Poor correlation between delayed neuronal death induced by transient forebrain ischemia,and immunoreactivity for parvalbumin and calbindin D-28k in developing gerbil hippocampus

In the normal developing hippocampus of the gerbil, parvalbumin-immunoreactive neurons first appear in the stratum pyramidale of CA3 at postnatal day 15 (P15), and in CA2 and hilus of the dentate gyrus from P21 onwards. Immunoreactive terminals also follow the same sequence from CA3 to CA1 to reach adult patterns by the end of the 1st month. Calbindin D-28k immunoreactivity is seen in the external part of the upper blade of the dentate gyrus at P5, and progresses to the granule cell and molecular layers of the whole gyrus by P15, except for a thin band of immature cells located at the base of the granule cell layer which are calbindin negative. Calbindin immunoreactivity in mossy fibers progresses from the external to the hilar region of CA3 during the same period. A few immunoreactive cells are also found in the stratum radiatum/lacunare of the CA3, but no calbindin-immunoreactive cells are observed in the CA1 and CA2 subfields. The adult pattern of calbindin immunoreactivity is reached at P21. Vulnerability following transient forebrain ischemia for 20 min was examined in the hippocampal formation of gerbils during postnatal development. No cellular damage was seen in animals aged 7 days. Dying cells were observed at the base of the granule cell layer of the dentate gyrus in animals aged 15, 21 and 30 days. Pyramidal cells in the CA3 subfield were also sensitive to ischemia in gerbils aged 15 days, and less frequently in animals aged 21 days. The adult pattern of cellular damage, characterized by selective vulnerability of the CA1 subfield, was seen from day 30 onwards. These findings show that the pattern of selective vulnerability following transient forebrain ischemia is different in young and adult gerbils, and suggest that little, if any, correlation exists between resistance to delayed cellular damage and parvalbumin and calbindin D-28k content in the hippocampus of young gerbils.Supported in part by grant FIS 93-131 and a grant from the Fundacio Pi i Synyer (to A.T.)     

Transient increase of synapsin-I immunoreactivity in the mossy fiber layer of the hippocampus after transient forebrain ischemia in the mongolian gerbil

Synapsin-I is a vesicular phosphoprotein, which regulates neurotransmitter release, neurite development, and maturation of synaptic contacts during normal development and following various brain lesions in adulthood. In the present study, we have examined by immunohistochemistry possible modifications in the expression of synapsin-I in the hippocampus of Mongolian gerbils after transient forebrain ischemia. The animals were subjected to 5 min of transient forebrain ischemia through bilateral common carotid occlusion, and were examined at different time-points post-ischemia. Transient forebrain ischemia produces cell death of the majority of CA1 pyramidal neurons of the hippocampus and polymorphic hilar neurons of the dentate gyrus. This is followed by reactive changes, including synaptic reorganization and modifications in the expression of synaptic proteins, which provide the molecular bases of synaptic plasticity. Transient decrease of synapsin-I immunoreactivity was observed in the inner zone of the molecular layer of the dentate gyrus, thus suggesting denervation and posterior reinervation in this area. In addition, a strong increase in synapsin-I immunoreactivity was observed in the hilus of the dentate gyrus and in the mossy fiber layer of the hippocampus at 2, 4 and 7 days after ischemia. Parallel increases in synaptophysin immunoreactivity were not observed, thus suggesting a selective induction of synapsin-I after ischemia. The present results indicate that synapsin-I participates in the reactive response of granule cells to transient forebrain ischemia in the hippocampus of the gerbil, and suggest a role for this protein in the plastic adaptations of the hippocampus following injury.     

Reduction in brain-derived neurotrophic factor protein level in the hippocampal CA1 dendritic field precedes the delayed neuronal damage in the rat brain

Utilizing a specific polyclonal antibody against a peptide unique for brain-derived neurotrophic factor (BDNF), we investigated the regional and temporal profiles of immunoreactivity of the BDNF protein in the rat hippocampus after transient forebrain ischemia. The pattern of immunoreactivity for the BDNF receptor (TrkB) was also examined and compared with that for BDNF. In the early phase after ischemia, we observed a distinct regional difference in immunoreactivity between the pyramidal cell layer and the stratum radiatum of the CA1 subfield. In the pyramidal cell layer, there was a rapid and transient increase in the positive immunostaining for both BDNF and TrkB. By contrast, in the stratum radiatum there was a marked decrease in BDNF immunoreactivity, but not one in that of TrkB. One week after ischemia, high immunoreactivity for both BDNF and TrkB was observed in the reactive astrocytes in the dendritic field of the CA1 subfield. These findings suggest that a transport of BDNF from the neuronal soma to the dendrites of the stratum radiatum might be ceased after the ischemic insult. Thus, a dysfunctional autocrine mechanism of BDNF within the CA1 neuron may be involved in the pathogenesis of selective neuronal damage after ischemia. J. Neurosci. Res. 53:318–329, 1998. © 1998 Wiley-Liss, Inc.     

