Astragaloside IV improves homocysteine-induced acute phase endothelial dysfunction via antioxidation

H yperhomocysteinemia is an independent risk factor for cardiovascular disease.1)Elevation of homocysteine (H CY)in blood plasma (Ͼ15 mmol/l), resulting from metabolic ab-normalities in H CY remethylation or transsulfuration, is common in subjects with cardiovascular disease. In the last decade, insights from the animal models have established that experimental hyperhomocysteinemia produces several phenotypic effects, including endothelial dysfunction. The most common abnormality of the dysfunction observed dur-ing hyperhomocysteinemia is impaired relaxation of conduit vessels (aorta, carotid arteries, renal, or pulmonary) or im-paired dilatation of micro vessels (mesenteric, cremasteric,coronary, skeletal, or cerebral arterioles) in response to the endothelium-dependent vasodilators in rat, mice, monkey, or guinea pigs, whereas no impairment was seen in response to exogenous nitric oxide (NO) donors.2—8)Therefore, it was suggested that the underlying mechanism by which endo-thelium derived NO, which then diffuses into the adjacent vascular muscle layer and causes its relaxation, may play an important role in mediating these deleterious effects on endothelial function in hyperhomocysteinemia.2,5,8)Further-more, the emerging evidence suggest that hyperhomocys-teinemia leads to increased vascular superoxide production through the upregulation and activation of nicotinamide ade-nine dinucleotide phosphate (NADPH) oxidase, and that re-active oxygen species (ROS) generated by NADPH oxidase contribute to the vascular phenotype.4,9—11)Radix Astragali is the dried root of Astragalus mem-branaceus (F ISCH .) B GE .var. mongholicus (B GE .) H SIAO or As-tragalus membranaceus (F ISCH .) B GE . It has been widely usedin Chinese medicine since ancient times for the improvement of immune disorders and management of cardiovascular dis-eases with an excellent safety.12)Research suggests that this herb was able to inhibit free radicals, decreases lipid peroxi-dation, and increases antioxidant enzymes.13)Many medici-nally active compounds have been isolated from this plant,including polysaccharides, flavones and astragalosides. As-tragalosides is the major active component extracted from the root of Astragalus membranaceus .12)Previous studies from our laboratory demonstrated that crude astragalosides frac-tion can potently protect endothelium-dependent relaxation against the acute injury from HCY through nitric oxide regu-latory pathways, in which antioxidation played a key role.14)Astragaloside IV (AST -IV), 3-o -beta-D -xylopyranosyl-6-o -beta-D -glucopyranosyl-cycloastragenol (Fig. 1), was a small molecular saponin. AST -IV can exert multipotent activities under pathophysiological conditions, such as anti-hyperten-sion,15)positive inotropic action,16)anti-inflammation,17)and anti-infarction.18)Up to now, however, whether AST -IV has an exact protective effect on vessels in hyperhomocysteine-mia and the mechanism are not known. The aim of the pres-ent study is to examine the effect of AST -IV on vasomotor dysfunction induced by HCY in rat aorta and explore the un-derlying mechanism.April 2010641Astragaloside IV Improves Homocysteine-Induced Acute Phase Endothelial Dysfunction via AntioxidationLi-Hong Q IU ,a Xian-Ji X IE ,b and Bi-Qi Z HANG *,aaDepartment of Cardiology, The First Affiliated Hospital, College of Medicine, Zhejiang University; and b Department of Pharmacy, The First Affiliated Hospital, College of Medicine, Zhejiang University; Qingchun road 79#, Hangzhou 310003,P . R. China.Received November 6, 2009; accepted January 22, 2010; published online January 27, 2010Astragaloside IV , the major active component extracted from Astragalus membranaceus , exerts multipotentactivities under pathophysiological conditions. Hyperhomocysteinemia, an independent risk factor for