Role of superoxide dismutase in ischemic brain injury: a study using SOD-1 transgenic mice

1. Nitric oxide radicals (NO) play an important role in the pathophysiology of focal cerebral ischemia. 2. Vascular NO can reduce ischemic brain injury by increasing CBF, whereas neuronal NO may mediate neurotoxicity following brain ischemia, mainly by its reaction with superoxide to generate peroxynitrite. 3. These findings could contribute to a strategy for the treatment of cerebral ischemia.

Dural arteriovenous shunts at the craniocervical junction

OBJECT

A retrospective analysis was conducted of 10 patients (three women and seven men) who were treated for spinal dural arteriovenous shunts (AVSs) located at the craniocervical junction. This analysis was performed to evaluate the characteristics of this unusual location in contrast with those of the more common thoracic and lumbar AVSs.

METHODS

Seven patients presented with subarachnoid hemorrhage (SAH) and one with slowly progressive quadriparesis and dyspnea due to myelopathy. The other two cases were detected incidentally and included a transverse-sigmoid dural AVS and a cerebellar arteriovenous malformation. Angiographic studies revealed that the spinal dural AVSs at the C-1 and/or C-2 levels were fed by the dural branches of the radicular arteries that coursed from the vertebral artery and drained into the medullary veins. Venous drainage was caudally directed in the patient with myelopathy. In contrast, the shunt flow drained mainly into the intracranial venous system in patients with SAH. Furthermore, in four of these patients a varix was found on the draining vein. In all patients, the draining vein was interrupted surgically at the point at which this vessel entered the intradural space, using intraoperative digital subtraction angiography to monitor flow. The postoperative course was uneventful in all patients and no recurrence was confirmed on follow-up angiographic studies obtained in seven patients at 6 months after discharge.

CONCLUSIONS

If computerized tomography scanning shows SAH predominantly in the posterior fossa and no abnormalities are found on intracranial four-vessel angiographic study, proximal vertebral angiography should be performed to detect dural AVS at the craniocervical junction. The results of surgical intervention for this disease are quite satisfactory.

[Differential diagnosis of hyponatremia following subarachnoid hemorrhage]

Hyponatremia is a common complication after subarachnoid hemorrhage (SAH). Although the mechanism of hyponatremia is still controversial, cerebral salt-wasting syndrome (CSNS) is currently regarded as being more responsible than the syndrome of inappropriate secretion of antidiuretic hormone (SIADH). The aim of our study was to assess the plasma volume status of a patient with hyponatremia following subarachnoid hemorrhage. In doing this it may be possible to indirectly differentiate its pathogenesis. Fifty patients with SAH were studied. Twenty patients demonstrated hyponatremia (serum sodium < 135 mEq/L) during day 7 to 13 after subarachnoid hemorrhage. Patients with hyponatremia were categorized on the basis of their daily body weight, and central venous pressure. Group A consisted of patients with hypovolemia (16 patients), with the onset time of hyponatremia being day 7 to 9. Group B included those with hypervolemia (4 patients); hyponatremia was observed during day 10 to 11 and was corrected in all patients within 72 hours after induction of fluid restriction. Our findings suggest that hyponatremia following subarachnoid hemorrhage usually occurs due to CSWS, although SIADH remains as a minor pathogenesis. We conclude that the combination of daily body weight and CVP measurements is a simple and practical method to distinguish promptly SIADH from CSWS.

[Anesthesia in two patients with essential thrombocythemia]

Case 1. The patient was a 69-year-old man with essential thrombocythemia (ET), who underwent urgent laparotomy. On admission he was dehydrated and the platelet count was more than 160 x 10(4).microliter-1, with hematocrit of 50%. Anesthesia was induced with ketamine i.v. and maintained with nitrous oxide and sevoflurane in oxygen. Postoperative care included the administration of gabexate mesilate (GM) which is an antiplatelet agent. Case 2. An 84-year-old woman with ET was diagnosed as gastric cancer and elective gastrectomy was scheduled. The platelet count was more than 100 x 10(4).microliter-1. The patient was anesthetized with nitrous oxide and oxygen supplemented with fentanyl and mepivacaine via epidural catheter. Intravenous infusion of GM was performed at a rate of 1 mg.kg-1.hr-1 during surgery. PF-4 and beta-TG were measured. These are platelet releasing factors. The level of PF-4 decreased to normal level during this procedure. In conclusions, it will be important to use GM during anesthesia in order to avoid the complications such as myocardial or pulmonary infarction caused by thrombocythemia.

