t

1gmP

Experimental Neurology 163, 39–45 (2000)doi:10.1006/exnr.2000.7364, available online at http://www.idealibrary.com on

Neuroprotection by 2-h Postischemia Administration of Two FreeRadical Scavengers, a-phenyl-n-tert-butyl-nitrone (PBN) and

N-tert-butyl-(2-sulfophenyl)-nitrone (S-PBN), in RatsSubjected to Focal Embolic Cerebral Ischemia

Yi Yang, Qiu Li, and Ashfaq ShuaibDivision of Neurology, Department of Medicine, University of Alberta, Edmonton, Alberta, Canada T6G 2B7

Received September 7, 1999; accepted January 18, 2000

fcib

Oxygen free radical generation may have importantsecondary damaging effects after the onset of cerebralischemia. Free radical scavengers have been used suc-cessfully in attenuating neuronal damage in the reper-fusion period in transient forebrain ischemia. Thereare limited data on effectiveness in models of focalischemia. Two free radical scavengers, a-phenyl-n-ert-butyl-nitrone (PBN) and N-tert-butyl-(2-sulfophe-

nyl)-nitrone (S-PBN), have been shown to reduce oxi-dative-stress-induced neuronal injury. Whereas PBNhas been demonstrated to reduce infarct volume infocal ischemia, neuroprotection has not been evalu-ated with S-PBN. The present study was designed toevaluate the neuroprotective effect of PBN and S-PBNcompared to vehicle in a focal embolic middle cerebralartery (MCA) cerebral ischemia model in rats. Wistarrats were randomly divided into three groups (n 5 10each group). Animals in the control group receivedvehicle and those in the treatment groups weretreated with PBN or S-PBN (both 100 mg/kg/day 3 3days, intraperitoneally) starting 2 h after the intro-duction of an autologous thrombus into the right-sideMCA. The neurological outcome was observed andcompared before and after treatment and betweengroups. The percentage of cerebral infarct volume wasestimated from 2,3,5-triphenyltetrazolium chloridestained coronal slices 72 h after the ischemic insult.Two-hour postischemia administration of PBN or S-PBN significantly improved neurobehavioral scores at24 h following MCA embolization (both P < 0.01). Thepercentage of infarct volume for animals receivingvehicle was 32.8 6 9.4%. Two-hour delayed administra-tion of PBN and S-PBN achieved a 35.4% reduction ininfarct volume in treatment groups when comparedwith animals receiving vehicle (PBN vs control, 21.2 610.9% vs 32.8 6 9.4%; P < 0.05; S-PBN vs control, 21.2 63.1%, (P < 0.05). These data indicate that free radicaleneration may be involved in brain damage in thisodel and 2-h delayed postischemia treatment withBN and S-PBN may have neuroprotective effects in

39

ocal cerebral ischemia. As S-PBN does not normallyross the blood–brain barrier, the neuroprotection ev-dent in this study may be explained by entry into therain via damaged vessels. © 2000 Academic Press

Key Words: neuroprotection; focal cerebral ischemia;free radical scavengers; PBN; S-PBN; rats.

INTRODUCTION

Most of the early work in understanding of the ef-fects of ischemia on cerebral circulation was completedin models of transient forebrain (or global) ischemia.More recently, especially in the past 10 years, therehas been considerable interest in studying how thebrain responds to focal ischemia. With a better under-standing of the mechanisms of neuronal damage aftercerebral ischemia, a number of therapies are beingtested in humans and in laboratory models of focalcerebral ischemia. Currently, therapy of acute isch-emic stroke has two basic principles: dissolving theintravascular occlusion (reperfusion) and preservingthe brain from the harmful cellular and metabolic cas-cade (neuroprotection).

