Neuroprotective Effects of Free Radical Scavengers in Stroke
Chen X. Wang and Ashfaq Shuaib
Stroke Research Laboratory, University of Alberta, Edmonton, Alberta, Canada

Abstract
Acute ischaemic stroke is a leading cause of death in the majority of industrial- ised countries and also in many developing countries. Free radicals are generated in the brain during ischaemic injury and these radicals are involved in the secondary injury processes. Several free radical scavengers have been developed and some of them have progressed into clinical trials. One of them, edaravone, has been approved by the regulatory authority in Japan for the treatment of stroke patients. Another scavenger, disodium 4-[(tert-butylimino)methyl] benzene-1,3- disulfonate N-oxide (NXY-059; disufenton), has demonstrated efficacy in a phase III clinical trial (SAINT [Stroke Acute Ischaemic NXY-059 Treatment study]-I) involving a large number of stroke patients. Unfortunately, SAINT II did not show efficacy in the treatment of stroke patients. The purpose of this article is to review the current development of antioxidant strategies, update recent findings for NXY-059 in the treatment of stroke patients, and discuss the future develop- ment of neuroprotective agents. Although the development of neuroprotective strategies for the treatment of stroke is challenging, progress in molecular and cellular neuroscience will uncover new information about stroke mechanisms, which should result in the realisation of neuroprotective therapy for this disease.

Stroke is a disorder that affects approximately thrombus or embolism. This occlusion results in loss 15 million people worldwide and is the third leading of blood flow and a major decrease in the supply of cause of death in the majority of industrialised coun- oxygen and nutrients to the affected region. tries and in many developing countries.[1,2] Stroke is Haemorrhagic injury, on the other hand, is a result of also an important cause of long-term disability. blood vessel bursts either in the brain or on its More than 30% of patients who survive the initial surface.
stroke incident have severe disability and 25% of The hypothesis that tissue oxidation resulting such patients become unable to walk independently. from processes initiated by free radicals plays an Clearly, stroke causes a severe physical and eco- essential role in cell death during ischaemic brain nomic burden to patients, their families and society. injury was originally examined in the late 1970s.[3] As the aging population increases, stroke will con- Since then, the role of oxidative stress in ischaemic tinue to remain a major problem in society. injury has been extensively studied and several lines Ischaemic brain injury accounts for >85% of all of evidence now support this hypothesis. First, in- strokes, while haemorrhagic brain injury is responsi- creased production of free radicals has been demon- ble for the remaining cases.[1] Ischaemic brain injury strated in the ischaemic injured brain,[4-6] and results from occlusion of a major cerebral artery by a reperfusion of the ischaemic injured brain worsens

oxidative stress.[7-9] Secondly, treatment with anti- produced by brain tissues in physiological condi- oxidant compounds has been shown to reduce is- tions, their production and elimination rates are chaemic brain injury in a variety of animal mod- equal and, therefore, they are not noxious under els.[3,10-13] Thirdly, experiments with transgenic normal circumstances. The natural elimination sys- mice support that idea that oxidative stress contrib- tems in the brain include endogenous enzymatic utes to injury. For example, overexpression of superoxide dismutase, glutathione peroxidase and superoxide dismutase, a natural free radical scaven- catalase; and nonenzymatic antioxidative mole- ger, has been shown to reduce ischaemic brain inju- cules, such as glutathione, uric acid, ascorbic acid ry.[14,15] (vitamin C) and tocopherol (vitamin E). However, Knowledge about free radicals in ischaemic brain after cerebral ischaemia, free radical production is injury has accumulated rapidly over the past few dramatically increased and overwhelms the endoge- decades.[16-19] The production and pathophysiologi- nous antioxidant systems, leading to disruption of
cal roles of free radicals in ischaemic brain injury the equilibrium and consequent cell damage.
have been discussed in detail recently.[17,18,20] There- During cerebral ischaemia, elevated intracellular fore, the present article focuses on the development Ca2+ activates phospholipase A2.[20,21] This enzyme of free radical scavengers that have already liberates the unsaturated fatty acid arachidonic acid progressed into clinical trials. In a very recent and thus initiates the formation of free radicals via phase III clinical trial, disodium 4-[(tert-butylimi- the cyclo-oxygenase and lipoxygenase pathways. no)methyl] benzene-1,3-disulfonate N-oxide (NXY- Cyclo-oxygenase catalyses the addition of O2 mole- 059; disufenton), a promising antioxidant agent, cules to arachidonic acid to produce prostaglandin failed to demonstrate a protective action in stroke (PG) G2, which is rapidly peroxidised to PGH2 with
patients. Thus, the scientific community once again simultaneous release of O2–. Other pro-oxidant en-
has reached a point where it needs to think carefully zymes that generate O2– are xanthine oxidase and about the future development of neuroprotective nicotinamide adenine dinucleotide phosphate hydro- strategies for the treatment of acute stroke. With this gen (NADPH) oxidase.[22] Degradation of adenine in mind, ideas for future research are also discussed nucleotides during ischaemia leads to an elevated
at the end of this article. production of hypoxanthine, which is then

