Calcium influx through the TRPV1 channel of endothelial cells (ECs) correlates with a stronger adhesion between monocytes and ECs.

PURPOSE: Atherosclerosis is thought to be initiated by the transendothelial migration of monocytes. In the early stage of this process, the adhesion of monocytes to endothelial cells is supported by an increase in the intracellular concentration of calcium ion ([Ca(2+)]i) in endothelial cells. However, the main source of Ca(2+) has been unclear. In this study, the changes in ionic transmittance and [Ca(2+)]i due to the adhesion of monocytes were continuously measured by an electrophysiological technique and fluorescent imaging. Especially, we focused on transient receptor potential vanilloid channel 1 (TRPV1) as a Ca(2+) channel that could influence the adhesion of monocytes. MATERIAL AND METHODS: Whole-cell current was continuously recorded in human umbilical vein endothelial cells (HUVECs) by a patch electrode. RESULTS: The adhesion of monocytes (THP-1) induced a transient inward current in HUVECs, as well as an elevation of [Ca(2+)]i. This inward element was abolished by the application of 100 nM SB366,791, a selective antagonist of TRPV1 channel. Furthermore, SB366,791 significantly decreased the number of THP-1 cells that adhered to HUVECs (control: 231 ± 38, SB366,791: 96 ± 16 cells/mm2). CONCLUSION: These results suggest that an inward calcium current via the TRPV1 channels of endothelial cells correlates with a stronger adhesion between monocytes and endothelial cells.

Delayed precursor cell marker response in hippocampus following cold injury-induced brain edema.

The purpose of this study was to examine the possibility of neuronal remodeling and repair after cold injury-induced brain edema using immunoassays of nestin, 3CB2, and TUC-4. Male ddN strain mice were subjected to cold-induced cortical injury. Animals were divided into the following 6 groups: 1) 1-day after injury, 2) 1-week after injury, 3) 2-weeks after injury, 4) 1-month after injury, 5) sham, and 6) normal controls. Brain water content measurement, Western blot analysis, histological examination, and neurobehavioral examination were performed. Brain water content was significantly increased in the ipsilateral cortex at 1-day after injury. At 1-day and 1-week after injury, immunoreactivity of nestin, 3CB2, and TUC-4 were absent. Nestin was expressed in 3CB2-positive astrocytes at 1-month after injury, and nestin expression with TUC-4 was present in the hippocampal cell layer. Neurobehavioral function of the 1-month after injury group was significantly improved compared with function 1-day after injury. These results suggest that delayed precursor cell marker expression in glia and neuron-like cells might be part of adaptation to the injury. Although brain injury causes brain edema and neuronal death, there is the possibility of remodeling.

Post-traumatic moderate systemic hypothermia reduces TUNEL positive cells following spinal cord injury in rat.

STUDY DESIGN: A standardized animal model of contusive spinal cord injury (SCI) with incomplete paraplegia was used to test the hypothesis that moderate systemic hypothermia reduces neural cell death. Terminal deoxynucleotidyl transferase [TdT]-mediated deoxyuridine triphosphate [dUTP] nick-end labeling (TUNEL) staining was used as a marker of apoptosis or cell damage. OBJECTIVE: To determine whether or not moderate hypothermia could have a neuroprotective effect in neural cell death following spinal cord injury in rats. SETTING: Kagawa Medical University, Japan. METHODS: Male Sprague-Dawley (SD) rats (n=39) weighing on average 300 g (280-320 g) were used to prepare SCI models. After receiving contusive injury at T11/12, rats were killed at 24 h, 72 h, or 7 days after injury. The spinal cord was removed en bloc and of examined at five segments: 5 and 10 mm rostral to the center of injury, center of injury, and 5 and 10 mm caudal to the center of injury. Rats that received hypothermia (32 degrees C/4 h) were killed at the same time points as those that received normothermia (37 degrees C/3 h). The specimens were stained with hematoxylin and eosin, and subjected to in situ nick-end labeling (TUNEL), a specific method for visualizing cell death in the spinal cord. RESULTS: At 24 h postinjury, TUNEL positive cells (TPC) decreased significantly 10 mm rostral to center of injury in hypothermic animals compared to the normothermia group. At 72 h post-SCI, TPC also decreased significantly at 5 mm rostral, and 5 and 10 mm caudal to the lesion center compared to normothermic animals. At 7 days postinjury, a significant decrease of TPC was observed at the 5 mm rostral and 5 mm caudal sites compared to normothermic animals. CONCLUSION: These results indicate that systemic hypothermia has a neuroprotective effect following SCI by attenuating post-traumatic TPC.

