SEARCH

SEARCH BY CITATION

Keywords:

  • emodin;
  • tyrosine kinase inhibitor;
  • angiogenesis;
  • vascular endothelial growth factor-A;
  • KDR/Flk-1

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Emodin (1,3,8-trihydroxy-6-methylanthraquinone), an active component in the root and rhizome of Rheum palmatum, is a tyrosine kinase inhibitor with a number of biological activities, including antitumor effects. Here, we examine the effects of emodin on vascular endothelial growth factor (VEGF)-A-induced angiogenesis, both in vitro and in vivo. In vitro, emodin dose-dependently inhibits proliferation, migration into the denuded area, invasion through a layer of Matrigel and tube formation of human umbilical vein endothelial cells (HUVECs) stimulated with VEGF-A. Emodin also inhibits basic fibroblast growth factor-induced proliferation and migration of HUVECs and VEGF-A-induced tube formation of human dermal microvascular endothelial cells. Specifically, emodin induces the cell cycle arrest of HUVECs in the G0/G1 phase by suppressing cyclin D1 and E expression and retinoblastoma protein phosphorylation, and suppresses Matrigel invasion by inhibiting the basal secretion of matrix metalloproteinase-2 and VEGF-A-stimulated urokinase plasminogen activator receptor expression. Additionally, emodin effectively inhibits phosphorylation of VEGF-A receptor-2 (KDR/Flk-1) and downstream effector molecules, including focal adhesion kinase, extracellular signal-regulated kinase 1/2, p38 mitogen-activated protein kinase, Akt and endothelial nitric oxide synthase. In vivo, emodin strongly suppresses neovessel formation in the chorioallantoic membrane of chick and VEGF-A-induced angiogenesis of the Matrigel plug in mice. Our data collectively demonstrate that emodin effectively inhibits VEGF-A-induced angiogenesis in vitro and in vivo. Moreover, inhibition of phosphorylation of KDR/Flk-1 and downstream effector molecules is a possible underlying mechanism of the anti-angiogenic activity of emodin. Based on these data, we propose that an interaction of emodin with KDR/Flk-1 may be involved in the inhibitory function of emodin toward VEGF-A-induced angiogenesis in vitro and responsible for its potent anti-angiogenic in vivo. © 2006 Wiley-Liss, Inc.

Angiogenesis, the formation of new vessels from preexisting vasculature, is an essential process in a variety of physiological and pathological conditions, including wound healing, embryonic development, chronic inflammation, cancer and metastasis.1, 2, 3, 4, 5 Complex sequential steps are involved in angiogenesis, such as basement membrane degradation by proteases, endothelial cell proliferation and migration, formation of capillary tubes and survival of newly formed blood vessels.6 Angiogenesis is tightly regulated by an intricate balance between stimulators and inhibitors.6 Among these, vascular endothelial growth factor (VEGF)-A, a soluble angiogenic factor produced by many tumors as well as normal cell lines, including vascular smooth muscle cells, lung epithelium and pituitary folliculo-stellate cells, plays a key role in regulating normal and pathologic angiogenesis.7

VEGF-A is a potent mitogen for endothelial cells, and an important mediator of angiogenesis, inducing endothelial cell proliferation, protease expression and migration, as well as subsequent organization of cells to form a capillary tube.8, 9, 10 In particular, a number of studies have shown that VEGF-A is the most important angiogenic factor closely associated with neovascularization in human tumors. Recent reports have disclosed increased VEGF-A mRNA in tumor cell lines. The VEGF-A level is an important prognostic marker of tumor angiogenesis.10, 11 VEGF-A induces angiogenesis via binding to its two receptor tyrosine kinases, KDR/Flk-1 and Flt-1, expressed on endothelial cells. KDR/Flk-1 is required mainly for mitogenic and chemotactic responses, whereas Flt-1 contributes to endothelial cell morphogenesis.12, 13 Recently, the intracellular signaling pathways mediating these effects downstream of KDR/Flk-1 activation have been identified. KDR/Flk-1 induces proliferation through activation of the extracellular signal-regulated kinase (ERK) 1/2 pathway leading to gene transcription14 and endothelial cell survival through phosphatidylinositol 3-kinase activation, resulting in increased lipid phosphatidylinositol (3,4,5)P3 and activation of several important intracellular molecules, such as Akt and the small GTP-binding protein, Rac.15 The Akt pathway additionally triggers endothelial nitric oxide synthase (eNOS) activity16, 17 to generate NO, leading to vascular permeability and cellular migration. Other signal transduction molecules implicated in KDR/Flk-1-dependent cytoskeletal regulation and cell migration include p38 mitogen-activated protein kinase (MAPK) and p125 focal adhesion kinase (FAK).18, 19

Emodin (1,3,8-trihydroxy-6-methylanthraquinone), a tyrosine kinase inhibitor isolated from Rheum palmatum, is an active constituent of Chinese herbs.20 The compound sensitizes HER-2/neu-overexpressing lung cancer cells, has antitumor effects on neuroectodermal and represses transformation and metastasis-associated properties of HER-2/neu-overexprssing breast cancer cells.20, 21, 22 Emodin additionally exerts anti-inflammatory effects on endothelial cells by inhibiting tumor necrosis factor -induced activation of nuclear factor-kappa B in human umbilical vein endothelial cells (HUVECs).23 Given that numerous endogenous factors and pharmacological agents that regulate cancer progression and inflammation additionally affect angiogenesis, it seems quite likely that emodin would also affect angiogenesis. In the present study, we investigated whether emodin has anti-angiogenic activity especially on the VEGF-A-induced angiogenesis, and if so, the critical mechanism of action.