Free Radicals and Ischaemic Brain Injury
Free radicals are reactive chemical species that have an odd number of electrons. In the brain, the main free radicals are superoxide anion radical (O –), hydrogen peroxide (H O ), hydroxyl radical
metabolised by xanthine oxidase in a reaction that
forms O2–. Oxidation of NADPH by NADPH oxi- dase constitutes an additional source of O2–. Moreo- ver, O2– is also produced in the mitochondrial elec- tron transport chain. In isolated brain mitochondria, 2–5% of the electron flow leaks in the form of superoxide and hydrogen peroxide.[20] Hydrogen

2 2 2
peroxide is also the precursor of reactive OH. NO is

(OH), nitric oxide (NO) and peroxynitrite anion a physiological messenger in the CNS that is formed (ONOO–) [figure 1]. Although these radicals are from L-arginine in a reaction catalysed by NO

O2
NO L-Arginine
synthase.[17,20] NO is an unstable gas that reacts with O2– to form ONOO–.[17,20]

O2 

ONOO
These highly reactive radicals damage the cells by several mechanisms, including lipid peroxida- tion, tyrosine nitration, sulfhydryl oxidation and ni- trosylation and DNA breakage.[23,24] Recent studies

H2O2 OH
Fig. 1. Major source of free radicals in the ischaemic injured brain.
have also demonstrated that these radicals inhibit the mitochondrial respiratory chain and are involved in

apoptotic cell death through the cytochrome c medi- cy reported in subgroups of stroke patients.[36,37] ated pathway.[25] Cytochrome c is normally confined Both clinical trials involved small numbers of pa- in the intermembrane space of the organelle. Re- tients, that is, 150 patients in one trial[37] and 99 lease of cytochrome c can lead to necrosis through patients in the other.[36] Since no clear evidence for irreversible mitochondrial damage and collapse of the efficacy of ebselen in the treatment of stroke the energetic capacity of the cell. Alternatively, re- patients was obtained, development of this drug was lease of cytochrome c can also cause cell death recently terminated.[1]
through the intrinsic pathway by activation of the

apoptosome.[26,27] The initiation and execution
Tirilazad

phases of apoptosis are both dependent on caspases, The lazaroid compound tirilazad also possesses the cytosolic cysteine proteases that are responsible free radical scavenging activity. In animal models, for proteolytic cleavage of many intracellular pro- tirilazad treatment reduced infarct volume and im- teins, resulting in the morphological and biochemi- proved functional recovery after temporary or per- cal changes associated with apoptosis. More infor- manent occlusion of the middle cerebral artery.[10,38] mation about the production of the free radicals in No clinical trials have demonstrated a beneficial the ischaemic brain injury and the mechanisms un- effect of tirilazad on functional recovery in stroke
derlying radical-induced cell death can be found in patients.[39] No effect on infarct volume following
several review papers.[17,20,28,29] treatment of stroke patients with tirilazad has been

Development of Therapeutic Antioxidant Medications
shown either.[40] Indeed, a systemic review of four
published and two unpublished phase III clinical trials found an increase in the combined endpoint of

Strategies have been developed to reduce free death or disability in acute ischaemic stroke patients radical-induced damage processes in ischaemic treated with tirilazad.[39] Therefore, development of brain injury. These strategies include inhibiting the this medication for the treatment of acute stroke has production of free radicals, increasing the removal also been terminated.[39,40]
of free radicals or fostering the degradation of these

radicals. Several compounds have been developed
Edaravone

and few have also progressed into clinical trials, Edaravone inhibits lipid peroxidation by scav- including ebselen, tirilazad, edaravone and enging free radicals, such as O2–, NO and NXY-059. ONOO–.[41-43] The neuroprotective action of

2.1 Ebselen
edaravone has been investigated in animal models of
ischaemic brain injury. Accumulated evidence dem-

Ebselen is a selenium compound that possesses onstrates that edaravone markedly reduces infarct glutathione peroxidase-like activity and, therefore, volume and neurological deficits in both transient its use may eliminate free radicals during ischaemic and permanent ischaemic brain injury models.[43-45] brain injury.[30] Experimental studies in animals sug- In a hypoxic-ischaemic model, treatment with gest that ebselen may delay ischaemia-induced inju- edaravone significantly improved functional recov- ry processes. Treatment with ebselen reduced brain ery and also decreased ischaemia-induced morpho- damage in both transient and permanent focal mod- logical changes.[42] Furthermore, an experimental els of ischaemic brain injury when animals were study showed that treatment with edaravone re- sacrificed between 4–24 hours after ischaemia.[31-34] stored the antioxidant defence mechanisms compro- However, ebselen did not show protective effects mised by transgenic modification and protected the when the animals were sacrificed on day 7 after brain from ischaemic injury.[41] A clinical trial has ischaemia.[35] Two clinical trials of ebselen were also been undertaken to examine the therapeutic also carried out with limited neuroprotective effica- efficacy of edaravone in 252 patients with acute

ischaemic stroke.[46] In this trial, edaravone was infused at a dosage of 30mg twice daily for 14 days. Functional outcome was evaluated within 3 months after the onset of stroke using a modified Rankin Scale. The study showed that treatment with edaravone significantly improved functional out- come compared with the placebo group.
Edaravone has been approved by the regulatory authority in Japan for the treatment of stroke pa- tients. However, it should be mentioned that the regulatory process for drug licensing in Japan is different from that in Western countries, and the developmental status of this compound outside Ja- pan is unclear.[1]
Neuroprotective Action of NXY-059 in Animal Models of Ischaemic Stroke
The neuroprotective actions of NXY-059 have been examined in various stroke models in several species of animals. In a transient ischaemic injury model in rats, treatment with NXY-059 adminis- tered as a loading dose of 0.3–30 mg/kg followed by a continuing infusion of 0.3–30 mg/kg/hour for 48 hours reduced infarct volume in a dose-dependent manner, as evaluated on day 2 after the ischae- mia.[49] Neurological symptoms also improved at 24 and 48 hours after ischaemia in a dose-dependent manner. Furthermore, the therapeutic window for NXY-059 was wide because the drug substantially reduced infarct volume when it was given 5 hours after the ischaemia. Moreover, when the compound