Do rapid systemic changes of brain temperature have an influence on the brain?

BACKGROUND: The purpose of the present study was to examine the influence of cooling and rewarming conditions using an accurate brain temperature control system. METHOD: The brain temperature of animals was measured with a thermometer while feedback regulation was achieved with a cold (4( degrees )C) and hot (50( degrees )C) water on-off flow system. Brain temperature was well controlled throughout the experiment by using both cold water and hot water simultaneously. Three groups were studied, as follows: 1) the standard group (cooled to 24( degrees )C for 1 hour, kept at 24( degrees )C for 2 hours and rewarmed to 37( degrees )C for 1 hour), 2) the rapid-cooling group (cooled to 24( degrees )C for 30 min, kept at 24( degrees )C for 2 h, and rewarmed to 37( degrees )C for 1 h), 3) the rapid-rewarming group (cooled to 24( degrees )C for 1 h, kept at 24( degrees )C for 2 h, and rewarmed to 37( degrees )C for 30 min) and the normal-control group. FINDINGS: An increase of MAP-2 immunoreactivity of the CA1 neurons in the dorsal hippocampus was observed one week but not one month after hypothermia in the rapid-rewarming group. There was also a significant increase in the glutamate and lactate value at the end of rewarming compared with the baseline in the rapid-rewarming group (p<0.01). INTERPRETATION: Our results suggest that rapid rewarming after hypothermia triggered an uncoupling of cerebral circulation and metabolism, inducing an increase of extracellular glutamate and lactate, consequently reversible neuronal cell damage.

Embryonic intermediate filament, nestin, expression following traumatic spinal cord injury in adult rats.

Precursor cells in the ependyma of the lateral ventricles of adult mammalian brain have been reported in brain, and also in the spinal cord. The present study used antibody to the intermediate filament protein (nestin) as an immunohistochemical marker for neural stem cells and precursor cells in a rat model of spinal cord trauma. Male Sprague-Dawley rats (n=25) had a laminectomy at Thll-Thl2, and spinal cord contusion was created by compression with 30 g of force for 10 min. The rats were killed at 24 h, 1 week and 4 weeks after injury, and four levels of the spinal cord were examined: 5 mm and 10 mm, both rostral and caudal region to the injury center. Time- and region-dependent alterations of nestin immunoreactivity were analyzed. Revealed at 24 h post-injury, 5 mm rostral and caudal to the lesions, nestin expression was observed in ependymal cells and around the hemorrhagic and necrotic lesion located in dorsal spinal cord, peaking at 1 week after injury. Moreover, nestin expression was also observed in the white matter of ventral spinal cord, extending into arborizing processes centripetally from the pial surface toward the central canal. At 4 weeks after injury, nestin expression in ependyma decreased 10 mm from the injury site. But nestin expression in white matter increased dramatically with a 100-fold increase in nestin originating from the pial surface, and extension now to all the white matter. The latter was accompanied by glial fibrillary acidic protein positivity into very long arborizing processes, morphologically compatible with radial glia. The findings suggest two possible sources of precursor cells in adult mammalian spinal cord; ependyma of the central canal and subpial astrocytes. Subpial astrocytes may be associated with neural repair and regeneration after spinal cord injury.

Long-term activation of the glutamatergic system associated with N-methyl-D-aspartate receptors after postischemic hypothermia in gerbils.

