Arch. Pharm. Res. DOI 10.1007/s12272-014-0340-6
RESEARCH ARTICLE
Ginsenoside compound K inhibits angiogenesis via regulation of sphingosine kinase-1 in human umbilical vein endothelial cells Kyong-Oh Shin • Cho-Hee Seo • Hyo-Hyun Cho Seikwan Oh • Seon-Pyo Hong • Hwan-Soo Yoo • Jin-Tae Hong • Ki-Wan Oh • Yong-Moon Lee
•
Received: 10 November 2013 / Accepted: 16 January 2014 Ó The Pharmaceutical Society of Korea 2014
Abstract Ginsenoside compound K (CK) is a metabolite of the protopanaxadiol-type saponins of Panax ginseng C.A. Meyer (Araliaceae), has long been used to treat against the development of cancer, inflammation, allergies, and diabetes. This study examined the anti-angiogenic properties of CK against sphingosine 1-phosphate (S1P)-induced cell migration via regulation of sphingosine kinase 1 (SPHK1) in human umbilical vein endothelial cells (HUVEC). Studies on S1P-induced cell migration, expression of SPHK1 and MMPs and analysis of sphingolipid metabolites by LC–MS/MS were examined after the treatment of CK (2.5, 5, 10 lg/mL) in HUVEC. S1P produced by SPHK1 is also involved in cell growth, migration, and protection of apoptosis; therefore, we sought to investigate whether ginsenosides are able to regulate SPHK1. For this purpose, we developed an inhibitory assay of SPHK1 activity and an analytical method for detection of S1P and other sphingolipid metabolites in HUVEC. Ginsenoside CK inhibited 100 nM S1P-induced cell migrations in a dose-dependent manner. Among tested ginsenosides, CK exclusively inhibited S1P production, SPHK1 activity and SPHK1 expression in HUVEC, whereas expression of the pro-apoptotic sphingolipids, sphingosine and ceramide, was increased in response to CK. The major K.-O. Shin C.-H. Seo H.-H. Cho H.-S. Yoo J.-T. Hong K.-W. Oh Y.-M. Lee (&) College of Pharmacy and MRC, Chungbuk National University, Cheongju 361-763, Korea e-mail: [email protected]; [email protected] S. Oh College of Medicine, Ehwa Women’s University, Seoul 158-710, Korea S.-P. Hong Department of Oriental Pharmaceutical Sciences, Kyung Hee University, Seoul 130-701, Korea
subspecies of the increased ceramide was C24:0-ceramide. CK also disrupted the sphingolipid rheostat, which ultimately influences cell fate, and dose-dependently inhibited HUVEC migration by reducing expression of metalloproteinases (MMPs). Ginsenoside CK acts as a unique HUVEC migration inhibitor by regulating MMP expression, as well as the activity of SPHK1 and its related sphingolipid metabolites. Keywords Panax ginseng Araliaceae Ginsenoside Compound K Sphingosine kinase Cell migration Human umbilical vein endothelial cells
Introduction Angiogenesis, proliferation of the endothelium and formation of new blood vessels, play important roles in metastasis and tumor growth, resulting in an increase in the size of solid tumors. Therefore, inhibition of angiogenic signaling is expected to be an efficient therapeutic approach for many tumor types. As drug targets indeed, vascular endothelium growth factor (VEGF) and its receptors, VEGFR-1, -2, and -3, constitute a key signaling system that regulate proliferation and migration of vascular endothelial cells. Ginsenoside Rg1 mediates hypoxia-independent upregulation of hypoxia inducible factor 1 alpha (HIF-1a) (Leung et al. 2011), and non-genomic crosstalk of the glucocorticoid receptor and fibroblast growth factor receptor-1, which promote angiogenesis (Cheung et al. 2011). Conversely, ginsenosides Rb2 and Rg3 inhibit tumor angiogenesis and metastasis (Sato et al. 1994; Xu et al. 2007). Ginsenoside compound K (CK), a major metabolite of ginsenoside Rb1 from Panax ginseng root that is generated by intestinal bacterial flora, is rapidly
123
K.-O. Shin et al.
absorbed from the gastrointestinal tract following oral administration and then slowly metabolized (Akao et al. 1998). CK inhibits interleukin-1 receptor-associated kinase 1 activation, the key step of inflammation (Joh et al. 2011), as well as lipopolysaccharide-induced biosynthesis of nitric oxide and prostaglandin E2 in RAW264.7 cells (Park et al. 2005). A hepatoprotective effect of CK, resulting from activation of AMP-activated protein kinase and subsequent attenuation of hepatic lipid accumulation, has also been reported (Lee et al. 2005; Kim Do et al. 2009). Sphingosine kinases (SPHKs) catalyze the phosphorylation of sphingosine into sphingosine 1-phosphate (S1P). Sphingosine kinase type 1 (SPHK1) has been shown to regulate several processes involved in cancer progression and SPHK1 is overexpressed in various types of cancers. Upregulation of SPHK1 is associated with tumor angiogenesis and resistance to radiation and chemotherapy. Many growth factors stimulate SPHK1 via their tyrosine kinase receptors, leading to a rapid increase in S1P production (Shida et al. 2008). S1P is a pro-angiogenic factor that mediates diverse biological processes, including endothelial cell growth, migration, invasion, angiogenesis, vascular maturation, and lymphocyte trafficking (Berdyshev et al. 2011). The newly synthesized S1P can release out of cellular membrane and act as an extracellular signaling factor, by binding to a family of five related G-protein-coupled S1P receptors (S1PRs), previously known as the endothelial differentiation genes. The amount of intracellular S1P available for signaling through various S1PRs is tightly regulated, primarily by the SPHK1, S1P lyase, and S1P phosphatase enzymes (Leong and Saba 2010). A complex interactive network of signaling cascades downstream from the G-protein-coupled S1PRs acts as a prime effector of angiogenesis that occurs in association with various pathologies (Kluk and Hla 2002). Recently, CK was reported effectively to disrupt basic fibroblast growth factor-induced neo-vascularization in Matrigel plugs excised from mice in vivo (Jeong et al. 2010). Nuclear export of NF-kappaB-p65 and reduction of metalloproteinases(MMP)2 and MMP9 expression are associated with metastatic inhibition induced by CK (Ming et al. 2011). In this study, we examined the effect of CK on SPHK1 activity and S1P levels in HUVEC. CK inhibited SPHK1 activity and impaired sphingolipid metabolic balance by reducing synthesis and extracellular release of S1P, resulting the inhibition of HUVEC migration.
Materials and methods Cell lines, chemicals and biochemicals HUVEC were purchased from Lonza (Basel, Switzerland). S1P, sphingosine, ceramides (fatty acid lengths C12, C16,
123
C18, C22, C24, and C24:1), 2S-amino-4E-heptadecene-1, 3R-diol (C17-sphingosine), C17-sphingosine-1-phosphate (C17-S1P), C17-sphinganine-1-phosphate (C17-DHS1P), and C17-ceramide (d18:1/C17:0) were obtained from Avanti Polar Lipid (Alabaster, AL, USA). Ginsenosides [CK (98.4 %), Rg1 (98.7 %), Rh1 (98.3 %), Rb1 (99.1 %), Rd (98.7 %), Re (99.3 %), Rf (98.6 %), and F1 (99.3 %)] were purchased from Chengdu Biopurify Phytochemical Company (Chengdu, China).GM6001 (95 %) and SKI II (98 %) were obtain from Sigma-aldrich (St. Louis, MO, USA). Endothelial cell growth medium-2 (EGM-2) was purchased from Lonza (Basel, Switzerland); fetal bovine serum (FBS) was purchased from Gibco-BRL (Grand Island, NY, USA). The monoclonal anti-human SPHK1 antibody (Cat. No. 3297) and polyclonal anti-human b-actin antibody (Cat. No. 4967) were purchased from Cell Signaling Technology (Danvers, MA, USA) and polyclonal antihuman-MMP2 antibody (Cat. No. SC-8830) and polyclonal anti-human-MMP9 antibody (Cat. No. SC-6840) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Organic solvents for sphingolipid extraction and HPLC analysis were purchased from Merck (Darmstadt, Germany). Unless otherwise stated, all other chemicals were obtained from Sigma (St Louis, MO, USA). Cell culture conditions HUVEC (passage 5–7) were grown on 1 % gelatin-coated culture plates in EGM-2 medium supplemented with 10 % FBS, 1 % antibiotics (100 U/mL penicillin, 100 lg/mL streptomycin), VEGF (1 ng/mL), bFGF (2 ng/mL), and heparin (10 U/mL). In vitro wound-healing assay An in vitro wound-healing assay was performed to measure unidirectional migration of HUVEC. HUVEC (1 9 105 cells/well) were seeded into 12-well plates coated with collagen, and incubated at 37 °C until they attached. The cells were then washed twice with phosphate-buffered saline and were incubated in EGM-2 containing 1 % FBS for 24 h at 37 °C. Monolayers of HUVEC were scratchwounded to a 1 mm depth in a straight line using a 200 lL pipette-tip, and then incubated for 24 h with S1P (100 nM), and either with or without GM6001 (25 lM) or various concentrations (1–10 lg/mL) of CK. To measure the number of migrated endothelial cells from the edge of the injured monolayer, images were photographed both immediately after wounding and after 24 h incubation, using a phase-contrast microscope (Olympus, Tokyo, Japan). At least four points in each of three fields were examined at random for two independent wounds.