Ischemic neuronal damage specific to monkey hippocampus: Histological investigation

We previously reported lesions confined specifically to the hippocampus when produced by occluding eight vessels (the bilateral vertebral, common, internal, and external carotid arteries), which supply blood to the brain. However, histopathological changes in the primate brain, caused by ischemic injury, have not previously been thoroughly investigated. In the present study, macaque monkeys were subjected to 5–18-min ischemia by occluding the eight vessels. After the brains were perfused and fixed 5 days after the occlusion, all regions were histologically investigated for ischemic cell changes. Ischemia for 5 min produced no ischemic cell change. Ischemia for 10–15 min produced cell death limited to the deeper portion of the pyramidal cell layer of the CA1 subfield in the hippocampus. In most monkeys, no cell death was observed in any brain region outside of the hippocampus after ischemia for up to 15 min. Ischemia for 18 min produced more widespread cell death in the CA1 subfield of the hippocampus, and cell death was no longer confined to the hippocampus, but was observed in layers III, V, and VI of the neocortices, the striatum, and some other regions. Brains that were perfused and fixed 1 year after 15-min ischemic insult revealed no ischemic cell morphological change in any region, but the number of pyramidal cells in the CA1 subfield was decreased to about half. The results indicate that the CA1 subfield of the monkey hippocampus is the precise region of the brain most susceptible to ischemic insult in the primate forebrain, and after a critical time (15-min ischemia in this procedure) ischemic cell changes occur suddenly and extensively. Ischemia due to occlusion of eight arteries for 10–15 min could produce a model of human amnesia caused by transient ischemic insult.     

Upregulation of phospholipase D in astrocytes in response to transient forebrain ischemia

Previous in vitro studies using cell cultures or brain slices have demonstrated that phospholipase D (PLD) in the nervous system is involved in the signaling mechanism in response to a variety of agonists. However, little is known about the pathophysiological role of PLD-mediated signaling in the adult brain. We examined the changes in the expression of a PLD isozyme, PLD1, in the adult rat hippocampus, using immunological approaches and an assay for PLD activity after transient forebrain ischemia (four-vessel occlusion model) that results in the selective delayed death of CA1 pyramidal cells and induces reactive astrocytes in the CA1 subfield. In the control hippocampus, PLD1 the level of immunoreactivity was very low. After ischemia, in parallel with the results of Western blot analysis and the PLD activity assay, immunohistochemical analysis of PLD1 demonstrated that the immunoreactive proteins peaked at 7-14 days and were most prominent in the CA1 and the dentate hilar region. The temporal and spatial patterns of immunoreactivity of both PLD1 and glial fibrillary acidic protein (GFAP) were very similar, indicating that reactive astrocytes express PLD1, confirmed by double staining for PLD1 and GFAP. These results demonstrate that reactive astrocytes upregulate PLD in vivo after injury in the adult rat hippocampus.     

The role of GTP binding proteins in ischemic brain damage: autoradiographic and histophatological study

Cerebral ischemia produces perturbation of signal transduction systems in neurons. In order to estimate the contribution of guanine nucleotide-binding protein (G-protein) to hippocampal neuronal death, the effect of pertussis toxin (PTX) on the CA1 pyramidal cell damage after transient forebrain ischemia in rats was examined. PTX was administered 3 days before 20 min of transient forebrain ischemia. PTX injection into the CA1 failed subfield to alter the number of ischemic-damaged CA1 pyramidal cells. In contrast, ventricular PTX injection exacerbated CA1 pyramidal cell damage. We also studied postischemic alteration of GTP binding sites in the hippocampal formation using quantitative in vitro autoradiography. Autoradiographic imaging demonstrated predominant distribution of GTP binding sites in synaptic areas in the hippocampus. No significant change of GTP binding activity was observed in the hippocampus until 2 days after recirculation. Seven days after ischemia, when the CA1 pyramidal cells were depleted, the GTP binding sites of the strata oriens and radiatum in the CA1 subfield had reduced by 32% and 31%, respectively. In contrast, GTP binding in the CA3 subfield and the dentate gyrus remained unaltered throughout the reperfusion period. These results suggest that the amount of G-proteins as estimated by GTP binding remained unaltered in the hippocampus during the early recirculation period, when the CA1 pyramidal cells were morphologically intact, and that signal transduction pathways mediated by G_i and G_o do not play a major role in delayed death of the CA1 pyramidal cells.     