OBJECTIVE: The objective of this study was to investigate whether hypothermia would suppress secondary damage in the chronic postischemic stage, in terms of glutamate excitotoxicity. METHODS: Gerbils underwent 5 minutes of ischemia via bilateral common carotid artery occlusion. Seven groups were studied, as follows: 1) ischemia without treatment group; 2) intraischemic hypothermia group; 3) postischemic hypothermia group (32 degrees C for 4 h); 4) MK-801 treatment group (2 mg/kg, every other day for 1 mo); 5) postischemic hypothermia with MK-801 treatment for 1 week group (2 mg/kg, every other day); 6) postischemic hypothermia with MK-801 treatment for 1 month group (2 mg/kg, every other day); and 7) sham-treated control group. One month after ischemia, histological changes in hippocampal CA1 neurons (assessed using hematoxylin and eosin staining) and memory function (assessed using an eight-arm radial maze) were studied. Extracellular glutamate concentrations were monitored by microdialysis during ischemia and hypothermia. Staining of microglia was performed 1 week and 1 month after ischemia. RESULTS: MK-801 alone, postischemic hypothermia alone, and postischemic hypothermia with MK-801 treatment for 1 week failed to prevent ischemic neuronal damage and memory function decreases 1 month after the insult (P < 0.05 versus control). However, the postischemic hypothermia with MK-801 treatment for 1 month group exhibited significant protective effects (not significant [P > 0.05] compared with the control group). Extracellular glutamate levels for the intraischemic hypothermia group were significantly low, compared with the postischemic hypothermia group. There was no microglial activation in the postischemic hypothermia at 1 week and 1 month after ischemia groups. CONCLUSION: Postischemic hypothermia and long-term intermittent administration of MK-801 demonstrated significant neuronal protection, indicating that long-term glutamatergic activation, with changes in N-methyl-D-aspartate receptors, plays a role in neuronal damage in the chronic postischemic stage.

Expression of an embryonic intermediate filament protein in amygdaloid kindled rats.

Amygdaloid kindling is well known as an experimental model of temporal lobe epilepsy. However, the mechanism of kindling epileptogenesis remains unclear. To examine the remodelling process in kindling, we performed immunohistochemistry of nestin, an embryonic intermediate neurofilament protein, in amygdaloid kindled rats. In rats expressing focal seizures (kindling stage C3), nestin immunoreactive cells (NIC) were detected at ipsilateral piriform cortex (PC) and ipsilateral perirhinal cortex (PRh), and at PC bilaterally in fully kindled rats expressing secondary generalized seizures (kindling stage C5). Double staining with glial fibrillary acidic protein revealed that almost all reactive astrocytes at PC express nestin immunoreactivity. These results suggest that glial NIC may participate in the remodelling process at the PC and PRh areas. This is the first report of nestin expression in kindling and suggests that glial nestin at PC and PRh may play a significant role in permanent epileptogenesis in kindling.

The contribution of low affinity NGF receptor (p75NGFR) to delayed neuronal death after ischemia in the gerbil hippocampus.

The implication of low affinity nerve growth factor receptor (p75NGFR), which is believed to play a pro-apoptotic role, in delayed neuronal death (DND) after ischemia in the gerbil hippocampus was investigated. Immunohistochemistry and Western blot analysis revealed that the presence of p75 NGFR immunoreactivity (IR) was negligible in the hippocampus of the sham control gerbil but appeared clearly in CA1 neurons 3 and 4 days after 5-min transient ischemia. Terminal deoxynucleotidyl transferase-mediated UTP nick end labeling (TUNEL) positive nuclei appeared when the level of p75NGFR IR increased. Furthermore, almost all TUNEL-positive CA1 neurons also costained for p75NGFR. These results suggest that p75NGFR contributes to DND after ischemia by an apoptotic mechanism.

Appearance and alteration of TUNEL positive cells through epileptogenesis in amygdaloid kindled rat.

To evaluate the role of neural cell death during and after kindling epileptogenesis, apoptotic cells were analyzed in amygdaloid kindled rat using TUNEL staining as a marker of programmed cell death. TUNEL positive cells (TPC) were stained and counted as apoptotic cells in hippocampus, white matter, diencephalon, and cortex at three stages; C0 (before kindling), C3 (during kindling) and C5 (fully kindled). The animals were studied 2 h after the last stimulation. In all regions except cortex, apoptotic cells in stage C3 were significantly increased as compared to those in stage C0. Especially, in white matter significantly more apoptotic cells in stage C3 were detected than in stage C5. The present data showed that in the course of getting the epileptogenesis, apoptosis had already occurred and this type of cell death may play a significant role in reaching stage C5 through kindling.

Depression of long term potentiation in gerbil hippocampus following postischemic hypothermia.

To investigate the mechanism of chronic cell death following postischemic hypothermia, the change of N-methyl-D-aspartate receptor (NMDAR) were examined by immunohistochemistry of NMDAR1 and long-term potentiation (LTP) in the CA1 subfield of the gerbil hippocampus. At 1 week following postischemic hypothermia (32 degrees Cx4 h), all CA1 neurons survived; however, immunoreactivity of NMDAR1 increased in neuronal perikarya whereas decreased in dendrites in the CA1 neurons. The abnormality was still observed in remaining CA1 neurons at 1 month after hypothermia. LTP was also significantly depressed at 1 week after hypothermia. These results suggest that some abnormalities in the glutamate receptor may be caused by ischemia; such abnormality would persist in spite of hypothermia treatment, resulting in the depression of LTP.

