Role of MC-1 alone and in combination with tissue plasminogen activator in focal ischemic brain injury in rats

Transcription

1 J Neurosurg 103: , 2005 Role of MC-1 alone and in combination with tissue plasminogen activator in focal ischemic brain injury in rats CHEN XU WANG, M.D., PH.D., TAO YANG, M.S., RAZA NOOR, B.S., AND ASHFAQ SHUAIB, M.D., F.R.C.P.(C) Stroke Research Laboratory, University of Alberta, Edmonton, Canada Object. Pyridoxal-5-phosphate (PLP), the biologically active form of pyridoxine, can rescue neurons from death in vitro and in vivo. In the present project, the authors have studied whether MC-1, an analog of PLP, alone or in combination with the thrombolytic agent tissue plasminogen activator (tpa), can protect the brains of rats injured by ischemia. Methods. Ischemic brain injury was induced in rats by injecting a preformed blood clot in the middle cerebral artery (MCA). Neurological deficits and infarct volumes caused by the embolus were measured to evaluate the effects of MC- 1 on the ischemic injury. Systemic blood pressure and local brain blood flow were also monitored. Administration of different doses of MC-1 1 hour after embolization significantly reduced the infarct volume and improved functional recovery. Injection of MC-1 (40 mg/kg) at 3 or 6 hours after embolization also reduced the volume of the infarct significantly and improved functional recovery. Combined treatment with MC-1 and tpa was also neuroprotective, although it was not superior to treatment involving either MC-1 or tpa alone. Treatment with MC-1 did not result in significant changes in either systemic blood pressure or local blood flow in the ischemic brain. Conclusions. These data support the hypothesis that in the focal embolic stroke model in rats MC-1 is a neuroprotective agent. The neuroprotection this compound provides still exists when MC-1 administration is delayed up to 6 hours after ischemic injury. KEY WORDS infarction stroke middle cerebral artery embolization rat P YRIDOXAL-5-phosphate, the biologically active form of pyridoxine, serves a vital role in metabolism as a coenzyme for a wide variety of metabolic transformations of amino acids, including decarboxylation, transamination, and racemization, as well as for enzymatic steps in the metabolism of tryptophan, sulfur-containing amino acids, and hydroxyl amino acids. 18 Pyridoxal-5-phosphate is also involved in neuronal differentiation, proliferation, migration, and maturation during brain development. 10 Moreover, PLP exerts neurotrophic effects on a wide range of neurons from different brain regions. 6 Recently, in both in vitro and in vivo studies investigators have demonstrated that PLP can also rescue neurons from death. For example, in an in vitro study, PLP was shown to attenuate cellular injury induced by glucose deprivation in cultured hippocampal neurons. 7 Application of PLP also protects cultured Abbreviations used in this paper: ATP = adenosine triphosphate; CA = carotid artery; ECA = external CA; GABA = -aminobutyric acid; LBF = local blood flow; MABP = mean arterial blood pressure; MCA = middle cerebral artery; PLP = pyridoxal-5-phosphate; SEM = standard error of the mean; tpa = tissue plasminogen activator; TTC = 2,3,5-triphenyltetrazolium chloride. retinal neurons against glutamate-induced cytotoxicity. 9 In a model of whole-brain complete ischemia, PLP treatment significantly reduced cell death in the retina and in the hippocampus in vivo. 21,25 In addition, in both in vitro and in vivo studies, PLP has been shown to be a potential protective agent in cases of regional myocardial ischemia. 8,14 In the present study, we examined the protective effects of MC-1, an analog of PLP, by using a focal embolic model of brain ischemia. In a preliminary study, we examined the protective effects of MC-1 in the presence of ischemic brain injury. In that study, MC-1 (10, 20, or 40 mg/kg) was administered 1 hour after ischemia and the volume of the infarct was measured 48 hours later. Our results demonstrated that treatment with 40 mg/kg MC-1 reduced infarct volume significantly. Based on the results from this preliminary study, we designed a series of experiments to examine the following: 1) the effects of higher doses of MC-1 on ischemic brain injury; 2) the effects of MC-1 on ischemic brain injury when administered at different time points after injury; and 3) whether MC-1 in combination with tpa can provide better protection for the brain with an ischemic injury. 165