Ginsenoside compound K inhibits angiogenesis
Measurement of SPHK1 activity HUVEC were seeded into 6-well plates at 1 9 106 cells/ well and incubated for 24 h to allow full attachment. The cells were then treated with each ginsenoside (10 lg/mL) dissolved in dimethyl sulfoxide (DMSO), for 24 h. For measurement of S1P released into the culture media, 1 lM sphingosine, which enhances S1P biosynthesis by SPHK1, was added to the culture media 30 min prior to cell harvesting. A specific inhibitor of SPHK1, SKI II was tested as a positive control. A total of 50 lg protein from cell lysate was incubated in a volume of 100 lL, containing 10 lL of 10 mM ATP in 200 mM MgCl2, and 10 lL of 200 lM C17-Sphingosine in 5 % Triton X-100. Incubations with cell lysates contained 5 mM NaF and Na3VO4, which were included as inhibitors of S1P phosphatase and lyase, respectively, to prevent potential degradation of C17-S1P. Reactions were incubated for 30 min at 37 °C and were terminated by addition of 0.8 mL CHCl3/MeOH/HCl (300:500:3, v/v/v) in cold ice. Following addition of 100 pmol C17-DHS1P as an internal standard, samples were vortexed vigorously for 10 min and then centrifuged at 20,0009g for 3 min. Next, 250 lL of CHCl3 and 250 lL of 1 M NaCl were added, and the mixture was vortexed vigorously for 10 min. Samples were centrifuged at 4 °C for 3 min at 20,0009g, and then the upper phase was removed. The organic phase was transferred to a fresh tube and stored on ice. C17-S1P was extracted by addition of 250 lL 1 M NaCl, 300 ll of MeOH, and 30 lL of 3 N NaOH. The tube was then vortexed for 10 min and centrifuged at 14,000 rpm for 4 min. The alkaline aqueous phase containing C17-S1P was mixed thoroughly with 100 lL of dephosphorylation reaction buffer (200 mM Tris–HCl (pH 7.4), 75 mM MgCl2 in 2 M glycine buffer, pH 9.0) and 10 units of alkaline phosphatase; the tube was then incubated at 37 °C for 90 min. The dephosphorylated C17-Sph was extracted twice with 500 lL of CHCl3 and 300 lL of CHCl3, and then washed three times with alkaline water (pH 10.0). The washed CHCl3 phase was transferred to a fresh tube and dried completely using a speed vacuum system (Vision Co., Seoul, Korea). S1P extraction A volume of 350 lL of MeOH, containing 0.6 lL of 35 % HCl, and 100 pmol of C17-S1P as an internal standard, was added to a 50 lg aliquot of cell lysate; the sample was then ultrasonicated for 5 min in ice-cold water. Lipids were extracted by Micromixer (Taitec, Tokyo, Japan) for 60 min at 750 rpm after adding 500 lL of CHCl3/1 M NaCl (1:1 v/v) and 35 lL of 3 N NaOH. The aqueous fractions
containing S1P were transferred to fresh tubes and then sphingosine was released from S1P using alkaline phosphatase. The aqueous phases obtained from the S1P extraction procedure were mixed thoroughly with 100 lL of reaction buffer (200 mM Tris–HCl (pH 7.4), 75 mM MgCl2 in 2 M glycine buffer, pH 9.0), containing 10 units of alkaline phosphatase, and were vortexed for 2 h at 37 °C. During this step, S1P efficiently released phosphate and was converted into sphingosine. Sphingosine was extracted twice with 500 lL CHCl3 and then dried completely using a speed vacuum system. HPLC analysis of S1P concentration and SPHK1 activity Residues were dissolved in 120 lL MeOH, incubated at 50 °C for 10 min, and then derivatized with 15 lL OPA reagent (50 mg ortho-phthalaldehyde, 1 mL EtOH, 100 lL 2-mercaptoethanol, 50 mL 3 % (w/v) boric acid solution). Following incubation of the tubes for 30 min in the dark at room temperature, 90 lL of each aliquot was injected into the HPLC system (Jasco, Tokyo, Japan), which consisted of a Model PU 980 pump, Nova-Pak C18 (4.6 mm i.d. 9 150 mm), and FP-920 fluorescence spectrophotometer (kex 340 nm, kem 455 nm) with autosampler. The isocratic mobile phase of 90 % MeOH was pumped at a flow rate of 1 mL/min. The resulting data was evaluated using the Borwin chromatographic system manager software (Jasco, Tokyo, Japan). LC–MS/MS analysis of ceramide species HUVEC were seeded into 6-well plates at 1 9 106 cells/ well. Following treatment with ginsenoside CK (10 lg/mL) for 24 h, the cells were collected as cell pellets. Total ceramide was extracted with the mixture of 50 lL cell lysates (100 lg), 100 pmol of internal standard C17-ceramide (d18:1/C17:0), and 2 mL of Chloroform/MeOH (2:1, v/v) solution. The low phase was collected after repeated extraction (final volume of 4 mL) and dried in a vacuum system. The dried residue was redissolved in 50 lL MeOH, and then 10 lL was injected into the LC–ESI–MS/ MS system. For optimization, a mixture of ceramide standards was infused directly into the mass spectrometer and all source parameters and ionization conditions were adjusted to improve the sensitivity of the assay. Extracted samples (10 lL) were injected into a HPLC (Agilent 1200 series, Agilent, CA, USA) and separated through a reverse phase KINETEX C18 column (2.1 9 50 mm, ID: 2.6 lm) (Phenomenex, St. Louis, MO, USA). The ceramides were resolved using a linear gradient from 8 % mobile phase A (water containing 0.2 % formic acid) at a flow rate of
123
K.-O. Shin et al.
0.3 mL/min for 1 min, to 100 % mobile phase B (MeOH containing 0.2 % formic acid) over 3 min, followed by 100 % mobile phase B for 12 min. The column was then equilibrated for 8 min with 92 % mobile phase B. The HPLC column effluent was introduced onto a Thermo LTQ-XL ESI-ION TRAP (Thermo Scientific, Fl, USA) and analyzed using electrospray ionization in positive mode. The configurations for mass spectrometry were: capillary voltage, 5.5 kV; cone voltage, 40 V; cone temperature, 300 °C. Analyses were performed using electrospray ionization in the positive-ion mode with multiple reaction monitoring to select both parent and characteristic daughter ions specific to each analyte simultaneously from a single injection. The MS/MS transitions (m/z) were 510 ? 264 for C14, 538 ? 264 for C16, 552 ? 264 for C17, 566 ? 264 for C18, 594 ? 264 for C20, 648 ? 264 for C24:1, and 650 ? 264 for C24. Data were acquired using X-calibur software.
A
Western blot analysis
BCA protein assay Cell lysate from the solubilized tissue was mixed with Pierce BCA reagents (Rockford, IL, USA), according to the manufacturer’s instructions, and incubated for 30 min. Protein content was quantified at 562 nm with an ELISA reader (Molecular Devices, Sunnyvale, CA, USA), based on a bovine serum albumin standard curve.
Statistical analysis All values are expressed as mean ± standard error of the mean. For single comparisons, differences between treatments were analyzed by unpaired Student’s t test. For multiple comparisons, one-way analysis of variance was
123
B Migrated cells density (fold increase)
SDS-PAGE was performed on HUVEC lysates using 10 % acrylamide gels. Proteins were transferred to PVDF membranes (Millipore Corporation, MA, USA) and blots were probed with anti-SPHK1 (1:1,000) anti-MMP2 (1:500), anti- MMP-9 (1:500), ß-actin (1:1,000), followed by incubation with horseradish peroxidase-conjugated mouse or rabbit immunoglobulin. Blots were then developed using West Pico Chemiluminescent Substrate western blotting detection reagent (Pierce, Woburn, MA, USA). The band intensity was determined by Quantity One 4.5.0 Image acquisition and Analysis software (BioRad).
12 10 8
** 6
** 4
***
***
2 0 DMSO
S1P GM6001 CK-1
CK-5
CK-10
+ 100nM S1P Fig. 1 HUVEC migration induced by 100 nM S1P in an in vitro wound-healing assay. (A) Migration was inhibited in a dosedependent manner by treatment of cells with CK at final concentrations of 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10).The inhibitory effect of 10 lg/mL CK was comparable to that of the positive control, 10 lM GM6001. Panel (B) shows the density of migrated cells, represented as fold change over the control group. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
Ginsenoside compound K inhibits angiogenesis b Fig. 2 The effects of each ginsenoside (10 lg/mL in DMSO) on S1P
A 350
S1P (pmol / mg protein)
300
*
** *
250
*
200
synthesis in HUVEC after pretreatment with a natural substrate, sphingosine (1 lM), for 30 min before cell harvesting (A), and SPHK1 activity, measured as the conversion rate of 2 nmol of C17sphingosine into C17-S1P at 37 °C during a 30 min incubation (B). Panel (C) shows the inhibition of SPHK1 activity by CK in a dosedependent manner. Control cells were treated with DMSO instead of CK. SKI II (1 lM) was used as a positive control. Data indicate the mean ? SEM for 3 samples. *P \ 0.05, **P \ 0.01
150 **
100
**
Results
50
Inhibitory effects of CK on HUVEC migration
0
I f SO SO KI I Rb1 CK F1 Rc Rd Re R Rg1 Rh1 Ro DM DM S
+ 1 µM sphingosine
B
1.6
C17-S1P (pmol / mg / min)
1.4 *
1.2 1.0 **
0.8 0.6 **
0.4 0.2 0.0
I f SO SO KI I Rb1 CK F1 Rc Rd Re R Rg1 Rh1 Ro DM DM S
+ 20 µM C 17-sphingosine
C17-S1P (pmol / mg / min)
C 1.2 1.0 **
0.8
**
0.6 **
0.4 0.2 0.0
SO MSO SKI II K-2.5 CK-5 CK-10 CK-20 D C
DM
+ 20 µM C 17-sphingosine
performed, followed by Newman-Keuls multiple comparison test. P values \ 0.05 were considered to be statistically significant.