The cellular mechanisms responsible for neuronaldeath following ischemic insult are still being clarified.Important factors include energy failure, loss of cell ionhomeostasis, excitotoxicity, acidosis, increased intra-cellular calcium, and free-radical-mediated toxicity (forliterature and review, see 9). Recent data from an invivo study also showed an interaction between gluta-mate-mediated neuroexcitoxicity and the generation ofoxygen-derived free radicals (32). Thus, therapeuticapproaches that limit oxidative stress may be poten-tially neuroprotective in ischemic stroke.

a-Phenyl-n-tert-butyl-nitrone (PBN) and N-tert-bu-tyl-(2-sulfophenyl)-nitrone (S-PBN) are the radicalscavengers that have shown promise as neuroprotec-tive agents (4, 5, 15, 27, 28, 31, 33, 40). These com-pounds react with free radicals to form more stable

0014-4886/00 $35.00Copyright © 2000 by Academic Press

All rights of reproduction in any form reserved.

otm(esioeosaa

pSrwiotah

40 YANG, LI, AND SHUAIB

adducts (18). PBN has been shown to ameliorate braindamage in models of global/forebrain ischemia in ger-bils (3, 25, 28) as well as in both permanent and tran-sient models of focal ischemia in rats (4, 40). Applica-tion of PBN also showed neuroprotection in other su-peroxide-involved situations by attenuating adjacentlesions of experimental hematoma (27) and epilepsy(15). In contrast, so far, few published data have beenrelated to neuroprotection of S-PBN in in vivo modelsf cerebral ischemia, although S-PBN has been showno reduce oxidative-stress-induced hydroxyl radical for-ation in a concentration-dependent manner in vitro

11). Most of the previous studies were conducted tovaluate the neuroprotective efficacy of free radicalcavenger on reperfusion damage in transient cerebralschemia models (29, 35, 38, 40) and there is a shortagef information regarding prolonged focal cerebral isch-mia. Additionally, in therapeutical protocols of previ-usly published experiments (16, 32, 33), free radicalcavengers were given prior to ischemia, which mayppear to be therapeutically impractical since mostcute strokes occur without warning.The present study was designed to define and com-

are a potential neuroprotective effect of PBN and-PBN on focal cerebral ischemia rats subjected toight-side middle cerebral artery (MCA) embolization,hich is considered closest to the clinical setting of

schemic stroke (39). Using this model, we have dem-nstrated that injection of an autologous thrombus intohe right-side MCA could produce about 35–45% cortexnd striatal lesions in an affected side of the brainemisphere (36).

MATERIALS AND METHODS

All experiments in this study were approved by theHealth Sciences Animal Welfare Committee of the Uni-versity of Alberta and performed in strict accordancewith the related guidelines.

Cerebral Focal Ischemia

Embolic focal cerebral ischemia was induced in anes-thetized rats with the technique detailed in our previ-ous report (37). Male Wistar rats (350–450 g) wereanesthetized with 3.0% halothane and then the anes-thesia was maintained with 1.5% halothane in a 70%N2O and O2 (vol/vol) mixture with a face mask. Therectal temperature of animals was monitored and keptat approximately 37°C using a heating pad and anoverhead lamp. The studied animals were kept at 37°Cafter surgery by placing an overhead lamp above thecage before and 1 h after drug treatment. Other phys-iological parameters including mean arterial bloodpressure, O2 saturation, and pulse were also monitoredthroughout the experiment using a Biopac MP100 sys-tem (Biopac System, Santa Barbara, CA).

In brief, after induction of anesthesia, a longitudinalincision of approximately 1.5 cm was made in the mid-dle cervical skin to expose both sides of the commoncarotid arteries (CCA), right internal carotid artery(ICA), and the right external carotid artery (ECA). Amodified PE-50 catheter (Becton–Dickinson, Ruther-ford, NJ) (0.3-mm OD at the tip of the catheter and0.97-mm OD in the catheter body), filled with bovinethrombin (Thrombostat, Parke-Davis, TM Warner-Lambert Co., Scarborough, Canada), was introducedinto the right ICA lumen via a small puncture on theright ECA. Ten microliters of blood was withdrawninto the catheter and retained for 15 min to allowformation of a clot before a 17-mm length (calculatedfrom the puncture point at the ECA) of catheter wasgently advanced into the intracranial section of theICA through the right ECA and extracranial section ofICA. The tip of the catheter at this depth was 1–2 mmaway from the origin of the MCA (39). Finally, bothCCA were temporarily clipped to reduce cerebral bloodflow and the clot in the catheter was injected into theICA. The catheter was withdrawn from the right ECA10 min after the injection. Then, after the incision wasclosed, the experimental animals were allowed to wakeup and were free to access food.

