Therapeutic window of Qingkailing injection for focal cerebral ischemia/reperfusion injury＊☆

Fafeng Cheng, Wenting Song, Xianggen Zhong, Yi Lu, Shaoying Guo, Dong Wang,Weipeng Zhao, Qingguo Wang

1College of Basic Medicine, Beijing University of Chinese Medicine, Beijing 100029, China

2Xiyuan Hospital of China Academy of Chinese Medical Sciences, Beijing 100091, China

lNTRODUCTlON

The committee of the Stroke Therapy Academic Industry Roundtable proposed that studies of drugs for brain ischemia should pay attention to their pharmaceutical window[1-2]. In clinical practice, treatment for stroke patients is always delayed after onset,and the time at which treatment is administered is highly correlated with therapeutic effect.

Qingkailing (QKL) injection was originally prepared by the Beijing University of Chinese Medicine in the 1970s, by modifying a traditional Chinese medicine,Angongniuhuang pills, composed of baicalin,jasminoidin, cholic acid and Cornu Bubali[3].

It has been extensively used to treat the acute stages of cerebrovascular disease. A previous meta-analysis demonstrated that QKL injection significantly improved neurological function compared with a control group[4].

Animal experiments have shown that QKL injection can promote endothelial nitric oxide synthase expression, reduce calcium overload, regulate matrix metallopeptidase 9 expression and inhibit inflammation in a murine model of cerebral ischemia/reperfusion[5-8].

The present study first aimed to establish a mouse model of middle cerebral artery occlusion (MCAO). These animals were then injected with QKL to verify whether this treatment has a prolonged therapeutic window for focal cerebral ischemia/reperfusion in mice.

RESULTS

Quantitative analysis and grouping of experimental animals

We used 48 Kunming mice to establish MCAO models. After eliminating animals that died or in which the model did not establish properly, 36 were randomly assigned to four groups of equal size, and injected via the tail vein with saline (0.9 mL/100 g) or 1.5, 3 or 6 mL/kg QKL. In therapeutic window experiments, 80 Kunming mice were equally and randomly assigned to model, 0, 1, 3, 4, 6, 9, and 12 hours groups. These animals were injected with 3 mL/kg QKL (or 0.9 mL/100 g saline in model group). In anti-oxidative neuroprotection experiments, animals were divided into sham-surgery, model and QKL(3 mL/kg) groups, with 16 animals in each group. All mice were subjected to MCAO,and those that died, or in which the model failed to establish properly, were eliminated.

After 24 hours, five mice from each group were used to prepare sections of frozen whole brain for use in terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and in reactive oxygen species (ROS) detection experiments. Brain homogenate from seven animals in each group was collected to determine NADPH oxidase, superoxide dismutase (SOD) and malondialdehyde (MDA) levels.

Dose-response of QKL for focal cerebral ischemia/reperfusion

Dose-response effects on neurological function

(Figure 1)

Figure 1 Therapeutic efficacy of Qingkailing (QKL)injection in mice undergoing middle cerebral artery occlusion. QKL injection at different doses significantly improved focal neurological function (A). Moderate and low dose QKL injection also significantly improved general neurological function (B). The scores of focal neurological function and general neurological function were 0 in normal mice. High scores represent severe injury. Five brain coronal sections, 2 mm thick, were selected for tetrazolium chloride staining. Red stain represents normal tissues;white represents the infarct region (C). The infarct volume was quantified as a percentage of total volume, with large infarcts representing more severe injury. The percent of infarct volume was significantly less in QKL injection groups compared with model group (D). aP ＜ 0.05, bP ＜ 0.01, vs.model group. Data are expressed as mean ± SEM, n = 9.

Mice with focal cerebral ischemia/reperfusion were injected with different doses of QKL. Prior to MCAO,scores of focal neurological function and general neurological function were 0. Mice developed neurological functional injury after MCAO for 24 hours, but this neurological dysfunction was ameliorated in animals injected with QKL compared with the model group (P ＜ 0.05 or P＜ 0.01). The group receiving the moderate QKL dose (3 mL/kg) exhibited the greatest improvement (P ＜ 0.01;Figure 1A).

Moderate and low dose QKL injection was able to restore general neurological function (P ＜ 0.01 and P ＜ 0.05,respectively). However, no such effect was evident following high dose QKL injection (Figure 1B).