Our data show that emodin significantly inhibits VEGF-A-induced angiogenesis in vitro and in vivo. Emodin effectively blocks VEGF-A-induced proliferation, migration, invasion and tube formation of HUVECs in vitro, and markedly inhibits neo-angiogenesis of chick chorioallantoic membrane (CAM) and mouse Matrigel in vivo. Blockade of VEGF-A-induced tyrosine phosphorylation of KDR/Flk-1 and downstream signaling molecules by emodin may be responsible for its potent anti-angiogenic activity. The results strongly highlight the efficacy of this compound in controlling neovessel formation in diseases induced by angiogenesis.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Materials

Emodin, bovine serum albumin (BSA) and gelatin were purchased from Sigma Chemical Co. (St. Louis, MO). Human recombinant VEGF-A and basic fibroblast growth factor (bFGF) were purchased from R&D Systems (Minneapolis, MN). Antibodies against matrix metalloproteinase (MMP)-2, -9, and urokinase plasminogen activator (uPA) were obtained from Calbiochem (La Jolla, CA), and the uPA receptor (uPAR) antibody was from R&D Systems. Antibodies and phospho-specific antibodies against KDR/Flk-1, FAK, p38 MAPK, ERK 1/2 and retinoblastoma (Rb) were obtained from Cell Signaling Technology (Beverly, MA), while antibodies against cyclin D1, cyclin E, cyclin dependent kinase 4 (CDK4), and p21WAF1/CIP1 were from Santa Cruz Biotechnology (Santa Cruz, CA). Matrigel was acquired from Becton Dickinson (Bedford, MA). A stock solution of 100 mM emodin was prepared in DMSO and maintained at –20°C. For in vitro use, the stock solution was diluted to an appropriate concentration with DMSO (Sigma, St. Louis, MO).

Cell culture

HUVECs (obtained from ATCC, Rockville, MA) were cultured in gelatin-coated plates with M199 medium containing 20% fetal bovine serum (FBS), sodium heparin (100 units/ml, Sigma), endothelial cell growth supplement (50 μg/ml, Sigma) and antibiotics, and incubated at 37°C in 5% CO2 in air. Human dermal microvascular endothelial cells (HDMECs) (purchased from Clonetics, Walkersville, MD) were cultured in gelatin-coated plates with EBM-2 medium (Clonetics) containing 5% FBS and antibiotics, and incubated at 37°C in 5% CO2 in air.

Endothelial cell proliferation assay

HUVECs growing in gelatin-coated plates were dispersed in 0.05% trypsin solution and resuspended with M199 containing 20% FBS. Approximately 3,000 cells were added to each well of the gelatinized 96-well plates and incubated at 37°C under 5% CO2 for 24 hr. The medium was replaced with 0.1 ml of fresh M199 containing 2% FBS, and emodin (at concentrations of 5, 10 or 25 μM) were added to each well. After 1 hr of incubation, the medium was added to a final volume of 0.2 ml M199 containing 2% FBS with or without VEGF-A (10 ng/ml) and bFGF (10 ng/ml). After 2 days of incubation, cells were washed twice with PBS, and the DNA amount was measured with CyQUANT GR reagent, according to the manufacturer's protocol (CyQUANT® Cell Proliferation Assay Kit, Molecular Probes, Eugene, OR), using a fluorescence spectrometer equipped a microplate reader (Molecular Device, Sunnyvale, CA).

Cell cycle analysis

To determine the DNA content, HUVECs were grown in 60 mm culture dishes either in the presence or absence of VEGF-A (10 ng/ml), and treated with emodin (at concentrations of 5, 10, 25, or 75 μM) for 20 hr or left untreated. Cells were trypsinized, resuspended in PBS and stained with propidium iodide using CycleTest™ Plus (Becton Dickinson, San Jose, CA), as described in the manufacturer's protocol. The DNA content was determined using a FACStar flow cytometer (Becton Dickinson).

In vitro endothelial cell wound healing repair assay

To determine the effects of emodin on HUVEC migration, in vitro wound healing repair assay was performed. HUVECs were cultured on a 1% gelatin-coated 12-well plate. At confluence, monolayers were wounded with a 200–1,000 μl micropipette tip, washed with serum-free medium and incubated for 24 hr in serum-free medium with or without VEGF-A (10 ng/ml) or bFGF (10 ng/ml) in the presence or absence of emodin (at concentrations of 5, 10 or 25 μM). Next, cells were washed twice with PBS, and fixed with Diff-Quick (Fisher Scientific, Pittsburgh, PA) solution. Migration was quantified in duplicate cultures by counting the number of cells that migrated from the wound edge into the denuded area for a distance of 1 cm.