Clostridium perfringens epsilon toxin causes excessive release of glutamate in the mouse hippocampus.

The mechanism of neurotoxicity of Clostridium perfringens epsilon toxin to the mouse brain was investigated. Intravenous injection in mice with the toxin caused seizure and excited hippocampal neurons. Microdialysis revealed that epsilon toxin induced excessive glutamate release in the hippocampus. Both the seizure and glutamate release were attenuated by prior injection with riluzole, an inhibitor of pre-synaptic glutamate release, suggesting that this toxin enhances glutamate efflux, leading to seizure and hippocampal neuronal damage.

Hypoxia, hyperoxia, ischemia, and brain necrosis.

BACKGROUND: Human brains show widespread necrosis when death occurs after coma due to cardiac arrest, but not after hypoxic coma. It is unclear whether hypoxia alone can cause brain damage without ischemia. The relationship of blood oxygenation and vascular occlusion to brain necrosis is also incompletely defined. METHODS: We used physiologically monitored Wistar rats to explore the relationship among arterial blood oxygen levels, ischemia, and brain necrosis. Hypoxia alone (PaO2 = 25 mm Hg), even at a blood pressure (BP) of 30 mm Hg for 15 minutes, yielded no necrotic neurons. Ischemia alone (unilateral carotid ligation) caused necrosis in 4 of 12 rats, despite a PaO2 > 100 mm Hg. To reveal interactive effects of hypoxia and ischemia, groups were studied with finely graded levels of hypoxia at a fixed BP, and with controlled variation in BP at fixed PaO2. In separate series, focal ischemic stroke was mimicked with transient middle cerebral artery (MCA) occlusion, and the effect of low, normal, and high PaO2 was studied. RESULTS: Quantitated neuropathology worsened with every 10 mm Hg decrement in BP, but the effect of altering PaO2 by 10 mm Hg was not as great, nor as consistent. Autoradiographic study of cerebral blood flow with 14C-iodoantipyrine revealed no hypoxic vasodilatation during ischemia. In the MCA occlusion model, milder hypoxia than in the first series (PaO2 = 46.5 +/- 1.4 mm Hg) exacerbated necrosis to 24.3 +/- 4.7% of the hemisphere from 16.6 +/- 7.0% with normoxia (PaO2 = 120.5 +/- 4.1 mm Hg), whereas hyperoxia (PaO2 = 213.9 +/- 5.8 mm Hg) mitigated hemispheric damage to 7.50 +/- 1.86%. Cortical damage was strikingly sensitive to arterial PaO2, being 12.8 +/- 3.1% of the hemisphere with hypoxia, 7.97 +/-4.63% with normoxia, and only 0.3 +/- 0.2% of the hemisphere with hyperoxia (p < 0.01), and necrosis being eliminated completely in 8 of 10 animals. CONCLUSIONS: Hypoxia without ischemia does not cause brain necrosis but hypoxia exacerbates ischemic necrosis. Hyperoxia potently mitigates brain damage in this MCA occlusion model, especially in neocortex.

Influence of rewarming conditions after hypothermia in gerbils with transient forebrain ischemia.

OBJECT: Recently, several studies have demonstrated that hypothermia has a beneficial effect on clinical outcome; however, it is difficult to determine the appropriate rewarming conditions in clinical use. The purpose of the present study was to examine the influence of rewarming conditions in gerbils with transient forebrain ischemia. METHODS: Ischemia was induced in the gerbils by a 5-minute bilateral common carotid artery occlusion, after which the animals were immediately subjected to moderate or deep hypothermia. After moderate hypothermia (30.5 degrees C for 4 hours) the animals were rewarmed over standard, fast, or slow time periods. After deep hypothermia (24 degrees C for 2 hours) the animals were rewarmed in a standard, fast, slow, or stepwise manner. Cerebral blood flow (CBF), extracellular glutamate, and lactate were monitored. Hippocampal CA I cell damage was assessed 7 days after induction of ischemia. In animals treated with moderate hypothermia, the rewarming rate had no influence on the number of surviving neurons. However, fast rewarming from deep hypothermia (to 37 degrees C for 30 minutes) failed to provide the neuroprotective effect of hypothermia. Furthermore, this group showed a poor recovery of CBF (p < 0.01) and, consequently, an increase in extracellular glutamate (p < 0.01) and lactate (p < 0.01) in the hippocampus. CONCLUSIONS: The results of this study indicate a transient uncoupling of CBF and cerebral metabolism during fast rewarming from deep hypothermia, whereas slow and stepwise rewarming periods were found to be useful for protection against uncoupling of CBF and cerebral metabolism during rewarming.