Dose-response effects on infarct volume

Infarct foci were evident in the model group but, compared to this group, the infarction size was reduced by 54.4% (P ＜ 0.01), 63.5% (P ＜ 0.01) and 22.9% (P ＜ 0.05)in high, moderate, and low dose QKL injection groups,respectively (Figures 1C, D).

Dose-response effects on brain cell apoptosis

Cellular apoptosis is an important mechanism of nerve injury after brain ischemia[9-10]. The TUNEL method was used to investigate whether an anti-apoptotic effect was involved in the neuroprotection by QKL. TUNEL-positive neurons were not detected in normal brain tissues. A large number of TUNEL-positive neurons were observed in the prefrontal cortex and hippocampal CA1 region 24 hours post-ischemia. Following QKL injection (3 mL/kg),the number of TUNEL-positive neurons was reduced by 79% in hippocampal CA1 and 59% in prefrontal cortex compared with MCAO (Figure 2).

Figure 2 Effect of Qingkailing (QKL) injection on cell apoptosis in hippocampus and prefrontal cortex of mice(TUNEL staining). After middle cerebral artery occlusion for 24 hours, apoptotic cells (A, × 400) were detected in the prefrontal cortex and hippocampal CA1 region.Apoptotic cells were labeled with red fluorescence. Five animals were selected from each group; three sections were selected from each site; and five 400-fold fields of view were randomly selected from each section to quantify the mean of positive cells. Results are expressed as mean ± SEM (B). aP ＜ 0.01, vs. sham-surgery group;bP ＜ 0.01, vs. model group. TUNEL: Transferase dUTP nick end labeling.

Estimation of effective time window for QKL injection after focal cerebral ischemia/reperfusion

Neurological function scores after treatment with QKL at 0, 1, 3, 4, 6, 9, 12 hours post ischemia/reperfusion

Injection of 3 mL/kg QKL was most effective at reducing infarct volume and improving neurological function, so this dose was used in time window experiments. Consistent with dose-effect experiments, the general and focal neurological functions were significantly improved in the 0 hour group (P ＜ 0.01). At later time points, QKL injection improved focal, but not neurological, function.QKL injection at 1, 3, 4, 6 or 9 hours significantly enhanced focal neurological function (P ＜ 0.05), but administration at 12 hours was not effective (Figures 3A, B).

Effects on cerebral infarct volume

QKL injection at 1, 3, 4, 6, 9 hours after ischemia significantly reduced cerebral infarct volume (P ＜ 0.05). The infarct volume remained unchanged in the 12 hour group compared with the model group. The largest reduction of infarct volume (66%) was in the 0 hour group compared with model group (P ＜ 0.01). Percent reduction of infarct volume gradually decreased with increasing delay before QKL injection (infarct size: 61%, 55%, 55%, 35%, 24% in the 1, 3, 4, 6 and 9 hour groups, respectively) (Figure 3C).

Figure 3 Effects of Qingkailing(QKL) injection at different time points in mice undergoing middle cerebral artery occlusion. Neurological function scores (A, B) and cerebral infarct volume (C) and were evaluated 24 hours after brain ischemia. QKL injection at 0, 1, 3, 4, 6, 9 or 12 hours after infarction significantly reduced infarct volume and improved focal neurological function (A, B), but had no effects on general neurological function (C). In panels A and B, high scores represent serious injury. aP ＜ 0.05,bP ＜ 0.01, vs. model group. Data are expressed as mean ± SEM, n = 9.

Anti-oxidative effects of QKL injection (Figure 4)

Figure 4 Effects of Qingkailing (QKL) injection on oxidative stress in mice undergoing middle cerebral artery occlusion. Relative intensity of red fluorescence represents reactive oxygen species (ROS) content in the hippocampal CA1 and prefrontal cortex regions of the injured hemisphere (A, TUNEL staining × 200). Relative fluorescence intensity in five sites from one section was determined by fluorescence microscopy. The mean value of ROS content was calculated and expressed as mean ±SEM for the hippocampal CA1 (dotted column) and prefrontal cortex (grey column) regions (B).Malondialdehyde (MDA) content, NADPH oxidase activity and superoxide dismutase (SOD) activity are shown in C,D and E, respectively. aP ＜ 0.01, vs. sham-surgery group;bP ＜ 0.01, vs. model group.