Probable brain abscess presenting as a high uptake lesion on thallium-201 single photon emission computed tomography--case report

A 61-year-old male presented with a brain abscess manifesting as high fever and generalized convulsion attacks. Magnetic resonance (MR) imaging disclosed a ring-like enhanced lesion in the parietal lobe. Thallium-201 single photon emission computed tomography (201Tl SPECT) images demonstrated a high uptake lesion with a 201Tl uptake index on the early image of 2.21, which suggested malignant disease. The washout ratio was 0.73. His symptoms and the ring-like enhanced lesion on MR images disappeared after 2 months of antibiotic treatment. The final diagnosis was brain abscess, despite the 201Tl SPECT findings. 201Tl SPECT washout ratio may be a better indicator of brain abscess than uptake index.

[The use of low dose midazolam for the management of spinal anesthesia]

The purpose of this study was to evaluate the effects of low dose midazolam (MZ) on memories and ease of management of spinal anesthesia. The low doses of MZ were administered to 70 patients (ASA 1-2), of whom 37 patients were premedicated with atropine sulfate 0.5 mg and pethidine hydrochloride (group P), with 33 patients receiving no premedication (group N). Double blind randomized trials were conducted with the doses of MZ (0, 0.03, 0.06 mg.kg-1), and MZ was administered i.v. to patients just prior to spinal puncture. The short-term and long-term memories were impaired after administration of MZ in both groups. We conclude that MZ appears to be a suitable replacement for other benzodiazepines for relieving anxiety of patients during surgery.

Inducible nitric oxide synthase following hypoxia in rat cultured glial cells

Nitric oxide (NO) produced by inducible nitric oxide (iNOS) exerts inhibitory and cytotoxic effects on various cells including neuronal cells. In the present study, we examined the ability of rat glial cells to produce NO following hypoxia/reoxygenation in vitro by measuring nitrite. The levels of nitrite produced in the cultured media of glial cells significantly increased after 12-h hypoxia but not after 0- and 6-h hypoxia. The NOS inhibitor, NG-monomethyl-L-arginine, decreased hypoxia-induced nitrite formation. In glial cells after hypoxia/reoxygenation, the iNOS and mRNA and protein expressions were detected by reverse-transcription polymerase chain reaction and by immunocytochemical analysis, respectively. The present study provides the first evidence that hypoxia induces NO production from glial cells. This hypoxia-induced, glial cell-derived NO may play a critical role in the pathogenesis of cerebral ischemia in vivo.

Attenuation of acute and chronic damage following traumatic brain injury in copper, zinc-superoxide dismutase transgenic mice

To elucidate the role of oxygen-derived free radicals and superoxide dismutase in traumatic brain injury (TBI), blood-brain barrier (BBB) permeability, brain edema, behavioral function, and necrotic cavity volume (CV) were evaluated after TBI using nontransgenic (nTg) mice and heterozygous and homozygous transgenic (Tg) mice with a 1.5- (Tg 1.5x), 3.1-(Tg3.1x) and five- (Tg5x) fold increase in human copper, zinc-superoxide dismutase (CuZn-SOD) activity. Traumatic brain injury was produced by the weight-drop method. Evans blue dye leakage 4 hours after injury was attenuated in a CuZn-SOD dose-dependent manner with decreases of 18.6%, 40.9%, and 48.8%, in the Tg1.5x, Tg3.1x, and Tg5x groups, respectively. The water content 6 hours after injury in the Tg3.1x (79.64%) and Tg5x (79.45%) groups was significantly lower than in nTg mice (81.37%). There was an initial decrease in body weight and in motor performance, as measured by beam walk and beam balance tasks undertaken 1 day after TBI. However, the average reduction in beam balance and beam walk performance deficits and changes in body weight postinjury were significantly ameliorated in Tg mice. The CV was significantly smaller in Tg mice than in nTg mice (p < 0.01). These results indicate that superoxide radicals play a deleterious role following TBI. Furthermore, Tg mice provide a useful model for demonstrating the beneficial role of an antioxidant enzyme in TBI without the confounding effect of pharmacokinetics, toxicity, and BBB permeability associated with exogenous agents.