The effect of CK on HUVEC migration was evaluated using an in vitro scratch wound-healing assay. Compared with control cells treated with DMSO, cellular migration was significantly enhanced in cells treated with 100 nM S1P (Fig. 1A). However, S1P-induced migration was dosedependently attenuated by treatment of cells with CK at final concentrations of 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10). In this assay system, the positive control 10 lM GM6001, a reversible matrix metalloproteinase (MMP) inhibitor, also attenuated the S1P-induced cell migration, and its effect was comparable to that observed for cells treated with 10 lg/mL CK, suggesting the involvement of MMP enzymes in S1P-induced HUVEC migration (Fig. 1B). Inhibitory effects of CK on SPHK1 activity The concentration of S1P (pmol/mg protein) in the lipid fraction extracted from an aliquot of HUVEC cell lysate (*50–100 lg protein) was analyzed. Pretreatment of cells with 1 lM sphingosine, a SPHK1 substrate, for 30 min prior to cell collection increased the basal level of SPHK1 activity by 40-fold (Fig. 2A). All ginsenosides tested mildly activated cellular SPHK1 activity and produced a large amount of S1P from 1 lM sphingosine. Compared with the control group, in which cells were treated with DMSO, six ginsenoside compounds (Rb1, Rc, Rd, Rg1, Rh1, and Ro) significantly activated S1P production by phosphorylating the extracellularly added sphingosine. Rb1 and Rh1 had the most significant effect on S1P formation, indicating that structural specificity of ginsenosides may be required to activate the SPHK1 enzyme. Three ginsenosides (F1, Re, and Rf) did not show any effective enhancement of S1P synthesis. Interestingly, the hydrophobic ginsenoside CK inhibited S1P production by SPHK1 by 45 % (Fig. 2B). SPHK1 activity in cell lysates containing 50 lg of protein was measured; enzyme activity was determined as the conversion rate of 2 nmol of C17-sphingosine into C17-S1P at 37 °C during a 30 min incubation. Most of the ginsenosides tested mildly increased SPHK1 activity,
123
K.-O. Shin et al.
100 80
A
50 Sphingosine Sphinganine
**
40
**
30
** 20
10
0 DMSO
60 40 20
**
B
Control Rb1 Rg1 CK Rh1
***
0
1
5
SKI II
CK
14 12
10
Concentration ( µg / mL ) Fig. 3 Effects of ginsenosides on HUVEC viability, measured as the absorbance of cells in culture medium containing WST-8 assay solution at 450 nm. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
S1P (pmol / mg protein)
Cell survival ( % )
120
SPHK1 protein by 30 % (Fig. 4). However, SPHK1 expression was not significantly affected by treatment of cells with Rg1 or Rh1; therefore, we concluded that the increase in SPHK1 activity and S1P production by these
Sphingoid base (pmol / mg protein)
whereas CK inhibited SPHK1 activity, as indicated by a reduction in C17-S1P levels (Fig. 2B); this result is in agreement with the observed inhibition of S1P production by CK, as shown in Fig. 2A. SPHK1 activity was inhibited by CK in a dose-dependent manner from 2.5 to 20 lg/mL (Fig. 2C); SPHK1 activity was not decreased any further at 40 lg/mL CK (data not shown). Therefore, 20 lg/mL of CK displayed a maximal inhibitory effect on SPHK1 activity in our assay system. In addition, HUVEC population was greatly reduced by CK treatment (Fig. 3). Furthermore, treatment of HUVEC with 10 lg/mL CK for 24 h reduced expression of the
10 8
**
**
6 4 2 0 DMSO
2.5
C
CK
0.5
**
0.4 2.0
S1P (nM)
SPHK1 expression (fold increase)
SPHK1( 42kDa )
SKI II
1.5
1.0
*
*
** 0.2
0.1
0.5
0.0
0.0 DMSO SKI II CK-5 CK-10 Rg1
Rh1
Rb1
Fig. 4 Western blot analysis of SPHK1 expression (42 kDa) in HUVEC treated with ginsenosides for 24 h. Control cells were treated with DMSO solution. CK-5 and CK-10 refer to cells treated with 5 or 10 lg/mL CK in DMSO, respectively. All other ginsenosides were treated at 10 lg/mL in DMSO. Data indicate the mean ± SEM for 3 samples and are represented as fold increase compared with the control cells. SKI II (1 lM) was used as a positive control. *P \ 0.05, **P \ 0.01
123
0.3
DMSO
SKI II
CK
Fig. 5 CK (10 lg/mL in DMSO) alters the levels of sphingolipid metabolites by inhibiting SPHK1 activity. The concentration (pmol/mg protein) of pro-aptotic sphingoid bases (sphingosine and sphinganine) was increased by approximately 2- to 3-fold following treatment of cells with CK (A). By contrast, CK inhibited S1P synthesis (B) and thus reduced extracellular secretion of S1P (C). SKI II (1 lM) was used as a positive control. Data indicate the mean ± SEM for 3 samples
Ginsenoside compound K inhibits angiogenesis
Evaluation of the ability of CK to inhibit SPHK1 was also performed by analyzing sphingolipid metabolites. The concentrations of the substrates of SPHK1, sphingosine, and sphinganine in cell lysates were increased twofold following treatment of HUVEC with CK (Fig. 5A). By contrast, the concentration of both intracellular and extracellular S1P was reduced by CK treatment (Fig. 5B, C). Therefore, treatment of cells with CK resulted in accumulation of SPHK1 substrates and a reduction in the amount of S1P produced, via inhibition of SPHK1 activity. Sphingolipid rheostat (the ratio of sphingosine/S1P) was greatly increased from basal level of 1.64–5.20 by SKI II and to 6.13 by CK, suggesting that CK can change sphingolipid rheostat by inhibiting SPHK1 activity. To analyze the changes of total ceramides, the sphingolipid fraction in HUVEC were separated by HPTLC, and then cellular ceramides were recollected and alkaline-digested. The liberated sphingosine was derivatized with OPA and then quantified by HPLC. Compared with control cells treated with DMSO, treatment with CK resulted in a 1.5-fold increase in the amount of total ceramide (Fig. 6A), indicating propagation of pro-apoptotic signals by ceramide. As expected, sphingolipid rheostat (the ratio of ceramide/S1P) was also increased from 80.0 (basal) to 182.0 (SKI II) and to 212.9 (CK), respectively. The ceramide species in the lipid extracts were further analyzed by LC–MS/MS. A single ceramide with a saturated fatty acid chain, tetracosanoic acid (C24:0), was profoundly accumulated in HUVEC following CK treatment (Fig. 6B). Inhibitory effects of CK on MMP expressions In Fig. 1, GM6001, a broad spectrum MMP inhibitor, also attenuated the cell migration, and its effect was comparable to that observed for cells treated with 10 lg/mL CK. We also confirmed the ability of CK to reduce MMP2 and MMP9 expression in HUVEC (Fig. 7A). Cellular MMP expression was enhanced in cells treated with 100 nM S1P and treatment of cells for 24 h with CK (10 lg/mL in DMSO) reduced expression of MMP2 (Fig. 7B) and MMP9 (Fig. 7C) by 80 and 50 %, respectively.
Total ceramide (pmol / mg protein)
Change of sphingolipid composition by CK
A 1600 **
1400 1200 1000 800 600 400 200 0 DMSO
B Ceramide (pmol / mg protein)
ginsenosides does not originate from an increase in SPHK1 protein levels. Unexpectedly, treatment of cells with ginsenoside Rb1 increased expression of the SPHK1 protein twofold. Ginsenoside Rb1 is a firstly founded plant compound, inducing SPHK1 protein expression and SPHK1 activity.
SKI II
CK
250
**
DMSO CK
200
150
**
100
50
0 C16:0
C18:0
C20:0
C22:0
C24:1
C24:0
N-acyl chain length Fig. 6 Concentrations of total ceramide (A) and individual ceramide species (B) following treatment of HUVEC with CK (10 lg/mL in DMSO). Control cells were treated with DMSO. SKI II (1 lM) was used as a positive control. Data indicate the mean ± SEM for 3 samples
Discussion It was recently reported that S1P produced by SPHK1 promotes breast cancer progression by stimulating angiogenesis (Nagahashi et al. 2012). Furthermore, knockdown of SPHK1 expression by specific siRNA was shown to inhibit HUVEC growth and migration, whereas knockdown of SPHK2 expression had no effect (Yan et al. 2008). S1P avidly releases outside of cellular membrane and interacts with five G-protein coupled S1P receptors to generate multiple downstream signals including HUVEC migration. Therefore, the regulation of SPHK1 activity may represent a new therapeutic strategy for treatment of angiogenesisrelated diseases such as cancer. Ginsenoside CK, a major metabolite of ginsenoside Rb1 generated by intestinal bacteria, is observed in plasma, indicating that CK, a main bioactive metabolite of protopanaxadiol ginsenosides, is well absorbed in the intestine
123
K.-O. Shin et al.