Superoxide anion production was detected using dihydroethidium (DHE) staining. As shown in Figure 4, no fluorescence was evident in normal brain tissues. A large number of fluorescent nerve cells contributed to significantly enhanced fluorescence in the prefrontal cortex and hippocampal CA1 region after ischemia for 1.5 hours.

Following treatment with QKL, the number of fluorescent nerve cells was reduced, and the fluorescence intensity was weakened, indicating that QKL injection reduced ROS production. Quantitative analysis showed that fluorescence in the hippocampus and cortex was significantly decreased in the QKL groups compared with the model group (Figure 4A, B). MDA, a product of lipid peroxidation, was significantly increased in the brain tissues post-ischemia. MDA content was significantly lower in the QKL groups compared with the model group(Figure 4C), consistent with our ROS detection results.

This further supports the theory that QKL injection has anti-oxidative effects.

The anti-oxidative effects of QKL injection were associated with regulation of SOD and NADPH oxidase activities. SOD, an endogenous antioxidase, is constitutively active in normal brain tissues, but activity was significantly decreased after brain ischemia for 24 hours. QKL injection significantly increased SOD activity, although it remained lower than normal. NADPH oxidase activity is low in normal animal brain tissues, but was significantly increased (approximately five-fold) after brain ischemia for 24 hours. However, QKL injection significantly reversed this increase (Figures 4D, E).

DlSCUSSlON

Kunming mice were used here to establish a MCAO model and to study the effects of QKL. In accordance with the recommendations of the Stroke Therapy Academic Industry Roundtable[1-2], we measured cerebral infarct volume and neurological function because infarct volume and functional scores are important indices in evaluating pharmacodynamics in the MCAO model.

Results showed that QKL injection significantly reduced infarct volume and improved focal neurological function when administered up to 9 hours after ischemia. In contrast, the general neurological function of MCAO mice was significantly improved only in the 0 and 3 hour groups These results are consistent with our preliminary experiments[11], in which focal, but not general, neurological function scores were significantly correlated with infarct volume. Therefore, focal neurological function injury may better reflect the severity of, and recovery from, brain ischemia.

The time window in which a drug is effective varies between drugs, commonly ranging between 2-4 hours for ischemic stroke drugs[12-14], but occasionally extended to 12 hours[15]. Administration beyond the therapeutic window can reduce or even abolish the pharmacodynamic action. The present study has defined the therapeutic window for QKL injection in the treatment of brain ischemia. QKL injection at 9 hours after ischemia significantly reduced infarct volume and improved focal neurological function (P ＜ 0.05), so the therapeutic window for QKL injection for brain ischemia in mice can be said to extend to over 9 hours.

Brain ischemia induces serial pathological changes.Oxidative stress is involved in pathological mechanisms of brain ischemia. In vitro studies have demonstrated that jasminoidin and baicalin, the main components of QKL,have anti-oxidant properties[16-17]. In this study, QKL injection significantly reduced ROS production and decreased MDA content, indicating that QKL injection attenuates cerebral ischemia/reperfusion-induced lipid peroxidation injury.

After ischemia/reperfusion injury, serum SOD concentration was decreased. Antioxidants can increase SOD content, ameliorating the severity of ischemic stroke[18-19].

This study observed significantly increased SOD content in brain tissues and serum after QKL injection at 0, 1, 3 and 4 hours after injury, with greatest effect at 1 hour.

Amino acid excitotoxicity can over-activate the N-methyl-D-aspartate (NMDA) receptor, resulting in neuronal nitric oxide synthase activation and nitric oxide production, which are the main sources of ROS after brain ischemia[20-21]. Recent evidence indicates that NADPH oxidase plays a key role in NMDA receptor-induced ROS production, and some NADPH oxidase regulators have been shown to be cerebroprotective[22-24].

The present study demonstrates that QKL injection negatively regulates NADPH oxidase during cerebral ischemia/reperfusion-induced ROS production.

Injected QKL is composed of various components that may become therapeutically powerful when they interact.

For example, the anti-oxidative effects of QKL may correlate with baicalin and jasminoidin, which have been shown to synergistically influence inflammation, nerve growth factor, and neuronal apoptosis[25-27]. Brain ischemia involves excitotoxicity, oxidative stress, calcium overload, inflammation, and apoptosis. QKL contains effective components that have multiple targets and exhibit promising effects in treating brain ischemia.

MATERlALS AND METHODS

Design

A randomized, controlled, animal experiment.

