B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Cholinergic agonist physostigmine suppresses excessivesuperoxide anion radical generation in blood, oxidative stress,early inflammation, and endothelial injury in rats withforebrain ischemia/reperfusion☆

Satoshi Kutsunaa, Ryosuke Tsurutaa, Motoki Fujitaa,⁎, Masaki Todania, Takeshi Yagia,Yasuaki Oginoa, Masatsugu Igarashib, Koshiro Takahashib, Tomonori Izumia,Shunji Kasaokaa, Makoto Yuasab, Tsuyoshi Maekawaa

aAdvanced Medical Emergency and Critical Care Center, Yamaguchi University Hospital, 1-1-1, Minami-Kogushi, Ube,Yamaguchi 755–8505, JapanbDepartment of Pure and Applied Chemistry, Faculty of Science and Technology, Tokyo University of Science, Noda, Chiba 278–8510, Japan

A R T I C L E I N F O

☆ No author has a financial relationship wi⁎ Corresponding author. Fax: +81 836 22 2344

E-mail address: motoki-ygc@umin.ac.jp (MAbbreviations: I/R, ischemia/reperfusion; R

hydroxyl radical; ONOO−, peroxynitrite; CAPischemia/reperfusion; HMGB1, high-mobilityduring ischemia; QR, Q after reperfusion; MDApartial pressure of arterial oxygen; PaCO2,phosphate; NF-κB, nuclear factor kappaB; IL,

0006-8993/$ – see front matter © 2009 Elsevidoi:10.1016/j.brainres.2009.11.077

A B S T R A C T

Article history:Accepted 29 November 2009Available online 5 December 2009

The cholinergic anti-inflammatory pathway is reportedly important in modulating theinflammatory response in local and systemic diseases, including ischemia/reperfusionpathophysiology. In this study, we investigated the effects of the cholinergic agonist,physostigmine, on jugular venous superoxide radical (O2

−U) generation, oxidative stress,early inflammation, and endothelial activation during forebrain ischemia/reperfusion(FBI/R) in rats. Fourteen male Wistar rat were allocated to the control group (n=7) orphysostigmine group (n=7). The physostigmine group received 80 ng/g physostigmineintraperitoneally 24 h and 1 h before forebrain ischemiawas established. The jugular venousO2

−U current was measured for 10 min during forebrain ischemia and for 120 min afterreperfusion. The O2

−U current increased gradually during forebrain ischemia in both groups.The current increased markedly immediately after reperfusion in the control group but wassignificantly attenuated in the physostigmine group after reperfusion. Brain and plasmamalondialdehyde, high-mobility group box 1 (HMGB1) protein, and intercellular adhesionmolecule 1 (ICAM1) were significantly attenuated in the physostigmine group comparedwith the control group, except for brain HMGB1. The amount of O2

−U generated during FBI/Rcorrelated with malondialdehyde, HMGB1, and ICAM1 in both the brain and plasma. Inconclusion, the cholinergic agonist physostigmine suppressed jugular venous O2

−U

Keywords:Forebrain ischemia/reperfusionElectrochemical sensorSuperoxide anion radical (O2

−U)PhysostigmineCholinergic anti-inflammatorypathwayHigh-mobility group box 1(HMGB1) protein

th a commercial entity that has an interest in the subject of this manuscript... Fujita).

OS, reactive oxygen species; O2−U, superoxide anion radical; H2O2, hydrogen peroxide; OH•,

, cholinergic anti-inflammatory pathway; AChR, acetylcholine receptor; FBI/R, forebraingroup box 1; ΔI, difference in O2

−U current; Q, quantified partial value of electricity; QI, Q, malondialdehyde; ICAM1, intercellular adhesion molecule 1; sICAM1, soluble ICAM1; PaO2,partial pressure of arterial carbon dioxide; NADPH, nicotinamide adenine dinucleotideinterleukin; ANOVA, analysis of variance

er B.V. All rights reserved.

243B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

generation, oxidative stress, early inflammation, and endothelial activation in the brain andplasma in the acute phase of cerebral ischemia/reperfusion. Therefore, the suppression ofO2

−U is a key mechanism of the cholinergic anti-inflammatory pathway in thepathophysiology of cerebral ischemia/reperfusion.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

In the pathophysiology of cerebral ischemia/reperfusion (I/R),such as stroke, head injury, and after resuscitation fromcardiac arrest, the postischemic inflammatory response is oneof the mechanisms that exacerbates brain injury (Adrie et al.,2002; Offner et al., 2005; Huang et al., 2006; Doyle et al., 2008;Oda et al., 2008; Wong and Crack, 2008). In the neuroinflam-matory response, the oxidative stress induced by reactiveoxygen species (ROS) is an important process of inflammation(White et al., 2000; Warner et al., 2004; Doyle et al., 2008; Wongand Crack, 2008). Among ROS, the superoxide anion radical(O2

−U) is the key radical because it functions as a messenger insignaling pathways and as an effector of the oxidative stressattributable to many toxic ROS, such as hydrogen peroxide(H2O2), the hydroxyl radical (OH•), and peroxynitrite (ONOO−),both intracellularly and extracellularly (White et al., 2000;Warner et al., 2004).