Endothelial cell invasion assay

Invasion assays were performed using modified Boyden chambers with a polycarbonate Nucleopore membrane (Corning, Corning, NY). Precoated filters [6.5 mm in diameter; 8 μm pore size; Matrigel (BD Bioscience, Bedford, MD), 100 μg/cm2] were rehydrated with 100 μl of serum-free M199 medium containing 0.1% BSA. The same medium containing VEGF-A (10 ng/ml) was pipetted into the lower wells. Cells were trypsinized and suspended at a final concentration of 1 × 106 cells/ml in the above medium. Next, cells were pretreated with various concentrations of emodin for 30 min at room temperature before seeding, and 100 μl of the cell suspension was loaded into each of the upper wells. The chamber was incubated at 37°C under 10% CO2 for 12 hr. Non-invading cells were removed from the upper side of the filters with a cotton ball, and the lower side was stained with Diff-Quik solution. Invasiveness was determined by counting the cells in 5 microscopic fields per well. The extent of invasion was expressed as the average number of cells per microscopic field.

Endothelial cell tube formation assay

For the endothelial cell tube formation assay, 250 μl of growth factor-reduced Matrigel (Collaborative Biomedical Products, Bedford, MA) was pipetted into a well of a 48-well plate and polymerized for 30 min at 37°C. HUVECs and HDMECs incubated in 1% FBS-containing M199 and EGM-2, respectively, for 12 hr were harvested after trypsin treatment and suspended in 1% FBS containing M199 and EGM-2. Cells treated with various concentrations of emodin were incubated for 30 min at room temperature before seeding, followed by the addition of VEGF-A (10 ng/ml). After 18 hr, tube formation in each well was monitored, and photos were taken at 18 hr using an inverted microscope. The tubular length of the cells was measured using Image-Pro Plus (Media Cybermetics, Silver Spring, MD).

Gelatin Zymography

HUVECs in subconfluent culture (70–80% cell density of confluent culture) were washed with serum-free M199 and incubated with or without VEGF-A containing various concentrations of emodin for 20 hr. The activities and molecular weights of electrophoretically separated gelatinolytic enzymes in conditioned HUVEC medium were determined by SDS-PAGE. For each sample, 20 μl of serum-free culture medium was prepared in nondenaturing loading buffer (0.5 M Tris-HCl, pH 6.8, 10% SDS, 0.1% bromophenol blue and 10% glycerol) and size-fractionated on a 10% SDS-polyacrylamide gel impregnated with 0.1% gelatin. Next, gels were washed with 2.5% Triton X-100 for 1 hr at room temperature to remove SDS, rinsed twice with water and incubated in developing buffer [50 mM Tris-HCl buffer (pH 7.4), 20 mM NaCl, 10 mM CaCl2 and 0.1 NaN3] for 18 hr at 37°C. Gels were fixed, stained with 10% 2-propanol and 10% acetic acid containing 0.5% Coomassie Blue R250 and destained in the same solution without Coomassie Blue R250. Gelatinase activity was visualized as clear bands within the stained gel.

In vivo Matrigel plug assay

Specific pathogen-free, 6-week-old male C57BL/6 mice (Charles River, Tokyo, Japan) were employed for these experiments. For the Matrigel plug assay,24 mice were injected subcutaneously with 0.5 ml of Matrigel containing heparin (10 U/ml), 100 ng VEGF-A or VEGF-A plus 50 μM emodin. The injected Matrigel rapidly formed a single solid gel plug. After 7 days, mouse skin was easily pulled back to expose the Matrigel plug, which remained intact. To quantify the formation of functional neovessels in Matrigel, the amount of hemoglobin in each plug was assayed according to the manufacturer's protocol (Drabkin reagent kit 525, Sigma-Aldrich, Louis, MO).

In vivo CAM assay

For the CAM assay,25 fertilized chick embryos were preincubated for 9 days at 8°C in 70% humidity. A hole was drilled over the air sac at the end of the eggs and an avascular zone was identified on the CAM. A 1 × 1 cm window in the shell was sectioned to expose the CAM. Thermanox discs were sterilized and loaded with control buffer or 50 μM emodin. After air-drying under a laminar flow hood, discs were applied to the CAM surface. Windows were sealed with clear tapes and eggs were incubated for 60 hr. Blood vessels were viewed and photographed.

Immunoprecipitation

HUVECs were incubated for 6 hr in M199 containing 1% FBS. Cells were stimulated by the addition of VEGF-A (10 ng/ml) in the presence or absence of various concentrations of emodin (at concentrations of 5, 10 or 25 μM). Next, cells were incubated in 1 ml of lysis buffer [20 mM Tris/HCl (pH 8.0), 2 mM EDTA, 137 mM NaCl, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 10% glycerol and 1% Triton X-100]. Lysates were clarified by centrifugation at 15,000g for 10 min, and the resulting supernatant fractions were immunoprecipitated with 1 μg/ml anti-KDR/Flk-1 antibody at 4°C for 1 hr. Immunoprecipitates were washed 3 times with lysis buffer, solubilized in SDS-PAGE sample buffer containing β-mercaptoethanol and analyzed by Western blotting.

