (50 µg/mL)-treated cells increased by 2-fold, andalmost no cytotoxic effects or induced morphologicalchanges were noted in cells treated with SKE, indicat-ing that enhancement of the proliferation of endothelialcells occurred at non-cytotoxic concentrations.

Further, a Boyden chamber assay and wound scratchassay showed that SKE strongly induced endothelialcell migration on gelatin in a dose-dependent manner.Additionally, SKE strongly induced neovasculariza-tion from pre-existing blood vessels in vivo on a chickchorioallantoic membrane. This is important becauseneovascularization is a prerequisite for wound healingthat leads to the formation of granulation tissue. Thegranulation tissue provides the medium for migrationof surrounding cells from the wound margin to occur,leading to gap closure. Immunohistochemical analysisof skin punched wound of mice using antibody againstvon Willebrand Factors (vWFs), a marker for endothelialcells, demonstrated that SKE stimulated neovasculizationin the wound region, suggesting that Stewartia koreana

extracts may potentially be useful for the developmentof agents to accelerate vascular wound healing or topromote the growth of collateral blood vessels in ischemictissues.

A number of studies has reported that various plantextracts contain wound healing activities via angiogenesis(Kim et al., 2007; Lam et al., 2008). The angiogenicactivity of SKE may be attributed to phytoconstituentspresent in the methanol extracts of Stewartia koreana,and the promoted process of angiogenesis may be afunction of either an active compound or the synergisticeffects of multiple components of phytochemicals. Aphytochemical analysis of the extracts of leaves fromStewartia koreana showed the presence of a varietyof compounds including triterpenoids, alkaroids andflavonoids. Any one of the phytochemical constituentspresent in SKE may be responsible for the angiogenicactivity. Phytochemicals such as triterpenoids (Coldrenet al., 2003) and flavonoids (Hsu, 2005; Kimura et al.,2007) are known to promote the wound healing pro-cess via angiogenesis, primarily as a result of their anti-microbial and astringent properties. However, α-amyrinalone appears not to have a significant angiogenic acti-vity, although it belongs to the triterpenoids family andis known to promote healing processes by enhancingthe synthesis of collagen in fibroblast cells (Moon et al.,2005). The preliminary fractionation experiments of theSKE indicated that there are several peaks that may leadto angiogenic activities. Isolation and characterizationof the phytochemicals present in SKE is being processedfor identification of the phytoconstituents responsiblefor the angiogenic activities in vitro and in vivo.

In conclusion, this study demonstrated that extractsof leaves from Stewartia koreana enhanced the prolif-eration and migration on human endothelial cells. SKEalso promoted differentiation of human endothelial cells,which may be mainly attributed to the enhancement ofthe angiogenesis process as a result of SKE treatment.The results demonstrated for the first time the angiogenicproperties of Stewartia koreana, which may be associatedwith mitogenic properties on endothelial cells. The findingswarrant further investigation of Stewartia koreana toidentify the phytoconstituents responsible for woundhealing and for the development of agents for the topicaltreatment of wounds.

Acknowledgement

This research was supported by a grant of the Korea HealthcareTechnology R&D Project, Ministry for Health, Welfare and FamilyAffairs, Republic of Korea (A080339).

REFERENCES

expression changes in the human fibroblast inducedby Centella asiatica triterpenoids. Planta Med 69: 725–732.

Han SS, Lo SC, Choi Y, Kim JH, Baek SH. 2003. Antioxidantactivity of crude extract and pure compounds of Acer ginnalaMax. Bull Korean Chem Soc 25: 389–391.

Hsu S. 2005. Green tea and the skin. J Am Acad Dermatol 52:1049–1059.

Hwang J, Kim CW, Son KN et al. 2004. Angiogenic activity ofhuman CC chemokine CCL15 in vitro and in vivo. FEBS Lett570: 47–51.

Kim CW, Son KN, Choi SY, Kim J. 2006. Human lactoferrin

Bates DO, Jones RO. 2003. The role of vascular endothelialgrowth factor in wound healing. Int J Low Extrem Wounds2: 107–120.

Biswas TK, Mukherjee B. 2003. Plant medicines of Indian originfor wound healing activity: a review. Int J Low ExtremWounds 2: 25–39.

Broughton G, Janis JE, Attinger CE. 2006. The basic science ofwound healing. Plast Reconstr Surg 117: 12S–34S.

Choi YH, Han SS, Lee HO, Baek SH. 2005. Biological activity ofbioactive components from Acer ginnala max. Bull KoreanChem Soc 26: 1450–1452.

Coldren CD, Hashim P, Ali JM, Oh SK, Rha C. 2003. Gene



upregulates expression of KDR/Flk-1 and stimulates VEGF-A-mediated endothelial cell proliferation and migration.FEBS Lett 580: 4332–4336.

