Glial uptake of excitatory amino acids influences neuronal survival in cultures of mouse hippocampus

Defining the nature of human pluripotent stem cell-derived interneurons via single-cell analysis

The specification of inhibitory neurons has been described for the mouse and human brain, and many studies have shown that pluripotent stem cells (PSCs) can be used to create interneurons in vitro. It is unclear whether in vitro methods to produce human interneurons generate all the subtypes found in brain, and how similar in vitro and in vivo interneurons are. We applied single-nuclei and single-cell transcriptomics to model interneuron development from human cortex and interneurons derived from PSCs. We provide a direct comparison of various in vitro interneuron derivation methods to determine the homogeneity achieved. We find that PSC-derived interneurons capture stages of development prior to mid-gestation, and represent a minority of potential subtypes found in brain. Comparison with those found in fetal or adult brain highlighted decreased expression of synapse-related genes. These analyses highlight the potential to tailor the method of generation to drive formation of particular subtypes.

Brain inflammation and injury at 48 h is not altered by human amnion epithelial cells in ventilated preterm lambs

BACKGROUND

Mechanical ventilation of preterm neonates is associated with neuroinflammation and an increased risk of adverse neurological outcomes. Human amnion epithelial cells (hAECs) have anti-inflammatory and regenerative properties. We aimed to determine if intravenous administration of hAECs to preterm lambs would reduce neuroinflammation and injury at 2 days of age.

METHODS

Preterm lambs were delivered by cesarean section at 128-130 days' gestation (term is ~147 days) and either ventilated for 48 h or humanely killed at birth. Lambs received 3 mL surfactant (Curosurf) via endotracheal tube prior to delivery (either with or without 90 × 10⁶ hAECs) and 3 mL intravenous phosphate-buffered saline (with or without 90 × 10⁶ hAECs, consistent with intratracheal treatment) after birth.

RESULTS

Ventilation increased microglial activation, total oligodendrocyte cell number, cell proliferation and blood-brain barrier permeability (P < 0.05, PBS + ventilation and hAEC + ventilation vs. control), but did not affect numbers of immature and mature oligodendrocytes. Ventilation reduced astrocyte and neuron survival (P < 0.05, PBS + ventilation and hAEC + ventilation vs. control). hAEC administration did not alter markers of neuroinflammation or injury within the white or gray matter.

CONCLUSIONS

Mechanical ventilation for 48 h upregulated markers of neuroinflammation and injury in preterm lambs. Administration of hAECs did not affect markers of neuroinflammation or injury.

IMPACT

Mechanical ventilation of preterm lambs for 48 h, in a manner consistent with contemporary neonatal intensive care, causes neuroinflammation, neuronal loss and pathological changes in oligodendrocyte and astrocyte survival consistent with evolving neonatal brain injury.Intravenous administration of hAECs immediately after birth did not affect neonatal cardiorespiratory function and markers of neuroinflammation or injury.Reassuringly, our findings in a translational large animal model demonstrate that intravenous hAEC administration to the preterm neonate is safe.Considering that hAECs are being used in phase 1 trials for the treatment of BPD in preterm infants, with future trials planned for neonatal neuroprotection, we believe these observations are highly relevant.

Glutathione S-Transferase Alpha 4 Prevents Dopamine Neurodegeneration in a Rat Alpha-Synuclein Model of Parkinson's Disease

Parkinson’s disease (PD) is a common, progressive neurodegenerative disease, which typically presents itself with a range of motor symptoms, like resting tremor, bradykinesia, and rigidity, but also non-motor symptoms such as fatigue, constipation, and sleep disturbance. Neuropathologically, PD is characterized by loss of dopaminergic cells in the substantia nigra pars compacta (SNpc) and Lewy bodies, neuronal inclusions containing α-synuclein (α-syn). Mutations and copy number variations of SNCA, the gene encoding α-syn, are linked to familial PD and common SNCA gene variants are associated to idiopathic PD. Large-scale genome-wide association studies have identified risk variants across another 40 loci associated to idiopathic PD. These risk variants do not, however, explain all the genetic contribution to idiopathic PD. The rat Vra1 locus has been linked to neuroprotection after nerve- and brain injury in rats. Vra1 includes the glutathione S-transferase alpha 4 (Gsta4) gene, which encodes a protein involved in clearing lipid peroxidation by-products. The DA.VRA1 congenic rat strain, carrying PVG alleles in Vra1 on a DA strain background, was recently reported to express higher levels of Gsta4 transcripts and to display partial neuroprotection of SNpc dopaminergic neurons in a 6-hydroxydopamine (6-OHDA) induced model for PD. Since α-syn expression increases the risk for PD in a dose-dependent manner, we assessed the neuroprotective effects of Vra1 in an α-syn-induced PD model. Human wild-type α-syn was overexpressed by unilateral injections of the rAAV6-α-syn vector in the SNpc of DA and DA.VRA1 congenic rats. Gsta4 gene expression levels were significantly higher in the striatum and midbrain of DA.VRA1 compared to DA rats at 3 weeks post surgery, in both the ipsilateral and contralateral sides. At 8 weeks post surgery, DA.VRA1 rats suffered significantly lower fiber loss in the striatum and lower loss of dopaminergic neurons in the SNpc compared to DA. Immunofluorescent stainings showed co-expression of Gsta4 with Gfap at 8 weeks suggesting that astrocytic expression of Gsta4 underlies Vra1-mediated neuroprotection to α-syn induced pathology. This is the second PD model in which Vra1 is linked to protection of the nigrostriatal pathway, solidifying Gsta4 as a potential therapeutic target in PD.

Reactive oxygen species from human astrocytes induced functional impairment and oxidative damage

Reactive oxygen species (ROS) have been shown to be a contributor to aging and disease. ROS also serve as a trigger switch for signaling cascades leading to corresponding cellular and molecular events. In the central nervous system (CNS), microglial cells are likely the main source of ROS production. However, activated astrocytes also appear to be capable of generating ROS. In this study we investigated ROS production in human astrocytes stimulated with interleukin (IL)-1β and interferon (IFN)-γ and its potential harmful effects. Although IFN-γ alone had no effect, it potentiated IL-1β-induced ROS production in a time-dependent manner. One of the sources of ROS in IL-1β-activated astrocytes was from increased superoxide production in mitochondria accompanied by enhanced manganese superoxide dismutase and inhibited catalase expression. NADPH oxidase (NOX) may also contribute to ROS production as astrocytes express NOX isoforms. Glutamate uptake, which represents one of the most important methods of astrocytes to prevent excitotoxicity, was down-regulated in IL-1β-activated astrocytes, and was further suppressed in the presence of IFN-γ; IFN-γ itself exerted minimal effect. Elevated levels of 8-isoprostane in IL-1β ± IFN-γ-activated human astrocytes indicate downstream lipid peroxidation. Pretreatment with diphenyleneiodonium abolished the IL-1β ± IFN-γ-induced ROS production, restored glutamate uptake function and reduced 8-isoprostane to near control levels suggesting that ROS contributes to the dysfunction of activated astrocytes. These results support the notion that dampening activated human astrocytes to maintain the redox homeostasis is vital to preserve their neuroprotective potential in the CNS.