The cholinergic anti-inflammatory pathway (CAP), whichhas been described as amechanism for the neuronal control ofinflammation via the efferent fiber of the vagus nerve, isimportant in modulating the inflammatory response in localand systemic diseases, including sepsis, hemorrhagic shock,myocardial I/R, pancreatitis, arthritis, ileus, and other inflam-matory syndromes (Borovikova et al., 2000; Mioni et al., 2005;Tracey, 2007; Hofer et al., 2008). CAP is activated by acholinergic brain network that is responsive to agonists ofM1, which is one of the subunits of the muscarinic receptor,and the efferent signals from the vagus nerve inhibit cytokineproduction via a pathway dependent on the α7 subunit of theacetylcholine receptor (AChR) on macrophages and other cells(Tracey, 2007). The neuroprotective effects of CAP in thepathophysiology of cerebral I/R have recently been reportedby several investigators (Wang et al., 2006a,b; Wang et al., 2008;Ottani et al., 2009). However, the in vivo effects of CAP on oxi-dative stress caused by ROS, especially O2

−U, are still unclear.Recently, we developed an in vivo real-time quantitative

O2−U analysis system with an all-synthetic electrochemical

O2−U sensor (Yuasa and Oyaizu, 2005; Yuasa et al., 2005a,b).

Using this apparatus, we have shown that O2−U is increased in

the jugular veins of rats with forebrain ischemia/reperfusion(FBI/R) during the ischemic and reperfusion periods (Aki et al.,2009), and that xanthine oxidase is one of themajor sources ofO2

−U in the pathophysiology of cerebral I/R (Ono et al., 2009).The generation of O2

−U also correlates with both brain andplasma high-mobility group box 1 (HMGB1), which is a keycytokine expressed during early inflammation in the patho-physiology of cerebral I/R (Kim et al., 2006, 2008; Qiu et al.,2008). Furthermore, the excessive production of O2

−U after I/Ris associated with early inflammation, oxidative stress, andendothelial activation in the brain and plasma (Aki et al., 2009;Ono et al., 2009). Therefore, we considered that O2

−Umight be atherapeutic target in the pathophysiology of cerebral I/R.

In this study, we confirmed the effects of the cholinergicagonist, physostigmine, on jugular venous O2

−U generation,oxidative stress, early inflammation, and endothelial injury inFBI/R rats.

2. Results

2.1. Jugular venous O2−U current and quantified partial

value of electricity (Q) of O2−U during FBI/R in rats

The baseline O2−U current was recorded before ischemia, and

the differences (ΔI) from baseline were recorded during thepreischemic, ischemic, and reperfusion periods (Fig. 1A).During the ischemic period, ΔI of O2

−U increased gradually inboth groups. Just after reperfusion, a further elevation of ΔIwas observed in the control group. In the physostigminegroup, ΔI of O2

−Uwas significantly attenuated after reperfusioncompared with that in the control group (P<0.05 from 48 to76min and from 94 to 118min after reperfusion; P<0.01 from 0to 46min, from 78 to 92min, and at 120min after reperfusion).

ΔI of O2−U was integrated during ischemia as QI and during

reperfusion as QR, reflecting the amount of O2−U generated

during each period (Aki et al., 2009). Figure panels 1B and 1Cshow the differences in QI and QR between the two groups,respectively. There was no difference in QI between the twogroups (P=0.403), but QR was significantly lower in thephysostigmine group than in the control group (P<0.01).

2.2. Malondialdehyde (MDA), HMGB1, and ICAM1 levelsin the forebrain tissue and plasma

The brain and plasma MDA levels were measured to evaluatethe degree of lipid peroxidation in the forebrain tissue and inthe circulating blood.MDA in the forebrain tissue 120min afterreperfusion is shown in Fig. 2A. Brain MDA was significantlylower in the physostigmine group than in the control group(P<0.01). Plasma MDA 120 min after reperfusion is shown inFig. 2B. Plasma MDA was significantly lower in the physostig-mine group than in the control group (P<0.01). The correlationcoefficients were calculated for total Q (the sum of QI and QR)and brain MDA (r=0.5873, P<0.05) and for total Q and plasmaMDA (r=0.8315, P<0.01).

HMGB1 in the forebrain cytoplasm 120 min after reperfu-sion is shown in Fig. 2C. There was no significant differencebetween the two groups (P=0.315). Plasma HMGB1 120 minafter reperfusion is shown in Fig. 2D. Plasma HMGB1 wassignificantly lower in the physostigmine group than in thecontrol group (P<0.05). The correlation coefficients werecalculated for total Q and brain HMGB1 (r=0.5707, P<0.05)and for total Q and plasma HMGB1 (r=0.5444, P<0.05).

ICAM1 in the forebrain tissue 120 min after reperfusion isshown in Fig. 2E. Brain ICAM1 was significantly lower in the

Fig. 1 – Changes in the superoxide anion radical (O2−U) current and differences in the quantified partial value of electricity (Q)

during forebrain ischemia/reperfusion in rats. (A) Mean O2−U current in the jugular vein during forebrain ischemia/reperfusion.

