Brain
Rrsearch
Bulletin.
Vol. 25. pp. 319-324. 0 Pergamon Press plc. 1990. Printed in the U.S A
Gerbil Hippocampal Extracellular Glutamate and Neuronal Activity After Transient Ischemia AKIRA MITANI,“’ HITOSHI IMON,t KOUZOU IGA,$ HIROHIKO KUBO§ AND KIYOSHI KATAOKA*
Departments of *Physiology, fAnesthesiology, iforthopedics and §Surger? The Universio of Ehime, School of Medicine, Ehime, 791-02, Japun Received 5 December
1989
MITANI, A., H. IMON, K. IGA, H. KUBO AND K. KATAOKA. Gerbil hippocampcll extracellularglurumc~eand neuroncrl acriviry rransienr ischemia. BRAIN RES BULL 25(2) 319-324, 1990.-m order to elucidate the role of glutamate in the pathogenesis of delayed neuronal death, we analyzed changes in extracellular levels of glutamate induced by transient ischemia in the Mongolian gerbil hippocampus by a new brain microdialysis method combined with an enzymatic cycling technique. We also studied the effect of this change in glutamate on CA1 spontaneous neuronal discharges. The level of glutamate significantly increased during the 5 min of ischemia and during the first 5 min of recirculation. However, neuronal hyperactivity anticipated as a result of the increased extracellular glutamate was not observed. Spike discharges disappeared during and shortly after 5 min of ischemia; CA1 spontaneous spike discharges reappeared about 15 min after the recirculation. The frequency and amplitude of the discharges of CA1 neurons returned to normal by 30 min of the recirculation. However, the pattern of discharges was different from that recorded before the ischemia. CA1 neurons were found dead 4 days after the ischemia. Brief exposure to toxic concentrations of glutamate may cause the delayed neuronal death. after
Ischemia Glutamate Neuronal death
Enzymatic
cycling
Single neuronal
GLUTAMATE has been proposed to be one of the major excitatory neurotransmitters and to be inherently neurotoxic in the mammalian central nervous system (20,21). Ischemia-induced neuronal death is frequently seen in regions where glutamate is involved in neurotransmission (9), and where elevation of extracellular levels of glutamate is observed during transient cerebral ischemia (2, 6, 7). These findings support the theory that the excitatory amino acid exerts toxic effects during ischemia. The hippocampus is one of the most vulnerable regions to ischemia (12, 13, 23). Following carotid arterial occlusion of Sminutes duration. the Mongolian gerbil has been reported to develop selective neuronal death; the pyramidal neurons in the CA1 sector of the hippocampus especially show delayed neuronal death 34 days after transient ischemia (12,30). According to previous reports ( 14. 28, 29), the pyramidal neurons in the CA1 sector receive strong glutamatergic innervations, namely the Schaffer collaterals from the CA3 sector, the commissural fibers from the contralateral hippocampus, and the perforant path fibers from the entorhinal cortex. Thus, glutamate appears to be abundantly released in this region and to exert a toxic effect on the CA1 neurons. In order to clarify glutamate’s role in the pathogenesis of delayed neuronal death, we measured extracellular levels of ‘Requests Onsen-Gun,
for reprints should be addressed Ehime, 791-02, Japan.
to Dr. A. Mitani, Department
activity
Mongolian
gerbil
Hippocampus
glutamate before, during and after ischemia and CA1 neuronal responses to the shift in extracellular levels of glutamate. Recently, a brain microdialysis method has been used to measure extracellular levels of glutamate ( 15.3 1). The extracellular levels of glutamate during and after ischemia were measured to investigate the neurotoxicity of glutamate (2, 6. 7). However, as Benveniste (1) mentioned in her review. previous studies (generally using HPLC analysis) had a disadvantage with respect to temporal resolution because the collection time for each sample had to be longer than 5 min to allow for a reliable measurement. The sampling duration of longer than 5 min is too long to reveal the dynamics of the rapid changes in extracellular levels of glutamate. The Mongolian gerbil is extensively used as a model for cerebral ischemia because of its unusual cerebral circulation which lacks connections between the carotid and vertebrobasilar circulations. Bilateral carotid occlusion causes complete forebrain ischemia, and transient carotid occlusion for 5 min induces delayed neuronal death in the hippocampal CA1 sector of this animal (12). If we intend to reveal the rapid temporal changes in extracellular levels of glutamate during a 5-min period of ischemia in this animal, a shorter duration of sampling is needed. Therefore. a new of Physiology.