2 C. X. Wang, et al. Animal Preparation Materials and Methods Male Wistar rats, each weighing between 300 and 350 g, were purchased from Charles River Canada, St. Constant, QB, Canada. The rats were housed in a 12-hour light/dark cycle and allowed free access to water and food. The animals were not forced to fast before surgery. Animal care and general protocols for animal use were approved by the Animal Ethics Committee of the University of Alberta. Animal Model A focal brain injury was created by injecting a preformed clot into the MCA as reported previously. 23,24 In brief, anesthesia was initially induced by administering 3% halothane in a mixture of O 2 and N 2 O and was then maintained by reducing the percentage of halothane to 1.5% for the duration of the surgery. Each animal s body temperature was maintained at 37 C by using a heating pad during surgery and in the immediate postoperative period until the animal recovered fully from the anesthesia. A 1.5-cm longitudinal incision was made in the midline of the ventral cervical skin. The right common CA, right internal CA, and right ECA were exposed, and the distal portion of the ECA was ligated and cut. A modified polyethylene (PE- 10) catheter filled with bovine thrombin (Warner-Lambert Canada, Inc., Scarborough, ON, Canada), was introduced into the lumen of the right ECA through a small puncture hole. Ten microliters of blood was withdrawn into the catheter and retained there for 15 minutes to allow formation of a clot. Once the clot had formed, the catheter was advanced 17 mm into the internal CA until its tip was 1 to 2 mm away from the origin of the MCA. The preformed clot in the catheter was then injected and the catheter was removed. The wound was closed and the animal was returned to its cage. Mean Arterial Blood Pressure and LBF in the Brain Mean arterial blood pressure and LBF in the brain were recorded using a Biopac system (Biopac Systems, Inc., Santa Barbara, CA). For the MABP measurement a catheter filled with saline and 0.6 U/ml heparin was introduced into the left femoral artery. Following this procedure, ischemia was induced, the animal was mounted on a stereotactic frame, and a laser Doppler probe, used for LBF measurement, was implanted into the right striatum, as described previously. 20 The coordinates for probe implantation were 0.2 mm anterior to the bregma, 3 mm lateral to the midline, and 5 mm below the dura mater. The MABP was measured before and after the ischemic injury. To avoid variations in flow caused by reposition of the Doppler probe, we measured LBF only after the ischemic injury. In this series of experiments, the animals were randomly assigned into a control group (four rats) or an MC-1 treated group (four animals). Behavioral Test Neurological deficits were investigated at 2 and 24 hours after embolization. The neurological deficits were evaluated using a modified Bederson system 2 with the following scores: 0, no observable deficit; 1, forelimb flexion; 2, forelimb flexion plus decreased resistance to lateral push; 3, unidirectional circling; and 4, unidirectional circling plus a decreased level of consciousness. Quantification of Infarct Volume The procedures we used to assess infarct volume have been detailed previously. 23,24 Briefly, at the end of each experiment (48 hours after onset of ischemia), the anesthetized animal was killed by decapitation and the brain was removed. For the morphometric study, 2-mm-thick coronal sections were cut using a rat brain matrix. A total of eight coronal sections were collected, and these sections were stained using a 2% TTC solution. We used a color flatbed scanner to scan the stained brain sections and analyzed the images by using Adobe PhotoShop (Adobe Systems, Inc., San Jose, CA). The total volume of each infarct was determined by integration of the areas from the sections. The infarct volume was expressed as a percentage of the total volume of the ipsilateral hemisphere. Therapeutic Regimen There were eight animal groups in this study. In the first group (control group, 10 rats), saline was administered. In three treatment groups (10 rats/group), MC-1 was infused at doses of 40, 80, or 120 mg/kg, respectively. In the first four groups, the infusion of either saline or MC-1 began 1 hour after the ischemic injury. In the fifth and sixth groups (10 rats/group), MC-1 (40 mg/kg) was infused at either 3 or 6 hours after the injury. In the seventh group (eight rats), tpa (10 mg/kg) was