Effects of nitric oxide synthase inhibition on brain infarction in SOD-1-transgenic mice following transient focal cerebral ischemia

To investigate the role of superoxide in the toxicity of nitric oxide (NO), we examined the effect of nitric oxide synthase (NOS) inhibition on brain infarction in transgenic mice overexpressing CuZn-superoxide dismutase (SOD-1). Male SOD-transgenic mice and non-transgenic littermates (30-35 g) were subjected to 60 min of middle cerebral artery occlusion followed by 24 h of reperfusion. Either NG-nitro-L-arginine methyl ester (L-NAME; 3 mg/kg), a mixed neuronal and endothelial NOS inhibitor, or 7-nitroindazole (7-NI; 25 mg/kg), a selective neuronal NOS inhibitor, was administered intraperitoneally 5 min after the onset of ischemia. At 24 h of reperfusion, the mice were decapitated and the infarct volume was evaluated in each group. In the nontransgenic mice, L-NAME significantly increased the infarct volume as compared with the vehicle, while 7-NI significantly decreased it. In the SOD-transgenic mice, L-NAME-treated animals showed a significantly larger infarct volume than vehicle-treated ones, whereas there were no significant differences between 7-NI- and vehicle-treated mice. Our findings suggest that selective inhibition of neuronal NOS ameliorates ischemic brain injury and that both neuronal and endothelial NOS inhibition may result in the deterioration of ischemic injury due to vasoconstriction of the brain. Since L-NAME increased infarct volume even in SOD-transgenic mice, the protective effect of SOD could result from the vasodilation by increased endothelial NO as well as the reduction of neuronal injury due to less production of peroxynitrite compared to wild-type mice. Moreover, the neurotoxic role of NO might not be dependent on NO itself, but the reaction with superoxide to form peroxynitrite, because of no additive effects of SOD and a neuronal NOS inhibitor.

Differential response in the release of hydrogen peroxide between astroglial cells and endothelial cells following hypoxia/reoxygenation

Hydrogen peroxide (H2O2) and lactate dehydrogenase (LDH) released from cultured astroglial cells (AGs) and brain endothelial cells (BECs) were measured following hypoxia/reoxygenation. AGs released H2O2 to the extracellular space from 8 h of hypoxia with the cellular injury indicated by LDH increase. In contrast, BECs released H2O2 at 4 h of hypoxia, prior to the LDH increase. These results suggest that AGs and BECs play different roles in production and metabolism of oxygen free radicals in the blood-brain barrier (BBB), and AGs may protect the BBB disruption from oxidative stress after hypoxia/reoxygenation.

Induction of brain-derived neurotrophic factor (BDNF) and the receptor trk B mRNA following middle cerebral artery occlusion in rat

Middle cerebral artery (MCA) occlusion in halothane-anesthetized rats induced brain-derived neurotrophic factor (BDNF) and the receptor, trk B mRNA, in brain. In situ hybridization studies showed that BDNF and trk B mRNAs were induced in a widespread region of the ipsilateral cortex outside the infarct at 4 h following MCA occlusion. They were also induced in the bilateral hippocampi which are remote from the ischemic MCA region. These data show that changes in neurotrophic factor and receptor gene expressions can occur in the areas outside the infarct which could survive.

Astroglial cells inhibit the increasing permeability of brain endothelial cell monolayer following hypoxia/reoxygenation

Blood-brain barrier (BBB) is known to be structured with astroglial cells (AGs) and brain endothelial cells (BECs), and it has been proposed that these cells play different roles in the BBB. We cultivated AGs and BECs from infant rats (2-week-old), and these cells were cultured on the opposing side of collagen membrane to produce a co-culture model of the BBB in vitro. Permeability of the cell layer was evaluated by the electrical resistance through the membrane. To clarify the role of AGs in the BBB disruption following ischemia/reperfusion, electrical resistance of the co-culture model was compared to that of BEC monolayer following hypoxia/reoxygenation. The electrical resistance through BEC monolayer showed 55.5 +/- 15.1 percent reduction at 4 h of hypoxia, and 93.3 +/- 5.4 percent reduction at 8 h of hypoxia (n = 8). However, the co-culture model showed attenuation of the reduction (24.8 +/- 14.2 percent) at 4 h of hypoxia (n = 8, P < 0.01), but not at 8 h of hypoxia (95.3 +/- 5.0 percent). These results indicate that AGs reduce the increasing permeability of the BEC monolayer following short duration of hypoxia/reoxygenation. It is suggested that AGs may have a protective effect to the BBB disruption following ischemia/reperfusion.