NXY-059 was infused for 48 hours at a dosage of 30 mg/kg/
hour, a marked reduction in infarct volume was also Nitrone-derived free radical trapping agents were observed on day 7 after ischaemia, indicating the originally developed as tools for studying free radi- long-lasting effects of the drug. Similar neuropro- cal chemistry. The nitrone compound reacts with the tective actions of NXY-059 were also reported by
free radical to form a compound called a spin ad- other groups.[50,51]
duct. Once the adduct is formed, it is relatively In a rabbit embolic model of stroke, the neuro- stable and the radical thus becomes inactivated and protective effects of NXY-059 were evaluated with unable to damage cellular tissues. -Phenyl-N-tert- behavioural tests at 2 and 24 hours after ischae- butylnitrone (PBN), a first-generation free radical mia.[34] Intravenous infusion of NXY-059 100 mg/ trapping agent, was shown to trap short-lived free kg 5 minutes after embolisation significantly re- radicals such as alkoxyl, O2– and OH radicals.[34,47] duced neurological deficits. However, if drug ad- Although PBN can easily penetrate the blood-brain ministration was delayed to 3 hours after embolisa- barrier and is effective in the protection of the is- tion, neuroprotection was no longer observed. More chaemic injured brain in experimental animals, few interestingly, if NXY-059 and tissue plasminogen human data are available.[12,13] Therefore, the sec- activator (tPA) were administered 5 minutes and ond-generation free radical trapping agent 3 hours, respectively, after embolisation, this combi- NXY-059 was developed.[48] NXY-059 is structural- nation significantly reduced neurological deficits, ly related to the parent compound PBN but contains whereas tPA treatment alone did not have this effect. two sulfonyl groups that confer greater water solu- These results support that idea that NXY-059 effec- bility (figure 2). NXY-059 cannot easily pass tively increases the therapeutic window for tPA in through the blood-brain barrier, probably because of the treatment of ischaemic brain injury.
its high solubility in water.[12,49] The effect of NXY-059 has also been examined

NaO3S
O
N+
in a permanent focal cerebral ischaemic model. In a
stroke model in spontaneously hypertensive rats,[52] treatment with NXY-059 60 mg/kg/hour significant-

H SO3Na
Fig. 2. Chemical structure of disodium 4-[(tert-butylimino)methyl] benzene-1,3-disulfonate N-oxide (NXY-059; disufenton).
ly reduced cortical infarct volume measured 24 hours after ischaemia. However, treatment with NXY-059 30 mg/kg/hour did not significantly atten- uate cortical infarct volume.

The protective actions of NXY-059 have been Marmosets received either NXY-059 77 mol/kg tested in nonhuman primates. The efficacy of NXY- intravenously plus 154 mol/kg by subcutaneous 059 in reducing long-term functional disability and injection or saline followed by a further 48 hours of damage arising from cerebral ischaemia was evalu- intravenous infusion of NXY-059 85 mol/kg/hour ated in two studies using a marmoset permanent or saline. The dose of NXY-059 used in this study middle cerebral artery occlusion model.[53,54] In both was higher than that used in the first study but still studies, functional outcome was measured using the below dosage levels known to be safely tolerated in Hill and Valley tests. These tests examine the use of stroke patients. In the Hill test, NXY-059-treated each arm reaching into hemispace and allow differ- marmosets were significantly better at reaching with entiation of the effects of unilateral motor impair- their affected left arm into neglected left hemispace ment (i.e. confined to one arm in either hemispace) compared with the control group at both 3 and 10 from those of unilateral perceptual spatial impair- weeks after ischaemia. In the Valley test, there was ment (i.e. confined to one hemispace with either no significant difference in reaching with the affect- arm). ed left arm into right unneglected (motor) hemispace

In the first study,[54] 12 marmosets received a loading dose of NXY-059 28 mg/kg or saline 5 min- utes after the onset of ischaemia, followed by con- tinuous infusion of NXY-059 at 16 mg/kg/hour or saline for 48 hours. Treatment with NXY-059 sig- nificantly reduced functional disability of the affect- ed arm at 3 weeks and 10 weeks after ischaemia, as measured on the Hill and Valley tests. At these assessments, four of the six marmosets in the NXY- 059 group had near-normal motor performance. These observations showed that NXY-059 treatment reduces long-term disability and that the beneficial effect results from neuroprotection rather than from a delay in the appearance of damage. In addition, NXY-059 significantly reduced the degree of spatial perceptual neglect compared with saline-treated marmosets. These results might be of clinical rele- vance because cognitive deficits such as spatial neg- lect are common in stroke patients. Histological analysis showed that treatment with NXY-059 re- duced the volume of damage by >50% compared with controls. Further histological assessment
at 3 weeks, but there was a clear effect at 10 weeks. At 3 weeks, NXY-059-treated marmosets were sig- nificantly better than control animals at reaching with their unaffected right arm into left neglected hemispace on the Valley test (figure 3). At 10 weeks, both the NXY-059 and control groups could reach into left neglected hemispace with their unaf- fected right arm without any significant deficit. In addition, data from both the Hill and Valley tests showed that NXY-059-treated marmosets demon- strated a significant increase in frequency of use of the affected left arm at 3 weeks and at 10 weeks after ischaemia (figure 4).
3 weeks 10 weeks
15