cardiovas-cular disease, induces oxidative stress leading to endothelial dysfunction. We investigated the effect of astragalo-side IV on acute phase endothelial dysfunction induced by homocysteine. In a concentration-dependent manner,endothelial dysfunction was induced by homocysteine. In organ bath experiment using rat aortic rings, treat-ment with astragaloside IV resulted in an improvement of the impaired endothelium-dependent vasorelaxation by homocysteine as reflected by the higher maximal vasorelaxation to acetylcholine. However, the presence of N w -nitro-L -arginine methyl ester hydrochloride could abolish the protective effect of astragaloside IV on homo-cysteine-induced vasomotor dysfunction. In human umbilical vein endothelial cells culture experiment, exposure to astragaloside IV significantly ameliorated the homocysteine-induced inactivation of nitric oxide–nitric oxide synthase signal pathway via reducing oxygen species and increasing the activity of superoxide dismutase. Addi-tionally, pretreatment with superoxide dismutase showed a similar effect to astragaloside IV on attenuation of the homocysteine-induced endothelial dysfunction. These data support the view that astragaloside IV might be ad-vantageous in the treatment of endothelial dysfunction induced by disturbed nitric oxide–nitric oxide synthase pathway due to oxidative stress in hyperhomocysteinemia.Key words astragaloside IV; homocysteine; oxidation; nitric oxide; endothelial functionBiol. Pharm. Bull.33(4) 641—646 (2010)© 2010 Pharmaceutical Society of Japan∗To whom correspondence should be addressed.e-mail: wzzbq@Fig.1.Chemical Structure of Astragaloside IV (C41H 68O14, Molecular Weight: 784)MATERIALS AND METHODSMaterials The AST-IV used in this study was purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China), with a high purity 99% by H PLC analysis. Phenylephrine (PE), acetylcholine(ACH), H CY, sodium nitroprusside (SNP), Nw -nitro-L-argi-nine methyl ester hydrochloride (L-NAME) and superoxide dismutase (SOD)-polyethylene were from Sigma Chemical Co. (St. Louis, MO, U.S.A.). 5-(6)-Chloromethyl-2Ј,7Ј-dichloro-dihydrofluorescein diacetate (CM-H2DCF-DA) wasfrom Molecular Probes Inc. (Eugene, U.S.A.). All chemicals were of the highest purity available.Organ Chamber Experiment Animals used in this study were experimentally naive male Sprague–Dawley (SD) rats obtained from Experiment Animal Center of Zhejiang Academy of Medical Sciences, China. The investigation con-forms to the Guide for the Care and Use of Laboratory Ani-mals published by the U.S. National Institutes of H ealth (NIH Publication No. 85-23, revised 1996).Rats were killed by cervical dislocation. Thoracic arterial rings were prepared according to the method described previ-ously by Zhang et al.19)For isometric force recording, aortic rings were mounted between two stainless steel hooks and suspended in the organ bath containing Krebs’ buffer, com-posed of (m M): NaCl, 118; KCl, 4.7; MgSO4·7H2O, 1.2;KH2PO4, 1.2; CaCl2, 2.5; NaH CO3, 25; and glucose, 11; at37°C bubbled with 95% O2ϩ5% CO2(pH 7.4). The tensionof the aortic ring was monitored by a force transducer (Nan-jing Medease Science and Technology Co., Ltd., China), connected to MedLab 5.0v, a computer-based data acquisi-tion system (Nanjing Medease Science and Technology Co., Ltd., China). After equilibrium for 60min at 2.0g resting tension, then rings were challenged with KCl (6.0ϫ10Ϫ2 mol/l) repeatedly until a reproducible maximal contractile re-sponse was obtained. The integrity of the endothelium was assessed in all preparations by determining the ability of ACH (10m mol/l) to induce more than 80% relaxation of rings pre-treated with PE (1m mol/l). Then rings was serially washed and used for the following experiment.Bioassay of Vasoreactivity To assess the effect of HCY on vasorelaxation, rings were incubated with Krebs buffer for 30min and then exposed to H CY (0.1—3mmol/l) for 