Ketamine appears associated with better word recall than etomidate after a course of 6 electroconvulsive therapies

Ten patients treated with electroconvulsive therapy (ECT) for depressive illness received anesthesia with either etomidate or ketamine. Three patients received both etomidate and ketamine anesthesia for ECT during separate episodes of depression. Patients anesthetized with ketamine for ECT had significantly less impairment of short-term memory function than did patients who received ECT with etomidate anesthesia. All patients who received both anesthetics for ECT during 2 different episodes had less memory loss during ECT with ketamine than with etomidate. These results show the importance of studying the effects of all anesthetic agents used during ECT on cognitive functions. The results imply that the effect of ECT on memory may be largely caused by effects mediated by glutamate at N-methyl-d-aspartate receptors and suggest that N-methyl-d-aspartate antagonists may offer protection from memory dysfunction during ECT.

Temporal development of hippocampal cell death is dependent on tissue strain but not strain rate

Deformation of brain tissue in response to mechanical loading of the head is the root-cause of traumatic brain injury (TBI). Even below ultimate failure limits, deformation activates pathophysiological cascades resulting in delayed cell death. Injury response of soft tissues, such as the chest and spinal cord, is dependent on the product of deformation and velocity, a parameter termed the viscous criterion. We set out to test if hippocampal cell death could be predicted by a similar combination of strain and strain rate and if the viscous criterion was valid for hippocampus. Quantitative prediction of the brain's biological response to mechanical stimuli is difficult to achieve in animal models of TBI, so we utilized an in vitro model of TBI based on hippocampal slice cultures. We quantified the temporal development of cell death after precisely controlled deformations for 30 combinations of strain (0.05-0.50) and strain rate (0.1-50s(-1)) relevant to TBI. Loading conditions for a subset of cultures were verified by analysis of high-speed video. Cell death was found to be significantly dependent on time-post injury, on strain magnitude, and to a lesser extent, on anatomical region by a repeated-measures, three-way ANOVA. The responses of the CA1 and CA3 regions of the hippocampus were not statistically different in contrast to some in vivo TBI studies. Surprisingly, cell death was not dependent on strain rate leading us to conclude that the viscous criterion is not a valid predictor for hippocampal tissue injury. Given the large data set and extensive combinations of biomechanical parameters, predictive mathematical functions relating independent variables (strain, region, and time post-injury) to the resultant cell death were defined. These functions can be used as tolerance criteria to equip finite element models of TBI with the added capability to predict biological consequences.

Development of novel selective cell ablation in the mammary gland and brain to study cell-cell interactions and chemoprevention

We have generated transgenic mice which express the gene encoding Escherichia coli nitroreductase (NTR) specifically in the luminal epithelial cells of the mammary gland and the glial cells of the brain. The enzyme activates an antitumour drug CB 1954, to produce a cross-linking agent that kills all cells expressing the enzyme. We have shown that administration of the antitumour drug CB 1954 rapidly and selectively kills these cells. Original experiments demonstrated the ability to ablate the luminal cells in the mammary gland with no apparent bystander effect. Subsequently, astrocytes expressing nitroreductase under the targeting of the GFAP promoter were selectively ablated following administration of the prodrug CB 1954 produces a degeneration of granular neurones due to changes in glutamate levels. Recent experiments demonstrated inhibition of myc-dependent mammary tumours using the same enzyme (nitroreductase)-prodrug (CB 1954), combination. Owing to the ease of control of NTR-mediated cell ablation, we anticipate that this system will supersede herpes simplex virus type 1 thymidine kinase. There are widespread potential applications for this approach in the dissection of complex cellular interactions during development and in the adult organism. The present transgenic models also have important applications for the study in vivo of novel prodrugs that can be selected for variable degrees of bystander effects. Such studies will have particular significance for those groups advocating the use of NTR as an appropriate enzyme for gene-directed enzyme prodrug therapy by providing models of a wide range of human disease for mechanistic and therapeutic experimentation. The results clearly demonstrate that the model has potential to study chemoprevention and fundamental questions on cell-cell interactions in cell biology.

Lactate and glucose as energy substrates during, and after, oxygen deprivation in rat hippocampal acute and cultured slices

The effects of raised brain lactate levels on neuronal survival following hypoxia or ischemia is still a source of controversy among basic and clinical scientists. We have sought to address this controversy by studying the effects of glucose and lactate on neuronal survival in acute and cultured hippocampal slices. Following a 1-h hypoxic episode, neuronal survival in cultured hippocampal slices was significantly higher if glucose was present in the medium compared with lactate. However, when the energy substrate during the hypoxic period was glucose and then switched to lactate during the normoxic recovery period, the level of cell damage in the CA1 region of organotypic cultures was significantly improved from 64.3 +/- 2.1 to 74.6 +/- 2.1% compared with cultures receiving glucose during and after hypoxia. Extracellular field potentials recorded from the CA1 region of acute slices were abolished during oxygen deprivation for 20 min, but recovered almost fully to baseline levels with either glucose (82.6 +/- 10.0%) or lactate present in the reperfusion medium (108.1 +/- 8.3%). These results suggest that lactate alone cannot support neuronal survival during oxygen deprivation, but a combination of glucose followed by lactate provides for better neuroprotection than either substrate alone.

Modulation of glutamatergic transmission by bergmann glial cells in rat cerebellum in situ

We obtained patch-clamp recordings from neuron-glial cell pairs in cerebellar brain slices to examine the contribution of glutamate (Glu) uptake by Bergmann glial cells to shaping excitatory postsynaptic currents (EPSCs) at the parallel fiber to Purkinje cell synapse. We show that electrical stimulation of parallel fibers not only activates EPSCs in Purkinje cells but also activates inward currents in antigenically identified Bergmann glial cells that invest Purkinje cell synapse with their processes. The inward current is partially due to 6-cyano-7-nitroquinoxalene-2,3-dione (CNQX)- and 2-amino-5-phosphonopentanoic acid (AP5)-sensitive ionotropic Glu receptors, but >/=70% of the current was mediated by D,L-threo-beta-hydroxyaspartate (THA)-sensitive Glu transporters. Glu inward currents were completely and reversibly inhibited by depolarization of Bergmann glial cells to positive membrane potentials allowing biophysical inhibition of Glu uptake into a single glial cell. Inhibition of Glu transport into Bergmann glial cells by voltage-clamping the cell to depolarized potentials caused a reversible increase in spontaneous EPSC frequency in the Purkinje cell. This increase could also be achieved by pharmacological inhibition of Glu transport with the Glu transport inhibitor THA, suggesting that inhibition of Glu uptake into Bergmann glial cells is responsible for the modulation of postsynaptic EPSCs. THA modulation of spontaneous EPSCs could only be observed in the absence of TTX, suggesting primarily a presynaptic effect. Taken together these data suggest that glial Glu uptake can profoundly affect excitatory transmission in the cerebellum, most likely by regulating presynaptic glutamate release.