Expression of c-fos and hsp70 mRNA after traumatic brain injury in transgenic mice overexpressing CuZn-superoxide dismutase

The aim of this study was to determine the role of oxidative stress on c-fos and hsp70 gene expression in transgenic (Tg) mice overexpressing CuZn-superoxide dismutase (SOD-1) following traumatic brain injury (TBI). hsp70 mRNA, as investigated using in situ hybridization, was induced around the lesion at 4 and 24 h, but not at 1 and 48 h, in both Tg and non-transgenic (nTg) mice littermates. The degree of hsp70 induction was somewhat greater in nTg than Tg mice at 4 and 24 h after TBI. c-fos mRNA was induced throughout cortex, hippocampus, caudate putamen and the ventricular wall in Tg and nTg mice. TBI induced c-fos bilaterally in the cortex in both animals. There was a time-dependent difference in cortical c-fos expression between nTg and Tg mice. The induction of c-fos mRNA in the striatum was greater in nTg at 24 h and decreased in both animals by 48 h. Edema of the injured cortex was significantly attenuated in Tg mice at all time points (1-48 h). These data show that the degree of hsp70 induction and the degree, extent, and duration of c-fos induction produced by TBI are affected by levels of superoxide dismutase activity. It is proposed that superoxide radicals affect spreading depression and brain edema produced by TBI and that this effect may either directly or indirectly modulate the expression of the c-fos and hsp70 genes after TBI.

Transgenic mice and knockout mutants in the study of oxidative stress in brain injury

A rapid increase in the need to explore the molecular basis of cellular function and injury in the central nervous system has led neuroscientists to employ transgenic mouse technology. The successful making of transgenic mice (Tg) overexpressing human CuZn-superoxide dismutase (SOD-1) activity has made it possible to investigate the role of oxygen free radicals in ischemic and traumatic brain injury in a molecular fashion. It has been demonstrated that the 3-fold increase in SOD-1 transgene activity in SOD-1 Tg mice offers protection against cerebral ischemia and reperfusion in two different models of focal cerebral ischemia, as compared to nontransgenic wild-type littermates. Studies involving traumatic brain injury have also demonstrated that acute injuries, including brain edema and blood-brain barrier permeability, are significantly reduced in SOD-1 Tg mice. Furthermore, chronic neurological deficits, such as beam walking, beam balance, and body weight, are significantly improved in these transgenic animals following traumatic brain injury. In addition to the SOD-1 Tg mice being a useful tool for the study of CNS injury, targeted disruption of the mouse gene for mitochondrial manganese SOD (SOD-2) has been successful. These SOD-2 knockout mutant mice, in addition to the recently developed knockout mutants of neuronal nitric oxide synthase (NOS), are believed to offer a unique opportunity to elucidate the oxidative mechanisms in brain injury following stroke and trauma.

Effect of phenytoin on cortical Na(+)-K(+)-ATPase activity in global ischemic rat brain

Na(+)K(+)-ATPase activity, water content, and Na+/K+ concentrations in the parietal cortex were measured in untreated and phenytoin-treated rats following global cerebral ischemia. Inhibitory effects of phenytoin treatment on brain edema and changes in Na(+)-K+ concentration with ischemia or ischemia followed by recirculation of varying intervals were assessed. The cortical Na(+)-K(+)-ATPase activity increased in the phenytoin-treated group during and after ischemia. Based on these results, we conclude that phenytoin provides ischemic brain protection by activating cortical Na(+)-K(+)-ATPase activity and by reducing intracellular Na+ and water content.

Expression of c-fos mRNA after a mild focal cerebral ischemia in SOD-1 transgenic mice

To clarify the role of oxygen free radicals in expression of the c-fos protooncogene, the distribution of c-fos mRNA was investigated in CuZn-superoxide dismutase (SOD-1) transgenic (Tg) mice compared to control nontransgenic (nTg) littermates after a mild (i.e. 10 min) focal cerebral ischemia. c-fos mRNA expression occurred at 1 to 6 h after reperfusion in the ipsilateral hippocampus and thalamus in Tg mice, whereas it did only at 1 h in the same regions in nTg mice. In the ipsilateral cortex, there were no significant differences in the pattern of the expression between nTg and Tg mice. These results suggest that oxygen radicals may suppress the expression of c-fos in the hippocampus and thalamus, the areas known to be without blood supply from the middle cerebral artery, following a mild focal cerebral ischemia and reperfusion.