60min. After the incubation, the endothelium-dependent or independent response curve was evoked by ACH (0.01—10m mol/l) or SNP (0.01—10m mol/l) to PE (1m mol/l)-in-duced contraction.In the subsequent experiments, the effect of AST-IV on va-somotor dysfunction induced by HCY in rat aorta and the un-derlying mechanism were determined. Some rings were pre-treated with different concentrations of AST-IV (5, 10, 50, and 100m g/ml) for 30min; some rings were preincubated with SOD (0.6, 1.2, and 1.8KU/l) alone or coincubated with SOD (1.2KU/l) and AST-IV (100m g/ml) for 30min; and the other rings were exposed to L-NAME (100m mol/l) with or without AST-IV (100m g/ml) for 30min. Continuously, HCY (1mmol/l) were administrated to co-incubate the rings in each group for another 60min. Then, the rings were recon-tracted with 1.0m mol/l PE, and endothelium-dependent re-sponses to ACH were repeated.Cellular Experiments H uman umbilical vein endothe-lial cells (HUVECs) were isolated and cultured as previouslydescribed.20)HUVECs from passages 3—4 were used in this study. For cell culture experiment, HUVECs were incubated with medium, AST-IV (5, 10, 50, and 100m g/ml), SOD (1.2, and 1.8KU/l) alone, and AST-IV (100m g/ml) plus SOD (1.2KU/l) for 30min before treatment with H CY (0.1 or 1mmol/l) for 60min. After above treatment, the supernatant and cultured cells were collected respectively. Cells were counted with microscopy and cellular protein concentrations were determined using a protein assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).Assay of NO Level The amount of NO released by cells was determined using a NO assay kit according to the manu-facturer’s protocol (Nanjing Jiancheng Bioengineering Insti-tute, Nanjing, China). The amount of NO metabolites (nitrate and nitrite), which were more stable than NO, was measured by nitric acid reductase method.21)Assay of NO Synthase (NOS) Activity The collected cells were disrupted with ultrasonic wave. NOS enzymatic activity in the cultured cells was determined from the rate of conversion of L-arginine to L-citrulline using a NOS assay kit according to the manufacturer’s protocol (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).22) Measurement of Intracellular ROS Levels For intra-cellular fluorescence measurement of ROS, some H UVECs were loaded with the probe CM-H2DCF-DA for 30min at37°C. CM-H2DCF-DA fluorescence was monitored using a fluorescence microscope (Nikon TE2000, Japan) (excitation 495nm, emission 520nm). The fluorescent intensities were recorded and analyzed by a charge-coupled device Camera (CoolSNAP HQ; Nippon Roper, Chiba, Japan) with an image analysis system (MetaMorph; Nippon Roper).Assay of SOD Activity SOD activity in the cultured cells was determined by inhibition of nitroblue tetrazolium reduction due to superoxide anion generation by xanthine–xanthine oxidase system using a SOD activity assay kit ac-cording to the manufacturer’s protocol (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).23)Statistical Analysis The ACH-induced maximal relax-ation (Emax) in aortic rings was calculated as a percentage of the contraction in response to PE (1m mol/l). The half maxi-mum effective concentration (EC50) was designated as the concentration of ACH that induced 50% of maximum relax-ation from the contraction elicited by PE (1m mol/l). These values were determined by nonlinear regression of the log concentration–response curves. Sensitivity was expressed aspD2(Ϫlog EC50). All values are expressed as meanϮS.E.M. Statistical analysis was performed with one-way ANOV A followed by Newman–Keuls test. Differences were accepted as statistically significant at p values Ͻ0.05 (Graph Pad Prism).RESULTSOrgan Chamber Experiment In each group, a sus-tained vascular contraction with a peak tension of about 2.65Ϯ0.31g was induced by PE (1m mol/l). In any group, no difference in relaxation responses to ACH (10m mol/l) in