Neurological Deficit

The method used in this study for the behavioralassessment of focal cerebral ischemic rats has beenpreviously described (36). Neurological deficit evalua-tion was conducted at 2 and 24 h, respectively, afterthe injection of an autogenous arterial thrombus. Theneurological findings were scored on a four-point gradescale: no observable deficit, 0; forelimb flexion, 1; fore-limb flexion and decreased resistance to lateral push,2; forelimb flexion, decreased resistance to lateralpush, and unilateral circling in three successive trials;3. Any other neurological behavior was also observed.

Quantification of Brain Infarct Volume

The details of the evaluation of the following meth-ods related to quantification of infarct volume werereported elsewhere (37). Briefly, 72 h after cerebralembolism, the rats received intracardiac perfusion of100 ml normal saline under deep anesthesia by injec-tion of an overdose of pentothal (100 mg/kg). The brainwas removed from the skull and cooled in ice-cold sa-line for about 5 min. For morphometric study, 2-mm-thick coronal sections were cut using a rat brain ma-trix. A total of eight coronal sections were prepared forestimation of the degree of infarct damage. Then thecoronal sections were stained using 2,3,5-triphenyltet-razolium chloride (TTC). The stained brain sectionswere placed directly on the scanning screen of a colorflatbed scanner (Scanjet 4p, Hewlett–Packard) within7 days. A colorless transparent glass cover and a piece

sIcpis

E

mdtwdritfPi

iMai

D

SaaSc

t

ino

41DELAYED PBN AND S-PBN INJECTION REDUCED BRAIN INFARCTION

of black cloth put on the glass cover were used toprovide a black background. Following image acquisi-tion, the images were analyzed blindly using a com-mercial image processing software program, Photo-Shop, version 4.0 (Adobe System) which was installedon the same computer. Measurements were made bymanually outlining the margins of infarct areas. Thetotal volume of infarction was determined by the inte-gration of the distance of the eight chosen sections.

Since brain edema may affect the accuracy of infarctestimation, to compensate for brain swelling in theischemic hemisphere we measured the corrected in-farct size instead of using the direct measurement.This method corrects infarct size and involves measur-ing both hemispheres and applying the following for-mula: Corrected infarct size 5 the size of the left hemi-phere 2 (the right hemisphere-measured infarct size).n this study, the infarct size was expressed as a per-entage of the volume in the region examined. Theercentage of infarct volume was obtained by calculat-ng the portion of the corrected infarct size in the totalize of the right hemisphere.

xperimental Protocols

Both PBN and S-PBN used in the current experi-ent were provided by Astra Arcus (Sodertalje, Swe-

en). The effective doses of PBN and S-PBN used inhis study were chosen from a published experimentith similar settings (32, 34). PBN and S-PBN wereissolved in normal saline to prepare a 2.5% solutioneady for injection. Animals were randomly dividednto three groups: PBN-treated (n 5 10), S-PBN-reated (n 5 10), and control group (n 5 10). For bothree radical scavenger treatment groups, either theBN or the S-PBN solution was systemically given

ntraperitoneally (ip) at a dose of 100 mg/kg 2 h after

FIG. 1. Neurological recovery score in the animals treated with Ps equal among all groups, there is significantly better outcome ineurobehavioral score almost equal to that achieved by administratiof animals at 24 h in control group; NS, no significant difference.

njection of an autologous thrombus into the rightCA. Animals in the control group received the same

mount of normal saline (about 2 ml) via the samenjection route.

ata Expression and Statistical Analysis

All data in this study were expressed as means 6D. Statistical analysis of more than two groups ofnimals was performed with one-way analysis of vari-nce, with subsequent individual comparisons usingcheffe’s test. The difference was considered signifi-ant when the P value was ,0.05.