Increased expression of glutamate binding protein mRNA in rat retina after ischemia-reperfusion injury

We investigated the distribution and expression of glutamate-binding protein (GBP) in the rat retinas after ischemia-reperfusion injury. Ischemia-reperfusion injury was induced in rats by clamping of the optic nerve for one hour. The distribution of GBP immunoreactivity was determined at 6, 24, 72, and 168 hours after reperfusion. Also, RT-PCR was performed to detect the change of GBP mRNA expression in the reperfused retinas. In untreated control retinas, GBP immunoreactivity was observed in the cells of ganglion cell layer, inner plexiform layer, and inner nuclear layer. At 6, 24, and 72 hours after reperfusion, GBP immunoreactivity was seen not only in the GCL, IPL, and INL, but also in the outer plexiform layer and photoreceptor outer segment. At 168 hours after reperfusion, GBP immunoreactivity in the OPL was decreased. Moreover, we found increased GBP mRNA expression at 24 hours after reperfusion. In this study, we demonstrated that ischemia-reperfusion induced increase of GBP immunoreactivity in the inner retina and increase of GBP mRNA expression in the rat retinas. Our results suggest that NMDA receptor-like complex may play some role in the ischemic cell death of the inner retina.

Changes in astroglial GLT-1 expression after neural transplantation or stab wounds

Uncontrolled release of glutamate from damaged brain initiates events that result in excitotoxic neuronal death. Glutamate uptake by specialized astroglial transporters is essential for control of extracellular glutamate levels. Many studies have demonstrated a reduction in astrocytic GLT-1 expression after different forms of injury. Because extensive neuronal death does not occur after direct cortical stab wounds and viable developing neurons populate fetal CNS grafts, we hypothesized that reactive astroglia associated with these procedures might maintain or up-regulate GLT-1. We examined the temporal and spatial distribution of GLT-1, GFAP and nestin proteins by confocal double-label immunohistochemistry combined with a new methodology in which precise brain areas are microdissected and analyzed for protein content by immunoaffinity chromatography. In stab wounds, GLT-1 protein content did not change compared to normal cortex, as determined by direct protein measurements; GLT-1 colocalized with nestin- and GFAP(+) astroglia adjacent to the lesion. In contrast, host reactive astroglia adjacent to grafts significantly upregulated GLT-1 by 3 days postoperative. The GFAP protein analysis suggests that increased GLT-1 is not the result of greater numbers of activated astroglia around grafts, but that developing graft tissue influences adjacent host astroglia to upregulate GLT-1. GLT-1 protein within grafts was rapidly accelerated to mature levels by just three days, and was expressed by the nestin(+) cell population. These data, which demonstrate immunoexpression of GLT-1 protein combined with a new method for protein measurement in situ indicate that, in contrast to other injury models, astroglial GLT-1 is upregulated or maintained following invasive CNS procedures. (c)2002 Elsevier Science (USA).

Ionotropic and metabotropic glutamate receptor protein expression in glioneuronal tumours from patients with intractable epilepsy

Glioneuronal tumours are an increasingly recognized cause of chronic pharmaco-resistant epilepsy. In the present study the immunocytochemical expression of various glutamate receptor (GluR) subtypes was investigated in 41 gangliogliomas (GG) and 16 dysembryoplastic neuroepithelial tumours (DNT) from patients with intractable epilepsy. Immunocytochemistry with antibodies specific for ionotropic NR1, NR2A/B (NMDA) GluR1, GluR2 (AMPA), GluR5-7 (kainate), and metabotropic mGluR1, mGluR2-3, mGluR5, mGluR7a subtypes demonstrated in both GG and DNT the presence of an highly differentiated neuronal population, containing subunits from each receptor class. More than 50% of tumours contained a high percentage of neuronal cells immunolabelled for NMDA, AMPA and kainate receptor subunits. A high percentage of neurones showed strong expression of NR2A-B, which co-localized with NR1. Group I mGluRs (mGluR1 and mGluR5) were highly represented in the neuronal component of the tumours. Immunolabelling for several GluRs was also present in the glial component. Increased expression of mGluR2-3, mGluR5 and GluR5-7 was observed in reactive astrocytes in the perilesional zone compared to normal cortex. The neurochemical profile of glioneuronal tumours, with high expression of specific GluR subtypes, supports the central role of glutamatergic transmission in the mechanisms underlying the intrinsic and high epileptogenicity of these lesions.

NMDA and glutamate evoke excitotoxicity at distinct cellular locations in rat cortical neurons in vitro

The development of cortical neurons in vivo and in vitro is accompanied by alterations in NMDA receptor subunit expression and concomitant modifications in the pharmacological profile of NMDA-activated ionic currents. For example, we observed that with decreasing NR2B/NR2A subunit expression ratio, the block of NMDA receptor-mediated whole-cell responses by the NR2B-selective antagonist haloperidol was also decreased. In mature cultures (>22 d in vitro), however, NMDA responses obtained from excised nucleated macropatches, which comprised a large portion of the soma, remained strongly antagonized by haloperidol. These results suggest that in more mature neurons NR1/NR2B receptors appear to be preferentially expressed in the cell body. As predicted from the whole-cell recording pharmacological profile, NMDA-induced toxicity was largely unaffected by haloperidol in mature cultures. However, haloperidol effectively blocked glutamate toxicity in the same cultures, suggesting that the neurotoxic actions of this amino acid were mostly due to the activation of somatic NMDA receptors. In experiments in which the potency of glutamate toxicity was increased by the transport inhibitor l-trans-pyrrolidine-2,4-dicarboxylic acid, the neuroprotective effects of haloperidol were significantly diminished. This was likely because of the fact that glutamate, now toxic at much lower concentrations, was able to reach and activate dendritic receptors under these conditions. These results strongly argue that exogenous glutamate and NMDA normally induce excitotoxicity at distinct cellular locations in mature mixed neuronal cultures and that NR1/NR2B receptors remain an important component in the expression of glutamate, but not NMDA-induced excitotoxicity.