巴曲酶对局灶性脑缺血再灌流大鼠成纤维细胞生长因子样免疫反应的影响

以往实验研究曾发现巴曲酶对脑血管病具有良好的神经保护作用。本文研究的目的为：巴曲酶是否还通过影响成纤维细胞生长因子起修复作用。用大鼠中大脑动脉（ＭＣＡ）线栓法脑缺血再灌注模型及免疫组化方法发现在缺血９０ｍｉｎ，再灌流４８ｈ后，缺血侧皮层、尾壳核及海马区ｂＦＧＦ样免疫反应细胞增加及神经细胞变性。在缺血后给予巴曲酶（８ＢＵ／ｋｇ），ｂＦＧＦ样免疫反应加强，并且缺血对侧相应ＭＣＡ供血脑区也可见轻度的ｂＦＧＦ样免疫反应改变，神经细胞变性之程度较轻。提示：巴曲酶对脑缺血再灌注的保护作用可能与它加强ｂＦＧＦ的修复作用有关。     

Tyrosine kinase A but not phosphacan/protein tyrosine phosphatase-zeta/beta immunoreactivity and protein level changes in neurons and astrocytes in the gerbil hippocampus proper after transient forebrain ischemia

In the present study, ischemia-related changes in tyrosine kinase A (trkA) and phosphacan/protein tyrosine phosphatase-zeta/beta (PTP-zeta/beta) immunoreactivities and protein contents were examined in the hippocampus proper after transient forebrain ischemia for 5 min in a gerbil model. Our investigations showed that ischemia-induced changes occurred in trkA immunoreactivity in the hippocampal CA1 region, but not in the CA2/3 region of the hippocampus proper. In the sham-operated group, trkA immunoreactivity was barely detectable. trkA immunoreactivity increased from 30 min after ischemia and peaked at 12 h. Four days after ischemic insult, trkA immunoreactivity was observed in GFAP-immunoreactive astrocytes in the strata oriens and radiatum. In addition, we found that ischemia-related changes in trkA protein content were similar to immunohistochemical changes. On the other hand, PTP-zeta/beta immunoreactivities in the hippocampus proper were unaltered by forebrain ischemia. These results suggest that chronological changes of trkA after transient forebrain ischemia may be associated with an ischemic damage compensatory mechanism in CA1 pyramidal cells.     

Extracellular signal-regulated kinase 1/2 activation in hippocampus after cerebral ischemia may not interfere with postischemic cell death

To investigate the effect of the activation of extracellular signal-regulated kinase 1/2 (ERK1/2) on cerebral ischemic injury, temporospatial alterations of active (diphosphorylated) ERK1/2 immunoreactivity in hippocampus was examined. Western blot showed that diphosphorylated ERK1/2 were decreased at 10 min of cerebral ischemia but increased rapidly (within 2 min) and transiently (within 4 h) during reperfusion. Immunohistochemistry showed that little diphosphorylated ERK1/2 immunoreactivity was seen in CA1 pyramidal cell bodies after ischemia, while strong immunoreactivity were seen in neuronal bodies in CA3/DG and in fiber systems in both CA1 and CA3 regions. Cerebral ventricular infusion of PD98059, a specific inhibitor of ERK kinase, completely prevented ERK1/2 activation after ischemia but had no effect on the survival of pyramidal cells in CA1 subfield. The results suggest that ERK1/2 activation in hippocampus after brain ischemia may not interfere with the postischemic cell death in CA1 region.     

Activation of mitogen-activated protein kinases after transient forebrain ischemia in gerbil hippocampus.