The vertical axis indicates changes in the O2−U current from the baseline current (ΔI). The baseline O2

−U current is defined as thestable state before ischemia. (B) Values of Q measured in the three groups during ischemia (QI). The generation of O2

−U wasevaluated as Q, which was calculated by the integration of ΔI. (C) Values of Q after reperfusion (QR). Values are the mean±SD ofseven measurements. **P<0.01 vs. the control group.

244 B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

physostigmine group than in the control group (P<0.01).Plasma soluble ICAM1 (sICAM1) 120 min after reperfusion isshown in Fig. 2F. Plasma sICAM1was significantly lower in thephysostigmine group than in the control group (P<0.05). Thecorrelation coefficients were calculated for total Q and brainICAM1 (r=0.6521, P<0.05) and for total Q and plasma ICAM1(r=0.8431, P<0.01).

2.3. Physiological parameters

Table 1 lists the physiological parameters for the rats duringthe experiments. Therewere no significant differences in bodyweight (data not shown) ormean arterial blood pressure in thetwo groups during the preischemic and reperfusion periods(P=0.903). The partial pressure of arterial oxygen (PaO2) andthe partial pressure of arterial carbon dioxide (PaCO2) werewell controlled and did not differ between the two groupsthroughout the experiments (P=0.652 and 0.680, respectively).Metabolic acidosis was observed with the elevation of plasmalactate after reperfusion in both groups.

3. Discussion

In this study, the cholinergic agonist physostigmine sup-pressed the generation of O2

−U in the jugular vein (Fig. 1),which in turn was associated with the suppression of MDA,HMGB1, and ICAM1 in the brain and plasma during FBI/R (Fig.2). This finding indicates that the activation of CAP suppressesoxidative stress and attenuates lipid peroxidation, early

inflammation, and endothelial injury in the pathophysiologyof cerebral I/R.

Physostigmine suppressed O2−U generation, as measured

with the O2−U sensor in the jugular vein (Fig. 1), and oxidative

stress in the brain and plasma, as evaluated by MDA (Fig. 2Aand B). This is the first study to report that the activation ofCAP suppresses O2

−U generation directly in vivo. Mioni et al.(2005) reported that the activation of CAP suppressed thegeneration of free radicals in the blood collected from rats withmyocardial I/R, as measured by electron spin resonance. In invitro experiments, Moon et al. (2008) reported that theactivation of nicotinic AChR prevented the production ofROS in fibrillar β-amyloid-peptide-stimulated microglia byblocking Ca2+ influx. It has been reported that the source ofROS is nicotinamide adenine dinucleotide phosphate (NADPH)oxidase in fibrillar β-amyloid-peptide-stimulated microglia(Bianca et al., 1999). Therefore, the activation of CAP mightsuppress O2

−U generation in the blood via the inhibition ofNADPH oxidase, which is considered one of the major sourcesof O2

−U in the pathophysiology of cerebral I/R (Abramov et al.,2007; Brennan et al., 2009; Chen et al., 2009).

HMGB1 is one of the earlymediators of inflammation in thepathophysiology of cerebral I/R, as described in our previousreport and those of others (Kim et al., 2006, 2008; Qiu et al.,2008; Aki et al., 2009). It has been reported that HMGB1 istranslocated from the nucleus to the cytoplasm and isreleased into the extracellular milieu as early as 1–2 h afterfocal cerebral ischemia, and that neurons are one of theprincipal sources of the HMGB1 released in the early phase ofischemic injury (Qiu et al., 2008). Extracellular HMGB1 is

Fig. 2 – Malondialdehyde (MDA), high-mobility group box 1 (HMGB1), and intercellular adhesion molecule 1 (ICAM1) levels inforebrain tissue and plasma 120 min after reperfusion. (A) MDA in forebrain tissue. (B) MDA levels in plasma. (C) HMGB1 inforebrain cytoplasm. (D) HMGB1 in plasma. (E) ICAM1 in forebrain tissue. (F) Soluble ICAM1 (sICAM1) in plasma. Values are themean±SD of seven measurements. *P<0.05 vs. the control group. **P<0.01 vs. the control group.

245B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

reported to function as a proinflammatory cytokine, in theactivation of microglia and the endothelium, and in stimulat-ing the release of other cytokines, thus aggravating braininjury (Kim et al., 2006, 2008; Qiu et al., 2008). In the presentstudy, physostigmine attenuated plasma HMGB1 (Fig. 2D), andbrain and plasma ICAM1 levels after reperfusion (Fig. 2E andF). It has been reported that the activation of CAP bycholinergic agonists or electrical stimulation inhibits HMGB1release via the attenuation of nuclear factor kappaB (NF-κB)activity (Wang et al., 2004; Huston et al., 2007). NF-κB regulatesthe expression of various genes, including those for tumornecrosis factor α, interleukin 1 (IL1), IL6, IL8, cell surfaceadhesion molecules such as E-selectin, vascular adhesionmolecule 1, and ICAM1 (Pahl, 1999). Furthermore, NF-κB isreported to be a redox-sensitive transcription factor, whichmeans that NF-κB activity is regulated by ROS (Bubici et al.,2006; Pantano et al., 2006; Bar-Shai et al., 2008). Therefore, thesuppression of O2

−U by the activation of CAP might haveattenuated plasma HMGB1 via the inhibition of NF-κB in thepresent study. These facts support our previous data (Aki etal., 2009) and suggest the existence of an O2

−U-mediated

HMGB1 loop, as indicated by the significant correlationbetween plasma HMGB1 and total Q.