The chronic cell death with DNA fragmentation after post-ischaemic hypothermia in the gerbil hippocampus.

The long-term effects of post-ischaemic hypothermia are controversial. The purpose of this study was to examine the long-term effects of post-ischaemic hypothermia on neuronal survival in gerbils in terms of morphology and function. Hypothermia was induced at 32 degrees C for 4 h immediately after ischaemia. Examination was performed at 1 week and at 1 month after ischaemia. Post-ischaemic hypothermia prevented CA1 neuronal damage 1 week after ischaemia. At 1 month after ischaemic insult, however, the degree of the protective effect of post-ischaemic hypothermia was reduced in the lateral and medial CA1 areas. DNA fragmentation was also observed at 1 month. The errors in the 8-arm radial maze trial were increased at 1 month. These data may indicate that cells in the CA1 area are very vulnerable to ischaemia and die after post-ischaemic hypothermia, and that their death is associated with apoptosis.

Re-evaluation of sexual dimorphism in human corpus callosum.

To study the sexual dimorphism of human corpus cauosum (CC), we analyzed the midsaggital magnetic resonance imaging (MRI) morphometry in 67 adults aged (mean+/-s.d.) 36.82+/-9.35 years. Four specific angles of the CC were determined. All four angles in 34 females and 33 age-matched males showed a significant difference between females and males. These morphometric findings confirm a gender difference in the orientation of corpus callosum.

The reversible change of GluR2 RNA editing in gerbil hippocampus in course of ischemic tolerance.

The ischemic tolerance is known to show protective effects on the neurons and the restricted Ca2+ influx through Ca2+ channels might be involved. In alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor, ribonucleic acid (RNA) editing of the GluR2 subunit determines receptor desensitization and Ca2+ permeability. The authors investigated the effect of ischemic tolerance on the messenger RNA editing of Q/R and R/G sites of GluR2 subunit in hippocampus. It was found that the rate of RNA editing in Q/R site showed no change (100% edited), whereas that in R/G site decreased significantly (83.3% normal editing level to 60.4%) at day 3 (preconditioning period) and returned to normal level at day 14 (after preconditioning period). Further investigation revealed that the decrease of editing rate in ischemic tolerance resulted mainly from the decrease of editing in CA1 area.

Calbrain, a novel two EF-hand calcium-binding protein that suppresses Ca2+/calmodulin-dependent protein kinase II activity in the brain.

A cDNA clone that encodes a novel Ca2+-binding protein was isolated from a human brain cDNA library. The gene for this clone, termed calbrain, encodes a 70-amino acid polypeptide with a predicted molecular mass of 8.06 kDa. The analysis of deduced amino acid sequence revealed that calbrain contains two putative EF-hand motifs that show significantly high homology to those of the calmodulin (CaM) family rather than two EF-hand protein families. By Northern hybridization analysis, an approximate 1.5-kilobase pair transcript of calbrain was detected exclusively in the brain, and in situ hybridization study revealed its abundant expression in the hippocampus, habenular area in the epithalamus, and in the cerebellum. A recombinant calbrain protein showed a Ca2+ binding capacity, suggesting the functional potency as a regulator of Ca2+-mediated cellular processes. Ca2+/calmodulin-dependent kinase II, the most abundant protein kinase in the hippocampus and strongly implicated in the basic neuronal functions, was used to evaluate the physiological roles of calbrain. Studies in vitro revealed that calbrain competitively inhibited CaM binding to Ca2+/calmodulin-dependent kinase II (Ki = 129 nM) and reduced its kinase activity and autophosphorylation.

Suppression of hyperemia after brain ischemia by L-threo-3,4-dihydroxyphenylserine.