infused; the infusion started 1 hour postinjury and lasted for 30 minutes. In the eighth group (eight rats), MC-1 (40 mg/kg) and tpa (10 mg/kg) were both infused. All drugs or saline were infused intravenously. For MC-1 infusions, one third of the dose was infused as a bolus and the remainder was infused over a period of 30 minutes. For the combination study, infusion of MC-1 and tpa started 1 hour postinjury, and the infusion period lasted 30 minutes. Statistical Analysis The differences in infarct volume were analyzed by performing a one-way analysis of variance followed by the Tukey test. Neurological deficit scores were reported as medians and interquartile ranges, taking into consideration the 25th to 75th percentiles. The neurological scores were analyzed by performing a Kruskal Wallis test when there were more than two groups. The Mann Whitney U-test was used when a comparison was made between two groups. The mortality rates following different treatments were compared using a chisquare test. Both MABP and LBF were analyzed by performing a repeated-measures analysis of variance. Differences were considered significant when the probability value was less than Blood Pressure and LBF Results The MABP and mean LBF values for saline- and MC- 1 treated groups are shown in Table 1. Values of MABP and LBF did not significantly differ between the control and MC-1 treated groups at any time point measured. Infarct Volume In a preliminary study, the effects of MC-1 (10, 20, or 40 mg/kg) on ischemic injury were examined and only MC- 1 at a dose of 40 mg/kg reduced the volume of the infarct significantly (data not shown). To characterize the neuroprotective actions of MC-1 further, we first examined the effects of this compound again by using higher doses (p 0.05; Figs. 1 and 2). Injection of the preformed blood clot resulted in an infarction in the territory irrigated by the MCA, which is mainly located in the cerebral cortex and striatum. In the control group, the infarct volume was % (mean SEM) 48 hours after the clot entered the MCA. In the animal groups given 40, 80, or 120 mg/kg MC-1 at 1 hour after embolization, the mean infarct volumes were , , and %, respectively; these values were significantly smaller than the mean value measured in the control group (p 0.05). Differences among the three groups of animals that received different doses of MC-1 were not significant. Second, we studied whether MC-1 was still an effective treatment for ischemic injury when it was administered later than 1 hour after embolization. The infarct volumes in groups that received MC-1 (40 mg/kg) at 3 and 6 hours postinjury were and %, respectively; these values were significantly smaller than the mean value measured in the control group (p 0.05; Fig. 1). Differences in infarct volumes among the three groups of animals 166

3 Ischemic brain injury and MC-1 TABLE 1 Blood pressure and LBF in rats with ischemic brain injury* After Drug or Vehicle Infusion Factor Before Ischemia Before Drug Treatment 1 Hr 2 Hrs 3 Hrs BP (mm Hg) saline-treated MC-1 treated LBF (BPU) saline-treated NM MC-1 treated NM * Values are expressed as means SEMs. There were four animals in each group. The MABP was measured before MCA occlusion, 30 minutes before saline or drug infusion, and 1, 2, and 3 hours after the treatment began; the LBF was measured 30 minutes before saline or drug infusion and 1, 2, and 3 hours afterward. Abbreviations: BPU = blood perfusion unit; NM = not measured. that received MC-1 at 1, 3, or 6 hours after embolization, however, were not significant. Third, we examined the effects of MC-1 in combination with tpa. Infarct volumes in groups that received tpa alone and MC-1 plus tpa were and %, respectively; these values were significantly smaller than the mean value measured in the control group (p 0.05; Figs. 1 and 2). Nevertheless, differences in infarct volumes among the groups of animals that received treatment with MC-1 alone, tpa alone, and MC-1 plus tpa were not significant. Neurological Deficits Neurological deficit scores following treatment with MC-1 alone or in combination with tpa are shown in Table 2. Two hours after MCA occlusion all animals displayed motor deficits, with mean scores of 3 in all eight groups. Twenty-four hours after MCA occlusion, differences in the neurological deficit scores were not significant among these groups. In the control group the animals neurological deficit scores obtained 24 hours after MCA occlusion were not