HMGB1 is also reported to upregulate the mRNA of ICAM1in endothelial cells (Treutiger et al., 2003; Qiu et al, 2008).ICAM1 is one of the most widely recognized adhesionmolecules on the endothelial surface. The upregulation ofICAM1 expression in endothelial cells might enhance inflam-mation, recruiting immune cells to the ischemic lesion andexacerbating tissue injury. In the present study, brain andplasma ICAM1 were significantly inhibited in the physostig-mine group compared with those in the control group (Fig. 2Eand F). These results suggest that the activation of CAPsuppresses both endothelial activation and endothelial injuryin the brain vascular endothelium, evaluated as brain ICAM1and plasma sICAM1 levels, respectively.

In brief, CAP suppressed neuroinflammation after FBI/R,which was demonstrated by the suppressions of the jugularvenous O2

−U generation, oxidative stress, HMGB1 expression,and ICAM1 expression, sequentially. CAP activation vianicotinic AChR activation might suppress O2

−U generation byNADPH oxidase inactivation (Moon et al., 2008). Thereafter,

Table 1 – Physiological parameters of rats with forebrain ischemia/reperfusion.

Preischemia After reperfusion

10 min 30 min 60 min 90 min 120 min

MAP (mmHg)Control 143±18 150±7 142±15 123±21 95±13 79±10Physostigmine 135±21 148±24 143±21 128±26 105±20 78±13

PaO2 (Torr)Control 108±27 108±17 102±13 88±21 81±13 82±13Physostigmine 102±14 114±23 105±20 95±16 90±12 85±10

PaCO2 (Torr)Control 40±3 49±7 43±4 41±6 42±4 39±6Physostigmine 39±4 51±5 44±3 42±4 42±3 42±4

pHControl 7.40±0.05 7.06±0.03 7.17±0.03 7.16±0.06 7.18±0.06 7.20±0.04Physostigmine 7.38±0.04 7.07±0.04 7.16±0.06 7.17±0.07 7.18±0.06 7.15±0.04

Lactate (mM)Control 1.4±0.6 10.6±1.3 6.9±1.2 6.8± 2.1 6.6±1.5 6.1±1.6Physostigmine 1.5±0.7 10.9±2.5 7.8±2.8 7.4±2.3 6.2±2.5 5.4±2.5

Values are the mean±SD of seven measurements. MAP: mean arterial blood pressure; PaO2, partial pressure of arterial oxygen; PaCO2, partialpressure of arterial carbon dioxide.

246 B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

the suppression of ROS might inhibit NF-κB activation, whichled to suppression of HMGB1, ICAM1, and other cytokines(Bubici et al., 2006; Pantano et al., 2006; Bar-Shai et al., 2008).

There are some limitations to the present study. First, theO2

−U-measuring device was placed in the bloodstream in thejugular vein. Therefore, the data analyzed may not perfectlyreflect O2

−U generation in the brain tissue. Second, our resultsreflect only the acute phase (120 min) of FBI/R pathophysiol-ogy, with and without physostigmine. The measurement ofO2

−U for long periods and the dose dependency of physostig-mine require future study.

In conclusion, the cholinergic agonist physostigminesuppressed the generation of O2

−U during reperfusion follow-ing forebrain ischemia and inhibited oxidative stress, asdemonstrated by evaluating lipid peroxidation (MDA), earlyinflammation (HMGB1), and endothelial injury (ICAM1). Theseresults suggest that the suppression of O2

−U is one of the keymechanisms of CAP in the pathophysiology of cerebral I/R.

4. Experimental procedures

4.1. Animals

The study protocol was approved by the Animal ExperimentCommittee of Yamaguchi University and all rats were handledaccording to National Institutes of Health Guidelines for theCare and Use of Laboratory Animals. Fourteen male Wistarrats, fasted overnight and weighing 250–350 g, were randomlyassigned to two groups: the control group (n=7) or thephysostigmine group (n=7). The physostigmine group re-ceived 80 ng/g physostigmine (MP Biomedicals LLC, SantaAna, CA, USA) dissolved in 0.3 mL of saline intraperitoneally24 h and 1 h before forebrain ischemia was established. Thecontrol group received an equivalent amount of saline over

the same time course. The dose of physostigmine was asdescribed in a previous report, which demonstrated improve-ment in survival during experimental sepsis (Hofer et al.,2008).