Score
10

5

showed that protection was not only evident in the
cortex, but also extended to the white matter, cau- date and putamen, demonstrating that NXY-059
0
Right (neglect)
Left (motor)
Right (neglect)
Arm
Left (motor)

prevents damage to structures of the brain that are at risk in human stroke.
In a second study, the neuroprotective effects of NXY-059 were examined in the same model with the drug being started at a more clinically relevant time, that is, 4 hours after the MCA occlusion.[53]
Fig. 3. Average scores achieved by marmosets that received treat- ment with either disodium 4-[(tert-butylimino)methyl] benzene-1,3- disulfonate N-oxide (NXY-059; disufenton) or saline, as measured by the Valley staircase test at 3 and 10 weeks after the ischaemic brain injury. Four hours after the injury, the marmosets received NXY-059 or saline followed by a further 48-hour infusion of NXY- 059 or saline (reproduced from Marshall et al.,[53] with permission).
* p < 0.05; ** p < 0.01.

50

40

Attempts
30

20

10

0
3 weeks 10 weeks
Fig. 4. Left arm use attempt scores in disodium 4-[(tert-butylimi- no)methyl] benzene-1,3-disulfonate N-oxide (NXY-059; dis- ufenton)- or saline-treated marmosets, as measured by the Valley staircase test. Four hours after the injury, the marmosets received NXY-059 or saline followed by a further 48-hour infusion of NXY- 059 or saline (reproduced from Marshall et al.,[53] with permission).
* p < 0.01.
scale for disability (range 0–5, with 0 indicating no residual symptoms and 5 indicating bedbound, re- quiring constant care). The patients and investiga- tors were unaware of the treatment assignments until data collection had been completed. The results of the study showed that NXY-059 treatment signifi- cantly improved the overall distribution of scores on the modified Rankin scale compared with placebo (figure 5). The common odds ratio for improvement across all categories of the scale with NXY-059 was 1.20 (95% CI 1.01, 1.42). However, NXY-059 treat-
ment did not improve neurological functioning, as
measured on the National Institutes of Health Stroke Scale and Barthel index. Mortality and rates of serious and nonserious adverse events were similar in both groups. In a post hoc analysis of patients

Histological data from this study also support the who also received tPA, NXY-059 treatment was beneficial effects of NXY-059 in this primate associated with a lower incidence of haemorrhagic model.[53] Quantitative histological analysis of brain transformation and symptomatic intracranial haem- tissue was conducted 11 weeks after the middle

cerebral artery occlusion. Saline-treated animals had substantial damage in the right hemisphere of their brains that extended to subcortical structures, with almost total loss of caudate and putamen. In the NXY-059-treated group, the total infarct volume
a

11.0 20.0 11.7 12.7 20.6 24.0

15.4 18.0 11.4 14.2 16.9 24.0

Placebo group NXY-059 group
Score 0
1
2

3
4
5 or death

was 28% lower than in the controls and less damage was evident in the cortex, the white matter, the

Improvement in
Proportion of patients in the efficacy population (%)
4.5 2.4 2.2 3.7 0.0

caudate and the putamen, although the decrease was significant only for the putamen. Overall, these re- sults suggest that treatment with NXY-059 not only improves function recovery but also protects both grey and white matter in the brain from damage when the drug is administered 4 hours after the onset
NXY-059 group (%)

b
10.5 20.3 12.2 12.5 20.8 23.7

15.4 18.5 11.5 14.6 17.6 22.4

Placebo group NXY-059 group
Proportion of patients in the per-protocol population (%)

of ischaemia in the primate model of stroke.
Improvement in NXY-059 group (%)
4.9 3.1 2.4 4.5 1.3

Protective Actions of NXY-059 in Stroke Patients
In a large clinical trial (SAINT [Stroke Acute Ischaemic NXY-059 Treatment study]-I), the effects of NXY-059 on the reduction of disability were examined in patients with acute ischaemic stroke.[55] In this phase III clinical trial, 1722 patients were randomly assigned to receive a 72-hour intravenous infusion of placebo or NXY-059 within 6 hours after the onset of the stroke. The effects of the treatments were measured 90 days later on a modified Rankin
Fig. 5. Effect of treatment with disodium 4-[(tert-butylimino)methyl] benzene-1,3-disulfonate N-oxide (NXY-059; disufenton) within 6 hours after the onset of ischaemic stroke on the primary outcome,
i.e. distribution of scores on the modified Rankin scale for disability at 90 days or the last rating. Scores on this scale range from 0, indicating no residual symptoms, to 5, indicating bedbound, requir- ing constant care. Odds ratios were calculated with the use of proportional-odds logistic regression, assuming a common odds ratio across all cut points of the modified Rankin scale, and are provided as an estimate of the treatment effect. Treatment with NXY-059 significantly improved the primary outcome in both (a) the efficacy population (p = 0.038) and (b) the per-protocol population (p = 0.028) [reproduced from Lees et al.,[55] with permission,
 2006, Massachusetts Medical Society. All rights reserved].