Time and setting

The experiments were performed at the Laboratory of Pharmacology, Beijing University of Chinese Medicine,China from October 2009 to February 2011.

Materials

Animals

We used 210 healthy, male, Kunming mice weighing 23-25 g, purchased from Vital River Laboratories, Beijing,China (No. SCXK (Beijing) 2006-0009) and housed in the Central Laboratory, Beijing University of Chinese Medi-cine on a 12 hour light:dark cycle at 22°C, with humidity 40-67%.

Drugs

QKL injection, composed of cholic acid, Concha Margaritifera Usta (powder), hyodeoxycholic acid, Fructus Gardeniae, Cornu Bubali (powder), Radix Isatidis, baicalin and Flos Lonicerae[28], was purchased from the Pharmaceutical Factory of Beijing University of Chinese Medicine (No. 813204A).

Methods

MCAO model establishment

Mice were anesthetized with 4% chloral hydrate(350 mg/kg) and placed in a supine position. Under a dissecting microscope (SXE-1, Shanghai Precision Instrument, Shanghai, China), the right carotid bifurcation was carefully exposed, and the external carotid artery coagulated distal to the bifurcation. A 0.16 mm diameter nylon filament (tip diameter 0.20 ± 0.01 mm; Beijing Sunbio Biotech, Beijing, China) was inserted through the external carotid artery stump and gently advanced 10 mm to occlude the origin of the middle cerebral artery[29-30]. The body temperature of the animals was maintained at 37°C. The filament was removed after 1.5 hours. Postoperation, the mice were housed separately.

Neurological function was evaluated when the mice were awake[31]; those with scores less than 2 were excluded.

QKL injection

In the dose-effect experiments, high, moderate, and low dose QKL injection and model groups were injected with 6, 3 and 1.5 mL/kg QKL (diluted in saline)[7]or normal saline (0.9 mL/100 g), respectively, via the tail veins. The first injection was performed immediately after model establishment, followed by administration after 4 hours,and once every 12 hours thereafter.

For time window experiments, the model group was injected with normal saline, and each QKL group with QKL diluted using normal saline, via the tail vein. The model and 0 hour groups were first injected while the middle

cerebral artery was occluding. The other groups received injections at 1, 3, 4, 6, 9 or 12 hours after MCAO, fol

lowed by a second injection after 4 hours, and every 12 hours thereafter.

Assessment of neurological function in mice after focal cerebral ischemia/reperfusion using Clark scores

Neurological function was evaluated using a blind method 24 hours after model establishment. Clark scores[32]include focal and general neurological function,which reflecting ischemia foci-induced neurological function injury and general function, respectively. The focal neurological function was scored from 0-28, and the general function ranged from 0-32. Normal mice had a score of 0. High scores reflect severe neurological functional injury.

Infarct volume changes in MCAO mice treated with QKL injection

Following neurological function evaluation, mice were sacrificed, and the brain was harvested for TTC staining(Nanjing Greensynthesis Biochemical Co., Ltd., Nanjing,Jiangsu, China). The percent of infarct volume out of the entire brain represented the degree of cerebral infarction.

Serial coronal sections (1 mm thickness) were prepared,and soaked in 2% TTC phosphate buffer at 37°C for 10 minutes in the dark. Normal brain tissues were stained red, while infarct tissues were not stained (white).

The sections were soaked in 4% paraformaldehyde phosphate buffer for 30 minutes, arranged in order and scanned (Tsinghua Unisplendour A688, Xi’an, China).

Areas of red and white staining were measured using a computer color multimedia image analysis system (Image-Pro Plus6.0, Media Cybernetics, Wyoming, USA).

The percent of infarction is given by the equation(Vc-VI)/(Vc×2)×100%: Vc=d×∑Ac where Ac is the normal hemisphere area; Vc is the normal hemisphere volume; Vl is the lesioned hemisphere volume; d is the length of the brain in sagittal plane (i.e.,the sum of the thickness of all coronal sections); and Vl=d×∑Al (where Al is the lesioned hemisphere area)[29-30].