4.2. Animal preparation

Transient forebrain ischemia was established as described inour previous studies (Aki et al., 2009; Ono et al., 2009; Fujita etal., in press; Koda et al., 2010; Tsuruta et al., 2009). In brief,under isoflurane anesthesia (3% during surgery) and mechan-ical ventilation (SN-480-7 respirator; Shimano ManufacturingCo. Ltd., Tokyo, Japan) through a tracheotomy tube, an arterialcatheter was inserted to measure blood pressure and tosample blood from the left femoral artery. A venous catheterwas inserted into the right atrium through the right externaljugular vein to administer drugs and to remove blood toinduce a state of hypotension. The distal side of the rightjugular vein was ligated. The tip of the O2

−U sensor wasinserted from the left anterior facial vein to the externaljugular vein. The distal side of the left anterior facial vein andthe proximal branches of the external jugular vein wereligated, and the left posterior external jugular vein and thecephalic vein were ligated under a surgical microscope toreduce the blood flow from the face and neck. The bilateralcarotid arteries were exposed. The isoflurane concentrationwas reduced to 0.7% with 40% oxygen, and the physiologicalparameters and the O2

−U current were stabilized for at least20 min. All surgical incision sites were anesthetized with0.25% bupivacaine (≤0.3 mL/rat).

4.3. Induction of FBI/R

After baseline measurements of the O2−U current were made,

forebrain ischemia was induced by the occlusion of the

247B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

bilateral common carotid arteries, and blood was removed toproduce a mean arterial blood pressure of 40–45 mmHg for10 min. Forebrain ischemia was confirmed by the completesuppression of electroencephalographic activity. Reperfusionwas achieved by releasing the bilateral carotid artery occlu-sion and returning the previously removed blood. Thegeneration of O2

−U was calculated as the difference betweenthe baseline current and the detected current recorded withthe O2

−U sensor during the preischemic, ischemic, andreperfusion periods, as described previously (Aki et al., 2009).PaO2, PaCO2, pH, and the base excess of the arterial blood,arterial blood pressure, and pharyngeal temperaturewere alsomeasured. The pharyngeal temperature was maintained at37.0±0.2 °C throughout the experiment. Blood was sampledand replaced with ice-cold saline 120 min after reperfusion.The brain was removed, frozen in liquid nitrogen, and storedat −80 °C until MDA, HMGB1, and ICAM1 analyses. The bloodwas centrifuged at 900 × g for 10 min at 4 °C and the plasmawas stored at −80 °C until analysis.

4.4. Measurement and evaluation of jugular venous O2−U

The O2−U generated was measured as a current, which was

recorded with an ROS analysis system using an electrochem-ical O2

−U sensor (Actiive Corp., Noda, Japan), as described inour previous studies (Yuasa and Oyaizu, 2005; Yuasa et al.,2005a,b; Fujita et al., 2009). This O2

−U sensor has a carbonworking electrode coated with a polymeric iron porphyrincomplex, bromo-iron(III) (5,10,15,20-tetra(3-thienyl)porphyrin)ligated two 1-methylimidazole as an axial ligand ([Fe(im)2(tpp)]Br), which mimics cytochrome c, and a stainless-steel counterelectrode (Yuasa and Oyaizu, 2005; Yuasa et al., 2005a,b). Thissensor has high catalytic activity for the oxidation of O2

−U, anda linear relationship exists between the current and the O2

−Uconcentration in phosphate-buffered saline and human blood(Yuasa et al., 2005a; Fujita et al., 2009). The sensor has beenshown to be sensitive and specific for extracellular O2

−U anddoes not respond to nitric oxide or H2O2 (Fujita et al., 2009).

The current data were recorded at two points per secondand smoothing procedures (i.e., moving averages) wereapplied during data analysis because the data containednoise and artifacts (Fujita et al., 2009). The current data arepresented as ΔI, which refers to the difference in the currentfrom baseline to the actual measured current, as described inour previous study (Aki et al., 2009).

The measured O2−U current was evaluated as Q, which

reflects the amount of O2−U generated (Fujita et al., 2009). ΔI

was integrated during the ischemic period as QI and duringthe reperfusion period as QR, as described in our previousstudy (Aki et al., 2009). Total Qwas calculated as the sum of QI

and QR, and reflects the amount of O2−U generated throughout

forebrain ischemia/reperfusion.

4.5. MDA analysis

The brain tissue of the left frontal lobe was homogenized inice-cold 50 mM Tris–HCl buffer (pH 7.4) with 5 mM butylatedhydroxytoluene (in acetonitrile) using a Polytron PT-MR3100homogenizer (Kinematica, Littau, Switzerland). MDA levels inthe forebrain homogenate and the plasma were analyzed

120 min after reperfusion with the Bioxytech® MDA-586™ Kit(OxisResearch, Foster, CA, USA). The final results are pre-sented as pmol/mg protein in the forebrain homogenate andas μM in the plasma.