orrhage. Moreover, therapeutic concentrations of chaemic injured brain through additional mecha- NXY-059 were well tolerated by patients. These nisms. For example, reduction of apoptotic cell results suggest that administration of NXY-059 death by NXY-059 has been suggested. Treatment within 6 hours after the onset of acute ischaemic with NXY-059 prevented the secondary decline in stroke significantly reduces disability but does not mitochondrial respiratory function and release of significantly improve neurological functioning. cytochrome c in ischaemic injured brain.[51] This

Results from the recently completed SAINT II clinical trial (Shuaib A., unpublished observations) showed that treatment with NXY-059 did not achieve the primary outcome of a statistically signif- icant reduction in stroke-related disability, as as- sessed by the modified Rankin Scale (p = 0.33, odds ratio 0.94) compared with the placebo group. Sub- group analyses, including time to treatment, also did not demonstrate a treatment benefit. In addition, treatment with NXY-059 did not result in a statisti- cally significant improvement in neurological status compared with placebo on the National Institutes of Health Stroke Scale (p = 0.70), which is consistent
treatment also prevented a decrease in anti-apoptotic kinase Akt during ischaemia.[50] Moreover, treat- ment with NXY-059 also reduced the number of terminal deoxynucleotidyl transferase biotin-d-uri- dine triphosphate nick end labelling (TUNEL)-posi- tive cells at the haematoma margin.[56] In addition, NXY-059 may contribute to the recovery of is- chaemic brain injury by inhibiting inflammatory reactions. It has been shown that treatment with NXY-059 reduces neutrophil infiltrate at the vicini- ty of the haematoma in a haemorrhagic stroke model.[56]

with the findings of the SAINT I trial. There was no 3. Is the Concept of Neuroprotection for evidence that NXY-059 lowered the incidence of the Treatment of Ischaemic Brain symptomatic intracranial haemorrhage when admin- Injury Dead?
istered with tPA (p = 0.56). Mortality and the inci-

dence and profile of adverse events in patients re- ceiving NXY-059 were similar to those in the place- bo group. Detailed results of this trial are not available yet, but should be published soon. At the present time, the manufacturer plans no further de- velopment of NXY-059 in acute ischaemic brain stroke. The lack of efficacy of NXY-059 in acute stroke patients in the SAINT II is unfortunate, and also suggests that development of new treatments for stroke patients is extremely challenging. Howev- er, the scientific community will gain experience from the development of NXY-059 that will benefit further stroke research.

Additional Mechanisms of Neuroprotective Actions of NXY-059
The negative results of clinical trials of free radi- cal scavengers in acute stoke have been very disap- pointing. However, with each of these trials we have continued to learn more about the complexity of the nature of translational research from animal models to the bedside. The important question at this stage is: is the concept of neuroprotection as a mode of treatment in acute stroke dead?
The concept of neuroprotection involves the ter- mination of one or more ischaemia-initiated destruc- tive cascades in order to reduce secondary injury. To date, >11 000 patients have participated in >65 clinical trials of neuroprotective therapies.[57] De- spite this enormous effort, all compounds that have reached clinical trials have failed because of lack of demonstrable efficacy or problems with toxicity. At time of writing, no neuroprotective drugs have been

Free radicals play a critical role in the secondary approved for the treatment of acute stroke. Recently, injury processes after ischaemia, and it is logical to the second NXY-059 phase III clinical trial was hypothesise that the neuroprotective actions of completed (SAINT II) and treatment with this drug NXY-059 are mediated by removal of these radicals did not show significant improvement in the out- from the injured brain. However, it is also worth come of stroke. This failure may give rise to a mentioning that NXY-059 may protect the is- pessimistic view of neuroprotective strategies in

clinical and basic research. Nevertheless, the search Although clinical trials have produced many neg-

for new treatments for this disease will continue, simply because patients urgently need it. However, we may need to rethink our strategies for future studies.
The following are some key elements that we may need to address in the future development of neuroprotective strategies. First, despite ever-in- creasing knowledge of the biochemical changes that
ative results, newer trials are taking treatment into the ambulance so that the drug may reach the brain as soon as possible after the injury, for example, FASTMAG (Field Administration of Stroke Ther- apy – Magnesium).[59] Other clinical trials of neuroprotective agents for the treatment of acute stroke are also ongoing.[60] We hope such innovative strategies may improve our odds of success. The fact that there are several clinical trials of neuroprotec-

occur in the brain following an ischaemic insult, the tion underway would suggest that the concept of
exact mechanism whereby neurons die, especially in neuroprotection is not dead at this time. human cerebral ischaemia, remains elusive. Future
studies aimed at understanding cell death pathways, 4. Conclusions
with elucidation of underlying molecular cellular
and mechanisms, may facilitate the development of Stroke is the third most common cause of death new therapeutic strategies for arresting such dam- in both industrialised and many developing coun- age. Secondly, ischaemic cell death is the result of tries. At present, the two main approaches for the complex processes compromising many noxious treatment of acute ischaemic stroke are thrombol- cascades, for example, oxidative stress, excitotoxici- ysis therapy and neuroprotection. Thrombolysis