Brain cell apoptosis in MCAO mice injected with QKL

After 24 hours of recovery, the brain was removed, frozen and cut into 20 μm slices. TUNEL staining was performed using a kit for programmed cell death (In Situ Cell Death Detection Kit, TMR Red, Roche, USA) according to the manufacturer’s directions[33]. Three sections were selected from each mouse. Five areas of each section were examined by fluorescence microscope (ZEISS,LSM510 meta, Germany) in the prefrontal cortex and the hippocampus of the ischemic hemisphere[9-10]and TUNEL-positive cells were quantified[9].

ROS content in MCAO mice treated with QKL

Brain ROS production was determined using DHE microfluorography[34-35]. DHE is a cell permeable dye, which can be oxidized into ethidium and other products by superoxide[36]. Animals were sacrificed and the brains removed, frozen, and sectioned (20 μm thickness) on a cryostat. Sections of prefrontal cortex and hippocampus were collected. A ROS fluorescence detection kit (Genmed, Wyoming, USA) was used. DHE solution was superfused on the brain sections for 60 minutes and fluorescence intensity was detected by fluorescence microscopy (Zeiss). The fluorescence intensities of five different fields of the brain section were averaged and expressed as RFU[37-38].

Determination of MDA content and SOD activity in MCAO mice treated with QKL

The brain was removed 24 hours after ischemia, and the cortex was isolated from the lesioned hemisphere, homogenized and the supernatant collected[9]. MDA production reflects the degree of lipid peroxidation injury[39-40]. The kit was provided by Nanjing Jiancheng Bioengineering Institute, Nanjing, China, and MDA was determined according to the manufacturer’s directions.

SOD activity was determined using the xanthine oxidase method[39]using a kit from Nanjing Jiancheng Bioengineering Institute according to the manufacturer’s directions.

Measurement of NADPH oxidase activity in MCAO mice treated with QKL

Prefrontal cortex from the ischemic hemisphere was homogenized with saline, the cytosolic (supernatant) and membrane (pellet) fractions separated using a Protein Extraction Kit (Transmembrane Protein Extraction Kit,Novagen, Darmstadt, Germany), and the membrane fraction used for analysis of NADPH oxidase enzymatic activity[38], determined according to previously described protocols[41-42]. Aliquots of the brain homogenate were incubated with NADPH at 37°C. NADPH oxidase enzymatic activity was determined every 10 minutes by measuring reduction of NADPH using a Radical Detector kit (Genmed) and a plate reader spectrophotometer(450 nm). NADPH oxidase activity was normalized by the amount of protein in each sample and the increase in absorbance between 10 and 20 minutes. Activity was calculated as mean absorbance/μg protein/min.

Statistical analysis

Data were analyzed using SPSS 17.0 (SPSS, Chicago,IL, USA). One-way analysis of variance was used followed by post hoc analysis for significance with the Student-Newman-Keuls multiple comparison test. All values are expressed as mean ± SEM. A value of P ＜ 0.05 was considered statistically significant.

Author contributions:Qingguo Wang was in charge of funding and authorized this study. Fafeng Cheng conducted experiments, conceived and designed the study, and wrote the draft of the manuscript. Wenting Song conducted animal experiments, conceived and designed this study, and participated in manuscript writing. Xianggen Zhong revised the manuscript and provided technical support. Yi Lu contributed to evaluation of the study. Shaoying Guo, Dong Wang, Weipeng Zhao participated in animal experiments and index detections.

Conflicts of interest:None declared.

Funding:This study was supported by the Science and Technology Major Projects for Major New Drugs (Qingkailing injection for treatment of ischemic stroke), No.2009ZX09102-136.

Ethical approval:This study received permission from the Animal Care and Research Committee of Beijing University of Chinese Medicine, China.

[1]Fisher M, Feuerstein G, Howells DW, et al. Update of the stroke therapy academic industry roundtable preclinical recommendations. Stroke. 2009;40(6):2244-2250.

[2]Fisher M, Hanley DF, Howard G, et al. Recommendations from the STAIR V meeting on acute stroke trials, technology and outcomes.Stroke. 2007;38(2):245-248.

[3]Beijing University of Chinese Medicine. The study of novel dosage form of An Gong Niu Huang Wan. Xinyiyao Xue Zazhi.1975(8):12.

[4]Cheng FF, Song WT, Guo SY, et al. Meta-analysis of Clearing Heat and Removing Toxicity Therapy on ischemic stroke.Zhongyao Yaoli yu Linchuang. 2011;27(1):106-109.