A
B Fold increase in MMP2 expression (%)
Fig. 7 Western blot analysis showing the effect of 100 nM S1P and 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10) CK on MMP2 and 9 expression. A typical western blot image showing expression of MMP and b-actin as a control (A). CK inhibits MMP2 (B) and MMP9 expression (C). GM6001 (25 lM) was used as a positive control. Data represent fold increase relative to control cells. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
140 120 100 80
** **
60 40
***
***
20 0 DMSO S1P GM6001 CK-1
CK-5 CK-10
+ 100nM S1P
Fold increase in MMP9 expression ( % )
C
140 120 100 *
80 **
60 40
**
20 0 DMSO S1P GM6001 CK-1
CK-5 CK-10
+ 100nM S1P
following oral administration of ginseng (Akao et al. 1998). In this study, the effect of ten putative ginsenosides on SPHK1 activation was investigated by adding a SPHK1 substrate, 1 lM sphingosine, to the culture medium of HUVEC. CK exclusively inhibited both S1P production and SPHK1 activation in a dose-dependent manner (Fig. 1). Although we did not find any relationship between the number of sugar moieties and SPHK1 inhibition, CK contains a glucose moiety at C-20 and is therefore less
123
polar than the other ginsenosides tested. Inhibition of SPHK1 by CK (5–10 lg/mL) also resulted in reduced HUVEC viability (Fig. 3), which is likely due to the function of SPHK as a critical regulator of the sphingolipid rheostat maintaining the concentration balance between the pro-growth, anti-apoptotic messenger S1P, and pro-apoptotic ceramide and sphingosine (Maceyka et al. 2002). Interestingly, a reduction in SPHK1 expression was observed following CK treatment, suggesting post-transcriptional regulation by CK (Fig. 4). Ginsenoside Rb1
Ginsenoside compound K inhibits angiogenesis
induced a twofold increase in SPHK1 expression (Fig. 4) and S1P production was also increased by Rb1 (data not shown). The sphingolipid rheostat is a concept that refers to modulation of the balance and interconversion of S1P, ceramide, and sphingosine, which ultimately determines cell fate (Loh et al. 2011). Our data suggest that CK alters the sphingolipid rheostat balance by increasing the amounts of pro-apoptotic ceramide and sphingosine, and by decreasing intracellular and extracellular levels of the anti-apoptotic messenger S1P (Figs. 4, 5A). Secreted S1P may bind to S1PRs expressed on outer membrane surfaces in an autocrine or paracrine manner, and transduce signals to intracellular targets. Therefore, CK-induced reduction in S1P production will attenuate chemotactic HUVEC migration via the S1PRs-S1P axis. Notably, the amount of sphinganine, which is an intermediate in the de novo sphingolipid biosynthetic pathway, was also increased by CK treatment, suggesting new ceramide synthesis via the de novo pathway. Further analysis by LC–MS/MS demonstrated that a specific ceramide species containing a saturated fatty acid chain (C24:0) was significantly increased by CK treatment of cells (Fig. 5B). The increase in C24:0 ceramide indicates the activation of ceramide synthase type 2 proteins. When ceramide synthase type 2 is downregulated, ceramide-bearing palmitate (C16:0) is accumulated instead of C24:0-ceramide or C24:1-ceramide, and can induce autophagy but not apoptotic cell death (Spassieva et al. 2009). Therefore, the accumulation of C24:0-ceramide by CK is directly associated with cytotoxicity in HUVEC. S1P was previously shown to induce angiogenesis by activating extracellular signal-regulated kinases and p38 mitogen-activated protein kinase in a pertussis toxinsensitive manner (Lee et al. 1999). Our data show that 100 nM S1P enhanced HUVEC migration in a woundhealing assay by ninefold; however, this migration was potently inhibited by CK, which was effective at a dose as low as 10 lg/mL (Fig. 6B). In addition, GM6001 (10 lM) strongly attenuated HUVEC migration via inhibition of MMPs, which are thought to play a major role in cell growth, migration, angiogenesis, and apoptosis. Very recently, CK treatment was also shown to inhibit MMP2 and MMP9 expressions, which are both associated with angiogenesis and metastasis in hepatocellular carcinoma (Ming et al. 2011). In summary, the present study demonstrates that CK displays an anti-angiogenic property that is capable of perturbing S1P-induced angiogenesis via SPHK1 inhibition, resulting in disruption of the sphingolipid rheostat. CK, which can be produced by hydrolyzing the sugar moieties of the major ginsenosides Rb1, Rb2, and Rc, has more hydrophobic properties than the other ginsenosides
included in this study; the additional hydrophobicity of CK may contribute to suppression of SPHK1 activity and inhibition of MMPs. We do not exclude the possibility that CK-induced regulation of S1P receptors and their downstream signaling, or specific responses of protein kinase C isoenzymes associated with SPHK1 activity, may occur. Further investigation to identify the targets of CK will help to decipher the inhibitory mechanisms of S1P-induced HUVEC migration and may open a new possibility for anti-angiogenic therapy.
Conclusion We found that Ginsenoside CK regulates the migration of human umbilical vein endothelial cells which is an essential factor in cancer angiogenesis via the inhibition of sphingosine kinase-1. Also, ginsenoside CK increases proapoptotic lipids, sphingoid bases and ceramides, specifically accumulates C24:0 fatty acid acylation ceramide in HUVEC. Importantly, CK blocks S1P-induced HUVEC migration and the expressions of MMP enzymes which are responsible to cells motility in dose-dependent manner. Acknowledgments This study was supported by a NRF Grants funded by the Korean Government (MRC, 2010-0029483) and (KRF-2010-0025271).
References Akao, T., M. Kanaoka, and K. Kobashi. 1998. Appearance of compound K, a major metabolite of ginsenoside Rb1 by intestinal bacteria, in rat plasma after oral administration– measurement of compound K by enzyme immunoassay. Biological and Pharmaceutical Bulletin 21: 245–249. Berdyshev, E.V., I. Gorshkova, P. Usatyuk, S. Kalari, Y. Zhao, N.J. Pyne, S. Pyne, R.A. Sabbadini, J.G. Garcia, and V. Natarajan. 2011. Intracellular S1P generation is essential for S1P-induced motility of human lung endothelial cells: Role of sphingosine kinase 1 and S1P lyase. PLoS ONE 6: e16571. Cheung, L.W., K.W. Leung, C.K. Wong, R.N. Wong, and A.S. Wong. 2011. Ginsenoside-Rg1 induces angiogenesis via non-genomic crosstalk of glucocorticoid receptor and fibroblast growth factor receptor-1. Cardiovascular Research 89: 419–425. Jeong, A., H.J. Lee, S.J. Jeong, E.O. Lee, H. Bae, and S.H. Kim. 2010. Compound K inhibits basic fibroblast growth factor-induced angiogenesis via regulation of p38 mitogen activated protein kinase and AKT in human umbilical vein endothelial cells. Biological and Pharmaceutical Bulletin 33: 945–950. Joh, E.H., I.A. Lee, I.H. Jung, and D.H. Kim. 2011. Ginsenoside Rb1 and its metabolite compound K inhibit IRAK-1 activation–the key step of inflammation. Biochemical Pharmacology 82: 278–286. Kim Do, Y., H.D. Yuan, I.K. Chung, and S.H. Chung. 2009. Compound K, intestinal metabolite of ginsenoside, attenuates hepatic lipid accumulation via AMPK activation in human hepatoma cells. Journal of Agriculture and Food Chemistry 57: 1532–1537.
123
K.-O. Shin et al. Kluk, M.J., and T. Hla. 2002. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochimica et Biophysica Acta 1582: 72–80. Lee, H.U., E.A. Bae, M.J. Han, N.J. Kim, and D.H. Kim. 2005. Hepatoprotective effect of ginsenoside Rb1 and compound K on tert-butyl hydroperoxide-induced liver injury. Liver Int 25: 1069–1073. Lee, O.H., Y.M. Kim, Y.M. Lee, E.J. Moon, D.J. Lee, J.H. Kim, K.W. Kim, and Y.G. Kwon. 1999. Sphingosine 1-phosphate induces angiogenesis: Its angiogenic action and signaling mechanism in human umbilical vein endothelial cells. Biochemical and Biophysical Research Communications 264: 743–750. Leong, W.I., and J.D. Saba. 2010. S1P metabolism in cancer and other pathological conditions. Biochimie 92: 716–723. Leung, K.W., H.M. Ng, M.K. Tang, C.C. Wong, R.N. Wong, and A.S. Wong. 2011. Ginsenoside-Rg1 mediates a hypoxia-independent upregulation of hypoxia-inducible factor-1alpha to promote angiogenesis. Angiogenesis 14: 515–522. Loh, K.C., D. Baldwin, and J.D. Saba. 2011. Sphingolipid signaling and hematopoietic malignancies: To the rheostat and beyond. Anti-Cancer Agents in Medicinal Chemistry 11: 782–793. Maceyka, M., S.G. Payne, S. Milstien, and S. Spiegel. 2002. Sphingosine kinase, sphingosine-1-phosphate, and apoptosis. Biochimica et Biophysica Acta 1585: 193–201. Ming, Y., Z. Chen, L. Chen, D. Lin, Q. Tong, Z. Zheng, and G. Song. 2011. Ginsenoside compound K attenuates metastatic growth of hepatocellular carcinoma, which is associated with the translocation of nuclear factor-kappaB p65 and reduction of matrix metalloproteinase-2/9. Planta Medica 77: 428–433.
123
Nagahashi, M., S. Ramachandran, E.Y. Kim, J.C. Allegood, O.M. Rashid, A. Yamada, R. Zhao, S. Milstien, H. Zhou, S. Spiegel, and K. Takabe. 2012. Sphingosine-1-phosphate produced by sphingosine kinase 1 promotes breast cancer progression by stimulating angiogenesis and lymphangiogenesis. Cancer Research 72: 726–735. Park, E.K., Y.W. Shin, H.U. Lee, S.S. Kim, Y.C. Lee, B.Y. Lee, and D.H. Kim. 2005. Inhibitory effect of ginsenoside Rb1 and compound K on NO and prostaglandin E2 biosyntheses of RAW264.7 cells induced by lipopolysaccharide. Biological and Pharmaceutical Bulletin 28: 652–656. Sato, K., M. Mochizuki, I. Saiki, Y.C. Yoo, K. Samukawa, and I. Azuma. 1994. Inhibition of tumor angiogenesis and metastasis by a saponin of Panax ginseng, ginsenoside-Rb2. Biological and Pharmaceutical Bulletin 17: 635–639. Shida, D., K. Takabe, D. Kapitonov, S. Milstien, and S. Spiegel. 2008. Targeting SphK1 as a new strategy against cancer. Current Drug Targets 9: 662–673. Spassieva, S.D., T.D. Mullen, D.M. Townsend, and L.M. Obeid. 2009. Disruption of ceramide synthesis by CerS2 down-regulation leads to autophagy and the unfolded protein response. The Biochemical Journal 424: 273–283. Xu, T.M., Y. Xin, M.H. Cui, X. Jiang, and L.P. Gu. 2007. Inhibitory effect of ginsenoside Rg3 combined with cyclophosphamide on growth and angiogenesis of ovarian cancer. Chinese Medical Journal (English) 120: 584–588. Yan, G., S. Chen, B. You, and J. Sun. 2008. Activation of sphingosine kinase-1 mediates induction of endothelial cell proliferation and angiogenesis by epoxyeicosatrienoic acids. Cardiovascular Research 78: 308–314.