4.6. HMGB1 analysis

The cytoplasmic fraction of the brain tissue was prepared asdescribed in our previous reports (Aki et al., 2009; Ono et al.,2009; Fujita et al., in press; Koda et al., 2010; Tsuruta et al.,2009). In brief, the brain tissue of the left frontal lobe wasgently homogenized in 10mMN-2-hydroxyethylpiperazine-N′-ethanesulfonic acid/10mMKCl bufferwith 0.08%NP-40, 0.1mMethylenediaminetetraacetic acid, 0.5 mM dithiothreitol, and0.5 mM phenylmethylsulfonyl fluoride; the soluble fractionderived from the cytoplasm was stored at − 80 °C until furthertesting. The HMGB1 levels in the brain cytoplasm and theplasmawere analyzed 120minafter reperfusionwith anHMGB1ELISA Kit II™ (Shino-test Corporation, Kanagawa, Japan). Thefinal results are presented as ngHMGB1/mg protein in the braincytoplasm and as ng/mL in the plasma.

4.7. ICAM1 analysis

ICAM1 levels in the forebrain homogenate and the plasma(soluble ICAM1, sICAM1) were analyzed 120 min after reperfu-sion using the Quantikine® Rat sICAM-1 (CD54) Immunoas-say™ (R&D Systems Inc., Minneapolis, MN, USA). The finalresults are presented as pg ICAM1/mg protein in the brainhomogenate and as ng/mL in the plasma.

4.8. Arterial blood gas and lactate analyses

Arterial blood gas and lactate were analyzed with the ABL™System 555 (Radiome A/S, Copenhagen, Denmark) during thepreischemic and reperfusion periods.

4.9. Statistical analysis of the data

The data were analyzed using the SPSS 10.0 statisticalsoftware package (SPSS Inc., Chicago, IL, USA). The statisticalsignificance of ΔI and the physiological parameters wasdetermined with two-way analysis of variance (ANOVA).When the results of ANOVA were significant, the Bonferronipost-hoc test was applied to determine specific groupdifferences. The statistical significance of differences in QI,QR, MDA, HMGB1, and ICAM1 was determined by t-test. Thecorrelation between total Q and MDA, HMGB1, or ICAM1 wasanalyzed statistically using Pearson's correlation coefficient.All data are expressed as themean±standard deviation (SD) ofseven measurements. A value of P<0.05 is consideredstatistically significant.

Acknowledgments

The authors thank Ms Chihiro Kobayashi, Mr Tetsuya Aoki,and Mr Masahiro Nanba (Tokyo University of Science) for theirassistance with the O2

−U sensor; Mrs. Hitomi Ikemoto, Dr.Takahiro Yamamoto, and Dr. Yohei Otsuka for their valuable

248 B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

technical assistance; and Ms. Masako Ueda for her patience inpreparing the original manuscript. The work described in thisreport was supported by a Grant-in-Aid for Young Scientists(grant 19791328) from the Ministry of Education, Science,Sports and Culture of Japan.

R E F E R E N C E S

Abramov, A.Y., Scorziello, A., Duchen, M., 2007. Three distinctmechanisms generate oxygen free radicals in neurons andcontribute to cell death during anoxia and reoxygenation.J. Neurosci. 27, 1129–1138.

Adrie, C., Adib-Conquy, M., Laurent, I., Monchi, M., Vinsonneau, C.,Fitting, C., Fraisse, F., Dinh-Xuan, A.T., Carli, P., Spaulding, C.,Dhainaut, J.F., Cavaillon, J.M., 2002. Successfulcardiopulmonary resuscitation after cardiac arrest as a“sepsis-like” syndrome. Circulation 106, 562–568.

Aki, S.H., Fujita, M., Yamashita, S., Fujimoto, K., Kumagai, K.,Tsuruta, R., Kasaoka, S., Aoki, T., Nanba, M., Murata, H., Yuasa,M., Maruyama, I., Maekawa, T., 2009. Elevation of jugularvenous superoxide anion radical is associated with earlyinflammation, oxidative stress, and endothelial injury inforebrain ischemia–reperfusion rats. Brain Res 1292, 180–190.

Bar-Shai, M., Carmeli, E., Ljubuncic, P., Reznick, A.Z., 2008. Exerciseand immobilization in aging animals: the involvement ofoxidative stress and NF-κB activation. Free Radic. Biol. Med. 44,202–214.

Bianca, V.D., Dusi, S., Bianchini, E., Dal Prà, I., Rossi, F., 1999.β-Amyloid activates the O2 forming NADPH oxidase inmicroglia, monocytes, and neutrophils. A possibleinflammatory mechanism of neuronal damage in Alzheimer'sdisease. J. Biol. Chem. 274, 15493–15499.

Borovikova, L.V., Ivanova, S., Zhang, M., Yang, H., Botchkina, G.I.,Watkins, L.R., Wang, H., Abumrad, N., Eaton, J.W., Tracey, K.J.,2000. Vagus nerve stimulation attenuates the systemicinflammatory response to endotoxin. Nature 405, 458–462.

Brennan, A.M., Suh, S.W., Won, S.J., Narasimhan, P., Kauppinen,T.M., Lee, H., Edling, Y., Chan, P.H., Swanson, R.A., 2009. NADPHoxidase is the primary source of superoxide induced by NMDAreceptor activation. Nat. Neurosci. 12, 857–863.

Bubici, C., Papa, S., Dean, K., Franzoso, G., 2006. Mutual cross-talkbetween reactive oxygen species and nuclear factor-kappa B:molecular basis and biological significance. Oncogene 25,6731–6748.