ty, functional failure of ionic pumps, inflammatory reactions and activation of apoptotic death path- ways. Termination of one of these cascades may not be sufficient to reduce ischaemia-induced injury processes. A broader-spectrum compound blocking several ischaemia-initiated cascades will have a bet- ter chance of inhibiting the cell death processes. Thirdly, combination therapy with different com- pounds targeting different death pathways, for ex- ample, free radical scavengers plus anti-inflammato- ry agents, may offer a better chance than single medications. Fourthly, Schaller et al.[58] have pro- posed an interesting idea following identification of a common DNA regulatory element in the promoter
therapy is aimed at restoring blood flow to the compromised region, which is reasonable as this disorder is caused by the obstruction of major arter- ies. Improvement in outcome has been shown fol- lowing therapeutic interventions that restore blood flow by thrombolysis, primarily with tPA; however, only a very small fraction of stroke patients are eligible for this thrombolytic therapy, and this treat- ment also causes haemorrhage transformation in the ischaemic injured brain. In addition, there is also a high risk of death and permanent disability for pa- tients who receive thrombolytic therapy. The second approach of neuroprotection involves the use of neuroprotective agents to terminate one or more ischaemia-initiated cascades and thus reduce secon-

regions of genes that are upregulated by ischaemic dary injury processes. A variety of neuroprotective
tolerance in cells. Small molecules capable of acti- strategies, including antioxidants, have been devel- vating this common DNA regulatory element could oped and have been associated with very encourag- augment an endogenous protective response, and ing results in numerous experimental stroke studies. could therefore become a class of novel neuropro- In contrast, clinical studies so far have failed to tective medications. Finally, we may need to test demonstrate similar efficacy. Among these, compounds in a small number of patients with acute NXY-059, a newer-generation free radical trapping ischaemic stroke, using newer technologies (espe- agent, has shown neuroprotection not only in a small
cially magnetic resonance imaging-based tech- animal model, but also in a non-human primate
niques) to determine if there is any signal for success model of stroke. However, a recently completed prior to embarking on large trials. second phase III clinical trial was unable to show

significant protection with this compound in stroke
and magnesium in rats subjected to reversible focal cerebral ischemia. Neurosurgery 1999 Jan; 44 (1): 163-71

patients. These findings indicate that the develop- 11. Hallenbeck JM, Dutka AJ, Tanishima T, et al. Polymorphonu-

ment of neuroprotective therapy for treatment of acute ischaemic stroke is extremely challenging.
clear leukocyte accumulation in brain regions with low blood flow during the early postischemic period. Stroke 1986 Mar- Apr; 17 (2): 246-53

This finding also indicates that the mechanisms re- 12. Green AR, Ashwood T, Odergren T, et al. Nitrones as neuropro-

sponsible for cell death during ischaemia are not well understood and that further research is needed
tective agents in cerebral ischemia, with particular reference to NXY-059. Pharmacol Ther 2003 Dec; 100 (3): 195-214
13. Yang Y, Li Q, Shuaib A. Neuroprotection by 2-h postischemia

to define the mechanisms underlying ischaemic cell administration of two free radical scavengers, alpha-phenyl-n-

death. Additionally, neuroprotective agents that more effectively terminate cell death pathways may
tert-butyl-nitrone (PBN) and N-tert-butyl-(2-sulfophenyl)-ni- trone (S-PBN), in rats subjected to focal embolic cerebral ischemia. Exp Neurol 2000 May; 163 (1): 39-45

be needed for treatment of this disorder. This, to- 14. Yang G, Chan PH, Chen J, et al. Human copper-zinc superoxide

gether with better animal models simulating stroke patients, may be essential if neuroprotection is to
dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia. Stroke 1994 Jan; 25 (1): 165-70

succeed as an effective treatment in stroke. 15. Kinouchi H, Epstein CJ, Mizui T, et al. Attenuation of focal
cerebral ischemic injury in transgenic mice overexpressing

Acknowledgements
This work is supported by grants from the Canadian Institutes of Health Research, the Heart and Stroke Founda- tion of Canada and AstraZeneca. Dr Wang and Dr Shuaib have received grants from AstraZeneca. Dr Shuaib has also acted as a consultant to AstraZeneca.

References
CuZn superoxide dismutase. Proc Natl Acad Sci U S A 1991 Dec 15; 88 (24): 11158-62
Siesjo BK, Agardh CD, Bengtsson F. Free radicals and brain damage. Cerebrovasc Brain Metab Rev 1989 Fall; 1 (3): 165- 211
Rodrigo J, Fernandez AP, Serrano J, et al. The role of free radicals in cerebral hypoxia and ischemia. Free Radic Biol Med 2005 Jul 1; 39 (1): 26-50
Alexandrova ML, Bochev PG. Oxidative stress during the chronic phase after stroke. Free Radic Biol Med 2005 Aug 1; 39 (3): 297-316
Schaller B. Introduction to serial reviews on free radicals and