significantly different from their scores obtained 2 hours after the occlusion. In rats given 40 mg/kg MC-1 and in those given 80 mg/kg MC-1 the neurological deficit scores were significantly improved after 24 hours, compared with the scores obtained 2 hours after MCA occlusion. In rats given MC-1 3 hours after injury the neurological deficits were also significantly improved at 24 hours, compared with the scores obtained 2 hours after MCA occlusion. There were no significant changes in neurological deficit scores in rats in the other treatment groups between 2 and 24 hours after MCA occlusion (Table 2). Mortality Rates In the control group, four rats died prematurely. In the 80 mg/kg MC-1 treatment group three rats died and in the 120 mg/kg MC-1 treatment group one rat died prematurely. In FIG. 1. Representative TTC-stained brain sections from different animal groups. Injection of a preformed clot into the MCA resulted in a large infarction in the ipsilateral brain. Treatment with MC-1 alone (40, 80, or 120 mg/kg [40, 80, and 120, respectively] given 1 hour postinjury or in combination with tpa reduced the infarction significantly. This effect was also observed when 40 mg/kg MC-1 was given 3 or 6 hours postinjury (3 h and 6 h, respectively). 167

4 C. X. Wang, et al. TABLE 2 Neurological deficit scores in control rats and rats treated with MC-1 and/or tpa* NDS at 2 Hrs NDS at 24 Hrs Group No. of Rats Median (1QR) Median (IQR) FIG. 2. Bar graph depicting the effects of MC-1 alone or in combination with a thrombolytic agent, tpa, on infarct volume following ischemic brain injury. Error bars represent means plus SEMs. Compared with the control group, infarct volumes in all drug-treated groups were significantly smaller. rats given MC-1 3 hours after ischemia, one rat died prematurely. All the rats that received 40 mg/kg MC-1 at 1 or 6 hours after ischemic injury survived to the end of the experiment as did all the rats that received tpa alone or tpa plus MC-1. Compared with the control group, there was no significant difference in the mortality rates in the drug-treated groups. Discussion In the present study, we examined the effects of MC-1, a PLP analog, in ischemic brain injury by using a focal embolic model of stroke. From this study we learned that MC- 1 was effective in protecting the brain following an injury when 40 to 120 mg/kg of the compound was administered. This neuroprotective action of MC-1 still existed when it was administered 6 hours after injury. Treatment with MC- 1 also improved functional recovery. Although treatment with MC-1 in combination with tpa was not superior to treatment with either MC-1 or tpa alone, the combined treatment also reduced infarct volume significantly. It is not clear why together MC-1 and tpa did not produce additive or synergistic neuroprotection in this embolic model. Some rats died before the end of the experiments. The major cause of this premature death in the rats that died before the end of the experiements was ischemia-induced edema, which occurred during infarction evolution. To study whether the neuroprotective effects of MC-1 are due to its action on changes in systemic blood pressure or LBF in the brain, these parameters were monitored after treatment with MC-1 in an additional series of experiments. Our results demonstrated that there were no significant changes in MABP and LBF in animals that received MC-1 treatment compared with control animals. These results indicate that additional mechanisms are involved in the protection given by MC-1 to the ischemic brain. There are several possibilities of how MC-1 can protect a brain following ischemic injury. First, prevention of a calcium overload may be one explanation. Increases in intracellular calcium in the injured brain have been observed in saline-treated 10 3 (4 3) 3 (3 2) MC-1 given 1 hr postinjury 40 mg/kg 10 3 (3 3) 2 (3 2) 80 mg/kg 10 3 (3 3) 2 (3 2) 120 mg/kg 10 3 (3 3) 2.5 (3 2) MC-1 (40 mg/kg) 10 3 (3 3) 2 (3 2) given 3 hrs postinjury MC-1 (40 mg/kg) 10 3 (3 3) 2 (3 2) given 6 hrs postinjury tpa (10 mg/kg), 8 3 (3 3) 2.5 (3 2) treatment beginning 1 hr postinjury MC-1 (40 mg/kg) 8 3 (3 3) 2.5 (3 2) & tpa (10 mg/kg) * Each neurological deficit score (NDS) is expressed as the median and interquartile range (IQR), the 25 to 75 th percentile is shown in parenthesis. Significantly different from the value 2 hours postinjury in the same group. instances of both global and focal ischemia. For example, a very large increase in cytosolic calcium occurs in CA1 and CA3 hippocampal neurons when ischemia is induced by four-vessel occlusion. 