RESULTS

No significant difference in physiological variableswas found between the animals before and after intro-duction of the thrombus into the MCA. In a similarmanner, the physiological status did not show signifi-cant changes between the animals in free-scavenger-treated groups and those in the vehicle-treated group.

Neurologic Deficits

As shown in Fig. 1, animals treated with vehicle didnot show a significant improvement in neurobehav-ioral score between 2 and 24 h following cerebral em-bolization (2 h, 3.6 6 0.5 vs 24 h, 3.4 6 0.5). In contrast,postischemia treatment with PBN and S-PBN signifi-cantly improved neurological outcome at 24 h afterfocal cerebral ischemia when compared to the scores at2 h after cerebral ischemia (PBN-treated group, 2 h,3.8 6 0.4 vs 24 h, 2.3 6 0.5, P 5 0.0002; S-PBN-reated group, 2 h, 3.6 6 0.5 vs 24 h, 2.5 6 0.7, P 5

0.001). However, we did not find a significant differ-ence in neurobehavioral improvement between the two

and S-PBN. Whereas the neurological damage at 2 h after the insultated animals. Treatment with S-PBN showed an improvement inf PBN. Note. * P , 0.05 when compared with the neurological scores

BNtre

n o

42 YANG, LI, AND SHUAIB

free-radical-treated groups in the vehicle-treatedgroup (P 5 0.23).

Pathological Evaluation

The extent of morphological damage and percentagevolume calculation in the studied groups of treatedanimals compared to controls were shown in Figs. 2and 3. Maximum damage was seen in the vehicle-treated animals with the cerebral infarction involving32.8 6 9.4% of the brain tissue on the affected side.Compared to this, the animals receiving PBN or S-PBN2 h postocclusion demonstrated a significant neuropro-tection and both treatments reduced brain lesions byalmost 35% (PBN, 21.2 6 10.9% of infarction in theaffected brain hemisphere, P , 0.05; S-PBN, 21.2 613.1%, P , 0.05). The neuroprotective efficacy

FIG. 2. Representative TTC-stained images of brain coronal sec-tions from animals in different groups. Embolization of the MCAwith the thrombus resulted in extensive brain infarction in theaffected hemisphere. Compared to placebo-treated animals, therewas significantly better protection in the animals treated with PBNand S-PBN at 2 h after the ischemic insult. A few hemorrhagictransformations within the ischemic region were also shown in theS-PBN-treated group (as arrows denote).

achieved by the two free radical scavengers was almostequal. Moreover, we have also observed that therewere incidences of intercerebral hemorrhagic transfor-mation in animals in the control group (3/10), the PBN-treated group (1/10), and the S-PBN-treated group (4/10), but the difference between the groups was notsignificantly different (P . 0.05).

DISCUSSION

In this study, we describe experiments in which thefree radical spin trap agents PBN and S-PBN weregiven, respectively, 2 h after cerebral embolization. Toour knowledge, the present experiment is the firststudy to compare the neuroprotective efficacy betweenPBN and S-PBN in a focal cerebral ischemia model.Neuroprotection was evident with the use of both PBNand S-PBN resulting in a significant attenuation ofabout 35% infarct volume. This is in accordance withprevious reports, in which the neuroprotective effect ofPBN was observed in focal cerebral ischemia rats sub-jected to permanent MCA and ipsilateral CCA (4).

The extent to which oxygen free radical generationmay promote neuronal injury during cerebral ischemiais not fully understood. Its damaging effects may varyaccording to the nature of the insult (for example,transient versus permanent occlusion or focal versusglobal ischemia). Free radicals are any atom or mole-cule containing one or more unpaired electrons. Forthis reason they are highly reactive with a great vari-ety of biomolecules, changing their physical–chemicalproperties. The brain contains high concentrations ofpolyunsaturated fatty acids, which can be easily oxi-dized by reactive oxygen species (ROS), rendering thecentral nervous system extremely vulnerable to oxida-

FIG. 3. Statistical evaluation of brain infarct volume in animalstreated with PBN and S-PBN. Compared with placebo-treated ani-mals, animals treated with PBN or S-PBN at 2 h following ischemicinsult showed about 35% reduction of the brain lesion. The neuro-protective effect of PBN was similar with that of S-PBN (P . 0.05).Note. * P , 0.05; NS, no significant difference.