Green AR, Shuaib A. Therapeutic strategies for the treatment of stroke. Free Radic Biol Med 2005 Feb 15; 38 (4): 409-10 stroke. Drug Discov Today 2006 Aug; 11 (15-16): 681-93 20. Moro MA, Almeida A, Bolanos JP, et al. Mitochondrial respira-
He J, Klag MJ, Wu Z, et al. Stroke in the People’s Republic of tory chain and free radical generation in stroke. Free Radic
China I: geographic variations in incidence and risk factors. Biol Med 2005 Nov 15; 39 (10): 1291-304
Stroke 1995 Dec; 26 (12): 2222-7 21. Chan PH. Role of oxidants in ischemic brain damage. Stroke
Flamm ES, Demopoulos HB, Seligman ML, et al. Free radicals 1996 Jun; 27 (6): 1124-9
in cerebral ischemia. Stroke 1978 Sep-Oct; 9 (5): 445-7 22. Stadtman ER, Oliver CN. Metal-catalyzed oxidation of proteins:
Morimoto T, Globus MY, Busto R, et al. Simultaneous mea- physiological consequences. J Biol Chem 1991 Feb 5; 266 (4): surement of salicylate hydroxylation and glutamate release in 2005-8
the penumbral cortex following transient middle cerebral ar- 23. Freeman BA, Crapo JD. Biology of disease: free radicals and
(1): 92-9
tery occlusion in rats. J Cereb Blood Flow Metab 1996 Jan; 16 tissue injury. Lab Invest 1982 Nov; 47 (5): 412-26

Watson BD, Busto R, Goldberg WJ, et al. Lipid peroxidation in vivo induced by reversible global ischemia in rat brain. J Neurochem 1984 Jan; 42 (1): 268-74
Nelson CW, Wei EP, Povlishock JT, et al. Oxygen radicals in cerebral ischemia. Am J Physiol 1992 Nov; 263 (5 Pt 2): H1356-62
Cao W, Carney JM, Duchon A, et al. Oxygen free radical involvement in ischemia and reperfusion injury to brain. Neurosci Lett 1988 May 26; 88 (2): 233-8
Kato N, Yanaka K, Nagase S, et al. The antioxidant EPC-K1 ameliorates brain injury by inhibiting lipid peroxidation in a rat model of transient focal cerebral ischaemia. Acta Neurochir (Wien) 2003 Jun; 145 (6): 489-93
Lancelot E, Callebert J, Revaud ML, et al. Detection of hydrox- yl radicals in rat striatum during transient focal cerebral ische- mia: possible implication in tissue damage. Neurosci Lett 1995
Kehrer JP. Free radicals as mediators of tissue injury and
disease. Crit Rev Toxicol 1993; 23 (1): 21-48
Lo EH, Dalkara T, Moskowitz MA. Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci 2003 May; 4 (5): 399-415
Hengartner MO. The biochemistry of apoptosis. Nature 2000 Oct 12; 407 (6805): 770-6
Yuan J, Yankner BA. Apoptosis in the nervous system. Nature 2000 Oct 12; 407 (6805): 802-9
Beckman JS. Interactions of oxidants, nitric oxide and antioxi- dant defences in cerebral ischemia and injury. In: Ginsberg M, Bogousslavsky J, editors. Malden: Blackwell Science, 1998: 455-70
Margaill I, Plotkine M, Lerouet D. Antioxidant strategies in the treatment of stroke. Free Radic Biol Med 2005 Aug 15; 39 (4): 429-43

Sep 8; 197 (2): 85-8 30. Noguchi N, Yoshida Y, Kaneda H, et al. Action of ebselen as an
10. Schmid-Elsaesser R, Zausinger S, Hungerhuber E, et al. antioxidant against lipid peroxidation. Biochem Pharmacol Neuroprotective effects of combination therapy with tirilazad 1992 Jul 7; 44 (1): 39-44

Dawson DA, Masayasu H, Graham DI, et al. The neuroprotec- 47. Maples KR, Ma F, Zhang YK. Comparison of the radical tive efficacy of ebselen (a glutathione peroxidase mimic) on trapping ability of PBN, S-PPBN and NXY-059. Free Radic brain damage induced by transient focal cerebral ischaemia in Res 2001 Apr; 34 (4): 417-26
the rat. Neurosci Lett 1995 Feb 6; 185 (1): 65-9 48. Green AR, Ashwood T. Free radical trapping as a therapeutic
Takasago T, Peters EE, Graham DI, et al. Neuroprotective approach to neuroprotection in stroke: experimental and efficacy of ebselen, an anti-oxidant with anti-inflammatory clinical studies with NXY-059 and free radical scavengers. actions, in a rodent model of permanent middle cerebral artery Curr Drug Targets CNS Neurol Disord 2005 Apr; 4 (2): 109- occlusion. Br J Pharmacol 1997 Nov; 122 (6): 1251-6 18
Imai H, Masayasu H, Dewar D, et al. Ebselen protects both gray 49. Kuroda S, Tsuchidate R, Smith ML, et al. Neuroprotective and white matter in a rodent model of focal cerebral ischemia. effects of a novel nitrone, NXY-059, after transient focal Stroke 2001 Sep; 32 (9): 2149-54 cerebral ischemia in the rat. J Cereb Blood Flow Metab 1999
Lapchak PA, Chapman DF, Zivin JA. Metalloproteinase inhibi- Jul; 19 (7): 778-87
tion reduces thrombolytic (tissue plasminogen activator)-in-

duced hemorrhage after thromboembolic stroke. Stroke 2000 Dec; 31 (12): 3034-40
50. Yoshimoto T, Kanakaraj P, Ying Ma J, et al. NXY-059 main- tains Akt activation and inhibits release of cytochrome C after focal cerebral ischemia. Brain Res 2002 Aug 30; 947 (2):