Kim RG, Shin KM, Kim YK et al. 2003a. Inhibition of methanolextract from the aerial parts of Saururus chinensis onlipopolysaccharide-induced nitric oxide and prostaglandinE2 production from murine macrophage RAW 264.7 cells.Biol Pharm Bull 26: 481–486.

Kim Y, Min HY, Park HJ et al. 2003b. Suppressive effects ofnitric oxide production and inducible nitric oxide synthase(iNOS) gene expression by Calystegia soldanella methanolextract on lipopolysaccharide-activated RAW 264.7 cells. EurJ Cancer Prev 13: 419–424.

Kim YM, Namkoong S, Yun YG et al. 2007. Water extract ofKorean red ginseng stimulates angiogenesis by activatingthe PI3K/Akt-dependent ERK1/2 and eNOS pathways inhuman umbilical vein endothelial cells. Biol Pharm Bull 30:1674–1679.

Kimura Y, Sumiyoshi M, Sakanaka M. 2007. Effects of Astilbethunbergii rhizomes on wound healing. Part 1. Isolationof promotional effectors from Astilbe thunbergii rhizomeson burn wound healing. J Ethnopharmacol 109: 72–77.

Kumara Swamy HM, Krishna V, Shankarmurthy K et al. 2007.Wound healing activity of embelin isolated from the ethanolextract of leaves of Embelia ribes Burm. J Ethnopharmacol109: 529–534.

Lam HW, Lin HC, Lao SC et al. 2008. The angiogenic effects ofAngelica sinensis extract on HUVEC in vitro and zebrafishin vivo. J Cell Biochem 103: 195–211.

Lee TH, Kwak HB, Kim HH et al. 2007. Methanol extracts ofStewartia koreana inhibit cyclooxygenase-2 (COX-2) and induc-ible nitric oxide synthase (iNOS) gene expression by blockingNF-kappaB transactivation in LPS-activated RAW 264.7 cells.Mol Cells 23: 398–404.

Maquart FX, Chastang F, Simeon A, Birembaut P, Gillery P,Wegrowski Y. 1999. Triterpenes from Centella asiatica stimu-late extracellular matrix accumulation in rat experimentalwounds. Eur J Dermatol 9: 289–296.

Min BS, Bae KH, Kim YH, Miyashiro H, Hattori M, ShimotohnoK. 1999. Screening of Korean plants against humanimmunodeficiency virus type 1 protease. Phytother Res 13:680–682.

Moon HI, Kim MR, Woo ER, Chung JH. 2005. Triterpenoid fromStyrax japonica Sieb. et Zucc, and its effects on the expres-sion of matrix metalloproteinases-1 and type 1 procollagencaused by ultraviolet irradiated cultured primary human skinfibroblasts. Biol Pharm Bull 27: 2003–2006.

Nayak BS, Pinto Pereira LM. 2006. Catharanthus roseus flowerextract has wound-healing activity in Sprague Dawley rats.BMC Complement Altern Med 21: 6–41.

Pandya NM, Dhalla NS, Santani DD. 2006. Angiogenesis – anew target for future therapy. Vasc Pharmacol 44: 265–274.

Phan TT, Lee ST, Chan SY, Hughes MA, Cherry GW. 2000.Investigating plant-based medicines for wound healing withthe use of cell culture technologies and in vitro models: areview. Ann Acad Med Singapore 29: 27–36.

Pokharel YR, Liu QH, Oh JW, Woo ER, Kang KW. 2007. 4-Hydroxykobusin inhibits the induction of nitric oxide synthaseby inhibiting NF-κB and AP-1 activation. Biol Pharm Bull 30:1097–1101.

Roesel JF, Nanney LB. 1995. Assessment of differential cytokineeffects on angiogenesis using an in vivo model of cutane-ous wound repair. J Surg Res 58: 449–459.

Saitoh Y, Koizumi K, Minami T, Sekine K, Sakurai H, Saiki I.2006. A derivative of aminopeptidase inhibitor (BE15) has adual inhibitory effect of invasion and motility on tumor andendothelial cells. Biol Pharm Bull 29: 709–712.

Scharpfenecker M, van Dinther M, Liu Z et al. 2007. BMP-9signals via ALK1 and inhibits bFGF-induced endothelial cellproliferation and VEGF-stimulated angiogenesis. J Cell Sci120: 964–972.

Schultz G, Rotatori DS, Clark W. 1991. EGF and TGF-alpha inwound healing and repair. J Cell Biochem 45: 346–352.

Tsuboi R, Rifkin DB. 1990. Recombinant basic fibroblast growthfactor stimulates wound healing in healing-impaired db/dbmice. J Exp Med 172: 245–251.

Documents

Stewartia koreana extract stimulates proliferation and migration of human endothelial cells and induces neovasculization in vivo