43DELAYED PBN AND S-PBN INJECTION REDUCED BRAIN INFARCTION

tive attack (9). Nitric oxide (NO) has been clearlylinked to ischemic neurodegeneration in both animalmodels and cell culture systems, although the finalcellular pathways that lead from the generation of NOto eventual neuronal death require further investiga-tion (for literature, see 20). It is assumed that en-hanced production of ROS is triggered by an excessiveinflux of calcium following ischemia. The rise of ROScauses oxidative damage to lipids, proteins, and DNAand finally leads to neuronal death (12, 14). Similarly,the mechanism of PBN and S-PBN neuroprotectionhas not been completely understood. Because of theirlipophilicity, they may be able to efficiently scavengethe intracranial free radicals that are generated duringbrain ischemia. Spin trapping agents are able to catchreactive free radicals and change them into relativelystable radicals. Therefore, they are regarded as a kindof radical scavenger, so these pharmacological effectswere thought to be derived from scavenging of radicalsformed in vivo under oxidative stress. This contentionis supported by results showing that PBN reduces ox-idation of soluble proteins produced by ischemia oraging (4, 5, 25). Additionally, a link between excitotox-icity and oxidative stress has been under investigationrecently. It has been suggested that excitotoxicity andoxidative stress may be sequential and interactivemechanisms leading to neuronal degeneration (7). Asubsequent study showed that S-PBN, as an antioxi-dant, was effective in treating neurologic illnesses inwhich excitotoxic mechanisms have been implicated(32). Recent animal model studies have also demon-strated that PBN administration causes transient hy-pothermia (38). As hypothermia has neuroprotectiveproperties, the attenuation of neuronal damage withthe use of PBN may be secondary to this effect. In ourexperiments, the role of hypothermia is limited as thebody temperature of studied animals was maintainedat 37°C throughout the study.

Direct evidence of neuroprotection by very selectivefree radical scavengers could also provide indirect ev-idence for free radical toxicity contributing to ischemicbrain damage as it has been difficult to obtain defini-tive evidence of the participation of free radicals inischemic damage (40). Cerebral ischemia/reperfusionis accompanied by enhanced production of free radicals(21, 24, 39) and infarct size is reduced by the adminis-tration of free radical scavengers (2, 29, 35). However,information about the neuroprotection of free radicalscavengers on permanent cerebral ischemia is limited.In present study, systemic administration of free rad-ical scavengers at a 2-h delay following cerebral embo-lization still achieved a significant improvement inneurological outcome and, correspondingly, a markedreduction in brain lesion size. Our data provide indi-rect evidence that free radical toxicity may be alsoinvolved in prolonged cerebral ischemia. Our resultssupport a recent suggestion that the underlying bio-

chemical processes, such as energy failure, loss of cellion homeostasis, acidosis, increased intracellular cal-cium, and excitotoxicity, as well as free-radical-medi-ated toxicity in cerebral ischemia, are similar, regard-less of the amount of brain that is made ischemic or theduration of ischemia (19). Reperfusion of ischemic tis-sue produces an influx of inflammatory cells and ofoxygen that can cause increases in oxygen-derived freeradicals. Under conditions of prolonged ischemia, as isevident in our study, free radicals may play an impor-tant role by interaction with excitotoxicity (32).

The therapeutic time window for N-methyl-D-aspar-tate (NMDA) antagonists, non-NMDA antagonists,and glutamate release inhibitors in focal models ofischemia appears to be about 1–2 h (1, 13, 22, 26). Incontrast, a free radical spin trap was found to have animproved therapeutic window. It has been reportedthat PBN still showed neuroprotective efficacy evenwhen it was administrated 12 h following focal cerebralischemia (4) or 6 h after mitochondrial toxin malonate-introduced striatal lesions due to histotoxic hypoxia(34). Furthermore, there is no significant difference inattenuation of cerebral ischemic lesion between vari-ous groups with administration of PBN with an initialtime of 0.5 h prior to ischemia and 0.5, 5, and 12 h afterischemia.