17 An increase in cytosolic calcium has also been seen in the injured brain in the focal ischemic model, and this increase occurs more rapidly in the ischemic core than in the penumbra. 5 Cytosolic calcium levels can rise as a result of a net entry of calcium across the plasmalemma or due to liberation of calcium from intracellular stores. Several mechanisms may be involved in cytosolic changes in calcium levels such as activation of N-methyl- D-aspartate receptors, 3 L-channels, and the Na Ca exchanger. 11 Recently. it has been reported that ATP-gated ion channels are present on many visceral and vascular smooth muscle cells, as well as on neuronal and glial cells. 1 These channels are believed to play an important role in the mobilization of calcium in endothelial cells in the MCA. 27 Pyridoxal-5-phosphate is a specific antagonist of ATP-gated ion channels and is able to inhibit ATP-induced calcium influx in cells via these channels. 21 In the present study, the infusion of MC-1 that was started 6 hours after MCA occlusion was still effective in protecting the injured brain, which is in agreement with the finding that there is a persistent elevation of cystolic calcium in the ischemic core throughout a 24-hour reperfusion period. 5 Second, MC-1 may protect the brain injured by ischemia by balancing the activity of the glutamate GABA system. In previous studies, we and other groups have demonstrated that excessive excitatory stimulation during ischemia plays an important role in neuronal cell death. Antagonizing this excessive stimulation by administering glutamate antagonists or GABA agonists has been shown to reduce ischemic brain injury significantly. 4,13,26 In vitro studies have shown that PLP can promote cell survival in cultured neurons by activation of the GABAergic system. 6 PLP can also reduce cell death in cultured neurons by inhibition of the excitatory activity of glutamate. 7 Third, MC-1 may protect brain injured by ischemia by inhibiting the decline of energy-yielding products. 168

5 Ischemic brain injury and MC-1 In cultured neurons, PLP treatment significantly suppressed the glucose deprivation induced reduction of pyruvate and ATP. 7 Tissue plasminogen activator was used to treat focal ischemic injury in the present model with satisfactory results. In combination with a glycoprotein IIb/IIIa antagonist, tpa has demonstrated better neuroprotective effects than the glycoprotein IIb/IIIa antagonist alone. 15 Because MC-1 can protect a brain that has suffered an ischemic injury through different mechanisms from those associated with tpa, we examined whether the combination of these two agents would provide synergistic or additive actions for neuroprotection. Although the combined MC-1 tpa treatment significantly reduced the volume of the infarct, it did not produce a greater protective effect than either MC-1 or tpa when used alone, nor did the combination produce additional side effects. The reason why the use of MC-1 and tpa together produced no additive or synergistic neuroprotective effects in the present model is unclear. Because between 80 and 90% of strokes are caused by thromboembolism, we used an embolic model of t stroke in the present study 12,16. The success rate in stroke induction is very high in our laboratory, and MCA occlusion occurred in all the rats immediately after the blood clot injection. The clots spontaneously dissolve between 1 and 24 hours after MCA occlusion, 22,23 and the volume of the infarct in this embolic stroke model is relatively consistent. This consistency may be attributable to several factors. Although the spontaneous lysis of the clots is not homogeneous, cell death already may have occurred before the clot was dissolved. Ischemic injury is not only caused by the clot injected into the larger arteries but also by fragments formed from the injected clots, which can move to downstream arteries and cause brain injury. 