aor-tic rings was observed at the initial experiment.Compared with the control group (Emax: 98.37Ϯ0.65%,pD2: 6.88Ϯ0.05), H CY (0.1—3mmol/l) treatment inhibited642V ol. 33, No. 4endothelium-dependent relaxation evoked by ACH , and the E max fell to 84.90Ϯ1.01%, 76.99Ϯ1.56%, 65.20Ϯ1.50%, and 51.18Ϯ1.89%, with a reduction of pD 2(6.74Ϯ0.06, 6.62Ϯ0.05, 6.53Ϯ0.05, and 6.50Ϯ0.05). There was no significant difference of pD 2was found between 1mmol/l and 3mmol/l HCY group, although E max exerted a further reduction (Fig.2). Moreover, H CY (0.1—3mmol/l) failed to deteriorate the endothelium-independent relaxation to SNP (data not shown). Therefore, 1mmol/l HCY treatment-induced inhibi-tion of endothelium-dependent relaxation to ACH was used in the subsequent experiment as a control.Treatment with AST -IV (50 or 100m g/ml) was able to improve the inhibition of endothelium-dependent relaxation induced by H CY (1mmol/l). E max and pD 2were shown as (74.67Ϯ2.10% or 81.73Ϯ1.87% versus 65.20Ϯ1.50%,p Ͻ0.05) and (6.78Ϯ0.06 or 6.87Ϯ0.06 versus 6.53Ϯ0.05,p Ͻ0.05), respectively. Additionally, due to the presence of L -NAME (100m mol/l), the inhibitor of nitric oxide synthase (NOS), AST -IV (100m g/ml) failed to show any ameliorating effect on endothelium-dependent vasomotor response caused by HCY (1mmol/l) (Fig. 3).To confirm the oxidative effect of H CY on endothelium-dependent relaxation, SOD was used. With the increase of the concentration of SOD in organ bath (0.6, 1.2, and 1.8KU/l), a more significant effect was observed (E max :70.59Ϯ1.72%, 82.18Ϯ1.26%, and 93.51Ϯ1.30% versus 65.20Ϯ1.50%; pD 2: 6.68Ϯ0.06, 6.91Ϯ0.06, and 9.94Ϯ0.06versus 6.53Ϯ0.05) and the inhibitory effect of HCY was al-most completely abolished. In an analogous manner, com-bined treatment of the rings with AST -IV (100m g/ml) and SOD (1.2KU/l) totally reversed the H CY -induced impair-ment of endothelium-dependent relaxation (E max : 95.10Ϯ2.04%; pD 2: 6.91Ϯ0.06), which was very similar to the ef-fect of 1.8KU/l SOD group (Fig. 4).Cellular Experiments Incubation of H UVECs with H CY (0.1 or 1mmol/l) for 60min resulted in the reduction both of NO content and NOS activity. Treatment with AST -IV (50 and 100m g/ml) was able to attenuate the inhibition both of NO content and NOS activity. Furthermore, SOD (1.8KU/l) exerted more significant ameliorating effects onthese parameters than AST -IV (100m g/ml). Meanwhile,combined treatment with SOD (1.2KU/l) and AST -IV (100m g/ml) induced a similar effect with SOD (1.8KU/l)(Figs. 5, 6).Compared to the control group, treatment with H CY (1mmol/l) for 60min significantly increased the ROS level and decreased the SOD activity in H UVECs. AST -IV (10,50, and 100m g/ml) markedly prevented this accumulation of ROS and reduction of SOD activity. H owever, AST -IV showed a lower potency than SOD (Figs. 7, 8).April 2010643Fig.2.Cumulative Concentration–Response Curves to Acetylcholine in Phenylephrine-Precontracted Endothelium-Intact Aorta Rings Incubated with Homocysteine (HCY) for 60minData are expressed as means ϮS.E.M. n ϭ10. ∗∗p Ͻ0.01 vs.control group.Fig.3.Effect of Astragaloside IV (AST -IV) and/or N w -Nitro-L -arginine Methyl Ester H ydrochloride (L -NAME) on Cumulative Concentration–Re-sponse Curves to Acetylcholine in Endothelium-Intact Aorta Rings Incu-bated with Homocysteine (HCY) for 60minData are expressed as means ϮS.E.M. n ϭ10. ∗∗p Ͻ0.01 vs.1mmol/l HCY group.Fig.4.Effect of Superoxide Dismutase (SOD) with or without Astragalo-side IV (AST -IV) on Cumulative Concentration–Response Curves to Acetyl-choline in Endothelium-Intact Aorta Rings Incubated with H omocysteine (HCY) for 60minData are expressed as means ϮS.E.M. n ϭ10. ∗p Ͻ0.05, ∗∗p Ͻ0.01 vs.1mmol/l HCY group.DISCUSSIONThe present study indicated that AST -IV improved the endothelium dysfunction induced by H CY . Furthermore, a favorable mechanism on activation of NO pathway through the antioxidant defense has