An in vitro study suggested that a single treatmentwith S-PBN at 100 mg/kg does not influence the ratretinal ganglion cell survival rate (17). Our study eval-uated neuronal damage 72 h after insult. The robustneuroprotection with the two spin trap agents whenused hours after the insult is promising. PBN has anexcellent safety profile and a favorable pharmacoki-netical profile. It has a good bioavailability and itshalf-life is more than 3 h after systemic administrationin the rat (unpublished information from Astra Arcus).PBN is widely distributed and penetrates the brainreadily in rats after systemic administration (6). Theconcentration of PBN in the brain is significantlyhigher than that in the blood (6). In contrast, withrespect to blood–brain barrier (BBB) penetration, S-PBN is assumed to have a poor permeability comparedto that of PBN based on its difference in structure, onemolecule being an anion instead of being neutral underphysiological conditions (personal communication withAstra Arcus). However, together with our current ob-servations, results from other variable experimentalsettings when S-PBN was used singly or in combina-tion with other neuroprotective agents (10, 32, 34)demonstrated that this structure difference does notlimit its neuroprotection efficacy via peripheral (ip)administration. It has been clearly documented thatthere is postischemic BBB breakdown or leakage (8,30). This pathological outcome may result in openingswhich allow poorly permeable S-PBN to reach the cen-tral nervous system. Furthermore, glutamate and ni-tric oxide have been indicated in increasing permeabil-

1

1

44 YANG, LI, AND SHUAIB

ity of the BBB and dilating cerebral arterioles via anitric-oxide-dependent mechanism (23). Therefore,treatment with a free radical scavenger may reason-ably help in reducing the ischemia-induced BBB dam-age and brain edema. All of these mechanisms mayenable S-PBN to act as a neuroprotectant.

In conclusion, our study shows that spin trap agentsPBN and S-PBN are useful and safe neuroprotectiveagents in an embolic model of focal cerebral ischemia.The extent of neuroprotection with the two agents wasvery similar, suggesting that, despite differences intheir pharmacokinetics and absorption through theblood–brain barrier, they appear to have similar prop-erties.

ACKNOWLEDGMENT

We are grateful to Professor David Jackson for his invaluableinformation about the PBN and S-PBN and his consistent support inpreparation of this study.

REFERENCES

1. Bakker, M. H., and A. C. Foster. 1991. An investigation of themechanisms of delayed neurodegeneration caused by direct in-jection of quinolinate into the rat striatum in vivo. Neuroscience42: 387–395.

2. Canevari, L., S. Kuroda, T. E. Bates, J. B. Clark, and B. K.Siesjo. 1997. Activity of mitochondrial respiratory chain en-zymes after transient focal ischemia in the rat. J. Cereb. BloodFlow Metab. 17: 1166–1169.

3. Cao, W., J. M. Carney, A. Duchon, R. Floyd, and M. Chevion.1988. Oxygen free radical involvement in ischemia and reper-fusion injury to brain. Neurosci. Lett. 88: 233–238.

4. Cao, X., and J. Phillis. 1994. a-Phenyl-tert-butyl-nitrone re-duces cortical infarct and edema in rats subjected to focal isch-emia. Brain Res. 644: 267–272.

5. Carney, J. M., and R. A. Floyd. 1991. Protection against oxida-tive damage to CNS by a-phenyl-tert-butyl-nitrone (PBN) andother spin-trapping agents: A novel series of nonlipid free rad-ical scavengers. J. Mol. Neurosci. 3: 47–57.

6. Cheng, H. Y., T. Liu, G. Feuerstein, and F. C. Barone. 1993.Distribution of spin-trapping compounds in rat blood and brain:In vivo microdialysis determination. Free Radical Biol. Med. 14:243–250.

7. Coyle, J. T., and P. Puttfarcken. 1993. Oxidative stress, gluta-mate and neurodegenerative disorders. Science 262: 689–695.

8. Dietrich, W. D., R. Prado, M. Halley, and B. D. Watson. 1993.Microvascular and neuronal consequences of common carotidartery thrombosis and platelet embolization in rats. J. Neuro-pathol. Exp. Neurol. 52: 351–360.