We investigated the expression, activation, and distribution of c-Jun N-terminal kinases (JNKs), p38 mitogen-activated protein kinases (p38s) and extracellular signal-regulated kinases (ERKs) using Western blotting and immunohistochemistry in gerbil hippocampus after transient forebrain ischemia to clarify the role of these kinases in delayed neuronal death (DND) in the CA1 subfield. Immunoblot analysis demonstrated that activities of JNK, p38, and ERK in whole hippocampus were increased after 5 min of global ischemia. We used an immunohistochemical study to elucidate the temporal and spatial expression of these kinases after transient global ischemia. The immunohistochemical study showed that active JNK and p38 immunoreactivities were enhanced at 15 min of reperfusion and then gradually reduced and disappeared in the hippocampal CA1 region. On the other hand, in CA3 neurons, active JNK and p38 immunoreactivities were enhanced at 15 min of reperfusion and peaked at 6 hr of reperfusion and then gradually reduced but was continuously detected 72 hr after ischemia. Active ERK immunoreactivity was observed transiently in CA3 fibers and dentate gyrus. Pretreatment with SB203580, a p38 inhibitor, but not with PD98059, an ERK kinase 1/2 inhibitor, reduced ischemic cell death in the CA1 region after transient global ischemia by inhibiting the activity of p38. These findings indicate that the p38 pathway may play an important role in DND during brain ischemia in gerbil. Components of the pathway are important target molecules for clarifying the mechanism of neuronal death.     

Expression of neuron specific phosphatase, striatal enriched phosphatase (STEP) in reactive astrocytes after transient forebrain ischemia

We studied the distribution and change of striatal enriched phosphatase (STEP) in the gerbil hippocampus after transient forebrain ischemia. STEP was expressed in the perikarya and in neuronal processes; it was not detected in non-neuronal cells of control animals. After 5-min forebrain ischemia, STEP immunoreactivity (STEP-IR) was preserved for 2 days; it disappeared 4 and more days after ischemia with completion of delayed neuronal death (DND) in the CA1 subfield. Furthermore, only in the CA1 after ischemia, STEP was expressed in reactive astrocytes for 4 to 28 days, showing different patterns of glial fibrillary acidic protein (GFAP)-positive reactive astrocytes. After non-or less-than lethal ischemia, STEP expression in reactive astrocytes corresponded with the degree of neuronal degeneration. Immunoblot analysis of the CA1 subfield revealed the expression of three isoforms, STEP45, -56 and -61; their expression patterns changed with time after ischemia. These data suggest that neuronal STEP is preserved until cell degeneration after ischemia and that STEP is expressed in reactive astrocytes only after lethal ischemia, with different expression patterns for its isoforms. Of STEP45, -56 and -61, STEP61 was the most strongly expressed in the reactive astrocytes; both STEP45 and -61 were expressed in neurons and the expression of STEP56 was weak. STEP may play an important role not only in neurons but also in reactive astrocytes after ischemia, depending on neuronal degeneration.     

c-Jun N-terminal kinase activation in hippocampal CA1 region was involved in ischemic injury

To clarify the role of c-Jun N-terminal kinase (NK) activation in brain ischemia, temporospatial alteration of active (diphosphorylated) JNK1/2 immunoreactivity in hippocampus after brain ischemia in rat was investigated. Western immunoblot study showed that JNK1/2 diphosphorylation level was increased biphasically in CA1 but not CA3/dentate gyrus (DG) after 10 min of ischemia. Cerebral ventricular infusion of JNK1/2 antisense oligonucleotides not only significantly decreased JNK1/2 protein expression and the activation level but also significantly decreased CA1 pyramidal cell death (demonstrated by cresyl violet staining) and DNA fragmentation (demonstrated by in situ end-labeling of DNA). These results suggest that JNK1/2 were selectively activated and involved in the selective cell death in hippocampal CA1 subfield after cerebral ischemia.     

Calbindin-D28K and ischemic damage of pyramidal cells in rat hippocampus.

An antibody against rat calbindin-D28K, a calcium-binding protein present at high concentration in certain neurons of the central and peripheral nervous systems, was used to determine the progression of the pathological events in the rat hippocampus following experimental cerebral ischemia. Calbindin-D28K immunoreactivity is present in dentate granule cells and in the CA1-CA2 pyramidal cells. CA1 subfield contains a higher proportion of calbindin-D28K-positive pyramidal cells than does the CA2 subfield and CA1 cells are more immunoreactive than the CA2 cells. The pyramidal cells of the CA1 and CA2 subfields are vulnerable to ischemia. The cells in the CA1 became necrotic within 3-4 days after ischemia while those of the CA2 became necrotic within 2 days. There was a concomitant decrease in calbindin-D28K immunoreactivity in the whole hippocampal regio superior after ischemia which peaked 3 days postischemia. The difference in CA2 and CA1 vulnerability seemed to be inversely correlated with the calbindin-D28K contents of the CA2 and CA1 pyramidal cells. The decrease in the calbindin-D28K contents of these neurons was accompanied by cell damage. We therefore suggest that calbindin-D28K is an important factor for the survival of pyramidal cells in the hippocampal formation after ischemia.     