[5]Chen X, Howard OM, Yang X, et al. Effects of Shuanghuanglian and Qingkailing, two multi-components of traditional Chinese medicinal preparations, on human leukocyte function. Life Sci.2002;70(24):2897-2913.

[6]Yue S, Li Q, Liu S, et al. Mechanism of neuroprotective effect induced by QingKaiLing as an adjuvant drug in rabbits with E. coli bacterial meningitis. Acta Neurochir Suppl. 2006;96:413-418.

[7]Hua Q, Zhu X, Li P, et al. Refined Qing Kai Ling, traditional Chinese medicinal preparation, reduces ischemic stroke-induced infarct size and neurological deficits and increases expression of endothelial nitric oxide synthase. Biol Pharm Bull. 2008;31(4):633-637.

[8]Lv L, Liu Y, Shi HF, et al. Qingkailing injection attenuates apoptosis and neurologic deficits in a rat model of intracerebral hemorrhage. J Ethnopharmacol. 2009;125(2):269-273.

[9]Qi X, Hosoi T, Okuma Y, et al. Sodium 4-phenylbutyrate protects against cerebral ischemic injury. Mol Pharmacol. 2004;66(4):899-908.

[10]Albéri L, Chi Z, Kadam SD, et al. Neonatal stroke in mice causes long-term changes in neuronal Notch-2 expression that may contribute to prolonged injury. Stroke. 2010;41(10 Suppl):S64-71.

[11]Cheng FF, Song WT, Guo SY, et al. Comparison of three kinds of neurological function rating systems on focal cerebral ischemia models of rodent animals. Zhongguo Kangfu Yixue Zazhi. 2011;26(4):337-341.

[12]David HN, Haelewyn B, Rouillon C, et al. Neuroprotective effects of xenon: a therapeutic window of opportunity in rats subjected to transient cerebral ischemia. FASEB J. 2008;22(4):1275-1286.

[13]Lee EJ, Hung YC, Chen HY, et al. Delayed treatment with carboxy-PTIO permits a 4-h therapeutic window of opportunity and prevents against ischemia-induced energy depletion following permanent focal cerebral ischemia in mice. Neurochem Res.2009;34(6):1157-1166.

[14]Jia J, Zhang X, Hu YS, et al. Protective effect of tetraethyl pyrazine against focal cerebral ischemia/reperfusion injury in rats:therapeutic time window and its mechanism. Thromb Res. 2009;123(5):727-730.

[15]Wang C, Liu JL, Sang HF, et al. Therapeutic time window of flurbiprofen axetil's neuroprotective effect in a rat model of transient focal cerebral ischemia. Chin Med J (Engl). 2008;121(24):2572-2577.

[16]Gao Z, Huang K, Xu H. Protective effects of flavonoids in the roots of Scutellaria baicalensis Georgi against hydrogen peroxideinduced oxidative stress in HS-SY5Y cells. Pharmacol Res. 2001;43(2):173-178.

[17]Liu J, Yin F, Zheng X, et al. Geniposide, a novel agonist for GLP-1 receptor, prevents PC12 cells from oxidative damage via MAP kinase pathway. Neurochem Int. 2007;51(6-7):361-369.

[18]Kinouchi H, Kamii H, Mikawa S, et al. Role of superoxide dismutase in ischemic brain injury: a study using SOD-1 transgenic mice. Cell Mol Neurobiol. 1998;18(6):609-620.

[19]Wakade C, Khan MM, De Sevilla LM, et al. Tamoxifen neuroprotection in cerebral ischemia involves attenuation of kinase activation and superoxide production and potentiation of mitochondrial superoxide dismutase. Endocrinology. 2008;149(1):367-379.

[20]Brenman JE, Chao DS, Gee SH, et al. Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and alpha1-syntrophin mediated by PDZ domains. Cell. 1996;84(5):757-767.

[21]Aarts M, Liu Y, Liu L, et al. Treatment of ischemic brain damage by perturbing NMDA receptor- PSD-95 protein interactions. Science.2002;298(5594):846-850.

[22]Fabian RH, Perez-Polo JR, Kent TA. Perivascular nitric oxide and superoxide in neonatal cerebral hypoxia-ischemia. Am J Physiol Heart Circ Physiol. 2008;295(4):H1809-1814.

[23]Brennan AM, Suh SW, Won SJ, et al. NADPH oxidase is the primary source of superoxide induced by NMDA receptor activation. Nat Neurosci. 2009;12(7):857-863.