Western blotting

HUVECs were lysed in lysis buffer. After brief sonication, lysates were clarified by centrifugation at 12,000g for 15 min at 4°C, and the protein content in the supernatant was measured using Bradford's method. Aliquots (30–50 μg protein per lane) of the total protein or immunoprecipitates were resolved by 10 or 12% SDS-PAGE and blotted onto a nitrocellulose transfer membrane (0.2 μm) (Amersham, Arlington Heights, IL). The membrane was blocked with 5% nonfat skimmed milk in TBST (20 mM Tris-HCl, pH 7.6, 137 mM NaCl and 0.01% Tween-20) for 1 hr at room temperature, followed by incubation with specific primary antibodies. After extensive washing with TBST, the membrane was reprobed with horseradish peroxidase-linked antimouse or antirabbit immunoglobulin (1:3,000 dilution) in 5% nonfat skimmed milk in TBST for 40 min at room temperature. Immunoblots were visualized by enhanced chemiluminescence (Amersham), according to the manufacturer's protocol.

To measure the levels of secreted uPA, MMP-2 and -9 proteins, conditioned medium from each sample was subjected to protein analysis. For this purpose, culture medium in each tissue culture dish was collected and concentrated 10-fold using a Centricon 10 microconcentrator (Amicon, Beverly, MA). Concentrated samples were subjected to SDS-PAGE analysis.

Immunohistochemistry

HUVECs were grown directly on glass slides with M199 medium containing 20% FBS. For routine immunocytochemistry, slides were rinsed 3 times in PBS and fixed in 4% formaldehyde. Immunostaining was performed with the DAB method using the Cap-Plus™ DAB system (Zymed Laboratories, San Francisco, CA), as described by the manufacturer.

Statistical analysis

The data are presented as means ± SE. Statistical comparisons between groups were performed using one-way ANOVA, followed by the Student's t test.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Emodin inhibits VEGF-A and bFGF-induced proliferation of HUVECs

Angiogenesis involves the local proliferation of endothelial cells. To determine whether emodin affects VEGF-A- and bFGF-induced proliferation, HUVECs were incubated with various concentrations of the compound for 1 hr prior to the addition of VEGF-A and bFGF (10 ng/ml, respectively). As shown in Figures 1a and 1b, emodin significantly inhibited the VEGF-A and bFGF-induced proliferation of HUVECs in a dose-dependent manner. No distinct cellular, morphological changes typically associated with apoptosis (such as cell detachment, rounding or chromosomal fragmentation) were detected after 2 days of incubation with emodin at a concentration below 25 μM. However, significant cell death was observed at emodin concentrations of above 50 μM (data not shown). Accordingly, we employed the noncytotoxic concentration of emodin (below 25 μM) and focused on the effect of emodin on VEGF-A-induced angiogenesis in subsequent experiments.

thumbnail image

Figure 1. Antiproliferative activity of emodin. HUVECs were pretreated with 5, 10 and 25 μM emodin for 1 hr. Cells were stimulated with 10 ng/ml of VEGF-A (a) or bFGF (b) and allowed to proliferate for 2 days. The DNA amount was measured with CyQUANT GR reagent. Bars represent the mean ± standard deviation (SD) from five independent experiments. *, p < 0.05 vs control and **, p < 0.05 vs VEGF-A alone.

Download figure to PowerPoint

Emodin arrests VEGF-A-stimulated HUVECs in G0/G1 phase of cell cycle through down-regulation of cyclin D1 and E.

The next step was to determine whether the anti-proliferation activity of emodin involves cell cycle regulation. The effect of emodin on the cell cycle progression of VEGF-A-stimulated HUVECs was investigated using FACS. Treatment of HUVECs with VEGF-A (10 ng/ml) for 20 hr triggered the transition of cells from the G0/G1 phase to S phase (Fig. 2a). In contrast, pretreatment of cells with emodin significantly arrested to G0/G1 phase and reduced S phase entry in a dose-dependent manner (Fig. 2a). Emodin also induced G2/M phase arrest in HUVECs at 75 μM concentration, which might induce cell death (data not shown). These results indicate that emodin specifically induces cell cycle arrest of VEGF-A-stimulated HUVECs at the G0/G1 phase at nontoxic dose, which may be responsible for the inhibition of VEGF-A-stimulated HUVEC proliferation. The transition of cells from the G0/G1 to S phase is mainly regulated by cyclin D1.26 To determine whether the anti-proliferative effects of emodin are mediated through cyclin D1 regulation, HUVECs were treated with VEGF-A in the presence or absence of various concentrations of emodin for 20 hr, and cyclin D1 expression was examined. Western blot analysis revealed that emodin suppressed the VEGF-A-induced expression of cyclin D1 as well as cyclin E, which is also involved in G0/G1 to S phase entry, in a dose-dependent manner (Fig. 2b). We next examined the phosphorylation state of Rb in endothelial cells, which reflects the ability of a cell to exit the G0/G1 phase. Emodin significantly inhibited the phosphorylation of Rb induced by VEGF-A (Fig. 2b), but had no and marginal effect on the expression of the CDK inhibitor, p21WAF1/CIP1 and CDK4, respectively. Thus, emodin specifically induces the cell cycle arrest of VEGF-A-stimulated HUVECs at the G0/G1 phase through downregulation of cyclin D1 and E expression and blocking Rb phosphorylation, which may be responsible for the inhibition of HUVEC proliferation.

thumbnail image

Figure 2. Emodin induces G0/G1 arrest of HUVECs through downregulation of cyclin D1 and E. (a) HUVECs were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr before exposure to VEGF-A (10 ng/ml). After 20 hr, cell cycle analysis was performed. (b) HUVECs were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr and then stimulated with VEGF-A (10 ng/ml) for 20 hr. Cells were harvested and lysed, and Western blotting was performed. All experiments were done in triplicate. Bars represent the mean ± SD. *, p < 0.05 vs control and **, p < 0.05 vs VEGF-A alone.