Glucose metabolism and neurogenesis in the gerbil hippocampus after transient forebrain ischemia

Recent evidence exists that glucose transporter 3 (GLUT3) plays an important role in the energy metabo-lism in the brain. Most previous studies have been conducted using focal or hypoxic ischemia models and have focused on changes in GLUT3 expression based on protein and mRNA levels rather than tissue levels. In the present study, we observed change in GLUT3 immunoreactivity in the adult gerbil hippocampus at various time points after 5 minutes of transient forebrain ischemia. In the sham-operated group, GLUT3 immunoreactivity in the hippocampal CA1 region was weak, in the pyramidal cells of the CA1 region in-creased in a time-dependent fashion 24 hours after ischemia, and in the hippocampal CA1 region decreased signiifcantly between 2 and 5 days after ischemia, with high level of GLUT3 immunoreactivity observed in the CA1 region 10 days after ischemia. In a double immunolfuorescence study using GLUT3 and gli-al-ifbrillary acidic protein (GFAP), we observed strong GLUT3 immunoreactivity in the astrocytes. GLUT3 immunoreactivity increased after ischemia and peaked 7 days in the dentate gyrus after ischemia/reperfu-sion. In a double immunolfuorescence study using GLUT3 and doublecortin (DCX), we observed low level of GLUT3 immunoreactivity in the differentiated neuroblasts of the subgranular zone of the dentate gyrus after ischemia. GLUT3 immunoreactivity in the sham-operated group was mainly detected in the subgran-ular zone of the dentate gyrus. These results suggest that the increase in GLUT3 immunoreactivity may be a compensatory mechanism to modulate glucose level in the hippocampal CA1 region and to promote adult neurogenesis in the dentate gyrus.     

Chronological alterations of neurofilament 150 immunoreactivity in the gerbil hippocampus and dentate gyrus after transient forebrain ischemia

In this study, we observed the chronological alterations of neurofilament 150 (NF-150) immunoreactivity in the gerbil hippocampus and dentate gyrus after 5 min transient forebrain ischemia. NF-150 immunoreactivity in the sham-operated group was mainly detected in mossy fibers and in the hilar region of the dentate gyrus. NF-150 immunoreactivity and protein contents of NF-150 and RT 97 (polyphosphorylation epitopes of neurofilament) were significantly decreased at 15 min after ischemic insult. Between 30 min and 12 h after ischemic insult, NF-150 immunoreactivity and protein content were significantly increased as compared with the sham-operated group. Thereafter, NF-150 immunoreactivity and protein content started to decrease. At 12 h after ischemic insult, unlike dentate gyrus, NF-150 immunoreactivity increased in pyramidal cells of the CA1 region. Thereafter, NF-150 immunoreactivity in the CA1 region started to decrease, and 4 days after ischemic insult, NF-150 immunoreactivity nearly was similar to that of the sham-operated group. These biphasic patterns of NF-150 immunoreactivity in the hippocampus and dentate gyrus are reverse correlated with that of the intracellular calcium influx. For calcium detection in the CA1 region, we also conducted alizarin red staining. Alizarin red positive neurons were detected in some neurons at 15-30 min after ischemic insult. At 12 h after ischemia, alizarin red positive neurons were decreased. Thereafter, alizarin red positive neurons started to decrease, but alizarin positive neurons were significantly increased in dying neurons 4 days after ischemia. These results suggest that ischemia-related changes of NF-150 expression may be caused by the calcium following transient forebrain ischemia.     