319
The University
of Ehime, School of Medicine.
Shipenobu,
method of measuring low levels of glutamate in a small amount of dialysate is required. In the present study we used a newly developed brain microdialysis method combined with an enzymatic cycling technique (17). This permitted the measuring of extracellular levels of glutamate concentrations in the Mongolian gerbil hippocampus before, during, and after a S-min ischemia at 100-set intervals. METHOD
Preparation Male Mongolian gerbils (60-80 g) were anesthetized with a mixture of nitrous oxide [oxygen (7:3) and 2% halothane]. Through a midline cervical incision, both common carotid arteries were exposed and dissected free of surrounding tissues, and 4-O silk sutures were looped around them. Body temperature was maintained at 37-38°C with a heating blanket. The animals were mounted on a stereotaxic apparatus (David Kopf) and a small bone opening was made by a dental drill at 1.5-2.5 mm posterior and 1.5-2.5 mm lateral to the bregma. The dura was carefully incised. Brain Dialysis Procedure A microdialysis probe (l-mm dialysis membrane, 0.22-mm o.d.; molwt. cut-off=50,000; Eicom, Japan) was attached to the micromanipulator of a Kopf stereotaxic device and positioned perpendicularly into the hippocampus (2.2 mm ventral to the cortical surface, 2.0 mm posterior and 2.0 mm lateral to the bregma). Subsequently, it was perfused with Ringer’s solution at a flow rate of 2 $/min by means of a microinfusion pump (BRC, Japan). Following a 2-hr stabilization period, a bilateral transient forebrain ischemia was performed. The sutures around the two common carotid arteries were pulled by 10-g weights to occlude the circulation (16,30). Following .5-min ischemia, the sutures were cut and removed to restore blood flow. Collections of dialysates began 500 set before the ischemia. One-hundredsecond samples (=3.3 ~1) of the dialysate were collected in paraffin oil in polyethylene sampling tubes in an ice bath: the end of a small teflon outlet tube for dialysate was dipped in the paraffin oil. Glutamate Assay This method is in principal the same as described in our previous study (17). Immediately after collection, all sampling tubes were centrifuged at 200 X g for 5 set at 4°C to make a bolus of dialysate at the bottom of the tube. Glutamate was analyzed by the following serial enzymatic reactions. In brief, the dialysate in the paraffin oil was combined first with 66 p.1of enzymatic reagent and NAD+ for 30 minutes to form NADH. The reaction was stopped by the addition of 15 p.1of 1 M NaOH followed by heating at 60°C for 20 min. Subsequently, for triplicate determinations, three 5-)~l aliquots were transferred to fluorometer tubes, and used for NAD+-NADH cycling by means of the enzymatic cycling reactions described by Kato et al. (11) and others (19). The fluorescence of NADH was measured with a Farrand’s fluorometer. L-Glutamic acid standards of zero and 0.5-50 X 10-i’ mol/l were quantified in parallel with the samples throughout the assay in each experiment, and glutamate concentrations of the samples were read from the standard curve; mean values and standard errors were calculated for all animals studied. Differences were evaluated by means of the two-tailed Student’s t-test. Unit Recording Procedure Two-barrelled
electrodes with a common tip of diameter 1.5-2
pm were used: one barrel was filled with 4 M NaCl (impedance 3-10 Mohms) and used to record extracellular single neuronal activities, and another barrel was filled with ‘1.5% pontamine shy blue dye in a solution of 0.2 M sodium acetate and used to confirm the site of recording. The recording microelectrode was perpendicularly advanced into the hippocampal CAI sector through the cortex. The spontaneous extracellular single neuronal discharges were observed on an oscilloscope and recorded by FM on magnetic tape. The recorded data were replayed through a small computer to produce histograms of the number of action potentials per second. After obtaining a stable recording, 5-min ischemia was induced as described above. Recording was continued for 60-90 min after the start of recirculation. At the end of the recording, the dye was injected iontophoretically ( -5 PA. 250 msec on-off, 2 Hz, 30 min). This relatively large current allowed visualization of the dye at the end of the 4-S day survival time necessary for development of neuronal death. Histology After the microdialysis or unit recording procedures all surgical incisions were carefully sutured. Animals were treated with antibiotics, removed from the stereotaxic apparatus and brought into a comfortable position on a warming blanket. After awakening, the gerbils were returned to individual cages in a room maintained at constant temperature (30°C) and allowed access to food and water ad lib. At four days post op, the animals were given an overdose of pentobarbital and perfused transcardially with 10% formalin in 0.1 M phosphate buffer (pH 7.4). Subsequently, brains were removed and saturated with a cold solution of 30% sucrose in 10% formalin. Frontal serial sections, 50 km in thickness, were then made on a freezing microtome and mounted on to gelatin-coated slides and counterstained with 0.3% neutral red or 0.5% cresyl violet to examine histological changes in CA1 neurons and the position of the microdialysis probe. RESULTS