Download figure to PowerPoint

Emodin suppresses VEGF-A-induced migration and invasion of HUVECs

The migration and invasion of endothelial cells through the basement membrane are crucial steps in the establishment of new blood vessels.27, 28, 29 To determine the effects of emodin on in vitro endothelial cell migration stimulated by VEGF-A and bFGF, confluent monolayers of HUVECs were scraped to remove a section of monolayer and cultured for 24 hr with control buffer, VEGF-A (10 ng/ml) or bFGF (10 ng/ml), or VEGF-A or bFGF plus various concentrations of emodin. Emodin significantly inhibited the migration of HUVECs in a dose-dependent manner (Figs. 3a and 3b, and supplement 1). Next, the effect of emodin on the invasion of HUVECs was examined using the Boyden chamber assay, which measured the ability of cells to pass through a layer of extracellular matrix on a Matrigel-coated filter. As shown in Figures 4a and 4b, VEGF-A significantly induced the invasiveness of HUVECs. However, this stimulation was blocked in a dose-dependent manner in cells pretreated with various concentrations of emodin for 1 hr before the addition of VEGF-A to the lower chamber. Since a crucial step of invasion is the degradation of extracellular matrix components, which allows cells to efficiently traverse the basement membrane,27, 28, 29 we examined protease secretion using zymography and immunoblotting. Gelatin zymography of serum-free conditioned medium revealed that HUVECs constitutively secreted MMP-2. VEGF-A treatment increased the level of protease secretion slightly, and 5 μM emodin inhibited MMP-2 secretion by these cells (Figs. 5a and 5b). However, no MMP-9 secretion was evident in HUVEC culture supernatant fractions. The major role of the uPA system is tumor progression in the stromal compartment, particularly neovascularization. Accordingly, we examined the effects of emodin on uPA and uPAR expression in HUVECs induced by VEGF-A. As depicted in Figure 5b, VEGF-A increased uPA and uPAR expression. Moreover, emodin effectively inhibited uPAR expression, but not uPA secretion. Our data collectively show that emodin effectively inhibits the migration and invasion of HUVECs stimulated by VEGF-A, and that the suppression of VEGF-A-induced secretion of MMP-2 and expression of uPAR is a result of the anti-invasive activity of emodin.

thumbnail image

Figure 3. Emodin inhibits VEGF-A-induced migration of HUVECs. Cells were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr. Subsequently, cells were scratched and incubated in serum-free media in the absence (Control) or presence of VEGF-A (10 ng/ml). After 24 hr, microphotographs were taken (×40). (a) Representative photographs of endothelial migration. (b) Migration was quantified in duplicate cultures by counting the number of cells that translocated from the wound edge into the denuded area for a distance of 1 cm. All Expeiments were done in tripicates. Bars represent the mean ± SD. *, p < 0.05 vs control and **, p < 0.05 vs VEGF-A alone.

Download figure to PowerPoint

thumbnail image

Figure 4. Emodin inhibits VEGF-A-induced invasion of HUVECs. Cells were pretreated for 1 hr with or without various concentrations (5, 10 and 25 μM) of emodin in the absence (control) or presence of VEGF-A (10 ng/ml). (a) Photomicrographs of HUVECs invading under the membrane after 12 hr. (b) Invasiveness were determined by counting cells in 4 microscopic fields per sample. All experiments were done in triplicate. Bars represent the mean ± SD. *, p < 0.05 vs control and **, p < 0.05 vs VEGF-A alone.

Download figure to PowerPoint

thumbnail image

Figure 5. Emodin inhibits the secretion and expression of MMP-2 and uPAR, respectively, induced by VEGF-A (10 ng/ml). HUVECs were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr before exposure to VEGF-A (10 ng/ml). After 20 hr, the conditioned media and cells were collected to perform zymography (a) and Western blotting (b). All experiments were done in triplicate.

Download figure to PowerPoint

Emodin inhibits tube formation of HUVECs and HDMECs induced by VEGF-A

Since the maturation of angiogenesis is characterized by the formation of tubular structures from capillary endothelial cells, we performed a tube formation assay to investigate the effect of emodin on the capillary-like structure formation of HUVECs and HDMECs. Following stimulation by VEGF-A on the Matrigel, HUVECs became aligned into cords, and the tube-like structure formation was maximal within 18 hr (Fig. 6a). Treatment of cells with various concentrations of emodin resulted in significant inhibition of VEGF-A-stimulated tube formation (Fig. 6a). Also, emodin effectively suppressed tube formation of HDMECs on Matrigel dose-dependently indicating that inhibition of tube formation was not restricted to HUVECs (Fig. 6b). Our results clearly demonstrate that emodin is effective in controlling the VEGF-A-stimulated tube formation of endothelial cells in vitro.

thumbnail image

Figure 6. Emodin inhibits VEGF-A-induced tube formation of endothelial cells. HUVECs (a) and HDMECs (b) were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr. Cells were then collected and replated on Matrigel-coated plates at a density of 1 × 105 cells/well and incubated in the absence (Control) or presence of VEGF-A (10 ng/ml). After 18 hr, microphotographs were taken (×40). (Upper panel) Representative photographs of endothelial tube formation. (Lower panel) The area covered by the tube network was quantified using Image-Pro Plus software. All experiments were done in triplicate. Bars represent the mean ± SD. *, p < 0.05 vs control and **, P < 0.05 vs VEGF-A alone.