Behavioral and morphological consequences of primary astrocytes transplanted into the rat cortex immediately after nucleus basalis ibotenic lesion

Adult male rats received transplants of dissociated 30-day old cultured cortical astrocytes into the ipsilateral frontal and parietal cortex immediately after unilateral ibotenic acid lesion of the NBM or after sham injury. We hypothesized that transplants of astrocytes into the acetylcholine-deprived cortex might provide trophic support to terminals arising from damaged NBM neurons. Twenty four hours after transplantation and every other day for 11 days post surgery, the animals were tested for locomotion and habituation in an open field. NBM lesion reduced vertical movements only as compared to no lesion and no transplant counterparts. Nine days after surgery rats with NBM lesion and astrocyte-transplants into the cortex were as impaired in the acquisition of a passive avoidance (PA) task as untreated counterparts. Animals with no lesions and transplants into the cortex also had significant PA acquisition deficits. All rats with ibotenic lesion were significantly impaired on PA retention as compared to rats with no lesions. Astrocyte-transplants survived up to 2 months after cortical implantation but these transplants produced severe laminar disruption and gliosis. This effect was greater in rats with NBM lesion than in intact animals with transplants into the cortex. These data show that astrocyte-transplants do not promote functional recovery after NBM lesion and suggest an immune rejection of the astrocyte transplants by the host brain.

Inducible ablation of astrocytes shows that these cells are required for neuronal survival in the adult brain

To study the function of astrocytes in the adult brain, we have targeted the expression of E. coli nitroreductase (NTR) to the astrocytes of transgenic mice under the control of the GFAP promoter. The astrocytes expressing NTR were selectively ablated after administration of the prodrug CB1954, resulting in motor discoordination. Histological examination showed that the region most affected in the brain was the cerebellum, in which the Bergmann glia were eliminated and the granular neurons had degenerated. Specific effects were also noted on the dendrites of the Purkinje cells, and the junction between these neurons and granular layer was disrupted. Astrocyte ablation was associated with a dramatic decrease in the expression of glutamate transporters, which may account for the degeneration of granular neurons since the excitotoxic effects of glutamate result in a similar phenotype. These results provide the first evidence that astrocytes are important for the survival of neurons in the adult brain in vivo.

Astroglial trophic support and neuronal cell death: influence of cellular energy level on type of cell death induced by mitochondrial toxin in cultured rat cortical neurons

Previous in vivo and in vitro analyses have shown that both necrosis and apoptosis are involved in neuronal cell death induced by energy impairment caused by mitochondrial dysfunction. However, little is known about the key factors that determine whether the cells undergo necrosis or apoptosis. In the present study, we analyzed neuronal cell death induced by 3-nitropropionic acid (3-NP), an irreversible inhibitor of mitochondrial complex II, in a primary culture system of rat cortical neurons. The neurons were maintained for a week in coculture with astroglial cells, and then they were treated with 3-NP in the presence or absence of astroglial cells. As judged from morphological (Hoechst 33258 staining) and biochemical (DNA fragmentation and caspase activation) analyses, the cortical neurons appeared to die through an apoptotic process after 3-NP treatment in the presence of astroglial cells. However, caspase inhibitors did not suppress the 3-NP-induced cell death, suggesting the involvement of a caspase-independent pathway of 3-NP-induced neuronal cell death in the presence of astroglial cells. On the other hand, 3-NP induced necrotic cell death within 1 day in the absence of astroglial cells, following a rapid decrease in intracellular ATP level. These changes were attenuated by the presence of astroglial cells or the addition of astroglial conditioned medium. These results suggest that astroglial trophic support influences the alteration of the intracellular energy state in 3-NP-treated neurons and consequently determines the type of neuronal cell death, apoptosis or necrosis.

Upregulation of metabotropic glutamate receptor subtype mGluR3 and mGluR5 in reactive astrocytes in a rat model of mesial temporal lobe epilepsy

Reactive gliosis is a prominent morphological feature of mesial temporal lobe epilepsy. Because astrocytes express glutamate receptors, we examined changes in metabotropic glutamate receptor (mGluR) 2/3, mGluR5 and transforming growth factor (TGF)-beta in glial cells of the hippocampal regions in an experimental rat model of spontaneous seizures. Rats that exhibited behavioural status epilepticus (SE) directly after 1 h of electrical angular bundle stimulation, displayed chronic spontaneous seizures after a latent period of 1-2 weeks as observed using continuous electrographic monitoring. SE resulted in hypertrophy of astrocytes and microglia activation throughout the hippocampus as revealed by immunolabelling studies. A dramatic, seizure intensity-dependent increase in vimentin immunoreactivity (a marker for reactive astrocytes) was revealed in CA3 and hilar regions where prominent neuronal loss occurs. Increased vimentin labelling was first apparent 24 h after onset of SE and persisted up to 3 months. mGluR2/3 and mGluR5 protein expression increased markedly in glial cells of CA3 and hilus by 1 week after SE, and persisted up to 3 months after SE. Double immunolabelling of brain sections with vimentin confirmed co-localization with glial fibrillary acidic protein (GFAP), mGluR2/3 and mGluR5 in reactive astrocytes. TGF-beta, a cytokine implicated in mGluR3-mediated neuroprotection, was also upregulated during the first 3 weeks after SE throughout the hippocampus. This study demonstrates seizure-induced upregulation of two mGluR subtypes in reactive astrocytes, which - together with the increased production of TGF-beta - may represent a novel mechanism for modulation of glial function and for changes in glial-neuronal communication in the course of epileptogenesis.

Cytokine effects on glutamate uptake by human astrocytes

Glutamate uptake by astrocytes has been postulated to play a neuroprotective role during brain inflammation. Using primary human fetal astrocyte cultures, we investigated the influence of selected cytokines on glutamate uptake activity. Interleukin (IL)-1beta and tumor necrosis factor-alpha dose-dependently inhibited astrocyte glutamate uptake, whereas interferon (IFN)-gamma alone stimulated this activity. The nitric oxide synthase inhibitor, N(G)-monomethyl-L-arginine, blocked IL-1beta-mediated inhibition of glutamate uptake, suggesting involvement of nitric oxide in the effect of IL-1beta. IL-1 receptor antagonist protein totally reversed the inhibitory effect of cytokines, suggesting a critical role of IL-1beta. The anti-inflammatory cytokine IFN-beta blocked cytokine (IL-1beta plus IFN-gamma)-induced inhibition of glutamate uptake with a corresponding reduction in nitric oxide generation. Taken together, these findings suggest that proinflammatory cytokines inhibit astrocyte glutamate uptake by a mechanism involving nitric oxide, and that IFN-beta may exert a therapeutically beneficial effect by blocking cytokine-induced nitric oxide production in inflammatory diseases of the brain.