Extracellular Glutamate Dialysates were collected from ten animals in which a microdialysis probe had been placed in the hippocampus. Location of the probe is shown in Fig. 1. The l-mm long dialysis membrane was situated in the CA1 sector and dentate gyrus (approximately 70 percent of the dialysis membrane was in the CA1 sector, and the remaining 30 percent was in the dentate gyrus). The temporal dynamics of glutamate release are illustrated in Fig. 2. Preischemic levels of extracellular glutamate were stable in five consecutive dialysates collected before ischemia. The average (-+ SEM) concentration of glutamate was 0.920.2 FM in ten animals. After the onset of ischemia, an extracellular increase in glutamate was recorded in samples collected during the first 100 set and a gradual increase persisted during the ischemia. The levels of glutamate were maximal at the end of &hernia (an 18.7-fold increase, average concentration in the dialysates = 16.8 FM). After recirculation, the increased levels of glutamate returned almost to preischemic levels by 5 min of recirculation. Spontaneous Single Neuronal Activiry Spontaneous single neuronal activities were recorded for fifteen CA1 pyramidal neurons which showed spike discharges during the recirculation. The recording sites were confirmed from sections showing iontophoretically injected dyes (Fig. 3). Fairly constant spontaneous discharges were recorded from CA1 neurons under anesthesia, and the average frequency of the spike dis-
ISCHEMIA-INDUCED GLUTAMATE AND DISCHARGE
321
FlG. I. Track of microdialysis probe in the Mongolian gerbil hippocampus
(frontal section: 50 km, stained with cresyl violet). A portion of dialysis membrane (l-mm long, corresponding to the distance between two single-headed arrows: the tip of probe does not have a dialysis membrane) was placed in the hippocampus. A double-headed arrow indicates the boundary between the CAI sector and the dentate gyrus. Note: widespread neuronal death is seen in the CA1 sector (right from arrowhead). Survival time was 3 days. Scale bar=0.5 mm.
was 6.2 2 1S/set (mean? SD) in the present study (Table I). Immediately after the onset of ischemia, abrupt increases in the frequency of spike discharges persisted for 2-8 set in many charges
2000
p
1500
2
E
P :: 1000 G
&
5
E
z
neurons (nine neurons; 60%). Spontaneous spike discharges disappeared 8-2.5 set after the onset of ischemia (Fig. 4): the average time of the disappearance was 14.554.8 set after the onset of ischemic insult. This silent period continued after the start of recirculation. Spike discharges reappeared about 15 min after recirculation. Continuous discharges were observed immediately after the first spike discharge in many neurons ( 11 neurons; 73%) (Fig. 4). In the others (four neurons; 27%). groups of a few spike discharges were recorded every lo-30 set for 5-10 min after the appearance of the first single spike; thereafter the frequency gradually increased. All of the recorded CA1 neurons recovered by 30 min after recirculation, in that spike discharges were similar to those before ischemia in frequency and amplitude (Table 1 and Figs. 4, 5). In contrast the discharge pattern was different. Spike discharges appeared mainly as bursts of a few spike discharges during the preischemic period (Fig. 5A); after recirculation discharges tended to appear as a series of single spike potentials (Fig. 5B). The spike discharge pattern was continuously observed until the end of recordings.
SOC
DISCUSSION
3Q
1oc C 5
0
5
10
15
20
25
30
Mln before and after onset of whemla FIG. 2. Time course changes in the dialysate levels of glutamate sampled from the Mongolian gerbil hippocampus (n= 10). The hippocampal dialysates were collected at IOO-set intervals, beginning 500 set before ischemia, during 5 min of ischemia (stippled area), and during the first 30 min of recirculation. Glutamate concentrations in each dialysate were analyzed by an enzymatic cycling assay and calculated as percentages of the mean concentration in the five dialysates collected before the onset of ischemic insult. The data are represented as mean k SEM (bars). A significant increase in glutamate levels was observed during the ischemia and during the first 5 min of recirculation. Asterisks indicate statistically significant differences compared with the preischemic levels using Student’s r-test (**p
Rrsearch
Bulletin.