Download figure to PowerPoint

Emodin inhibits in vivo angiogenesis

To determine whether emodin is capable of suppressing blood vessel formation in vivo, we employed an in vivo angiogenesis model, mouse Matrigel plug assay and chick CAM assay. Matrigel containing VEGF-A (100 ng) with or without emodin (50 μM) was injected into C57BL/6 mice, and the Matrigel plugs formed in mice were excised and photographed 7 days later. Matrigel containing VEGF-A appeared dark red (Fig. 7a). However, Matrigel containing both VEGF-A and emodin displayed pale yellow, indicating less blood vessel formation compared to VEGF-A alone (Fig. 7a). The vessels were filled abundantly with intact red blood cells, indicating the formation of functional vasculature inside the Matrigel and blood circulation in newly formed vessels by angiogenesis. Accordingly, we measured the hemoglobin content in each Matrigel plug. The hemoglobin content of Matrigel containing VEGF-A was ∼2.8-fold higher than that of the control. A combination of emodin (50 μM) and VEGF-A markedly suppressed hemoglobin accumulation (Fig. 7b). We additionally performed the chick CAM assay to examine the effect of emodin on another in vivo angiogenesis model. As shown in Figure 7c, a themanox coverslip containing emodin effectively inhibited the formation of capillary vessels in CAM, compared with the coverslip containing DMSO alone, which had no visible effect on the preexisting blood vessels. The results indicate that emodin is capable of inhibiting neovessel formation in vivo in both natural and VEGF-A-induced conditions.

thumbnail image

Figure 7. Emodin inhibits VEGF-A-induced angiogenesis in vivo. C57BL6 mice were injected with 0.5 ml of Matrigel containing VEGF-A (100 ng) and emodin (50 μM). After 7 days, mice were killed and Matrigel plugs excised. (a) Representative Matrigel plugs that contained no VEGF-A (control), VEGF-A alone or VEGF-A plus emodin (50 μM) were photographed. (b) Quantification of neovessel formation by measurement of hemoglobin in the Matrigel. (c) Representative photographs of chick CAM assays. A themanox cover slip with or without emodin entrapped in type I collagen gel was loaded on chick CAMs. After 72 hr incubation, fat emulsion was injected under the CAMs for better visualization of the vessels. Circles show the location of the themanox cover slip. All experiments were done 5 times.

Download figure to PowerPoint

Emodin suppresses VEGF-A-induced tyrosine phosphorylation of VEGF-A receptor-2 (KDR/Flk-1) in HUVECs

VEGF-A induces endothelial cell proliferation, migration and survival through activation of its cell surface receptor, KDR/Flk-1.30 Phosphorylation in tyrosine residues is the earliest event occurring after VEGF-A-dependent KDR/Flk-1 dimerization.31 To evaluate the molecular mechanism by which emodin inhibits VEGF-A-induced angiogenesis, we investigated its effects on VEGF-A-induced KDR/Flk-1 phosphorylation. Following stimulation of HUVECs with VEGF-A (10 ng/ml) for 10 min, KDR/Flk-1 was strongly tyrosine phosphorylated. However, preincubation of HUVECs with various concentrations of emodin before VEGF-A stimulation led to dramatic suppression of KDR/Flk-1 tyrosine phosphorylation (Fig. 8a). We additionally examined the effects of emodin on VEGF-A-induced KDR/Flk-1 phosphorylation by immunohistochemistry, using phosphorylated KDR/Flk-1 antibody. VEGF-A-treated HUVECs were immunostained with antibodies against KDR/Flk-1, whereas VEGF-A together with various concentrations of emodin exhibited negligible staining (Fig. 8b), confirming that emodin inhibits VEGF-A-induced KDR/Flk-1 activation in endothelial cells.

thumbnail image

Figure 8. Emodin inhibits VEGF-A-induced phosphorylation of KDR/Flk-1. HUVECs were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr and stimulated with 10 ng/ml VEGF-A for 10 min. (a) Anti-KDR/Flk-1 immunoprecipitates were analyzed by SDS-PAGE. Immunoblot analysis was performed with antiphosphotyrosine antibody (anti-PY20). (b) Activation of KDR/Flk-1 by VEGF-A was determined by immunocytochemistry using antiphospho-KDR/Flk-1. All experiments were done at least in triplicate.