been postulated.H yperhomocysteinemia is considered an established risk factor for atherosclerosis. Increased HCY levels are found in 40% of patients with coronary, cerebral or peripheral artery diseases, and only in 15% of healthy individuals.24)Accumu-lating evidence suggests a significant role of altered cellular redox reactions in the vascular phenotype of hyperhomocys-teinemia.25)Redox effects are particularly important in medi-ating the adverse effects of hyperhomocysteinemia on the endothelium, leading to loss of endothelium-derived nitric oxide and vasomotor dysfunction.Consistent with the previous report, the endothelial speci-ficity of HCY was clearly demonstrated in that a significant effect on endothelium-dependent relaxation to ACH was seen both in our present and previous studies.14,26)The result of this study that HCY did not alter the relaxation responses to SNP and the other observation that SNP-induced relaxation in endothelium-denuded tissues was also unaffected by HCY 27)would tend to support the hypothesis that the HCY -induced vasomotor dysfunction occurs within endothelial rather than vascular smooth muscle cells.A reduction in both bioavailability and production of endothelium-derived NO is the hallmark of endothelial dys-function. NO produced through the calcium-dependent acti-vation of endothelial nitric oxide synthase in endothelial cells diffuses into the adjacent vascular muscle layer, and causes its relaxation by activity of cGMP pathway.28)In agree-ment with previous reports,21,29)our data suggested that both NOS activity and NO formation were inhibited by H CY in H UVECs. Thus, we have shown H CY impaired endothelial dysfunction mediated by NOS-NO pathway.Redox reactions resulting in increased vascular levels of ROS may play an important role in mediating endothelial dysfunction, particularly the loss of endothelium-derived NO, during hyperhomocysteinemia. The present study indi-cated that HCY significantly increased intracellular ROS lev-644Vol. 33, No. 4Fig.5.Effects of Astragaloside IV (AST -IV) and/or Superoxide Dismu-tase (SOD) on Content of Nitric Oxide (NO) Released from Human Umbili-cal Vein Endothelial Cells (HUVECs) Treated with Homocysteine (HCY)HUVECs were pretreated with AST -IV (5—100m g/ml) and/or SOD (1.2, 1.8KU/l)for 30min and then stimulated with 1mmol/l HCY for 60min. The content of NO in the supernatant was measured by nitric acid reductase method. The values are ex-pressed as a percentage of the control group from five individual experiments (means ϮS.E.M.). ∗p Ͻ0.05, ∗∗p Ͻ0.01, vs.control group; ϩp Ͻ0.05, ϩϩp Ͻ0.01, vs.1mmol/l HCY group; #p Ͻ0.05 ##p Ͻ0.01, vs.100m g/ml AST-IV group.Fig.6.Effects of Astragaloside IV (AST -IV) and/or Superoxide Dismu-tase (SOD) on the Activity of Nitric Oxide Synthase (NOS) in Human Um-bilical Vein Endothelial Cells (H UVECs) Treated with H omocysteine (HCY)HUVECs were pretreated with AST -IV (5—100m g/ml) and/or SOD (1.2, 1.8KU/l)for 30min and then stimulated with 1mmol/l HCY for 60min. NOS enzymatic activity in the cultured cells was determined from the rate of conversion of L -arginine to L -cit-rulline. The values are expressed as a percentage of the fluorescence intensity of the control group from five individual experiments (means ϮS.E.M.). ∗∗p Ͻ0.01 vs.control group; ϩϩp Ͻ0.01 vs.1mmol/l HCY group; #p Ͻ0.05 vs.100m g/ml AST-IV group.Fig.7.Effects of Astragaloside IV (AST -IV) and/or Superoxide Dismu-tase (SOD) on H omocysteine (H CY)-Induced Reactive Oxygen Species (ROS) ProductionH uman umbilical vein endothelial cells (H UVECs) were pretreated with AST -IV (5—100m g/ml) and/or SOD (1.2, 1.8KU/l) for 30min and then stimulated with HCY for 60min. The probe 5-(6)-chloromethyl-2Ј,7Ј-dichloro-dihydrofluorescein diacetate (CM-H 2DCF-DA) was used to detect the ROS production. The values are expressed as a percentage of the fluorescence intensity of the control group from five