Vol. 25. pp. 319-324. 0 Pergamon Press plc. 1990. Printed in the U.S A
Gerbil Hippocampal Extracellular Glutamate and Neuronal Activity After Transient Ischemia AKIRA MITANI,“’ HITOSHI IMON,t KOUZOU IGA,$ HIROHIKO KUBO§ AND KIYOSHI KATAOKA*
Departments of *Physiology, fAnesthesiology, iforthopedics and §Surger? The Universio of Ehime, School of Medicine, Ehime, 791-02, Japun Received 5 December
1989
MITANI, A., H. IMON, K. IGA, H. KUBO AND K. KATAOKA. Gerbil hippocampcll extracellularglurumc~eand neuroncrl acriviry rransienr ischemia. BRAIN RES BULL 25(2) 319-324, 1990.-m order to elucidate the role of glutamate in the pathogenesis of delayed neuronal death, we analyzed changes in extracellular levels of glutamate induced by transient ischemia in the Mongolian gerbil hippocampus by a new brain microdialysis method combined with an enzymatic cycling technique. We also studied the effect of this change in glutamate on CA1 spontaneous neuronal discharges. The level of glutamate significantly increased during the 5 min of ischemia and during the first 5 min of recirculation. However, neuronal hyperactivity anticipated as a result of the increased extracellular glutamate was not observed. Spike discharges disappeared during and shortly after 5 min of ischemia; CA1 spontaneous spike discharges reappeared about 15 min after the recirculation. The frequency and amplitude of the discharges of CA1 neurons returned to normal by 30 min of the recirculation. However, the pattern of discharges was different from that recorded before the ischemia. CA1 neurons were found dead 4 days after the ischemia. Brief exposure to toxic concentrations of glutamate may cause the delayed neuronal death. after
Ischemia Glutamate Neuronal death
Enzymatic
cycling
Single neuronal
GLUTAMATE has been proposed to be one of the major excitatory neurotransmitters and to be inherently neurotoxic in the mammalian central nervous system (20,21). Ischemia-induced neuronal death is frequently seen in regions where glutamate is involved in neurotransmission (9), and where elevation of extracellular levels of glutamate is observed during transient cerebral ischemia (2, 6, 7). These findings support the theory that the excitatory amino acid exerts toxic effects during ischemia. The hippocampus is one of the most vulnerable regions to ischemia (12, 13, 23). Following carotid arterial occlusion of Sminutes duration. the Mongolian gerbil has been reported to develop selective neuronal death; the pyramidal neurons in the CA1 sector of the hippocampus especially show delayed neuronal death 34 days after transient ischemia (12,30). According to previous reports ( 14. 28, 29), the pyramidal neurons in the CA1 sector receive strong glutamatergic innervations, namely the Schaffer collaterals from the CA3 sector, the commissural fibers from the contralateral hippocampus, and the perforant path fibers from the entorhinal cortex. Thus, glutamate appears to be abundantly released in this region and to exert a toxic effect on the CA1 neurons. In order to clarify glutamate’s role in the pathogenesis of delayed neuronal death, we measured extracellular levels of ‘Requests Onsen-Gun,
for reprints should be addressed Ehime, 791-02, Japan.
to Dr. A. Mitani, Department
activity
Mongolian
gerbil
Hippocampus
glutamate before, during and after ischemia and CA1 neuronal responses to the shift in extracellular levels of glutamate. Recently, a brain microdialysis method has been used to measure extracellular levels of glutamate ( 15.3 1). The extracellular levels of glutamate during and after ischemia were measured to investigate the neurotoxicity of glutamate (2, 6. 7). However, as Benveniste (1) mentioned in her review. previous studies (generally using HPLC analysis) had a disadvantage with respect to temporal resolution because the collection time for each sample had to be longer than 5 min to allow for a reliable measurement. The sampling duration of longer than 5 min is too long to reveal the dynamics of the rapid changes in extracellular levels of glutamate. The Mongolian gerbil is extensively used as a model for cerebral ischemia because of its unusual cerebral circulation which lacks connections between the carotid and vertebrobasilar circulations. Bilateral carotid occlusion causes complete forebrain ischemia, and transient carotid occlusion for 5 min induces delayed neuronal death in the hippocampal CA1 sector of this animal (12). If we intend to reveal the rapid temporal changes in extracellular levels of glutamate during a 5-min period of ischemia in this animal, a shorter duration of sampling is needed. Therefore. a new of Physiology.