RESEARCH ARTICLE
Ginsenoside compound K inhibits angiogenesis via regulation of sphingosine kinase-1 in human umbilical vein endothelial cells Kyong-Oh Shin • Cho-Hee Seo • Hyo-Hyun Cho Seikwan Oh • Seon-Pyo Hong • Hwan-Soo Yoo • Jin-Tae Hong • Ki-Wan Oh • Yong-Moon Lee
•
Received: 10 November 2013 / Accepted: 16 January 2014 Ó The Pharmaceutical Society of Korea 2014
Abstract Ginsenoside compound K (CK) is a metabolite of the protopanaxadiol-type saponins of Panax ginseng C.A. Meyer (Araliaceae), has long been used to treat against the development of cancer, inflammation, allergies, and diabetes. This study examined the anti-angiogenic properties of CK against sphingosine 1-phosphate (S1P)-induced cell migration via regulation of sphingosine kinase 1 (SPHK1) in human umbilical vein endothelial cells (HUVEC). Studies on S1P-induced cell migration, expression of SPHK1 and MMPs and analysis of sphingolipid metabolites by LC–MS/MS were examined after the treatment of CK (2.5, 5, 10 lg/mL) in HUVEC. S1P produced by SPHK1 is also involved in cell growth, migration, and protection of apoptosis; therefore, we sought to investigate whether ginsenosides are able to regulate SPHK1. For this purpose, we developed an inhibitory assay of SPHK1 activity and an analytical method for detection of S1P and other sphingolipid metabolites in HUVEC. Ginsenoside CK inhibited 100 nM S1P-induced cell migrations in a dose-dependent manner. Among tested ginsenosides, CK exclusively inhibited S1P production, SPHK1 activity and SPHK1 expression in HUVEC, whereas expression of the pro-apoptotic sphingolipids, sphingosine and ceramide, was increased in response to CK. The major K.-O. Shin C.-H. Seo H.-H. Cho H.-S. Yoo J.-T. Hong K.-W. Oh Y.-M. Lee (&) College of Pharmacy and MRC, Chungbuk National University, Cheongju 361-763, Korea e-mail: [email protected]; [email protected] S. Oh College of Medicine, Ehwa Women’s University, Seoul 158-710, Korea S.-P. Hong Department of Oriental Pharmaceutical Sciences, Kyung Hee University, Seoul 130-701, Korea
subspecies of the increased ceramide was C24:0-ceramide. CK also disrupted the sphingolipid rheostat, which ultimately influences cell fate, and dose-dependently inhibited HUVEC migration by reducing expression of metalloproteinases (MMPs). Ginsenoside CK acts as a unique HUVEC migration inhibitor by regulating MMP expression, as well as the activity of SPHK1 and its related sphingolipid metabolites. Keywords Panax ginseng Araliaceae Ginsenoside Compound K Sphingosine kinase Cell migration Human umbilical vein endothelial cells
Introduction Angiogenesis, proliferation of the endothelium and formation of new blood vessels, play important roles in metastasis and tumor growth, resulting in an increase in the size of solid tumors. Therefore, inhibition of angiogenic signaling is expected to be an efficient therapeutic approach for many tumor types. As drug targets indeed, vascular endothelium growth factor (VEGF) and its receptors, VEGFR-1, -2, and -3, constitute a key signaling system that regulate proliferation and migration of vascular endothelial cells. Ginsenoside Rg1 mediates hypoxia-independent upregulation of hypoxia inducible factor 1 alpha (HIF-1a) (Leung et al. 2011), and non-genomic crosstalk of the glucocorticoid receptor and fibroblast growth factor receptor-1, which promote angiogenesis (Cheung et al. 2011). Conversely, ginsenosides Rb2 and Rg3 inhibit tumor angiogenesis and metastasis (Sato et al. 1994; Xu et al. 2007). Ginsenoside compound K (CK), a major metabolite of ginsenoside Rb1 from Panax ginseng root that is generated by intestinal bacterial flora, is rapidly
123
K.-O. Shin et al.
absorbed from the gastrointestinal tract following oral administration and then slowly metabolized (Akao et al. 1998). CK inhibits interleukin-1 receptor-associated kinase 1 activation, the key step of inflammation (Joh et al. 2011), as well as lipopolysaccharide-induced biosynthesis of nitric oxide and prostaglandin E2 in RAW264.7 cells (Park et al. 2005). A hepatoprotective effect of CK, resulting from activation of AMP-activated protein kinase and subsequent attenuation of hepatic lipid accumulation, has also been reported (Lee et al. 2005; Kim Do et al. 2009). Sphingosine kinases (SPHKs) catalyze the phosphorylation of sphingosine into sphingosine 1-phosphate (S1P). Sphingosine kinase type 1 (SPHK1) has been shown to regulate several processes involved in cancer progression and SPHK1 is overexpressed in various types of cancers. Upregulation of SPHK1 is associated with tumor angiogenesis and resistance to radiation and chemotherapy. Many growth factors stimulate SPHK1 via their tyrosine kinase receptors, leading to a rapid increase in S1P production (Shida et al. 2008). S1P is a pro-angiogenic factor that mediates diverse biological processes, including endothelial cell growth, migration, invasion, angiogenesis, vascular maturation, and lymphocyte trafficking (Berdyshev et al. 2011). The newly synthesized S1P can release out of cellular membrane and act as an extracellular signaling factor, by binding to a family of five related G-protein-coupled S1P receptors (S1PRs), previously known as the endothelial differentiation genes. The amount of intracellular S1P available for signaling through various S1PRs is tightly regulated, primarily by the SPHK1, S1P lyase, and S1P phosphatase enzymes (Leong and Saba 2010). A complex interactive network of signaling cascades downstream from the G-protein-coupled S1PRs acts as a prime effector of angiogenesis that occurs in association with various pathologies (Kluk and Hla 2002). Recently, CK was reported effectively to disrupt basic fibroblast growth factor-induced neo-vascularization in Matrigel plugs excised from mice in vivo (Jeong et al. 2010). Nuclear export of NF-kappaB-p65 and reduction of metalloproteinases(MMP)2 and MMP9 expression are associated with metastatic inhibition induced by CK (Ming et al. 2011). In this study, we examined the effect of CK on SPHK1 activity and S1P levels in HUVEC. CK inhibited SPHK1 activity and impaired sphingolipid metabolic balance by reducing synthesis and extracellular release of S1P, resulting the inhibition of HUVEC migration.
Materials and methods Cell lines, chemicals and biochemicals HUVEC were purchased from Lonza (Basel, Switzerland). S1P, sphingosine, ceramides (fatty acid lengths C12, C16,
123
C18, C22, C24, and C24:1), 2S-amino-4E-heptadecene-1, 3R-diol (C17-sphingosine), C17-sphingosine-1-phosphate (C17-S1P), C17-sphinganine-1-phosphate (C17-DHS1P), and C17-ceramide (d18:1/C17:0) were obtained from Avanti Polar Lipid (Alabaster, AL, USA). Ginsenosides [CK (98.4 %), Rg1 (98.7 %), Rh1 (98.3 %), Rb1 (99.1 %), Rd (98.7 %), Re (99.3 %), Rf (98.6 %), and F1 (99.3 %)] were purchased from Chengdu Biopurify Phytochemical Company (Chengdu, China).GM6001 (95 %) and SKI II (98 %) were obtain from Sigma-aldrich (St. Louis, MO, USA). Endothelial cell growth medium-2 (EGM-2) was purchased from Lonza (Basel, Switzerland); fetal bovine serum (FBS) was purchased from Gibco-BRL (Grand Island, NY, USA). The monoclonal anti-human SPHK1 antibody (Cat. No. 3297) and polyclonal anti-human b-actin antibody (Cat. No. 4967) were purchased from Cell Signaling Technology (Danvers, MA, USA) and polyclonal antihuman-MMP2 antibody (Cat. No. SC-8830) and polyclonal anti-human-MMP9 antibody (Cat. No. SC-6840) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Organic solvents for sphingolipid extraction and HPLC analysis were purchased from Merck (Darmstadt, Germany). Unless otherwise stated, all other chemicals were obtained from Sigma (St Louis, MO, USA). Cell culture conditions HUVEC (passage 5–7) were grown on 1 % gelatin-coated culture plates in EGM-2 medium supplemented with 10 % FBS, 1 % antibiotics (100 U/mL penicillin, 100 lg/mL streptomycin), VEGF (1 ng/mL), bFGF (2 ng/mL), and heparin (10 U/mL). In vitro wound-healing assay An in vitro wound-healing assay was performed to measure unidirectional migration of HUVEC. HUVEC (1 9 105 cells/well) were seeded into 12-well plates coated with collagen, and incubated at 37 °C until they attached. The cells were then washed twice with phosphate-buffered saline and were incubated in EGM-2 containing 1 % FBS for 24 h at 37 °C. Monolayers of HUVEC were scratchwounded to a 1 mm depth in a straight line using a 200 lL pipette-tip, and then incubated for 24 h with S1P (100 nM), and either with or without GM6001 (25 lM) or various concentrations (1–10 lg/mL) of CK. To measure the number of migrated endothelial cells from the edge of the injured monolayer, images were photographed both immediately after wounding and after 24 h incubation, using a phase-contrast microscope (Olympus, Tokyo, Japan). At least four points in each of three fields were examined at random for two independent wounds.