MK-801 inhibits the induction of immediate early genes in cerebral cortex, thalamus, and hippocampus, but not in substantia nigra following middle cerebral artery occlusion

Middle cerebral artery (MCA) occlusion in rats induced c-fos and junB mRNA 4h later in all ipsilateral cortex outside the MCA distribution and in many subcortical structures: medial striatum; most of thalamus including medial and lateral geniculate nuclei: substantia nigra; and hippocampus. The N-methyl-D-aspartate (NMDA) antagonist, MK-801 (4 mg/kg, i.p.) inhibited c-fos and junB mRNA induction in the cortex, striatum, thalamus, and hippocampus but not in the substantia nigra. These data show that c-fos and junB mRNA induction in cortex, striatum, thalamus, hippocampus involves the activation of NMDA receptors whereas different receptors must be implicated in the induction in substantia nigra.

Induction of c-fos, junB, c-jun, and hsp70 mRNA in cortex, thalamus, basal ganglia, and hippocampus following middle cerebral artery occlusion

Middle cerebral artery (MCA) occlusion in halothane-anesthetized rats induced c-fos, junB, and c-jun immediate early gene mRNAs and hsp70 heat shock gene mRNA in brain. In situ hybridization studies showed that c-fos and junB were induced throughout all of the cortex at 1 and 4 h following MCA occlusion. hsp70 was induced in the core and margins of the MCA ischemia. By 24 h, there was little expression of c-fos, junB, c-jun, and hsp70 in the core of the MCA infarct; there was modest induction of hsp70 at the margins of the infarct; and there was diffuse induction of c-fos, junB, and c-jun in all of the cortex outside the infarct. MCA occlusion also induced these genes in subcortical structures. c-fos, junB, and hsp70 were induced in ipsilateral medial striatum, most of thalamus including medial and lateral geniculate nuclei, substantia nigra, and hippocampus. Most of these structures, except for the striatum, are not supplied by the MCA. These data show that changes in gene expression can occur in regions remote from an infarction.

Prolonged expression of hsp70 mRNA following transient focal cerebral ischemia in transgenic mice overexpressing CuZn-superoxide dismutase

The distribution of heat shock protein hsp70 mRNA after 10 min of middle cerebral artery (MCA) occlusion was investigated through in situ hybridization in transgenic (Tg) mice overexpressing CuZn-superoxide dismutase (CuZn-SOD) and in control nontransgenic (nTg) littermates. In the ischemic cortex of nTg mice, hsp70 mRNA was detected 1 h after reperfusion and was observed for up to 6 h. In Tg mice, however, it was still detectable within the cortex even at 24 h. In the caudate putamen, hsp70 mRNA appeared at 1 h and was present for up to 6 h in both nTg and Tg mice. Although hsp70 mRNA was detected in the thalamus only at 1 h in nTg mice, it was observed for up to 6 h in Tg mice. Similarly, hsp70 mRNA was detected in the hippocampus of nTg mice only at 1 h, whereas it was detected in Tg mice at 1 h and continued up to 24 h, with high intensity in the CA1 subfield. Despite the significant amounts of hsp70 mRNA in both Tg and nTg mice following ischemia, there was no observable neuronal necrosis (as assessed using hematoxylin and eosin staining) for up to 7 days. Cortical cerebral blood flow (CBF), measured by laser-Doppler flowmetry, did not differ between nTg and Tg mice during ischemia and reperfusion, despite exhibiting hyperemia following hypoperfusion. These results suggest that oxidative stress affects the expression of hsp70 following temporary focal ischemia. An alteration in oxidation stress, which resulted from reduced levels of superoxide radicals in the presence of the CuZn-SOD transgenes, may permit the prolonged expression of hsp70.(ABSTRACT TRUNCATED AT 250 WORDS)

Induction of NGFI-A mRNA following middle cerebral artery occlusion in rats: in situ hybridization study

Middle cerebral artery (MCA) occlusion in halothane-anesthetized rats induced the zinc finger gene, NGFI-A, in brain. In situ hybridization studies showed that NGFI-A was induced throughout all of the cortex following MCA occlusion. By 24 h after MCA occlusion there was little expression of NGFI-A mRNA in the core of the MCA infarct, but the mRNA was still induced in all of cortex outside the infarct. MCA occlusion also induced this gene in subcortical structures: ipsilateral medial striatum; most of thalamus including medial and lateral geniculate nuclei; substantia nigra; and hippocampus at 4 h of MCA occlusion which generally disappeared by 24 h of MCA occlusion. Most of these structures, except for the striatum, are not supplied by the MCA. These data show that changes in brain gene expression can occur in many regions remote from an infarction.