Chronic and acute compressive spinal cord lesions in dogs due to intervertebral disc herniation are associated with elevation in lumbar cerebrospinal fluid glutamate concentration

Acute injury to the central nervous system initiates a series of biochemical events that cause secondary tissue damage. The accumulation of excessive concentrations of glutamate in the extracellular space causes excitotoxic damage, and is incriminated as a mediator of this secondary tissue damage. The aim of this study was to measure the concentration of glutamate in cerebrospinal fluid (CSF) obtained from the cerebellomedullary cistern and lumbar subarachnoid space in dogs with acute and chronic compressive injuries of the cervical and thoracolumbar spinal cord, and to correlate the glutamate concentration with injury severity. The results demonstrate that focal injuries of the spinal cord do not affect the glutamate concentration in CSF taken from the cerebellomedullary cistern. However, dogs with severe, acute thoracolumbar disc herniations have two- to 10-fold increases in glutamate concentration in their lumbar CSF at intervals of >12 h after injury. Moreover, the severity of their clinical signs is directly related to the glutamate concentration. Dogs with chronic compressive thoracolumbar lesions have a two-fold elevation of CSF glutamate concentration, suggesting that excitotoxicity may also be a component of chronic spinal cord compression.

Cerebral cortical neuron apoptosis after mild excitotoxic injury in vitro: different roles of mesencephalic and cortical astrocytes

OBJECTIVE

Increasing evidence supports the presence of neuronal apoptosis after ischemic or excitotoxic brain injury. Astrocytes, which exhibit significant regional differences in function, may exert a protective effect on neurons exposed to ischemic injury. We examined the effects of astrocytes derived from different regions of the central nervous system on neuronal apoptosis after mild excitotoxic injury in tissue culture.

METHODS

Purified astrocyte cultures derived from P4 rat cerebral cortex or mesencephalon showed transient cell swelling but no cell death when exposed to 50 micromol/L glutamate for 5 minutes. When mixed neuronal/glial cocultures were exposed to the same glutamate dose, neuron death was observed. Necrotic and apoptotic cell death during 24 hours was examined using morphological criteria, nuclear staining, triphosphate nick end labeling, and trypan blue exclusion.

RESULTS

We found that cortical neurons that elaborate a more extensive dendritic arbor when grown on homotypic astrocytes are more likely to undergo apoptosis than neurons with a limited dendritic arbor grown on heterotypic astrocytes. By contrast, a similar number of neurons undergo necrotic cell death.

CONCLUSION

This finding may be associated with 1) increased vulnerability of neurons with a more elaborate dendrite structure to mild excitotoxic injury, or 2) regional differences in the ability of astrocytes to attenuate apoptosis.

Astroglia inhibit the proliferation of neocortical cells and prevent the generation of small GABAergic neurons in vitro

We quantitatively studied the dynamics of rat neocortical precursor proliferation in vitro, and additionally examined the effects of neuron-glia interactions on the proliferation and differentiation of neurons, and particularly of gamma-aminobutyric acid (GABA)-containing cells. In cultures grown on glia-free substrate, cellular proliferation was detected at least until the end of the second week in vitro, but most neurons which expressed detectable amounts of microtubule-associated protein at 12 days in vitro were generated early during the first week. Further double-labelling experiments, combining 5'-bromo-2'-deoxyuridine with GABA or beta-tubulin III immunohistochemistry, provided direct evidence that neuronal proliferation continued through the second week in vitro, and that a population of small GABAergic neurons was generated between 3 and 12 days in vitro. Culturing cells on a glial substrate significantly reduced the generation of small GABAergic cells and strongly inhibited the total cell proliferation. Inhibition also occurred if astrocytes were added to the culture after 6 days in vitro, but was significantly decreased if cells were grown on a fixed glial substrate, suggesting that the effect might be at least partially mediated by active interactions between neurons and glia. In conclusion, our results show that the sustained proliferation of precursor cells in neocortical cultures is necessary for the differentiation of small GABAergic neurons, and that mature astroglia effectively inhibit the proliferation of neocortical precursors thereby affecting the appearance of a population of GABAergic cells.

Pathogenesis of bacterial meningitis

Bacterial meningitis is fatal in 5% to 40% of patients and causes neurologic sequelae in up to 30% of survivors. Much has been learned recently about the mechanisms that lead to brain injury during meningitis. Once bacteria have gained access to the central nervous system, their multiplication triggers a complex host response consisting of humoral and cellular immune mediators, reactive oxygen intermediates, matrix-metalloproteinases, and other host-derived factors. Alterations of the cerebral vasculature, with disruption of the blood brain barrier and global and focal ischemia, ultimately lead to functional and structural brain damage. This article reviews current concepts of the pathophysiology of bacterial meningitis and emphasizes possible therapeutic strategies to prevent its harmful consequences.

Ability of retinal Müller glial cells to protect neurons against excitotoxicity in vitro depends upon maturation and neuron-glial interactions

Glutamate is the most abundant excitatory amino acid in the central nervous system. It has also been described as a potent toxin when present in high concentrations because excessive stimulation of its receptors leads to neuronal death. Glial influence on neuronal survival has already been shown in the central nervous system, but the mechanisms underlying glial neuroprotection are only partly known. When cells isolated from newborn rat retina were maintained in culture as enriched neuronal populations, 80% of the cells were destroyed by application of excitotoxic concentrations of glutamate. Massive neuronal death was also observed in newborn retinal cultures containing large numbers of glia, or when neurons were seeded onto feeder layers of purified cells prepared from immature (postnatal 8 day) rat retina. When newborn retinal neurons were seeded onto feeder layers of purified glial cells prepared from adult retinas, application of excitotoxic amino acids no longer led to neuronal death. Furthermore, neuronal death was not observed in mixed neuron/glial cultures prepared from adult retina. However, in all cases (newborn and adult) application of kainate led to amacrine cell-specific death. Activity of glutamine synthetase, a key glial enzyme involved in glutamate detoxification, was assayed in these cultures in the presence or absence of exogenous glutamate. Whereas pure glial cultures alone (from young or adult retina) showed low activity that was not stimulated by glutamate addition, mixed or co-cultured neurons and adult glia exhibited up to threefold higher levels of activity following glutamate treatment. These data indicate that two conditions must be satisfied to observe glial neuroprotection: maturation of glutamine synthetase expression, and neuron-glial signalling through glutamate-elicited responses.