Ginsenoside compound K inhibits angiogenesis
Measurement of SPHK1 activity HUVEC were seeded into 6-well plates at 1 9 106 cells/ well and incubated for 24 h to allow full attachment. The cells were then treated with each ginsenoside (10 lg/mL) dissolved in dimethyl sulfoxide (DMSO), for 24 h. For measurement of S1P released into the culture media, 1 lM sphingosine, which enhances S1P biosynthesis by SPHK1, was added to the culture media 30 min prior to cell harvesting. A specific inhibitor of SPHK1, SKI II was tested as a positive control. A total of 50 lg protein from cell lysate was incubated in a volume of 100 lL, containing 10 lL of 10 mM ATP in 200 mM MgCl2, and 10 lL of 200 lM C17-Sphingosine in 5 % Triton X-100. Incubations with cell lysates contained 5 mM NaF and Na3VO4, which were included as inhibitors of S1P phosphatase and lyase, respectively, to prevent potential degradation of C17-S1P. Reactions were incubated for 30 min at 37 °C and were terminated by addition of 0.8 mL CHCl3/MeOH/HCl (300:500:3, v/v/v) in cold ice. Following addition of 100 pmol C17-DHS1P as an internal standard, samples were vortexed vigorously for 10 min and then centrifuged at 20,0009g for 3 min. Next, 250 lL of CHCl3 and 250 lL of 1 M NaCl were added, and the mixture was vortexed vigorously for 10 min. Samples were centrifuged at 4 °C for 3 min at 20,0009g, and then the upper phase was removed. The organic phase was transferred to a fresh tube and stored on ice. C17-S1P was extracted by addition of 250 lL 1 M NaCl, 300 ll of MeOH, and 30 lL of 3 N NaOH. The tube was then vortexed for 10 min and centrifuged at 14,000 rpm for 4 min. The alkaline aqueous phase containing C17-S1P was mixed thoroughly with 100 lL of dephosphorylation reaction buffer (200 mM Tris–HCl (pH 7.4), 75 mM MgCl2 in 2 M glycine buffer, pH 9.0) and 10 units of alkaline phosphatase; the tube was then incubated at 37 °C for 90 min. The dephosphorylated C17-Sph was extracted twice with 500 lL of CHCl3 and 300 lL of CHCl3, and then washed three times with alkaline water (pH 10.0). The washed CHCl3 phase was transferred to a fresh tube and dried completely using a speed vacuum system (Vision Co., Seoul, Korea). S1P extraction A volume of 350 lL of MeOH, containing 0.6 lL of 35 % HCl, and 100 pmol of C17-S1P as an internal standard, was added to a 50 lg aliquot of cell lysate; the sample was then ultrasonicated for 5 min in ice-cold water. Lipids were extracted by Micromixer (Taitec, Tokyo, Japan) for 60 min at 750 rpm after adding 500 lL of CHCl3/1 M NaCl (1:1 v/v) and 35 lL of 3 N NaOH. The aqueous fractions
containing S1P were transferred to fresh tubes and then sphingosine was released from S1P using alkaline phosphatase. The aqueous phases obtained from the S1P extraction procedure were mixed thoroughly with 100 lL of reaction buffer (200 mM Tris–HCl (pH 7.4), 75 mM MgCl2 in 2 M glycine buffer, pH 9.0), containing 10 units of alkaline phosphatase, and were vortexed for 2 h at 37 °C. During this step, S1P efficiently released phosphate and was converted into sphingosine. Sphingosine was extracted twice with 500 lL CHCl3 and then dried completely using a speed vacuum system. HPLC analysis of S1P concentration and SPHK1 activity Residues were dissolved in 120 lL MeOH, incubated at 50 °C for 10 min, and then derivatized with 15 lL OPA reagent (50 mg ortho-phthalaldehyde, 1 mL EtOH, 100 lL 2-mercaptoethanol, 50 mL 3 % (w/v) boric acid solution). Following incubation of the tubes for 30 min in the dark at room temperature, 90 lL of each aliquot was injected into the HPLC system (Jasco, Tokyo, Japan), which consisted of a Model PU 980 pump, Nova-Pak C18 (4.6 mm i.d. 9 150 mm), and FP-920 fluorescence spectrophotometer (kex 340 nm, kem 455 nm) with autosampler. The isocratic mobile phase of 90 % MeOH was pumped at a flow rate of 1 mL/min. The resulting data was evaluated using the Borwin chromatographic system manager software (Jasco, Tokyo, Japan). LC–MS/MS analysis of ceramide species HUVEC were seeded into 6-well plates at 1 9 106 cells/ well. Following treatment with ginsenoside CK (10 lg/mL) for 24 h, the cells were collected as cell pellets. Total ceramide was extracted with the mixture of 50 lL cell lysates (100 lg), 100 pmol of internal standard C17-ceramide (d18:1/C17:0), and 2 mL of Chloroform/MeOH (2:1, v/v) solution. The low phase was collected after repeated extraction (final volume of 4 mL) and dried in a vacuum system. The dried residue was redissolved in 50 lL MeOH, and then 10 lL was injected into the LC–ESI–MS/ MS system. For optimization, a mixture of ceramide standards was infused directly into the mass spectrometer and all source parameters and ionization conditions were adjusted to improve the sensitivity of the assay. Extracted samples (10 lL) were injected into a HPLC (Agilent 1200 series, Agilent, CA, USA) and separated through a reverse phase KINETEX C18 column (2.1 9 50 mm, ID: 2.6 lm) (Phenomenex, St. Louis, MO, USA). The ceramides were resolved using a linear gradient from 8 % mobile phase A (water containing 0.2 % formic acid) at a flow rate of
123
K.-O. Shin et al.
0.3 mL/min for 1 min, to 100 % mobile phase B (MeOH containing 0.2 % formic acid) over 3 min, followed by 100 % mobile phase B for 12 min. The column was then equilibrated for 8 min with 92 % mobile phase B. The HPLC column effluent was introduced onto a Thermo LTQ-XL ESI-ION TRAP (Thermo Scientific, Fl, USA) and analyzed using electrospray ionization in positive mode. The configurations for mass spectrometry were: capillary voltage, 5.5 kV; cone voltage, 40 V; cone temperature, 300 °C. Analyses were performed using electrospray ionization in the positive-ion mode with multiple reaction monitoring to select both parent and characteristic daughter ions specific to each analyte simultaneously from a single injection. The MS/MS transitions (m/z) were 510 ? 264 for C14, 538 ? 264 for C16, 552 ? 264 for C17, 566 ? 264 for C18, 594 ? 264 for C20, 648 ? 264 for C24:1, and 650 ? 264 for C24. Data were acquired using X-calibur software.
A
Western blot analysis
BCA protein assay Cell lysate from the solubilized tissue was mixed with Pierce BCA reagents (Rockford, IL, USA), according to the manufacturer’s instructions, and incubated for 30 min. Protein content was quantified at 562 nm with an ELISA reader (Molecular Devices, Sunnyvale, CA, USA), based on a bovine serum albumin standard curve.
Statistical analysis All values are expressed as mean ± standard error of the mean. For single comparisons, differences between treatments were analyzed by unpaired Student’s t test. For multiple comparisons, one-way analysis of variance was
123
B Migrated cells density (fold increase)
SDS-PAGE was performed on HUVEC lysates using 10 % acrylamide gels. Proteins were transferred to PVDF membranes (Millipore Corporation, MA, USA) and blots were probed with anti-SPHK1 (1:1,000) anti-MMP2 (1:500), anti- MMP-9 (1:500), ß-actin (1:1,000), followed by incubation with horseradish peroxidase-conjugated mouse or rabbit immunoglobulin. Blots were then developed using West Pico Chemiluminescent Substrate western blotting detection reagent (Pierce, Woburn, MA, USA). The band intensity was determined by Quantity One 4.5.0 Image acquisition and Analysis software (BioRad).
12 10 8
** 6
** 4
***
***
2 0 DMSO
S1P GM6001 CK-1
CK-5
CK-10
+ 100nM S1P Fig. 1 HUVEC migration induced by 100 nM S1P in an in vitro wound-healing assay. (A) Migration was inhibited in a dosedependent manner by treatment of cells with CK at final concentrations of 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10).The inhibitory effect of 10 lg/mL CK was comparable to that of the positive control, 10 lM GM6001. Panel (B) shows the density of migrated cells, represented as fold change over the control group. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
Ginsenoside compound K inhibits angiogenesis b Fig. 2 The effects of each ginsenoside (10 lg/mL in DMSO) on S1P
A 350
S1P (pmol / mg protein)
300
*
** *
250
*
200
synthesis in HUVEC after pretreatment with a natural substrate, sphingosine (1 lM), for 30 min before cell harvesting (A), and SPHK1 activity, measured as the conversion rate of 2 nmol of C17sphingosine into C17-S1P at 37 °C during a 30 min incubation (B). Panel (C) shows the inhibition of SPHK1 activity by CK in a dosedependent manner. Control cells were treated with DMSO instead of CK. SKI II (1 lM) was used as a positive control. Data indicate the mean ? SEM for 3 samples. *P \ 0.05, **P \ 0.01
150 **
100
**
Results
50
Inhibitory effects of CK on HUVEC migration
0
I f SO SO KI I Rb1 CK F1 Rc Rd Re R Rg1 Rh1 Ro DM DM S
+ 1 µM sphingosine
B
1.6
C17-S1P (pmol / mg / min)
1.4 *
1.2 1.0 **
0.8 0.6 **
0.4 0.2 0.0
I f SO SO KI I Rb1 CK F1 Rc Rd Re R Rg1 Rh1 Ro DM DM S
+ 20 µM C 17-sphingosine
C17-S1P (pmol / mg / min)
C 1.2 1.0 **
0.8
**
0.6 **
0.4 0.2 0.0
SO MSO SKI II K-2.5 CK-5 CK-10 CK-20 D C
DM
+ 20 µM C 17-sphingosine
performed, followed by Newman-Keuls multiple comparison test. P values \ 0.05 were considered to be statistically significant.