Salom JB, Perez-Asensio FJ, Burguete MC, et al. Single-dose 191-8
ebselen does not afford sustained neuroprotection to rats sub-

jected to severe focal cerebral ischemia. Eur J Pharmacol 2004 Jul 8; 495 (1): 55-62
Ogawa A, Yoshimoto T, Kikuchi H, et al. Ebselen in acute middle cerebral artery occlusion: a placebo-controlled, double-
51. Yoshimoto T, Kristian T, Hu B, et al. Effect of NXY-059 on
secondary mitochondrial dysfunction after transient focal is- chemia; comparison with cyclosporin A. Brain Res 2002 Apr 5; 932 (1-2): 99-109

blind clinical trial. Cerebrovasc Dis 1999 Mar-Apr; 9 (2): 112- 52. Zhao Z, Cheng M, Maples KR, et al. NXY-059, a novel free
8 radical trapping compound, reduces cortical infarction after

Yamaguchi T, Sano K, Takakura K, et al. Ebselen in acute ischemic stroke: a placebo-controlled, double-blind clinical
permanent focal cerebral ischemia in the rat. Brain Res 2001 Aug 3; 909 (1-2): 46-50

trial. Ebselen Study Group. Stroke 1998 Jan; 29 (1): 12-7 53. Marshall JW, Cummings RM, Bowes LJ, et al. Functional and

Schmid-Elsaesser R, Zausinger S, Hungerhuber E, et al. Monotherapy with dextromethorphan or tirilazad – but not a combination of both – improves outcome after transient focal
histological evidence for the protective effect of NXY-059 in a primate model of stroke when given 4 hours after occlusion. Stroke 2003 Sep; 34 (9): 2228-33

cerebral ischemia in rats. Exp Brain Res 1998 Sep; 122 (1): 54. Marshall JW, Duffin KJ, Green AR, et al. NXY-059, a free 121-7 radical: trapping agent, substantially lessens the functional
Bath PM, Iddenden R, Bath FJ, et al. Tirilazad for acute is- disability resulting from cerebral ischemia in a primate spe-
chaemic stroke. Cochrane Database Syst Rev 2001; (4): cies. Stroke 2001 Jan; 32 (1): 190-8 CD002087

van der Worp HB, Kappelle LJ, Algra A, et al. The effect of tirilazad mesylate on infarct volume of patients with acute ischemic stroke. Neurology 2002 Jan 8; 58 (1): 133-5
Zhu C, Wang X, Huang Z, et al. Apoptosis-inducing factor is a major contributor to neuronal loss induced by neonatal cere- bral hypoxia-ischemia. Cell Death Differ 2007; 14: 775-84
Noor JI, Ikeda T, Mishima K, et al. Short-term administration of a new free radical scavenger, edaravone, is more effective than its long-term administration for the treatment of neonatal hyp-
Lees KR, Zivin JA, Ashwood T, et al. NXY-059 for acute
ischemic stroke. N Engl J Med 2006 Feb 9; 354 (6): 588-600
Peeling J, Del Bigio MR, Corbett D, et al. Efficacy of disodium 4-[(tert-butylimino)methyl]benzene-1,3-disulfonate N-oxide (NXY-059), a free radical trapping agent, in a rat model of hemorrhagic stroke. Neuropharmacology 2001 Mar; 40 (3): 433-9
Hankey GJ. Neuroprotection for acute ischaemic stroke: hope reignited. Lancet Neurol 2006 Apr; 5 (4): 287-8

oxic-ischemic encephalopathy. Stroke 2005; 36: 2468-74 58. Schaller B, Graf R, Jacobs AH. Ischaemic tolerance: a window
Shichinohe H, Kuroda S, Yasuda H, et al. Neuroprotective to endogenous neuroprotection? Lancet 2003 Sep 27; 362 effects of the free radical scavenger edaravone (MCI-186) in (9389): 1007-8
mice permanent focal brain ischemia. Brain Res 2004 Dec 17; 59. Saver JL, Kidwell C, Eckstein M, et al. Physician-investigator 1029 (2): 200-6 phone elicitation of consent in the field: a novel method to
Zhang N, Komine-Kobayashi M, Tanaka R, et al. Edaravone obtain explicit informed consent for prehospital clinical re-
reduces early accumulation of oxidative products and sequen- search. Prehosp Emerg Care 2006 Apr-Jun; 10 (2): 182-5 tial inflammatory responses after transient focal ischemia in

mice brain. Stroke 2005; 36: 2220-5
Amemiya S, Kamiya T, Nito C, et al. Anti-apoptotic and neuroprotective effects of edaravone following transient focal ischemia in rats. Eur J Pharmacol 2005 Jun 1; 516 (2): 125-30
Edaravone Acute Infarction Study Group. Randomized, place- bo-controlled, double-blind study at multicenters: effect of a novel free radical scavenger, edaravone (MCI-186), on acute brain infarction. Cerebrovasc Dis 2003; 15 (3): 222-9
60. American Stroke Association. Major ongoing stroke trials.
Stroke 2006 Feb; 37 (2): e18-26

Correspondence: Dr Chen X. Wang, Stroke Research Labo- ratory, 533 HMRC, University of Alberta, Edmonton, AB T6G 2S2, Canada.
E-mail: [email protected]