Although several studies have shown that L-threo3,4-dihydroxyphenylserine (DOPS) may provide a neuroprotective effect against ischemic brain damage, its protective mechanism is not fully understood. Glutamate release and hippocampal blood flow in ischemia with administration of DOPS were investigated to elucidate the neuroprotective mechanism of DOPS. Pre- (but not post-) ischemic administration of DOPS rescued 73% of hippocampal CA1 neurons (p < 0.001, compared with ischemia only) 1 week after transient global ischemia in gerbils. While glutamate release induced by ischemia was not affected, the increase of hippocampal blood flow during reperfusion was significantly suppressed by DOPS. These results demonstrate that DOPS may prevent reperfusion injury by suppression of hyperemia after ischemia, resulting in neuroprotection.

Specific inhibition of apoptosis after cold-induced brain injury by moderate postinjury hypothermia.

OBJECTIVE: Apoptosis of neuronal cells plays a key role in many developmental and pathological processes of the central nervous system. Deoxyribonucleic acid (DNA) of cells undergoing apoptosis is cleaved by an endonuclease into oligonucleosoma-sized fragments. These fragments can be labeled using in situ terminal deoxynucleotidyl transferase so that the apoptotic cells can be visualized by in situ apoptotic staining. The model of cold-induced rat brain edema was used to further examine this hypothesis. The protective effect of hypothermia was also studied in this model of cold-induced brain injury. METHODS: Using a terminal deoxynucleotidyl transferase-mediated deoxyuridine 5'-triphosphate-biotin nick end labeling technique, the neuronal cells with DNA fragmentation in different regions of the brains of rats subjected to cold-induced brain injury were detected. The internucleosomal fragments of DNA in apoptotic cells were examined using agarose gel electrophoresis. The animals were randomly divided into three groups: 1) sham (n = 8); 2) cold-induced brain injury, killed at 12, 24, 48, 72, and 168 hours after cold lesion (n = 10 for each time point); 3) hypothermia, both mean temporalis and rectal temperatures were reduced by surface cooling to 32 degrees C (standard deviation, 0.1 degrees C) for 3, 6, and 12 hours (n = 10 for each time point) beginning 1 hour after cold-induced brain injury. RESULTS: The apoptotic cells were detectable for up to 72 hours after the initial brain injury and reached a peak at approximately 24 to 48 hours, with a mean peak value of 24.29 +/- 5.26, 15.37 +/- 4.10, 15.81 +/- 3.56, 13.94 +/- 2.48, 10.46 +/- 2.23, and 7.68 +/- 2.48% in the cortex, subcortex, white matter, CA1, CA3, and dentate gyrus, respectively, and had a significant increase, compared with the control value (mean +/- standard error, P < 0.01). Agarose gel electrophoresis of DNA extracted from cortex and hippocampus containing apoptotic cells revealed a "DNA ladder" at 180- to 200-base pair intervals. In animals subjected to the same brain injury that underwent 32 degrees C hypothermia, the numbers of apoptotic cells were reduced evidently and DNA fragmentation was inhibited. CONCLUSION: The data suggest that apoptosis occurs after cold-induced brain injury and that DNA fragmentation may be associated with apoptotic cell death. Moderate hypothermia shows specific effect on inhibition of apoptotic cell death and cellular DNA fragmentation after cold-induced brain injury in rats.

Neurotoxicity of Clostridium perfringens epsilon-toxin for the rat hippocampus via the glutamatergic system.

The neurotoxicity of epsilon-toxin, one of the major lethal toxins produced by Clostridium perfringens type B, was studied by histological examination of the rat brain. When the toxin was injected intravenously at a lethal dose (100 ng/kg), neuronal damage was observed in many areas of the brain. Injection of the toxin at a sublethal dose (50 ng/kg) caused neuronal damage predominantly in the hippocampus: pyramidal cells in the hippocampus showed marked shrinkage and karyopyknosis, or so-called dark cells. The dark cells lost the immunoreactivity to microtubule-associated protein-2, a postsynaptic somal and dendric marker, while acetylcholinesterase-positive fibers were not affected. Timm's zinc staining revealed that zinc ions were depleted in the mossy layers of the CA3 subfield containing glutamate as a synaptic transmitter. The cerebral blood flow in the hippocampus was not altered significantly before or after administration of the toxin, as measured by laser-Doppler flowmetry, excluding the possibility that the observed histological change was due to a secondary effect of ischemia in the hippocampus. Prior injection of either a glutamate release inhibitor or a glutamate receptor antagonist protected the hippocampus from the neuronal damage caused by epsilon-toxin. These results suggest that epsilon-toxin acts on the glutamatergic system and evokes excessive release of glutamate, leading to neuronal damage.