HSP70 heat shock gene regulation during ischemia

The hsp70 gene is induced by denatured protein in injured cells and is an extremely sensitive and reliable marker of cells injured by ischemia, seizures, and toxins. Normal brains have little detectable hsp70 mRNA or HSP70 protein. After status epilepticus produced by systemic injections of kainic acid, however, HSP70 protein is induced in neurons but not glia in brain regions known to be injured by kainic acid. Global and focal ischemia also induce the hsp70 gene in brain. The induction of HSP70 protein in hippocampus following increasing durations of global ischemia correlates with the regional and cellular vulnerability to ischemia: CA1 neurons express HSP70 after the briefest periods of ischemia followed by CA4, CA3, dentate granule neurons, glia, and lastly, endothelial cells. Moreover, as the severity of ischemia worsens, a transcriptional and/or translational blockade of the hsp70 gene occurs in the same order so that moderate degrees of ischemia induce HSP70 in CA3 neurons and dentate granule neurons but not necrotic CA1 neurons, and severe ischemia induces HSP70 in capillary endothelial cells of hippocampus but not in any infarcted neurons or glia throughout the hippocampus. Brief periods of focal ischemia induce HSP70 primarily in neurons, suggesting that even focal ischemia can produce selective neuronal injury without infarction. In some instances, HSP70 immunoreactive astrocytes surround the HSP70 immunostained neurons. Focal ischemia that produces infarction induces HSP70 primarily in endothelial cells of cerebral blood vessels in the regions of infarction and in neurons and astrocytes on the perimeter or the penumbral area of infarction.(ABSTRACT TRUNCATED AT 250 WORDS)

Induction of heat shock hsp70 mRNA and HSP70 kDa protein in neurons in the 'penumbra' following focal cerebral ischemia in the rat

Induction of hsp70 heat shock protein (HSP70) and hsp70 mRNA was examined using adjacent sections in the same rat brain following permanent middle cerebral artery (MCA) occlusions, hsp70 mRNA was induced within 4 h of MCA occlusion and persisted for at least 24 h. Cellular resolution autoradiographs suggested that hsp70 mRNA was induced primarily in neurons in the periphery of ischemia both outside and inside of the infarction, with small amounts of hsp70 mRNA being induced in the core of the infarction. HSP70 protein was localized in neurons outside the infarction and in endothelial cells within the infarction at 24 h but not at 4 h following permanent MCA occlusions. It is proposed that the penumbra, one of the areas that can be rescued by pharmacological agents, can be defined anatomically as the volume of tissue outside the area of infarction in which HSP70 protein is expressed primarily in neurons.

Induction of 70-kDa heat shock protein and hsp70 mRNA following transient focal cerebral ischemia in the rat

Induction of the 70-kDa heat shock protein (HSP70) was demonstrated immunocytochemically in adult rats 4 h to 7 days following temporary middle cerebral artery (MCA) occlusions lasting 30, 60, or 90 min. Maximal HSP70 induction occurred approximately 24 h following ischemia. Thirty minutes of ischemia induced HSP70 in neurons throughout the cortex in the MCA distribution, whereas 90 min of ischemia induced HSP70 in neurons in the penumbra. HSP70 protein was induced in endothelial cells in infarcted neocortex following 60-90 min of MCA occlusion, and HSP70 was induced in endothelial cells in infarcted regions of lateral striatum following 30-90 min of MCA occlusion. hsp70 mRNA was induced in the MCA distribution in cortex and to a lesser extent in striatum at 2 h to 3 days following 60 min of ischemia. It is proposed that brief ischemia induces hsp70 mRNA and HSP70 protein in the cells most vulnerable to ischemia--the neurons. HSP70 protein is not induced in most neurons and glia following 60-90 min of ischemia in areas destined to infarct, whereas it is induced in vascular endothelial cells.