individual ex-periments (means ϮS.E.M.). ∗∗p Ͻ0.01, vs.control group; ϩp Ͻ0.05, ϩϩp Ͻ0.01, vs.1mmol/l HCY group; ##p Ͻ0.01, vs.100m g/ml AST-IV group.Fig.8.Effects of Astragaloside IV (AST -IV) on the Activity of Super-oxide Dismutase (SOD) in H uman Umbilical Vein Endothelial Cells (HUVECs) Treated with Homocysteine (HCY)HUVECs were pretreated with AST -IV (5—100m g/ml) for 30min and then stimu-lated with 1mmol/l HCY for 60min. SOD activity was determined by inhibition of ni-troblue tetrazolium reduction due to superoxide anion generation by xanthine–xanthine oxidase system. The values are expressed as units per milligram protein (means ϮS.E.M.). ∗p Ͻ0.05, ∗∗p Ͻ0.01 vs.control group; ϩp Ͻ0.05, ϩϩp Ͻ0.01 vs.1mmol/l HCY group.els and decreased the SOD activity in HUVECs, which was consistent with the reported by Lin et al.30)Treatment with SOD, an antioxidant enzyme, reversed the H CY-impaired endothelium-dependent vasomotor responses. Other re-searchers have suggested that SOD showed similar protective effects on injury aorta, pulmonary artery, or skeletal muscle arterioles.27)During these years, natural products with antioxidant activity have drawn the most attention. Astragalus mem-branaceus is widely used for prevention of ROS-mediated in-jury in pathological situation through its antioxidant proper-ties.13,31—33)In an in vitro study, it was demonstrated by elec-tron paramagnetic resonance imaging technique that Astra-galus membranaceus can potently inhibit ROS produced by the dimethyl sulphoxide system and scavenge over 90% of ROS.34)Recently, the emerging experiments reported that a herbal formulation, comprising Astragalus membranaceus, improved the pancreatic beta cell function via the antioxida-tive effect.35,36)Furthermore, total saponins extracted from Astragalus membranaceus had the effects on enhancing free radical removal and decreasing lipid peroxidation, thus pre-venting the calcium overload in isoproterenol-treated car-diomyocytes.37)According to our previous study, it has been shown that total saponins isolated from Astragalus mem-branaceus ameliorated endothelium-dependent relaxation in-jured by HCY via scavenging free radical species.14) Astragalus membranaceus contains a series of cycloartane triterpene glycosides denoted astragalosides I—VII (sapo-nins), which are based on the aglycone cycloastragenol and contain from one to three sugars attached at the 3-, 6-, and 25-positions.12)AST-IV is regarded as the main criterion of for quality control of Astragalus membranaceus in the Phar-macopoeia of the People’s Republic of China.38)There is emerging evidence that AST-IV has the biological property of eliminating ROS, due to which it protects cardiomyocytes from oxidative stress-mediated injury under hypoxic condi-tions, delays the aging effect in rats treated by hydrocorti-sone, and protects coxsackievirus B3-induced murine my-ocarditis.33,39,40)In the present study, we have shown that AST-IV is effective in protecting endothelium-dependent va-somotor function from injury by HCY, and this ameliorating effect of AST-IV was attenuated by L-NAME, a NOS in-hibitor. In order to clarify whether this effect of AST-IV was exerted through NO pathway, we tested the influence of AST-IV on NO pathway in endothelial cells. The result showed that AST-IV was able to improve content of NO and activity of NOS. However, whether the beneficial effects of AST-IV on vascular endothelial dysfunction injured by HCY are re-lated with scavenging ROS remain largely unknown. It is worth noting that AST-IV not only scavenged OHϪgenerated from the Fenton reaction in an in vitro reaction system,41)but also entered cells and upregulation of SOD content and activ-ity.40)This study provided detailed information that AST-IV inhibited the production of ROS and improved the SOD ac-tivity in endothelial cells incubated by HCY. Moreover, SOD, a scavenger of superoxide anions showed a similar but