Impairment of excitatory amino acid transport in astroglial cells infected with the human immunodeficiency virus type 1

Perturbation of astrocyte functions by HIV-1 infection may contribute to the pathogenesis of AIDS dementia complex (ADC). The present study investigated the possibility that astroglial transport of glutamate and aspartate, the major excitatory amino acids (EAAs) in the mammalian central nervous system (CNS), is altered by HIV-1 infection. Human U251 glioma cells were infected with the brain isolate SF162 of HIV-1. HIV-1 persisted in glial cells over several months. This nonproductive infection of glial cells was characterized by persistent expression of Nef over the time of the infection, and the transient presence of structural viral proteins, including the viral transmembrane glycoprotein gp41, which was detected during the initial 2 weeks following HIV-1 infection. The presence of gp41 in acutely HIV-1-infected glial cells coincided with a 36% decrease in D-[3H]aspartate uptake, owing to a reduction in the maximal transport capacity (vmax) for D-aspartate. The expression of typical astrocytic glutamate transporters EAAT1 and EAAT2 in U251 glioma cells was not altered by HIV-1 infection. To determine whether viral protein gp120, gp41, or Nef was involved in the impairment of EAA transport in acutely HIV-1-infected glial cells, effects of lentiviral lytic peptide type 1 (LLP-1) (corresponding to the carboxy terminus of gp41), recombinant SF2 gp120, and recombinant LAI Nef on D-[3H]aspartate uptake and the release of glutamate in glial cells were investigated. Only LLP-1 reduced D-[3H]aspartate uptake and facilitated the release of glutamate from glial cells in a concentration-dependent manner. These results suggest that the carboxy terminus of gp41 impairs EAA transport in glial cells, which may contribute to excitotoxic damage to neurons in HIV-1 infection of the CNS.

Glial cells potentiate kainate-induced neuronal death in a motoneuron-enriched spinal coculture system

AMPA/kainate receptor-mediated excitotoxicity is believed to play a pathogenic role in amyotrophic lateral sclerosis. To further characterize the mechanisms involved in AMPA/kainate receptor-mediated motoneuron injury, we investigated the influence of spinal glial cells on kainate-induced motoneuron death in vitro. A motoneuron-enriched neuronal population was obtained from embryonic mouse spinal cord by metrizamide density centrifugation. This population was cultured either on a pre-established glial feeder layer of ventral spinal origin (coculture) or in glia-free conditions (monoculture). Glial feeder layers significantly enhanced basal survival of neurons, and supported neuronal differentiation as judged by neuronal morphology and expression of the motoneuron markers peripherin and SMI-32. Neuronal vulnerability to kainate was two- to three-fold higher in coculture than in monoculture, and increased significantly with time in coculture. The effects of glial feeder layers on neuronal basal survival, differentiation and kainate vulnerability were not mimicked by conditioned medium from glial cells. The increase in neuronal kainate vulnerability with time in coculture was associated with a marked rise in the proportion of cocultured neurons possessing Ca2+-permeable AMPA/kainate receptors, as determined by kainate-activated Co2+-uptake. Neurons in monoculture were unstained by kainate-activated Co2+-uptake. Neurons were immunoreactive to specific antibodies against the AMPA receptor subunits GluR1 and GluR2 both in monoculture and coculture. This study indicates that motoneuron differentiation in coculture is associated with increased vulnerability to kainate and increased expression of Ca2+-permeable AMPA/kainate receptors. In this paradigm glial cells support basal survival and differentiation of neurons, but potentiate kainate-induced neuronal death.

Effects of microgravity and bone morphogenetic protein II on GFAP in rat brain

This study evaluated effects of bone morphogenetic protein II (BMP) on glial fibrillary acidic protein (GFAP) in the brain of female Fischer 344 rats during 14 days of spaceflight. GFAP mRNA decreased in vehicle-implanted rats flown on the space shuttle by 53 and 48% in the stratum moleculare and stratum lacunosum moleculare hippocampal subregions, respectively. GFAP mRNA was not significantly affected by BMP implantation during spaceflight. Rats returning from space exhibited a 56% increase in serum corticosterone. BMP treatment did not additively increase corticosterone elevations in microgravity but appeared to increase serum corticosterone and reduce GFAP mRNA in the stratum moleculare in control rats. These data suggest that exposure to microgravity reduces GFAP expression in hippocampal astrocytes.

Soluble macrophage factors trigger apoptosis in cultured hippocampal neurons

It is not clear whether macrophages which can phagocytose dead cells, may also contribute to death of potentially viable neurons when they enter brain lesion sites after insult. We have initially examined the effects of macrophage-conditioned medium on the integrity of hippocampal neurons in culture. We assessed qualitative and quantitative changes in neuronal status in terms of nuclear morphology, internucleosomal cleavage, cell membrane integrity and process density. Cell morphology with manual counts to quantitate findings showed that macrophage conditioned medium significantly increased the percentage of neurons with abnormal nuclei. Aurintricarboxylic acid attenuated this effect. Demonstration of laddering of DNA on agarose gels suggested an apoptosis-like event. A commercially available kit used to detect high concentrations of 3'-OH DNA ends showed marked increase in labelled cells. These combined findings confirmed that apoptosis was the main event triggered by conditioned medium. Although the number of cells with incompetent membranes also increased with conditioned medium application the majority of cells with apoptotic nuclei maintained membrane integrity. Conditioned medium also resulted in significant loss of cell processes. Conditioned medium from stimulated microglia showed a similar pattern of injury. The response of stressed neurons to conditioned medium was also tested. Exposure of cultures to mild hypoxia resulted in injury but did not significantly alter their subsequent vulnerability to macrophage-conditioned medium. Early experiments suggest that the documented changes in neuronal status are caused by relatively large and stable secreted macrophage proteins.

Hypoxia-ischemia causes abnormalities in glutamate transporters and death of astroglia and neurons in newborn striatum

The neonatal striatum degenerates after hypoxia-ischemia (H-I). We tested the hypothesis that damage to astrocytes and loss of glutamate transporters accompany striatal neurodegeneration after H-I. Newborn piglets were subjected to 30 minutes of hypoxia (arterial O2 saturation, 30%) and then 7 minutes of airway occlusion (O2 saturation, 5%), producing cardiac arrest, followed by cardiopulmonary resuscitation. Piglets recovered for 24, 48, or 96 hours. At 24 hours, 66% of putaminal neurons were injured, without differing significantly thereafter, but neuronal densities were reduced progressively (21-44%). By DNA nick-end labeling, the number of dying putaminal cells per square millimeter was increased maximally at 24 to 48 hours. Glial fibrillary acidic protein-positive cell body densities were reduced 48 to 55% at 24 to 48 hours but then recovered by 96 hours. Early postischemia, subsets of astrocytes had fragmented DNA; later postischemia, subsets of astrocytes proliferated. By immunocytochemistry, glutamate transporter 1 (GLT1) was lost after ischemia in the astroglial compartment but gained in cells appearing as neurons, whereas neuronal excitatory amino acid carrier 1 (EAAC1) dissipated. By immunoblotting, GLT1 and EAAC1 levels were 85% and 45% of control, respectively, at 24 hours of recovery. Thus, astroglial and neuronal injury occurs rapidly in H-I newborn striatum, with early gliodegeneration and glutamate transporter abnormalities possibly contributing to neurodegeneration.