319
The University
of Ehime, School of Medicine.
Shipenobu,
method of measuring low levels of glutamate in a small amount of dialysate is required. In the present study we used a newly developed brain microdialysis method combined with an enzymatic cycling technique (17). This permitted the measuring of extracellular levels of glutamate concentrations in the Mongolian gerbil hippocampus before, during, and after a S-min ischemia at 100-set intervals. METHOD
Preparation Male Mongolian gerbils (60-80 g) were anesthetized with a mixture of nitrous oxide [oxygen (7:3) and 2% halothane]. Through a midline cervical incision, both common carotid arteries were exposed and dissected free of surrounding tissues, and 4-O silk sutures were looped around them. Body temperature was maintained at 37-38°C with a heating blanket. The animals were mounted on a stereotaxic apparatus (David Kopf) and a small bone opening was made by a dental drill at 1.5-2.5 mm posterior and 1.5-2.5 mm lateral to the bregma. The dura was carefully incised. Brain Dialysis Procedure A microdialysis probe (l-mm dialysis membrane, 0.22-mm o.d.; molwt. cut-off=50,000; Eicom, Japan) was attached to the micromanipulator of a Kopf stereotaxic device and positioned perpendicularly into the hippocampus (2.2 mm ventral to the cortical surface, 2.0 mm posterior and 2.0 mm lateral to the bregma). Subsequently, it was perfused with Ringer’s solution at a flow rate of 2 $/min by means of a microinfusion pump (BRC, Japan). Following a 2-hr stabilization period, a bilateral transient forebrain ischemia was performed. The sutures around the two common carotid arteries were pulled by 10-g weights to occlude the circulation (16,30). Following .5-min ischemia, the sutures were cut and removed to restore blood flow. Collections of dialysates began 500 set before the ischemia. One-hundredsecond samples (=3.3 ~1) of the dialysate were collected in paraffin oil in polyethylene sampling tubes in an ice bath: the end of a small teflon outlet tube for dialysate was dipped in the paraffin oil. Glutamate Assay This method is in principal the same as described in our previous study (17). Immediately after collection, all sampling tubes were centrifuged at 200 X g for 5 set at 4°C to make a bolus of dialysate at the bottom of the tube. Glutamate was analyzed by the following serial enzymatic reactions. In brief, the dialysate in the paraffin oil was combined first with 66 p.1of enzymatic reagent and NAD+ for 30 minutes to form NADH. The reaction was stopped by the addition of 15 p.1of 1 M NaOH followed by heating at 60°C for 20 min. Subsequently, for triplicate determinations, three 5-)~l aliquots were transferred to fluorometer tubes, and used for NAD+-NADH cycling by means of the enzymatic cycling reactions described by Kato et al. (11) and others (19). The fluorescence of NADH was measured with a Farrand’s fluorometer. L-Glutamic acid standards of zero and 0.5-50 X 10-i’ mol/l were quantified in parallel with the samples throughout the assay in each experiment, and glutamate concentrations of the samples were read from the standard curve; mean values and standard errors were calculated for all animals studied. Differences were evaluated by means of the two-tailed Student’s t-test. Unit Recording Procedure Two-barrelled
electrodes with a common tip of diameter 1.5-2
pm were used: one barrel was filled with 4 M NaCl (impedance 3-10 Mohms) and used to record extracellular single neuronal activities, and another barrel was filled with ‘1.5% pontamine shy blue dye in a solution of 0.2 M sodium acetate and used to confirm the site of recording. The recording microelectrode was perpendicularly advanced into the hippocampal CAI sector through the cortex. The spontaneous extracellular single neuronal discharges were observed on an oscilloscope and recorded by FM on magnetic tape. The recorded data were replayed through a small computer to produce histograms of the number of action potentials per second. After obtaining a stable recording, 5-min ischemia was induced as described above. Recording was continued for 60-90 min after the start of recirculation. At the end of the recording, the dye was injected iontophoretically ( -5 PA. 250 msec on-off, 2 Hz, 30 min). This relatively large current allowed visualization of the dye at the end of the 4-S day survival time necessary for development of neuronal death. Histology After the microdialysis or unit recording procedures all surgical incisions were carefully sutured. Animals were treated with antibiotics, removed from the stereotaxic apparatus and brought into a comfortable position on a warming blanket. After awakening, the gerbils were returned to individual cages in a room maintained at constant temperature (30°C) and allowed access to food and water ad lib. At four days post op, the animals were given an overdose of pentobarbital and perfused transcardially with 10% formalin in 0.1 M phosphate buffer (pH 7.4). Subsequently, brains were removed and saturated with a cold solution of 30% sucrose in 10% formalin. Frontal serial sections, 50 km in thickness, were then made on a freezing microtome and mounted on to gelatin-coated slides and counterstained with 0.3% neutral red or 0.5% cresyl violet to examine histological changes in CA1 neurons and the position of the microdialysis probe. RESULTS