The effect of CK on HUVEC migration was evaluated using an in vitro scratch wound-healing assay. Compared with control cells treated with DMSO, cellular migration was significantly enhanced in cells treated with 100 nM S1P (Fig. 1A). However, S1P-induced migration was dosedependently attenuated by treatment of cells with CK at final concentrations of 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10). In this assay system, the positive control 10 lM GM6001, a reversible matrix metalloproteinase (MMP) inhibitor, also attenuated the S1P-induced cell migration, and its effect was comparable to that observed for cells treated with 10 lg/mL CK, suggesting the involvement of MMP enzymes in S1P-induced HUVEC migration (Fig. 1B). Inhibitory effects of CK on SPHK1 activity The concentration of S1P (pmol/mg protein) in the lipid fraction extracted from an aliquot of HUVEC cell lysate (*50–100 lg protein) was analyzed. Pretreatment of cells with 1 lM sphingosine, a SPHK1 substrate, for 30 min prior to cell collection increased the basal level of SPHK1 activity by 40-fold (Fig. 2A). All ginsenosides tested mildly activated cellular SPHK1 activity and produced a large amount of S1P from 1 lM sphingosine. Compared with the control group, in which cells were treated with DMSO, six ginsenoside compounds (Rb1, Rc, Rd, Rg1, Rh1, and Ro) significantly activated S1P production by phosphorylating the extracellularly added sphingosine. Rb1 and Rh1 had the most significant effect on S1P formation, indicating that structural specificity of ginsenosides may be required to activate the SPHK1 enzyme. Three ginsenosides (F1, Re, and Rf) did not show any effective enhancement of S1P synthesis. Interestingly, the hydrophobic ginsenoside CK inhibited S1P production by SPHK1 by 45 % (Fig. 2B). SPHK1 activity in cell lysates containing 50 lg of protein was measured; enzyme activity was determined as the conversion rate of 2 nmol of C17-sphingosine into C17-S1P at 37 °C during a 30 min incubation. Most of the ginsenosides tested mildly increased SPHK1 activity,
123
K.-O. Shin et al.
100 80
A
50 Sphingosine Sphinganine
**
40
**
30
** 20
10
0 DMSO
60 40 20
**
B
Control Rb1 Rg1 CK Rh1
***
0
1
5
SKI II
CK
14 12
10
Concentration ( µg / mL ) Fig. 3 Effects of ginsenosides on HUVEC viability, measured as the absorbance of cells in culture medium containing WST-8 assay solution at 450 nm. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
S1P (pmol / mg protein)
Cell survival ( % )
120
SPHK1 protein by 30 % (Fig. 4). However, SPHK1 expression was not significantly affected by treatment of cells with Rg1 or Rh1; therefore, we concluded that the increase in SPHK1 activity and S1P production by these
Sphingoid base (pmol / mg protein)
whereas CK inhibited SPHK1 activity, as indicated by a reduction in C17-S1P levels (Fig. 2B); this result is in agreement with the observed inhibition of S1P production by CK, as shown in Fig. 2A. SPHK1 activity was inhibited by CK in a dose-dependent manner from 2.5 to 20 lg/mL (Fig. 2C); SPHK1 activity was not decreased any further at 40 lg/mL CK (data not shown). Therefore, 20 lg/mL of CK displayed a maximal inhibitory effect on SPHK1 activity in our assay system. In addition, HUVEC population was greatly reduced by CK treatment (Fig. 3). Furthermore, treatment of HUVEC with 10 lg/mL CK for 24 h reduced expression of the
10 8
**
**
6 4 2 0 DMSO
2.5
C
CK
0.5
**
0.4 2.0
S1P (nM)
SPHK1 expression (fold increase)
SPHK1( 42kDa )
SKI II
1.5
1.0
*
*
** 0.2
0.1
0.5
0.0
0.0 DMSO SKI II CK-5 CK-10 Rg1
Rh1
Rb1
Fig. 4 Western blot analysis of SPHK1 expression (42 kDa) in HUVEC treated with ginsenosides for 24 h. Control cells were treated with DMSO solution. CK-5 and CK-10 refer to cells treated with 5 or 10 lg/mL CK in DMSO, respectively. All other ginsenosides were treated at 10 lg/mL in DMSO. Data indicate the mean ± SEM for 3 samples and are represented as fold increase compared with the control cells. SKI II (1 lM) was used as a positive control. *P \ 0.05, **P \ 0.01
123
0.3
DMSO
SKI II
CK
Fig. 5 CK (10 lg/mL in DMSO) alters the levels of sphingolipid metabolites by inhibiting SPHK1 activity. The concentration (pmol/mg protein) of pro-aptotic sphingoid bases (sphingosine and sphinganine) was increased by approximately 2- to 3-fold following treatment of cells with CK (A). By contrast, CK inhibited S1P synthesis (B) and thus reduced extracellular secretion of S1P (C). SKI II (1 lM) was used as a positive control. Data indicate the mean ± SEM for 3 samples
Ginsenoside compound K inhibits angiogenesis
Evaluation of the ability of CK to inhibit SPHK1 was also performed by analyzing sphingolipid metabolites. The concentrations of the substrates of SPHK1, sphingosine, and sphinganine in cell lysates were increased twofold following treatment of HUVEC with CK (Fig. 5A). By contrast, the concentration of both intracellular and extracellular S1P was reduced by CK treatment (Fig. 5B, C). Therefore, treatment of cells with CK resulted in accumulation of SPHK1 substrates and a reduction in the amount of S1P produced, via inhibition of SPHK1 activity. Sphingolipid rheostat (the ratio of sphingosine/S1P) was greatly increased from basal level of 1.64–5.20 by SKI II and to 6.13 by CK, suggesting that CK can change sphingolipid rheostat by inhibiting SPHK1 activity. To analyze the changes of total ceramides, the sphingolipid fraction in HUVEC were separated by HPTLC, and then cellular ceramides were recollected and alkaline-digested. The liberated sphingosine was derivatized with OPA and then quantified by HPLC. Compared with control cells treated with DMSO, treatment with CK resulted in a 1.5-fold increase in the amount of total ceramide (Fig. 6A), indicating propagation of pro-apoptotic signals by ceramide. As expected, sphingolipid rheostat (the ratio of ceramide/S1P) was also increased from 80.0 (basal) to 182.0 (SKI II) and to 212.9 (CK), respectively. The ceramide species in the lipid extracts were further analyzed by LC–MS/MS. A single ceramide with a saturated fatty acid chain, tetracosanoic acid (C24:0), was profoundly accumulated in HUVEC following CK treatment (Fig. 6B). Inhibitory effects of CK on MMP expressions In Fig. 1, GM6001, a broad spectrum MMP inhibitor, also attenuated the cell migration, and its effect was comparable to that observed for cells treated with 10 lg/mL CK. We also confirmed the ability of CK to reduce MMP2 and MMP9 expression in HUVEC (Fig. 7A). Cellular MMP expression was enhanced in cells treated with 100 nM S1P and treatment of cells for 24 h with CK (10 lg/mL in DMSO) reduced expression of MMP2 (Fig. 7B) and MMP9 (Fig. 7C) by 80 and 50 %, respectively.
Total ceramide (pmol / mg protein)
Change of sphingolipid composition by CK
A 1600 **
1400 1200 1000 800 600 400 200 0 DMSO
B Ceramide (pmol / mg protein)
ginsenosides does not originate from an increase in SPHK1 protein levels. Unexpectedly, treatment of cells with ginsenoside Rb1 increased expression of the SPHK1 protein twofold. Ginsenoside Rb1 is a firstly founded plant compound, inducing SPHK1 protein expression and SPHK1 activity.
SKI II
CK
250
**
DMSO CK
200
150
**
100
50
0 C16:0
C18:0
C20:0
C22:0
C24:1
C24:0
N-acyl chain length Fig. 6 Concentrations of total ceramide (A) and individual ceramide species (B) following treatment of HUVEC with CK (10 lg/mL in DMSO). Control cells were treated with DMSO. SKI II (1 lM) was used as a positive control. Data indicate the mean ± SEM for 3 samples
Discussion It was recently reported that S1P produced by SPHK1 promotes breast cancer progression by stimulating angiogenesis (Nagahashi et al. 2012). Furthermore, knockdown of SPHK1 expression by specific siRNA was shown to inhibit HUVEC growth and migration, whereas knockdown of SPHK2 expression had no effect (Yan et al. 2008). S1P avidly releases outside of cellular membrane and interacts with five G-protein coupled S1P receptors to generate multiple downstream signals including HUVEC migration. Therefore, the regulation of SPHK1 activity may represent a new therapeutic strategy for treatment of angiogenesisrelated diseases such as cancer. Ginsenoside CK, a major metabolite of ginsenoside Rb1 generated by intestinal bacteria, is observed in plasma, indicating that CK, a main bioactive metabolite of protopanaxadiol ginsenosides, is well absorbed in the intestine
123
K.-O. Shin et al.