Chen, H., Song, Y.S., Chan, P.H., 2009. Inhibition of NADPH oxidaseis neuroprotective after ischemia–reperfusion. J. Cereb. BloodFlow Metab. 29, 1262–1272.

Doyle, K.P., Simon, R.P., Stenzel-Poore, M.P., 2008. Mechanisms ofischemic brain damage. Neuropharmacology 55, 310–318.

Fujita, M., Tsuruta, R., Kaneko, T., Otsuka, Y., Kutsuna, S., Izumi, T.,Aoki, T., Shitara, M., Kasaoka, S., Maruyama, I., Yuasa, M.,Maekawa, T., in press. Hyperoxia suppresses excessivesuperoxide anion radical generation in blood, oxidative stress,early inflammation, and endothelial injury in forebrainischemia/reperfusion rats-Laboratory study. Shock.

Fujita, M., Tsuruta, R., Kasaoka, S., Fujimoto, K., Tanaka, R., Oda,Y., Nanba, M., Igarashi, M., Yuasa, M., Yoshikawa, T., Maekawa,T., 2009. In vivo real-time measurement of superoxide anionradical with a novel electrochemical sensor. Free Radic. Biol.Med. 47, 1039–1048.

Hofer, S., Eisenbach, C., Lukic, I.K., Schneider, L., Bode, K.,Brueckmann, M., Mautner, S., Wente, M.N., Encke, J., Werner, J.,Dalpke, A.H., Stremmel, W., Nawroth, P.P., Martin, E.,Krammer, P.H., Bierhaus, A., Weigand, M.A., 2008.Pharmacologic cholinesterase inhibition improves survival inexperimental sepsis. Crit. Care Med. 36, 404–408.

Huang, J., Upadhyay, U.M., Tamargo, R.J., 2006. Inflammation instroke and focal cerebral ischemia. Surg. Neurol. 66, 232–245.

Huston, J.M., Gallowitsch-Puerta, M., Ochani, M., Ochani, K., Yuan,R., Rosas-Ballina, M., Ashok, M., Goldstein, R.S., Chavan, S.,Pavlov, V.A., Metz, C.N., Yang, H., Czura, C.J., Wang, H., Tracey,K.J., 2007. Transcutaneous vagus nerve stimulation reducesserum high mobility group box 1 levels and improves survivalin murine sepsis. Crit. Care Med. 35, 2762–2768.

Kim, J.B., Sig Choi, J., Yu, Y.M., Nam, K., Piao, C.S., Kim, S.W., Lee,M.H., Han, P.L., Park, J.S., Lee, J.K., 2006. HMGB1, a novelcytokine-like mediator linking acute neuronal death anddelayed neuroinflammation in the postischemic brain.J. Neurosci. 26, 6413–6421.

Kim, J.B., Lim, C.M., Yu, Y.M., Lee, J.K., 2008. Induction andsubcellular localization of high-mobility group box-1 (HMGB1)in the postischemic rat brain. J. Neurosci. Res. 86, 1125–1131.

Koda, Y., Tsuruta, R., Fujita, M., Miyauchi, T., Kaneda, K., Todani,M., Aoki, T., Shitara, M., Izumi, T., Kasaoka, S., Yuasa, M.,Maekawa, T., 2010. Moderate hypothermia suppresses jugularvenous superoxide anion radical, oxidative stress, earlyinflammation, and endothelial injury in forebrain ischemia/reperfusion rats. Brain Res. 1311, 197–205.

Mioni, C., Bazzani, C., Giuliani, D., Altavilla, D., Leone, S., Ferrari,A., Minutoli, L., Bitto, A., Marini, H., Zaffe, D., Botticelli, A.R.,Iannone, A., Tomasi, A., Bigiani, A., Bertolini, A., Squadrito, F.,Guarini, S., 2005. Activation of an efferent cholinergic pathwayproduces strong protection against myocardial ischemia/reperfusion injury in rats. Crit. Care Med. 33, 2621–2628.

Moon, J.H., Kim, S.Y., Lee, H.G., Kim, S.U., Lee, Y.B., 2008. Activationof nicotinic acetylcholine receptor prevents the production ofreactive oxygen species in fibrillar β amyloid peptide(1–42)-stimulated microglia. Exp. Mol. Med. 40, 11–18.

Oda, Y., Tsuruta, R., Kasaoka, S., Inoue, T., Maekawa, T., 2008. Thecutoff values of intrathecal interleukin 8 and 6 for predictingthe neurological outcome in cardiac arrest victims.Resuscitation 80, 189–193.

Offner, H., Subramanian, S., Parker, S.M., Afentoulis, M.E.,Vandenbark, A.A., Hurn, P.D., 2005. Experimental strokeinduces massive, rapid activation of the peripheral immunesystem. J. Cereb. Blood Flow Metab. 26, 654–665.

Ono, T., Tsuruta, R., Fujita, M., Aki, S.H., Kutsuna, S., Kawamura, Y.,Wakatsuki, J., Aoki, T., Kobayashi, C., Kasaoka, S., Maruyama, I.,Yuasa, M., Maekawa, T., 2009. Xanthine oxidase is one of themajor sources of superoxide anion radicals in blood afterreperfusion in rats with forebrain ischemia/reperfusion. BrainRes. 1305, 158–167.