Download figure to PowerPoint

Emodin blocks VEGF-A-induced downstream signaling events

Following VEGF-A binding to KDR/Flk-1 in endothelial cells, several intracellular kinases are activated.30 To further evaluate the effects of emodin on downstream signal molecules of KDR/Flk-1 triggered by VEGF-A, we examined the activation of ERK 1/2, p38 MAPK and p125 FAK. HUVECs were preincubated with various concentrations of emodin for 30 min before stimulation with VEGF-A (10 ng/ml) for 10 min. Emodin significantly inhibited VEGF-A-induced p38 MAPK, p125 FAK and ERK1/2 activation in a dose-dependent manner (Fig. 9). It is well established that the Akt/NO pathway is important for VEGF-A-induced endothelial cell migration, proliferation and tube formation in vitro.32, 33 Specifically, VEGF-A stimulates the phosphorylation of Akt (Ser-473) and eNOS (Ser-1177), which play a key role in angiogenesis.17, 34 Emodin significantly inhibited VEGF-A-induced Akt and eNOS phosphorylation in a dose-dependent manner (Fig. 9). From the results, it is evident that emodin inhibits VEGF-A-induced activation of p38 MAPK, p125 FAK, pERK 1/2, Akt and eNOS, which are downstream signal molecules of KDR/Flk-1.

thumbnail image

Figure 9. Emodin inhibits VEGF-A-induced phosphorylation of downstream effector molecules, including ERK1/2, p38 MAPK, Akt, FAK and eNOS. HUVECs were pretreated with various concentrations (5, 10 and 25 μM) of emodin for 1 hr and stimulated with 10 ng/ml VEGF-A for 10 min. Activation of ERK 1/2, p38 MAPK, Akt, FAK and eNOS by VEGF-A was determined by Western blotting analysis using phospho-specific antibodies. All experiments were done at least in triplicate.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Emodin is an active component isolated from the rhizome of Rheum palmatum, a widely used traditional Chinese herb,35, 36 and is conventionally used as Chinese medicine. Recent studies have disclosed a number of pharmacological and toxicological properties of emodin, which are applicable for cancer treatment in humans. Antitumor effects of emodin or aloe-emodin in lung, breast and neuroectodermal cancer cells have been established.20, 22, 37 In view of the findings that emodin possesses anticancer activity and abnormal angiogenesis is an important prerequisite for uncontrolled cell growth in cancer, we investigated the effects of emodin on angiogenic processes, both in vivo and in vitro. Recently, Wang et al. reported that emodin inhibited angiogenesis by inducing apoptosis and G2/M phase arrest of cell cycle in cultured bovine aortic endothelial cells.38 However, critical molecular mechanisms of anti-angiogenic activity of emodin on the VEGF-A-induced angiogenesis were not demonstrated. We show in the present study that emodin exerts a potent anti-angiogenic effect in vivo, as verified from its inhibition of VEGF-A-induced neovascularization of Matrigel plug implanted in mice and naturally induced neovascularization of capillary vessel formation of CAM. Emodin also affects VEGF-A-stimulated endothelial cell functions relevant to the angiogenic process in vitro, possibly by blocking phosphorylation and consequently by the activation of the VEGF-A receptor-2, KDR/Flk-1.

We speculate whether the anti-angiogenic effect of emodin is specific to VEGF-A/KDR/Flk-1 system and HUVECs. To verify the possibility, we first examined the effect of emodin on endothelial cell proliferation and migration with another potent angiogenic factor, bFGF, and found that emodin also suppressed bFGF-induced proliferation (Fig. 1b) and migration (supplement 1) of HUVECs. In addition, emodin effectively inhibited the tube formation of HDMECs on Matrigel induced by VEGF-A (Fig. 6b). These results indicated that anti-angiogenic activity of emodin is not restricted to specific ligand and receptor system or specific endothelial cells.

A critical step in angiogenesis involves the local proliferation of endothelial cells.39 VEGF-A mediates the mitogenic activities of HUVECs almost 2-fold above the basal level. Emodin effectively inhibits proliferation of HUVECs in a concentration-dependent manner. Since emodin does not have a cytotoxic effect up to a concentration of 25 μM, the antiproliferative effect may not be due to apoptosis or necrosis of endothelial cells, but rather to the inhibition of cell cycle progression induced by VEGF-A. To confirm this theory, we examined the effect of emodin on the cell cycle progression of HUVECs stimulated by VEGF-A. Emodin arrested cells at the G0/G1 phase of the cell cycle, possibly through either inhibition of cyclin D1 expression or induction of a CDK inhibitor, p21WAF1/CIP1. Our results clearly indicate that emodin suppresses VEGF-A-induced cyclin D1 protein synthesis in HUVECs, but has a marginal effect on p21WAF1/CIP1 expression. Cyclin D1 is a critical mediator of the progression of endothelial cells through the G1/S phase of the cell cycle, forming a cyclin D1-CDK4 complex that phosphorylates and inactivates the RB protein, thereby releasing E2F to mediate the G1 to S transition.26 In our experiments, emodin inhibited VEGF-A-induced Rb phosphorylation in HUVECs. Consistent with the present results, earlier studies show that various anti-angiogenic molecules, such as endostatin, capsaicin and curcumin, inhibit Rb phosphorylation and DNA synthesis of endothelial cells through downregulation of cyclin D1.40, 41 On the contrary, treatment of emodin in bovine aortic endothelial cells induced G2/M phase arrest of cell cycle.38 Our experiment also showed G2/M phase arrest by emodin in HUVECs; however, its concentration was higher enough for inducing cell death. Therefore, it is possible that the antiproliferative effect of emodin on VEGF-A-stimulated HUVECs is accomplished by arresting cells at the G0/G1 phase of the cell cycle through the suppression of cyclin D1 expression and Rb phosphorylation.