Extracellular Glutamate Dialysates were collected from ten animals in which a microdialysis probe had been placed in the hippocampus. Location of the probe is shown in Fig. 1. The l-mm long dialysis membrane was situated in the CA1 sector and dentate gyrus (approximately 70 percent of the dialysis membrane was in the CA1 sector, and the remaining 30 percent was in the dentate gyrus). The temporal dynamics of glutamate release are illustrated in Fig. 2. Preischemic levels of extracellular glutamate were stable in five consecutive dialysates collected before ischemia. The average (-+ SEM) concentration of glutamate was 0.920.2 FM in ten animals. After the onset of ischemia, an extracellular increase in glutamate was recorded in samples collected during the first 100 set and a gradual increase persisted during the ischemia. The levels of glutamate were maximal at the end of &hernia (an 18.7-fold increase, average concentration in the dialysates = 16.8 FM). After recirculation, the increased levels of glutamate returned almost to preischemic levels by 5 min of recirculation. Spontaneous Single Neuronal Activiry Spontaneous single neuronal activities were recorded for fifteen CA1 pyramidal neurons which showed spike discharges during the recirculation. The recording sites were confirmed from sections showing iontophoretically injected dyes (Fig. 3). Fairly constant spontaneous discharges were recorded from CA1 neurons under anesthesia, and the average frequency of the spike dis-
ISCHEMIA-INDUCED GLUTAMATE AND DISCHARGE
321
FlG. I. Track of microdialysis probe in the Mongolian gerbil hippocampus
(frontal section: 50 km, stained with cresyl violet). A portion of dialysis membrane (l-mm long, corresponding to the distance between two single-headed arrows: the tip of probe does not have a dialysis membrane) was placed in the hippocampus. A double-headed arrow indicates the boundary between the CAI sector and the dentate gyrus. Note: widespread neuronal death is seen in the CA1 sector (right from arrowhead). Survival time was 3 days. Scale bar=0.5 mm.
was 6.2 2 1S/set (mean? SD) in the present study (Table I). Immediately after the onset of ischemia, abrupt increases in the frequency of spike discharges persisted for 2-8 set in many charges
2000
p
1500
2
E
P :: 1000 G
&
5
E
z
neurons (nine neurons; 60%). Spontaneous spike discharges disappeared 8-2.5 set after the onset of ischemia (Fig. 4): the average time of the disappearance was 14.554.8 set after the onset of ischemic insult. This silent period continued after the start of recirculation. Spike discharges reappeared about 15 min after recirculation. Continuous discharges were observed immediately after the first spike discharge in many neurons ( 11 neurons; 73%) (Fig. 4). In the others (four neurons; 27%). groups of a few spike discharges were recorded every lo-30 set for 5-10 min after the appearance of the first single spike; thereafter the frequency gradually increased. All of the recorded CA1 neurons recovered by 30 min after recirculation, in that spike discharges were similar to those before ischemia in frequency and amplitude (Table 1 and Figs. 4, 5). In contrast the discharge pattern was different. Spike discharges appeared mainly as bursts of a few spike discharges during the preischemic period (Fig. 5A); after recirculation discharges tended to appear as a series of single spike potentials (Fig. 5B). The spike discharge pattern was continuously observed until the end of recordings.
SOC
DISCUSSION
3Q
1oc C 5
0
5
10
15
20
25
30
Mln before and after onset of whemla FIG. 2. Time course changes in the dialysate levels of glutamate sampled from the Mongolian gerbil hippocampus (n= 10). The hippocampal dialysates were collected at IOO-set intervals, beginning 500 set before ischemia, during 5 min of ischemia (stippled area), and during the first 30 min of recirculation. Glutamate concentrations in each dialysate were analyzed by an enzymatic cycling assay and calculated as percentages of the mean concentration in the five dialysates collected before the onset of ischemic insult. The data are represented as mean k SEM (bars). A significant increase in glutamate levels was observed during the ischemia and during the first 5 min of recirculation. Asterisks indicate statistically significant differences compared with the preischemic levels using Student’s r-test (**p