Ottani, A., Giuliani, D., Mioni, C., Galantucci, M., Minutoli, L., Bitto,A., Altavilla, D., Zaffe, D., Botticelli, A.R., Squadrito, F., Guarini,S., 2009. Vagus nerve mediates the protective effects ofmelanocortins against cerebral and systemic damage afterischemic stroke. J. Cereb. Blood Flow Metab. 29, 512–523.

Pahl, H.L., 1999. Activators and target genes of Rel/NF-κBtranscription factors. Oncogene 18, 6853–6866.

Pantano, C., Reynaert, N.L., van der Vliet, A., Janssen-Heininger,Y.M., 2006. Redox-sensitive kinases of the nuclear factor-κBsignaling pathway. Antioxid. Redox Signal. 8, 1791–1806.

Qiu, J., Nishimura, M., Wang, Y., Sims, J.R., Qiu, S., Savitz, S.I.,Salomone, S., Moskowitz, M.A., 2008. Early release of HMGB1from neurons after the onset of brain ischemia. J. Cereb. BloodFlow Metab. 28, 927–938.

Tracey, K.J., 2007. Physiology and immunology of the cholinergicantiinflammatory pathway. J. Clin. Invest. 117, 289–296.

Treutiger, C.J., Mullins, G.E., Johansson, A.S., Rouhiainen, A.,Rauvala, H.M., Erlandsson-Harris, H., Andersson, U., Yang, H.,Tracey, K.J., Andersson, J., Palmblad, J.E., 2003. High mobilitygroup 1 B-box mediates activation of human endothelium.J. Intern. Med. 254, 375–385.

Tsuruta, R., Fujita, M., Ono, T., Koda, Y., Koga, Y., Yamamoto, T.,Nanba, M., Shitara, M., Kasaoka, S., Maruyama, I., Yuasa, M.,

249B R A I N R E S E A R C H 1 3 1 3 ( 2 0 1 0 ) 2 4 2 – 2 4 9

Maekawa, T., 2009. Hyperglycemia enhances excessivesuperoxide anion radical generation, oxidative stress, earlyinflammation, and endothelial injury in forebrain ischemia/reperfusion rats. Brain Res. 1309, 155–163.

Wang, H., Liao, H., Ochani, M., Justiniani, M., Lin, X., Yang, L.,Al-Abed, Y., Wang, H., Metz, C., Miller, E.J., Tracey, K.J., Ulloa, L.,2004. Cholinergic agonists inhibit HMGB1 release and improvesurvival in experimental sepsis. Nat. Med. 10, 1216–1221.

Wang, R., Yan, H., Tang, X.C., 2006a. Progress in studies ofhuperzine A, a natural cholinesterase inhibitor from Chineseherbal medicine. Acta Pharmacol. Sin. 27, 1–26.

Wang, Z.F., Tang, L.L., Yan, H., Wang, Y.J., Tang, X.C., 2006b. Effectsof huperzine A on memory deficits and neurotrophic factorsproduction after transient cerebral ischemia and reperfusion inmice. Pharmacol. Biochem. Behav. 83, 603–611.

Wang, Z.F., Wang, J., Zhang, H.Y., Tang, X.C., 2008. Huperzine Aexhibits anti-inflammatory and neuroprotective effects in a ratmodel of transient focal cerebral ischemia. J. Neurochem. 106,1594–1603.

Warner, D.S., Sheng, H., Batinić-Haberle, I., 2004. Oxidants,antioxidants and the ischemic brain. J. Exp. Biol. 207, 3221–3231.

White, B.C., Sullivan, J.M., DeGracia, D.J., O'Neil, B.J., Neumar, R.W.,Grossman, L.I., Rafols, J.A., Krause, G.S., 2000. Brain ischemiaand reperfusion: molecular mechanisms of neuronal injury.J. Neurol. Sci. 179, 1–33.

Wong, C.H., Crack, P.J., 2008. Modulation of neuro-inflammationand vascular response by oxidative stress following cerebralischemia–reperfusion injury. Curr. Med. Chem. 15, 1–14.

Yuasa, M., Oyaizu, K., 2005. Electrochemical detection andsensing of reactive oxygen species. Curr. Org. Chem. 9,1685–1697.

Yuasa, M., Oyaizu, K., Yamaguchi, A., Ishikawa, M., Eguchi, K.,Kobayashi, T., Toyoda, Y., Tsutsui, S., 2005a. Electrochemicalsensor for superoxide anion radical using polymeric ironporphyrin complexes containing axial 1-methylimidazoleligand as cytochrome c mimics. Polym. Adv. Technol. 16,287–292.

Yuasa, M., Oyaizu, K., Yamaguchi, A., Ishikawa, M., Eguchi, K.,Kobayashi, T., Toyoda, Y., Tsutsui, S., 2005b. Structure andredox properties of electropolymerized film obtained from ironmeso-tetrakis(3-thienyl)porphyrin. Polym. Adv. Technol. 16,616–621.