A
B Fold increase in MMP2 expression (%)
Fig. 7 Western blot analysis showing the effect of 100 nM S1P and 1 lg/mL (CK-1), 5 lg/mL (CK-5), or 10 lg/mL (CK-10) CK on MMP2 and 9 expression. A typical western blot image showing expression of MMP and b-actin as a control (A). CK inhibits MMP2 (B) and MMP9 expression (C). GM6001 (25 lM) was used as a positive control. Data represent fold increase relative to control cells. Data indicate the mean ± SEM for 3 samples. **P \ 0.01, ***P \ 0.001
140 120 100 80
** **
60 40
***
***
20 0 DMSO S1P GM6001 CK-1
CK-5 CK-10
+ 100nM S1P
Fold increase in MMP9 expression ( % )
C
140 120 100 *
80 **
60 40
**
20 0 DMSO S1P GM6001 CK-1
CK-5 CK-10
+ 100nM S1P
following oral administration of ginseng (Akao et al. 1998). In this study, the effect of ten putative ginsenosides on SPHK1 activation was investigated by adding a SPHK1 substrate, 1 lM sphingosine, to the culture medium of HUVEC. CK exclusively inhibited both S1P production and SPHK1 activation in a dose-dependent manner (Fig. 1). Although we did not find any relationship between the number of sugar moieties and SPHK1 inhibition, CK contains a glucose moiety at C-20 and is therefore less
123
polar than the other ginsenosides tested. Inhibition of SPHK1 by CK (5–10 lg/mL) also resulted in reduced HUVEC viability (Fig. 3), which is likely due to the function of SPHK as a critical regulator of the sphingolipid rheostat maintaining the concentration balance between the pro-growth, anti-apoptotic messenger S1P, and pro-apoptotic ceramide and sphingosine (Maceyka et al. 2002). Interestingly, a reduction in SPHK1 expression was observed following CK treatment, suggesting post-transcriptional regulation by CK (Fig. 4). Ginsenoside Rb1
Ginsenoside compound K inhibits angiogenesis
induced a twofold increase in SPHK1 expression (Fig. 4) and S1P production was also increased by Rb1 (data not shown). The sphingolipid rheostat is a concept that refers to modulation of the balance and interconversion of S1P, ceramide, and sphingosine, which ultimately determines cell fate (Loh et al. 2011). Our data suggest that CK alters the sphingolipid rheostat balance by increasing the amounts of pro-apoptotic ceramide and sphingosine, and by decreasing intracellular and extracellular levels of the anti-apoptotic messenger S1P (Figs. 4, 5A). Secreted S1P may bind to S1PRs expressed on outer membrane surfaces in an autocrine or paracrine manner, and transduce signals to intracellular targets. Therefore, CK-induced reduction in S1P production will attenuate chemotactic HUVEC migration via the S1PRs-S1P axis. Notably, the amount of sphinganine, which is an intermediate in the de novo sphingolipid biosynthetic pathway, was also increased by CK treatment, suggesting new ceramide synthesis via the de novo pathway. Further analysis by LC–MS/MS demonstrated that a specific ceramide species containing a saturated fatty acid chain (C24:0) was significantly increased by CK treatment of cells (Fig. 5B). The increase in C24:0 ceramide indicates the activation of ceramide synthase type 2 proteins. When ceramide synthase type 2 is downregulated, ceramide-bearing palmitate (C16:0) is accumulated instead of C24:0-ceramide or C24:1-ceramide, and can induce autophagy but not apoptotic cell death (Spassieva et al. 2009). Therefore, the accumulation of C24:0-ceramide by CK is directly associated with cytotoxicity in HUVEC. S1P was previously shown to induce angiogenesis by activating extracellular signal-regulated kinases and p38 mitogen-activated protein kinase in a pertussis toxinsensitive manner (Lee et al. 1999). Our data show that 100 nM S1P enhanced HUVEC migration in a woundhealing assay by ninefold; however, this migration was potently inhibited by CK, which was effective at a dose as low as 10 lg/mL (Fig. 6B). In addition, GM6001 (10 lM) strongly attenuated HUVEC migration via inhibition of MMPs, which are thought to play a major role in cell growth, migration, angiogenesis, and apoptosis. Very recently, CK treatment was also shown to inhibit MMP2 and MMP9 expressions, which are both associated with angiogenesis and metastasis in hepatocellular carcinoma (Ming et al. 2011). In summary, the present study demonstrates that CK displays an anti-angiogenic property that is capable of perturbing S1P-induced angiogenesis via SPHK1 inhibition, resulting in disruption of the sphingolipid rheostat. CK, which can be produced by hydrolyzing the sugar moieties of the major ginsenosides Rb1, Rb2, and Rc, has more hydrophobic properties than the other ginsenosides
included in this study; the additional hydrophobicity of CK may contribute to suppression of SPHK1 activity and inhibition of MMPs. We do not exclude the possibility that CK-induced regulation of S1P receptors and their downstream signaling, or specific responses of protein kinase C isoenzymes associated with SPHK1 activity, may occur. Further investigation to identify the targets of CK will help to decipher the inhibitory mechanisms of S1P-induced HUVEC migration and may open a new possibility for anti-angiogenic therapy.
Conclusion We found that Ginsenoside CK regulates the migration of human umbilical vein endothelial cells which is an essential factor in cancer angiogenesis via the inhibition of sphingosine kinase-1. Also, ginsenoside CK increases proapoptotic lipids, sphingoid bases and ceramides, specifically accumulates C24:0 fatty acid acylation ceramide in HUVEC. Importantly, CK blocks S1P-induced HUVEC migration and the expressions of MMP enzymes which are responsible to cells motility in dose-dependent manner. Acknowledgments This study was supported by a NRF Grants funded by the Korean Government (MRC, 2010-0029483) and (KRF-2010-0025271).
References Akao, T., M. Kanaoka, and K. Kobashi. 1998. Appearance of compound K, a major metabolite of ginsenoside Rb1 by intestinal bacteria, in rat plasma after oral administration– measurement of compound K by enzyme immunoassay. Biological and Pharmaceutical Bulletin 21: 245–249. Berdyshev, E.V., I. Gorshkova, P. Usatyuk, S. Kalari, Y. Zhao, N.J. Pyne, S. Pyne, R.A. Sabbadini, J.G. Garcia, and V. Natarajan. 2011. Intracellular S1P generation is essential for S1P-induced motility of human lung endothelial cells: Role of sphingosine kinase 1 and S1P lyase. PLoS ONE 6: e16571. Cheung, L.W., K.W. Leung, C.K. Wong, R.N. Wong, and A.S. Wong. 2011. Ginsenoside-Rg1 induces angiogenesis via non-genomic crosstalk of glucocorticoid receptor and fibroblast growth factor receptor-1. Cardiovascular Research 89: 419–425. Jeong, A., H.J. Lee, S.J. Jeong, E.O. Lee, H. Bae, and S.H. Kim. 2010. Compound K inhibits basic fibroblast growth factor-induced angiogenesis via regulation of p38 mitogen activated protein kinase and AKT in human umbilical vein endothelial cells. Biological and Pharmaceutical Bulletin 33: 945–950. Joh, E.H., I.A. Lee, I.H. Jung, and D.H. Kim. 2011. Ginsenoside Rb1 and its metabolite compound K inhibit IRAK-1 activation–the key step of inflammation. Biochemical Pharmacology 82: 278–286. Kim Do, Y., H.D. Yuan, I.K. Chung, and S.H. Chung. 2009. Compound K, intestinal metabolite of ginsenoside, attenuates hepatic lipid accumulation via AMPK activation in human hepatoma cells. Journal of Agriculture and Food Chemistry 57: 1532–1537.
123
K.-O. Shin et al. Kluk, M.J., and T. Hla. 2002. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochimica et Biophysica Acta 1582: 72–80. Lee, H.U., E.A. Bae, M.J. Han, N.J. Kim, and D.H. Kim. 2005. Hepatoprotective effect of ginsenoside Rb1 and compound K on tert-butyl hydroperoxide-induced liver injury. Liver Int 25: 1069–1073. Lee, O.H., Y.M. Kim, Y.M. Lee, E.J. Moon, D.J. Lee, J.H. Kim, K.W. Kim, and Y.G. Kwon. 1999. Sphingosine 1-phosphate induces angiogenesis: Its angiogenic action and signaling mechanism in human umbilical vein endothelial cells. Biochemical and Biophysical Research Communications 264: 743–750. Leong, W.I., and J.D. Saba. 2010. S1P metabolism in cancer and other pathological conditions. Biochimie 92: 716–723. Leung, K.W., H.M. Ng, M.K. Tang, C.C. Wong, R.N. Wong, and A.S. Wong. 2011. Ginsenoside-Rg1 mediates a hypoxia-independent upregulation of hypoxia-inducible factor-1alpha to promote angiogenesis. Angiogenesis 14: 515–522. Loh, K.C., D. Baldwin, and J.D. Saba. 2011. Sphingolipid signaling and hematopoietic malignancies: To the rheostat and beyond. Anti-Cancer Agents in Medicinal Chemistry 11: 782–793. Maceyka, M., S.G. Payne, S. Milstien, and S. Spiegel. 2002. Sphingosine kinase, sphingosine-1-phosphate, and apoptosis. Biochimica et Biophysica Acta 1585: 193–201. Ming, Y., Z. Chen, L. Chen, D. Lin, Q. Tong, Z. Zheng, and G. Song. 2011. Ginsenoside compound K attenuates metastatic growth of hepatocellular carcinoma, which is associated with the translocation of nuclear factor-kappaB p65 and reduction of matrix metalloproteinase-2/9. Planta Medica 77: 428–433.
123
Nagahashi, M., S. Ramachandran, E.Y. Kim, J.C. Allegood, O.M. Rashid, A. Yamada, R. Zhao, S. Milstien, H. Zhou, S. Spiegel, and K. Takabe. 2012. Sphingosine-1-phosphate produced by sphingosine kinase 1 promotes breast cancer progression by stimulating angiogenesis and lymphangiogenesis. Cancer Research 72: 726–735. Park, E.K., Y.W. Shin, H.U. Lee, S.S. Kim, Y.C. Lee, B.Y. Lee, and D.H. Kim. 2005. Inhibitory effect of ginsenoside Rb1 and compound K on NO and prostaglandin E2 biosyntheses of RAW264.7 cells induced by lipopolysaccharide. Biological and Pharmaceutical Bulletin 28: 652–656. Sato, K., M. Mochizuki, I. Saiki, Y.C. Yoo, K. Samukawa, and I. Azuma. 1994. Inhibition of tumor angiogenesis and metastasis by a saponin of Panax ginseng, ginsenoside-Rb2. Biological and Pharmaceutical Bulletin 17: 635–639. Shida, D., K. Takabe, D. Kapitonov, S. Milstien, and S. Spiegel. 2008. Targeting SphK1 as a new strategy against cancer. Current Drug Targets 9: 662–673. Spassieva, S.D., T.D. Mullen, D.M. Townsend, and L.M. Obeid. 2009. Disruption of ceramide synthesis by CerS2 down-regulation leads to autophagy and the unfolded protein response. The Biochemical Journal 424: 273–283. Xu, T.M., Y. Xin, M.H. Cui, X. Jiang, and L.P. Gu. 2007. Inhibitory effect of ginsenoside Rg3 combined with cyclophosphamide on growth and angiogenesis of ovarian cancer. Chinese Medical Journal (English) 120: 584–588. Yan, G., S. Chen, B. You, and J. Sun. 2008. Activation of sphingosine kinase-1 mediates induction of endothelial cell proliferation and angiogenesis by epoxyeicosatrienoic acids. Cardiovascular Research 78: 308–314.