[24]Zhu XL, Xiong LZ, Wang Q, et al. Therapeutic time window and mechanism of tetramethylpyrazine on transient focal cerebral ischemia/reperfusion injury in rats. Neurosci Lett. 2009;449(1):24-27.

[25]Gao YF, Yuan ZZ, Niu FL, et al. Influence of hyocholic acid on ICAM-1 of brain microvascular endothelial cell in rats after ischemia and reperfusion injury in vitro. Beijing Zhongyiyao Daxue Xuebao. 2007;30(10):670-673.

[26]Zhu XL, Zhang N, Li PT, et al. Effects of cholic acid and jasminoidin on the levels of cerebral TNF-α, IL-1 β and ICAM-1 in the rats undergone cerebral ischemia. Beijing Zhongyiyao Daxue Xuebao. 2004;27(6):26-29.

[27]Zhang ZJ, Li P, Wang Z, et al. A comparative study on the individual and combined effects of baicalin and jasminoidin on focal cerebral ischemia-reperfusion injury. Brain Res. 2006;1123(1):188-195.

[28]China Pharmacopoeia Committee. Pharmacopoeia of the People's Republic of China. 2005 ed. Beijing: Chemical Industry Press. 2005.

[29]Capone C, Fabrizi C, Piovesan P, et al. 2-Aminotetraline derivative protects from ischemia/reperfusion brain injury with a broad therapeutic window. Neuropsychopharmacology. 2007;32(6):1302-1311.

[30]De Simoni MG, Storini C, Barba M, et al. Neuroprotection by complement (C1) inhibitor in mouse transient brain ischemia. J Cereb Blood Flow Metab. 2003;23(2):232-239.

[31]Menzies SA, Hoff JT, Betz AL. Middle cerebral artery occlusion in rats: a neurological and pathological evaluation of a reproducible model. Neurosurgery. 1992;31(1):100-107.

[32]Clark WM, Lessov NS, Dixon MP, et al. Monofilament intraluminal middle cerebral artery occlusion in the mouse. Neurol Res. 1997;19(6):641-648.

[33]Pillet S, von Messling V. Canine distemper virus selectively inhibits apoptosis progression in infected immune cells. J Virol.2009;83(12):6279-6287.

[34]Bindokas VP, Jordán J, Lee CC, et al. Superoxide production in rat hippocampal neurons: selective imaging with hydroethidine. J Neurosci. 1996;16(4):1324-1336.

[35]Girouard H, Park L, Anrather J, et al. Cerebrovascular nitrosative stress mediates neurovascular and endothelial dysfunction induced by angiotensin II. Arterioscler Thromb Vasc Biol. 2007;27(2):303-309.

[36]Robinson KM, Janes MS, Pehar M, et al. Selective fluorescent imaging of superoxide in vivo using ethidium-based probes. Proc Natl Acad Sci U S A. 2006;103(41):15038-15043.

[37]Girouard H, Wang G, Gallo EF, et al. NMDA receptor activation increases free radical production through nitric oxide and NOX2. J Neurosci. 2009;29(8):2545-2552.

[38]Wang Q, Sun AY, Simonyi A, et al. Ethanol preconditioning protects against ischemia/reperfusion-induced brain damage: role of NADPH oxidase-derived ROS. Free Radic Biol Med. 2007;43(7):1048-1060.

[39]Gümü?ta? K, Meta Güzeyli FM, Atükeren P, et al. The effects of vitamin E on lipid peroxidation, nitric oxide production and superoxide dismutase expression in hyperglycemic rats with cerebral ischemia-reperfusion injury. Turk Neurosurg. 2007;17(2):78-82.

[40]Lazzarino G, Tavazzi B, Di Pierro D, et al. The relevance of malondialdehyde as a biochemical index of lipid peroxidation of postischemic tissues in the rat and human beings. Biol Trace Elem Res. 1995;47(1-3):165-170.

[41]Wei Y, Sowers JR, Nistala R, et al. Angiotensin II-induced NADPH oxidase activation impairs insulin signaling in skeletal muscle cells.J Biol Chem. 2006;281(46):35137-35146.

[42]Whaley-Connell AT, Morris EM, Rehmer N, et al. Albumin activation of NAD(P)H oxidase activity is mediated via Rac1 in proximal tubule cells. Am J Nephrol. 2007;27(1):15-23.