The pharmacology of amino-acid responses in septal neurons

Septal cholinergic neurons are known to play an important role in cognitive processes including learning and memory through afferent innervation of the hippocampal formation and cerebral cortex. The septum contains not only cholinergic neurons but also various types of neurons including GABA (gamma-aminobutyric acid)-ergic neurons. Although synaptic transmission in the septum is mediated primarily by the activation of excitatory and inhibitory amino-acid receptors, it is possible that a distinct phenotype of neuron is endowed with a different type for each of the amino-acid receptors and thus they play different roles from each other, since it has been demonstrated within the septum that there is a regional distribution of various types of amino-acid receptor subunits, their expression as different combinations within a specific cell may produce receptor channels with disparate functional properties. As a first step towards knowing the various functions of septal cholinergic neurons, we characterized the functional properties of glutamate, GABA (type A; GABAA) and glycine receptor channels on cultured rat septal neurons which were histologically identified to be cholinergic. These were similar to those of receptor channels on other types of neurons, except for the actions of some neuromodulators. The septal N-methyl-D-aspartate receptor channel was distinct in being less sensitive to Mg2+ and in a voltage-dependent action of Zn2+. The septal GABAA receptor channel exhibited a lanthanide site whose activation resulted in a positive allosteric interaction with a binding site of pentobarbital. The septal glycine receptor channel was only positively modulated by Zn2+; this action of Zn2+ was not accompanied by an inhibitory effect. Our data suggest that the amino-acid receptors on septal cholinergic neurons may play a distinct role compared to other types of neurons; this difference depends on the actions of neuromodulators and metal cations. It would be interesting to compare these effects recorded in tissue culture to those observed with septal cholinergic neurons in slice preparations.

Neuroprotection by both NMDA and non-NMDA receptor antagonists in in vitro ischemia

We have investigated the relative contributions of oxygen and glucose deprivation to ischaemic neurodegeneration in organotypic hippocampal slice cultures. Cultures prepared from 10-day-old rats were maintained in vitro for 14 days and then deprived of either oxygen (hypoxia), glucose (hypoglycaemia), or both oxygen and glucose (ischaemia). Hypoxia alone induced degeneration selectively in CA1 pyramidal cells and this was greatly potentiated if glucose was removed from the medium. We have also characterised the effects of both pre- and post-treatment using glutamate receptor antagonists and the sodium channel blocker tetrodotoxin (TTX). Neuronal death following either hypoxia or ischaemia was prevented by pre-incubation with CNQX, MK-801 or tetrodotoxin. MK-801 or CNQX also prevented death induced by either hypoxia or ischaemia if added immediately post-insult, however, post-insult addition of TTX prevented hypoxic but not ischaemic damage. Organotypic hippocampal slice cultures are sensitive to both NMDA and non-NMDA glutamate receptor blockade and thus represent a useful in vitro system for the study of ischaemic neurodegeneration paralleling results reported using in vivo models of ischaemia.

Long-term changes in glial fibrillary acidic protein and glutamine synthetase immunoreactivities in the supraoptic nucleus of portacaval shunted rats

The present study was undertaken to ascertain whether, and to what extent, glial fibrillary acidic protein (GFAP) and glutamine synthetase (GS) expressions in the supraoptic nucleus (SON) could be modulated after one month and six months of portacaval shunting (PCS) in rats. GFAP and GS immunoreactivities were significantly higher in PCS rats than in control rats at one and six months. The increased GFAP and GS immunoreactivities observed in the SON astrocytes were directly related to the duration of PCS. In PCS rats, the number and length of both GFAP and GS immunopositive astroglial processes increased not only in the hypothalamic nucleus but in the perinuclear zone, where glutamatergic pathways have been described, whereas GFAP and GS expressions decreased in the ventral glial lamina. Since GS is one of the glutamate metabolizing enzymes and the SON is one of the areas of glutamatergic activity, our results show that astrocytes respond differentially to glutamate toxicity. This suggests that overexpression of GFAP and GS immunoreactivities could be associated with glutamatergic neurotransmission disorders.

Glutamine synthetase and glutamate metabolism in the guinea pig cochlea

Glutamate is thought to act as a neurotransmitter of the sensory hair cells of the organ of Corti. Glutamine synthetase could be involved in a type of glutamate-glutamine cycle in the cochlea which could clear glutamate off the synaptic cleft and replenish the hair cell glutamate neurotransmitter store. Using both light and electron microscopic immunocytochemistry to localize this enzyme in the guinea pig cochlea, we have observed immunoreactive satellite glial cells surrounding parvalbumin-immunoreactive primary auditory neurons in the spiral ganglion. Glutamine synthetase was also detected in Schwann cells of the osseous spiral lamina which form the myelin sheath of nerve fibers. On the contrary, no immunoreactivity could be observed in the cochlear nerve and in the organ of Corti, although this organ contains structures able to take up glutamate. Although they confirm earlier works involving glutamine synthetase in the conversion of L-[3H]glutamate taken up by glial cells, our results suggest that the cochlear glutamate-glutamine cycle is not primarily involved in the recycling and replenishment of hair cell neurotransmitter glutamate. Alternatively, it is proposed that glutamine synthetase functions to limit the perilymphatic glutamate concentrations.

Reconstruction of the glial environment of a photochemically induced lesion in the rat spinal cord by transplantation of mixed glial cells

It is becoming increasingly apparent that the astrocytic environment is critical to the normal development and functioning of the CNS, and that acute injury to the spinal cord causes destruction of glial cells in addition to neurones and axons. The aims of this study were to assess the viability of reconstructing the astrocytic environment of a cystic spinal cord lesion by transplantation of glial cells and to examine the effect of the transplanted cells on meningeal cell invasion and revascularisation of the lesion and on axonal regeneration. Neonatal rat and kitten mixed glial cells and the CG-4 rat O-2A progenitor cell line were transplanted into a lesion produced in the dorsal funiculus of the rat spinal cord by photochemical infarction. The animals were killed 4 weeks after injury, their cords examined with light and electron microscopy and compared with control animals that were injected with medium alone. Transplantation of all three preparations resulted in increased numbers of astrocytes in the area of Wallerian degeneration cranial to the lesion and within the cyst. Mixed glial cell cultures prepared from neonatal rat forebrain contained cells with in vitro characteristics of type-1 astrocytes, and produced dense clusters of astrocytes that were surrounded by meningeal cells, resulting in a fragmented environment in the cyst. In contrast, glial cell cultures prepared from kitten forebrain and the CG-4 cell line produced cells that filled the cyst with a loose network of fine processes and reduced meningeal cell infiltration of the lesion. The CG-4 cell line significantly increased the density of blood vessels in the centre of the lesion and the number of spared axons present dorsal to the lesion, but none of the preparations significantly increased the number of axons regenerating at the caudal end of the lesion. We conclude that O-2A progenitor-derived astrocytes are more suitable for reconstruction of the glial environment of a cystic lesion in the rat spinal cord than 'type-1 like' astrocytes and would therefore be the cell of choice to engineer to produce factors that promote axonal regeneration.