Chemotaxis, local invasion and morphogenesis developed by migration, spreading and mutual alignment of endothelial cells are also critical steps for new vessel sprouting.39, 42 MMPs and uPA are clearly implicated in the initiation of angiogenesis,43 and endothelial cell differentiation and spreading during angiogenesis. These proteins selectively degrade stroma and extracellular basement proteins. In particular, the release of MMP-2 by endothelial cells represents an important step in neovascularization, because this major extracelluar matrix proteolytic enzyme is secreted when endothelial sprouting takes place, thus enhancing cell migration across the extracellular matrix and tube-like structure formation.44 During migration, VEGF-A induces pro-uPA activation via KDR/Flk-1, and uPAR internalization leads to increased fibrinolytic activity in endothelial cells.45 Furthermore, increased uPAR expression by hypoxia is enhanced in migrating endothelial cells in vitro,46 and the finding that neovascularization of transplanted tumors in vivo is inhibited by uPAR antagonists47, 48 suggests that hypoxia-stimulated uPAR expression plays an essential role in tumor metastasis as well as local invasion. In our experiments, HUVECs stimulated with VEGF-A displayed elevated levels of uPA and uPAR proteins, and emodin inhibited the VEGF-A-induced expression of uPAR, but not of uPA. Emodin additionally suppressed the MMP-2 secretion at both basal and VEGF-A-induced level by these cells. Our findings strongly indicate that the anti-angiogenic effect of emodin on migration, invasion and tube formation of HUVECs stimulated by VEGF-A is at least partly due to the suppression of MMP-2 secretion and uPAR expression.

Emodin was first isolated as a tyrosine kinase inhibitor from Polygonum cuspidatum.49 Emodin was reported to an inhibitor of the protein tyrosine kinase, p56lck, and additionally blocks the activity of p185neu receptor tyrosine kinase, which is critical for the growth of breast cancer cells that overexpress HER-2/neu.22 Therefore, it might be possible that emodin inhibits tyrosine kinase activity of KDR/Flk-1 by blocking VEGF-A-stimulated autophosphorylation. Consequently, we investigated whether the mechanism of anti-angiogenic effect of emodin involves the inhibition of receptor tyrosine kinase activity. As expected, tyrosine phosphorylation of KDR/Flk-1 (which plays an important role in proliferation, migration and differentiation of endothelial cells in response to VEGF-A)12, 13 was dose-dependently inhibited by emodin (Figs. 8a and 8b). Recently, emodin was found to inhibit casein kinase II (CK2) more potently than protein tyrosine kinases, as judged from lower Ki values.50 Emodin penetrates to the active site of CK2 and replaces ATP.51 Therefore, it also would be possible that emodin binds to active site of KDR/Flk-1 and blocks autophosphorylation induced by VEGF-A. In addition, emodin was capable of blocking downstream events of VEGF-A-induced KDR/Flk-1 signaling, such as activation of p38 MAPK, ERK1/2 and p125 FAK, which are requisites for angiogenic activity in HUVECs (Fig. 9). Emodin blocked VEGF-A-induced phosphorylation of AKT (Ser-473) and eNOS (Ser-1177), which are involved in the critical pathway of endothelial cell migration, proliferation and tube formation in vitro.32, 52 Although we cannot rule out the possibility that an unidentified tyrosine kinase is additionally inhibited by emodin, we strongly believe that suppression of KDR/Flk-1 is the major mechanism accounting for the anti-angiogenesis effect of emodin on the VEGF-A-induced angiogenesis. Emodin effectively blocks the in vivo angiogenesis of mouse-implanted VEGF-A containing Matrigel, consistent with its potent anti-angiogenic activity observed in vitro.

VEGF-A stimulation of endothelial cells is initiated by the interaction with its cognate receptor, VEGFR, a member of the receptor tyrosine kinase (RTK) family.31 RTK activation is evidenced by autophosphorylation of specific tyrosine residues within the homodimer subunits and this autophosphorylation/activation can be reversed by protein tyrosine phosphatase activities.53 HCPTPA, a low molecular weight protein tyrosine phosphatase, associated with VEGFR2 (KDR/Flk-1), which is a substrate for HCPTPA, results in regulating KDR/Flk-1 activation, VEGF-A-mediated signal transduction and VEGF-A-mediated proliferation and chemotaxis negatively.54 In addition, some SH2-containing phosphatases also have been shown to be involved in down-regulating KDR/Flk-1 activation.55, 56 Therefore, it is possible that blocking of VEGF-A-mediated KDR/Flk-1 activation by emodin is through the induction of phosphatases that affect KDR/Flk-1 phosphorylation by VEGF-A.

In conclusion, emodin, a natural anthraquinone, preferentially inhibits VEGF-A-induced angiogenesis in vivo and in vitro, possibly through blocking the phosphorylation of KDR/Flk-1 and downstream effector molecules. We propose that emodin is a potential anti-angiogenic agent for the effective treatment of various diseases, including cancer.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

This article contains supplementary material available via the Internet at http://www.interscience.wiley.com/jpages/0020-7136/suppmat .

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.