higher maximum effect than AST-IV.In summary, it was demonstrated for the first time that AST-IV has the ability to regulate NO pathway impaired by HCY through antioxidant defense. In light of these findings, it could lead to the development of a new therapy for the vas-cular lesion in the pathological condition characterized by elevation of HCY in blood plasma.Acknowledgements This study supported by grants from Province Administration of Traditional Chinese Medi-cine, Zhejiang, P.R. China (Project Number: 2009CA055). REFERENCES1)Fowler B., Semin. V asc. Med., 5, 77—86 (2005).2)Eberhardt R. T., Forgione M. A., Cap A., Leopold J. A., Rudd M. A.,Trolliet M., H eydrick S., Stark R., Klings E. S., Moldovan N. I., Y aghoubi M., Goldschmidt-Clermont P. J., Farber H. W., Cohen R., Loscalzo J., J. Clin. Invest., 106, 483—491 (2000).3)Dayal S., Bottiglieri T., Arning E., Maeda N., Malinow M. R., SigmunC. D., Heistad D. D., Faraci F. M., Lentz S. R., Circ. Res., 88, 1203—1209 (2001).4)Bagi Z., Ungvari Z., Koller A., Arterioscler. Thromb. V as. Biol., 22,28—33 (2002).5)Dayal S., Arning E., Bottiglieri T., Boger R. H., Sigmund C. D., FaraciF. M., Lentz S. R., Stroke, 35, 1957—1962 (2004).6)Jacobsen D. W., Catanescu O., Dibello P. M., Barbato J. C., Clin.Chem. Lab. Med., 43, 1076—1083 (2005).7)Symons J. D., Rutledge J. C., Simonsen U., Pattathu R. A., Am. J.Physiol., 290, H181—H191 (2006).8)Tasatargil A., Sadan G., Karasu E., Pulm. Pharmacol. Ther., 20, 265—272 (2007).9)Ungvari Z., Csiszar A., Edwards J. G., Kaminski P. M., Wolin M. S.,Kaley G., Koller A., Arterioscler. Thromb. V asc. Biol., 23, 418—424 (2003).10)Shukla N., Koupparis A., Jones R. A., Angelini G. D., Persad R., Je-remy J. Y., Eur. J. Pharmacol., 531, 201—208 (2006).11)Suematsu N., Ojaimi C., Kinugawa S., Wang Z., Xu X., Koller A.,Recchia F. A., Hintze T. H., Circulation, 115, 255—262 (2007).12)No authors listed, Altern. Med. Rev., 8, 72—77 (2003).13)Ko J. K., Lam F. Y., Cheung A. P., World J. Gastroenterol., 11, 5787—5794 (2005).14)Zhang B. Q., Hu S. J., Qiu L. H., Zhu J. H., Xie X. J., Sun J., Zhu Z.H., Xia Q., Bian K., V ascul. Pharmacol., 46, 278—285 (2007).15)Zhang W. D., Zhang C., Wang X. H., Gao P. J., Zhu D. L., Chen H.,Liu R. H., Li H. L., Planta Med., 72, 621—666 (2006).16)Li Z. P., Cao Q., Acta Pharmacol. Sin., 23, 898—904 (2002).17)Zhang W. J., Hufnagl P., Binder B. R., Wojta J., Thromb. Haemost., 90,904—914 (2003).18)Luo Y., Qin Z., Hong Z., Zhang X., Ding D., Fu J. H., Zhang W. D.,Chen J., Neurosci. Lett., 363, 218—223 (2004).19)Zhang B. Q., Hu S. J., Qiu L. H., Shan Q. X., Sun J., Xia Q., Bian K.,Biol. Pharm. Bull., 28, 1450—1454 (2005).20)Lin R., Liu J., Peng N., Y ang G., Gan W., Wang W., Biol. Pharm. Bull.,28, 1630—1634 (2005).21)Jin L., Caldwell R. B., Li-Masters T., Caldwell R. W., J. Physiol. Phar-macol., 58, 191—206 (2007).22)Ferro A., Queen L. R., Priest R. M., Xu B., Ritter J. M., Poston L.,Ward J. P. T., Br. J. Pharmacol., 126, 1872—1880 (1999).23)Liu H. T., Li W. M., Xu G., Li X. Y., Bai X. F., Wei P., Yu C., Du Y. G.,Pharmacol. Res., 59, 167—175 (2009).24)Welch G. N., Upchurch G. J. R., Loscalzo J., Ann. N.Y. Acad. Sci., 811,48—58 (1997).25)Dayal S., Lentz S. R., Antioxid. Redox Signal., 9, 1899—1909 (2007).26)Fu Y. F., Xiong Y., Fu S. H., J. Cardiovasc. Pharmacol., 42, 566—572(2003).27)Lang D., Kredan M. B., Moat S. J., Hussain S. A., Powell C. A., Bel-lamy M. F., Powers H. J., Lewis M. J., Arterioscler. Thromb. V asc.Biol., 20, 422—427 (2000).28)Linder L., Kiowski W., Buhler F. R., Luscher T. F., Circulation, 81,1762—1767 (1990).29)Jiang X., Y ang F., Tan H., Liao D., Bryan R. M. Jr., Randhawa J. K.,Rumbaut R. E., Durante W., Schafer A. I., Y ang X., Wang H., Arte-rioscler. Thromb. V asc. Biol., 25, 2515—2521 (2005).30)Lin R., Liu J., Gan W., Ding C., Basic Clin. Pharmacol. Toxicol., 101,197—202 (2007).31)Hei Z. Q., Huang H. Q., Zhang J. J., Cheng B. X., Li X. Y., World J.April 2010645。