Induction of tolerance to ischemia: Alterations in second-messenger systems in the gerbil hippocampus

Preconditioning the brain with sublethal ischemia protects against neuronal damage following subsequent ischemic insult. Using [3H]inositol 1,4,5-triphosphate (IP3), [3H]phorbol 12,13-dibutyrate (PDBu), [3H]cyclic adenosine monophosphate (cAMP) and [3H]rolipram, we performed quantitative autoradiography to determine postischemic alterations in second-messenger systems in the gerbil hippocampus following preconditioning the brain with sublethal ischemia. At 7 days of reperfusion, no alterations were observed in brains subjected to 2 min of forebrain ischemia which produced no neuronal damage. However, 3-min ischemia caused a 75% reduction in [3H]IP3 binding (p < 0.01 vs. control) and 15-25% reductions in [3H]forskolin (p < 0.01 vs. control), [3H]cAMP (p < 0.05 vs. control), and [3H]rolipram (p < 0.01 vs. control) binding in the CA1 subfield coincident with histopathological CA1 pyramidal cell destruction, but no significant alterations in [3H]PDBu binding. Preconditioning the brain with 2 min of ischemia followed by 4 days of reperfusion prevented both histopathological cell death and the reductions in binding following subsequent 3 min of ischemia. Interestingly, [3H]IP3 and [3H]rolipram binding in CA1 showed a transient reduction, by 30% and 20% (both p < 0.01 vs. control), respectively, in the early reperfusion period. This downregulation of the IP3 system may play a role in the protection against cell death.     

Na+/Ca2+ exchanger 1 alters in pyramidal cells and expresses in astrocytes of the gerbil hippocampal CA1 region after ischemia

Alterations of immunoreactivity and protein contents of Na(+)/Ca(2+) exchanger 1 (NCX1) were observed in the gerbil hippocampus proper after 5 min of transient forebrain ischemia. NCX1 immunoreactivity was significantly changed in the hippocampal CA1 region, but not in the CA2/3 region after ischemia/reperfusion. In the sham-operated group, NCX1 immunoreactivity was mainly detected in CA1 pyramidal cells. However, 30 min after ischemia/reperfusion, NCX1 immunoreactivity was significantly decreased and then increased at 1 day after ischemia. At this time, NCX1 immunoreactivity in CA1 pyramidal cells was similar to that of the sham-operated group. At 3 days after ischemia, NCX1 immunoreactivity was significantly reduced in the CA1 region compared to that of the sham-operated group and NCX1 immunoreactivity was significantly increased again 4 days after ischemia. Thereafter, NCX1 immunoreactivity was decreased time-dependently in ischemia groups. Between 15 min and 6 h post-ischemia, NCX1 immunoreactivity was expressed in astrocytes in the strata oriens and radiatum of the CA1 region. From 3 days post-ischemia, NCX1 immunoreactivity was expressed in astrocytes in the strata oriens and radiatum. Ischemia-induced changes in NCX1 protein contents in the hippocampus proper concurred with immunohistochemical data post-ischemia. Our results suggest that changes in NCX1 in CA1 pyramidal cells and astrocytes after ischemia are associated with intracellular Na(+) concentrations and that NCX1 may induce an intracellular calcium overload, which may be related to neuronal death.     

Excitatory amino acid binding sites in the rat hippocampus after transient forebrain ischemia

The influence of transient forebrain ischemia on the temporal alteration of glutamate receptors in the hippocampal formation was analyzed by means of in vitro quantitative receptor autoradiography. We compared the binding of N-methyl-D-aspartate (NMDA) receptors using [3H]3-[+/-)2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP), noncompetitive NMDA antagonist binding sites using [3H]N-(1-(2-thienyl)-cyclohexyl)-3,4-piperidine (TCP), and kainate (KA) receptors. In the CA1 subfield of the hippocampus, the number of NMDA receptors and noncompetitive NMDA antagonist binding sites remained constant during the early stage of recirculation when the CA1 pyramidal cells remained histologically intact. A significant reduction of these receptor densities was observed 7 days following ischemia, when NMDA receptors and noncompetitive NMDA antagonist binding sites lost 64 and 29% of their binding sites in the stratum radiatum of the CA1, respectively. The KA receptor density in the CA1 subfield decreased by 44% 7 days after ischemia. Marked loss of the above-mentioned receptors in the CA1 after selective depletion of the CA1 pyramidal cells indicated that NMDA receptors, noncompetitive NMDA antagonist binding sites, and KA receptors in the CA1 are predominantly localized on the CA1 pyramidal cells. NMDA receptor density in the CA3 gradually decreased during the recirculation period. The stratum moleculare of the dentate gyrus, whose structure was histologically intact after ischemic insult, also showed a reduction of NMDA receptors 7 days following ischemia. [3H]KA receptor density in the stratum lucidum of the CA3 and in the hilus also decreased during recirculation. These