9. Facchinetti, F., V. L. Dawson, and T. M. Dawson. 1998. Freeradicals as mediators of neuronal injury. Cell. Mol. Neurobiol.18: 667–682.

10. Fallon, J., R. T. Matthews, B. T. Hyman, and M. F. Beal. 1997.MPP1 produces progressive neuronal degeneration which ismediated by oxidative stress. Exp. Neurol. 144: 193–198.

11. Ferger, B., C. Spratt, P. Teismann, G. Seitz, and K. Kuschinsky.1998. Effects of cytisine on hydroxyl radicals in vitro andMPTP-induced dopamine depletion in vivo. Eur. J. Pharmacol.360: 155–163.

12. Floyd, R. A., and J. M. Carney. 1992. Free radical damage toprotein and DNA: Mechanisms involved relevant observationson brain undergoing oxidative stress. Ann. Neurol. 32: S22–S27.

13. Foster, A. C., R. Gill, and G. N. Woodruff. 1988. Neuroprotec-tive effects of MK-801 in vivo: Selectivity and evidence fordelayed degeneration mediated by NMDA receptor activation.J. Neurosci. 8: 4745–4754.

14. Halliwell, B. 1992. Reactive oxygen species and the centralnervous system. J. Neurochem. 59: 1609–1623.

15. He, Q. P., M. L. Smith, P. A. Li, and B. K. Siesjo. 1997. Necrosisof the substantia nigra, pars reticulate, in flurothyl-inducedstatus epilepticus is ameliorated by the spin trap a-phenyl-tert-butyl-nitrone. Free Radical Biol. Med. 22: 917–922.

6. Irving, E. A., K. Yatsushiro, J. McCulloch, and D. Dewar. 1997.Rapid alteration of tau in oligodendrocytes after focal ischemicinjury in the rat: Involvement of free radicals. J. Cereb. BloodFlow Metab. 17: 612–622.

7. Klocker, N., A. Cellerino, and M. Bahr. 1998. Free radicalscavenging and inhibition of nitric oxide synthase potentiatesthe neurotrophic effects of brain-derived neurotrophic factor onaxotomized retinal ganglion cells In vivo. J. Neurosci. 18: 1038–1046.

18. Knecht, K. T., and R. P. Mason. 1993. In vivo spin trapping ofxenobiotic free radical metabolites. Arch. Biochem. Biophys.303: 185–194.

19. MacDonald, R. L., and M. Stoodley. 1998. Pathophysiology ofcerebral ischemia. Neurol. Medico-Chirurg. 38: 1–11.

20. Maiese, K. 1998. From the bench to the bedside: The molecularmanagement of cerebral ischemia. Clin. Neuropharmacol. 21:1–7.

21. Malinski, T., F. Bailey, Z. G. Zhang, and M. Chopp. 1993. Nitricoxide measured by a porphyrinic microsensor in rat brain aftertransient middle cerebral artery occlusion. J. Cereb. Blood FlowMetab. 13: 355–358.

22. Massieu, L., K. H. Thedinga, M. McVey, and G. E. Fagg. 1993.A comparative analysis of the neuroprotective properties ofcompetitive and uncompetitive N-methyl-D-aspartate receptorantagonists in vivo: Implications for the process of excitotoxicdegeneration and its therapy. Neuroscience 55: 883–892.

23. Mayhan, W. G., and S. P. Didion. 1996. Glutamate-induceddisruption of the blood-brain barrier in rats. Role of nitric oxide.Stroke 27: 965–969.

24. Nelson, C. W., E. P. Wei, J. T. Povlishock, H. A. Kontos, andM. A. Moskowitz. 1992. Oxygen radicals in cerebral ischemia.Am. J. Physiol. 263: H1356–H1362.

25. Oliver, C., P. Starke-Reed, E. Stadtman, G. Liu, J. M. Carney,and R. A. Floyd. 1990. Oxidative damage to brain proteins, lossof glutamine synthetase activity, and production of free radicalsduring ischemia/reperfusion-induced injury to gerbil brain.Proc. Natl. Acad. Sci. USA 87: 5144–5147.