22 Conclusions In this study we showed that MC-1 can protect the brain during ischemia by reducing the volume of the infarction and improving functional recovery. Although the combined treatment of MC-1 and tpa did not produce results that were superior to either treatment alone, the combined treatment also reduced the extent of infarction in the present model. Given the fact that this compound was still effective when administered as late as 6 hours after ischemia, MC-1 may hold great promise as a treatment in ischemic brain injury in patients. References 1. Abbracchio MP, Burnstock G: Purinoceptors: are there families of P 2X and P 2Y purinoceptors? Pharmacol Ther 64: , Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H: Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke 17: , Benveniste H, Jorgensen MB, Diemer NH, Hansen AJ: Calcium accumulation by glutamate receptor activation is involved in hippocampal cell damage after ischemia. Acta Neurol Scand 78: , Chen Xu W, Yi Y, Qiu L, Shuaib A: Neuroprotective activity of tiagabine in a focal embolic model of cerebral ischemia. Brain Res 874:75 77, DeGraba TJ, Ostrow PT, Grotta JC: Threshold of calcium disturbances after focal cerebral ischemia in rats. Implications of the window of therapeutic opportunity. Stroke 24: , Geng MY, Satio H, Katsuki H: Effects of vitamin B6 and its related compounds on survival of cultured brain neurons. Neurosci Res 24:61 65, Geng MY, Satio H, Nishiyama N: Protective effects of pyridoxal- 5-phosphate against glucose deprivation-induced damage in cultured hippocampal neurons. J Neurochem 68: , Kandzari DE, Labinaz M, Cantor WJ, Madan M, Gallup DS, Hasselblad V, et al: Reduction of myocardial ischemic injury following coronary intervention (the MC-1 to Eliminate Necrosis and Damage trial). Am J Cardiol 92: , Kaneda K, Kikuchi M, Kashii S, Honda Y, Takehiko M, Kaneko S, et al: Effects of B vitamins on glutamate-induced neurotoxicity in retinal cultures. Eur J Pharmacol 322: , Kirksey A, Morre DM, Wasynczuk AZ: Neuronal development in vitamin B6 deficiency. Ann NY Acad Sci 585: , Lobner D, Lipton P: Intracellular calcium levels and calcium fluxes in the CA1 region of the rat hippocampal slice during in vitro ischemia: relationship to electrophysiological cell damage. J Neurosci 13: , Mohr JP, Caplan LR, Melski JW, Goldstein RJ, Duncan GW, Kistler JP, et al: The Harvard Cooperative Stroke Registry: a prospective registry. Neurology 28: , Palmer GC, Miller JA, Cregen EF Gendron P, Peeling J: Low-affinity NMDA receptor antagonists. The neuroprotective potential of ARL 15896AR. Ann N Y Acad Sci 825: , Pham V, Zhang W, Chen V, Whitney T, Yao J, Froese D, et al: Design and synthesis of novel pyridoxine 5 -phosphonates as potential antiischemic agents. J Med Chem 46: , Shuaib A, Yang Y, Nakada MT, Li Q, Yang T: Glycoprotein IIb/ IIIa antagonist, murine 7E3 F(ab ) 2, and tissue plasminogen activator in focal ischemia: evaluation of efficacy and risk of hemorrhage with combination therapy. J Cereb Blood Flow Metab 22: , Sloan MA: Thrombolysis and stroke. Past and future. Arch Neurol 44: , Silver IA, Erecinska M: Intracellular and extracellular changes of (Ca 2 ) in hypoxia and ischemia in rat brain in vivo. J Gen Physiol 95: , Stryer L: Biochemistry, ed 4. New York: WH Freeman, 1995, pp Trezise DJ, Bell NJ, Khakh BS, Michel AD, Humphrey PA: P 2 putinoceptor antagonist properties of pyridoxal-5-phosphate. Eur J Pharmacol 259: , Wang CX, Erecius LF, Beverly JL III, Gietzen DW: Essential amino acids affect interstitial dopamine metabolites in the anterior piriform cortex of rats. J Nutr 129: , Wang XD, Kashii S, Zhao L, Tonchev AB, Katsuki H, Akaike A, et al: Vitamin B6 protects primate retinal neurons from ischemic injury. Brain Res 940:36 43, Wang CX, Todd KG, Yang Y, Gordon T, Shuaib A: Patency of cerebral microvessels after focal embolic stroke in the rat. J Cereb Blood Flow Metab 21: , Wang CX, Yang T, Shuaib A: An improved version of embolic model of brain ischemic injury in the rat. J Neurosci Meth 109: , Wang CX, Yang Y, Yang T, Shuaib A: A focal embolic model of cerebral ischemia in rats. Brain Res Protoc 7: , Yamashima T, Zhao L, Wang XD, Tsukada T, Tonchev AB: Neuroprotective effects of pyridoxal-5-phosphate and pyridoxal against ischemia in monkeys. Nutr Neurosci 4: , Zhang L, Mitani A, Yanase H, Kataoka K: Continuous monitoring and regulating of brain temperature in the conscious and freely moving ischemic gerbil: effect of MK-801 on delayed neuronal death in hippocampal CA1. J Neurosci Res 47: , Zhang H, Weir BKA, Marton LS, Lee KS, Macdonald RL: P2 purinoceptors in cultured bovine middle cerebral artery endothelial cells. J Cardiovasc Pharmacol 30: , 1997 Manuscript received April 11, Accepted in final form March 15, Address reprint requests to: Ashfaq Shuaib, M.D., Stroke Research Laboratory, 533 HMRC, University of Alberta, Edmonton, Alberta T6G 2S2, Canada. ashfaq.shuaib@ualberta.ca. 169