26. Park, C. K., and E. D. Hall. 1994. Dose–response analysis of theeffect of 21-aminosteroid tirilazad mesylate (U-74006F) uponneurological outcome and ischemic brain damage in permanentfocal cerebral ischemia. Brain Res. 645: 157–163.

27. Peeling, J., H. J. Yan, S. G. Chen, M. Campbell, and M. R. DelBigio. 1998. Protective effects of free radical inhibitors in intra-cerebral hemorrhage in rat. Brain Res. 795: 63–70.

28. Phillis, J., and C. Clough-Helfman. 1990. Protection from cere-bral ischemic injury in gerbils with the spin trap agent N-tert-butyl-a-phenylnitrone (PBN). Neurosci. Lett. 116: 315–319.

29. Phillis, J. W., S. Sen, and X. Cao. 1994. Amflutizole, a xanthineoxidase inhibitor, inhibits free radical generation in the isch-emic/reperfused rat cerebral cortex. Neurosci. Lett. 169: 188–190.

3

45DELAYED PBN AND S-PBN INJECTION REDUCED BRAIN INFARCTION

30. Preston, E., and D. O. Foster. 1997. Evidence for pore-likeopening of the blood-brain barrier following forebrain ischemiain rats. Brain Res. 761: 4–10.

31. Schulz, J. B., D. R. Henshaw, U. MacGarvey, and M. F. Beal.1996. Involvement of oxidative stress in 3-nitropropionic acidneurotoxicity. Neurochem. Int. 29: 167–171.

32. Schulz, J. B., D. R. Henshaw, D. Siwek, B. G. Jenkins, R. J.Ferrante, P. B. Cipolloni, N. W. Kowall, B. R. Rosen, and M. F.Beal. 1995. Involvement of free radicals in excitotoxicity in vivo.J. Neurochem. 64: 2239–2247.

3. Schulz, J. B., R. T. Matthews, D. R. Henshaw, and M. F. Beal.1996. Neuroprotective strategies for treatment of lesions pro-duced by mitochondrial toxins: Implications for neurodegenera-tive diseases. Neuroscience 71: 1043–1048.

34. Schulz, J. B., R. T. Matthews, B. G. Jenkins, P. Brar, and M. F.Beal. 1995. Improved therapeutic window for treatment of his-totoxic hypoxia with a free radical spin trap. J. Cereb. BloodFlow Metab. 15: 948–952.

35. Yamamoto, M., T. Shima, T. Uozumi, T. Sogabe, K. Yamada,and T. Kawasaki. 1983. A possible role of lipid peroxidation incellular damages caused by cerebral ischemia and the protec-

tive effect of alpha-tocopherol administration. Stroke 14: 977–982.

36. Yang, Y., A. Shuaib, and Q. Li. 1998. Neuroprotection by two-hour delayed administration of topiramate in the middle cere-bral artery embolization model of rat. Brain Res. 804: 169–176.

37. Yang, Y., A. Shuaib, and Q. Li. 1998. Quantification of infarctsize on focal cerebral ischemia modal of rats using a simple andeconomical method. J. Neurosci. Methods 84: 9–16.

38. Yue, T. L., J. L. Gu, P. G. Lysko, H. Y. Cheng, F. C. Barone, andG. Feuerstein. 1992. Neuroprotective effects of phenyl-t-butyl-nitrone in gerbil global brain ischemia and in cultured ratcerebellar neurons. Brain Res. 574: 193–197.

39. Zhang, J., and C. A. Piantadosi. 1994. Prolonged production ofhydroxyl radical in rat hippocampus after brain ischemia–reperfusion is decreased by 21-aminosteroids. Neurosci. Lett.177: 127–130.

40. Zhao, Q., K. Pahlmark, M. L. Smith, and B. K. Siesjo. 1994.Delayed treatment with the spin trap a-phenyl-tert-butyl-ni-trone (PBN) reduces infarct size following trasient middle ce-rebral artery occlusion in rats. Acta Physiol. Scand. 152: 349–350.