Disturbances in the distribution of neurotransmitters in the rat retina after ischemia

PURPOSE

Disturbances in neurotransmitter distribution have been observed in cerebral ischemia in the pathophysiologic process of excitotoxicity. The goal of this study was to examine the effect of pressure-induced retinal ischemia on the distribution of the retinal neurotransmitters glutamate and gamma-aminobutyric acid (GABA) within the rat retina.

METHODS

Animals were subjected to increased intraocular pressure of 110 mm Hg for 45 min using an intracameral hydrostatic pressure device. The distribution of glutamate and GABA immunoreactivity (IR) was determined at 0, 2, 4, 8 and 24 hrs after reperfusion by immunogold with silver intensification.

RESULTS

Three phases of neurotransmitter immunoreactivity patterns were discernible following retinal ischemia. Immediately following reperfusion (Phase I), a shift of GABA-IR from inner retinal neurons to the Mueller cells and their processes was noted. In contrast, despite marked decreases in neuronal glutamate-IR, a less pronounced shift of glutamate-IR to the Muller cells was simultaneously noted. This shift of neurotransmitter IR to the Mueller cells was transient with the gradual reappearance of IR within the inner retinal neurons noted 2-8 hrs after reperfusion (Phase II). Phase III began at 8 hrs after reperfusion with progressive loss of GABA-IR noted in the inner retina; by 24 hrs, secondary loss of inner retinal glutamate-IR was evident with corresponding dropout and pyknosis of inner retinal neurons apparent.

CONCLUSIONS

The distribution of glutamate-IR and GABA-IR was significantly altered following retinal ischemia. The alteration noted in Phase I suggested that the regulation of glutamate by Mueller cells was disrupted by this ischemic insult leading to glutamate excitotoxicity, and delayed neuronal cell degeneration as evidenced by the subsequent loss of inner retinal immunoreactivity in Phase III.

Neurotoxic and neurotrophic effects of chronic N-methyl-D-aspartate exposure upon mesencephalic dopaminergic neurons in organotypic culture

Current theories regarding the mechanisms of degeneration of dopaminergic nigrostriatal neurons in Parkinson's disease suggest a pivotal role for excitotoxicity. In this study, the effects of chronic exposure of rat ventral mesencephalic slice cultures to the excititoxin N-methyl-D-aspartate, were investigated. Chronic (18 day) exposure to N-methyl-D-aspartate produced widely varying, dose-dependent effects. High doses (100 mu M) caused a pronounced toxicity upon tyrosine hydroxylase-positive neurons, with the surviving neurons possessing shrunken somata and stunted neurites: co-administration of the N-methyl-D-aspartate receptor antagonist MK-801, inhibited N-methyl-D-aspartate-induced toxicity. In contrast, exposure to a low concentration of N-methyl-D-aspartate (0.1 mu M), stimulated the outgrowth of tyrosine hydroxydase-positive neurites from the culture; this effect was abolished by MK-801. Chronic application of glutamate had similar, though not as pronounced, growth-promoting actions. However, the concentration of glutamate required was 1000 times that of N-methyl-D-aspartate, due to the presence ot high-affinity glutamate transport mechanisms. Cultures exposed to a submicromolar concentration of N-methyl-D-aspartate exhibited a significant resistance to subsequent exposure to a lethal (300 mu M) concentration of the toxin. It would thus appear that N-methyl-D-aspartate may have both trophic and toxic actions upon dopaminergic neurons in culture. Moreover, the ability of low doses of N-methyl-D-aspartate to protect neurons in this critical brain region may be of relevance to future attempts to arrest the degeneration associated with Parkinson's disease. The putative mechanisms of these phenomena are discussed.

Serum affects the characteristics of excitatory amino acid-binding sites on Müller cells

The binding of [3H]L-aspartate to membranes obtained from primary cultures of chick retinal Müller cells (glia) was studied Cells seeded in low-serum-containing medium (1%) and maintained in this condition showed an increased number of binding site from 1 to 5 days in vitro (DIV), when compared with controls cultured in medium containing 10% serum; these changes were not reversed by the addition of 10% serum after 48 h in vitro. Increased binding at this age was due to the expression of a low affinity binding system, competitively inhibited by the glutamate uptake blocker L-aspartate-beta-hydroxamate, suggesting that high serum might inhibit the expression of uptake sites at precise maturation stages. Experiments showed the effect was due to a thermolabile serum component. The increase in binding sites is parallel in time to both an increase in aspartate uptake and the initiation of synaptogenesis in the whole retina. Our results suggest that the presence of serum at defined stages in retinal development, could result in the elevation of extracellular glutamate and the concomitant excitotoxic death of neuronal cells, due to a decreased glutamate uptake by glial cells.

Traumatic neuroprotection with inhibitors of nitric oxide and ADP-ribosylation

N-Methyl-D-aspartate (NMDA) receptor activation is known to contribute to neuronal damage from head trauma. Additionally, NMDA neurotoxicity occurs in part through the generation of nitric oxide (NO), and injury from NO has been shown to be mediated by ADP-ribosylation. Therefore, we investigated whether inhibitors of NO and ADP-ribosylation would protect against acute CA1 traumatic neuronal injury in hippocampal slices subjected to fluid percussion. Treatment with the nitric oxide synthase (NOS) inhibitor, methyl-L-arginine 170 microM for 35 min after trauma injury, improved CA1 antidromic population spike (PS) recovery to 91 +/- 2%, compared to unmediated slices which recovered to only a mean of 20 +/- 4%, 90 min after trauma. Similarly, hemoglobin 50 microM, which directly binds NO, protected against traumatic neuronal injury and yielded a mean CA1 PS recovery of 92 +/- 1%. Treatment with inhibitors of poly-ADP-ribosylation was also strongly protective, with the vitamin nicotinamide 10 mM and 3-aminobenzamide 1 mM yielding PS recoveries of 98 +/- 2% and 90 +/- 3%, respectively. Protection was also seen with inhibitors of mono-ADP-ribosylation, including novobiocin 500 microM and meta-iodobenzylguanidine 20 microM which yielded recoveries of 89 +/- 6% and 96 +/- 26%. Novobiocin also protected against direct application of NO and NMDA. These findings suggest that NO and ADP-ribosylation are mediators of acute traumatic neuronal injury.

Infusion of glial maturation factor-beta reduces behavioral deficits after caudate nucleus injury in rats

Adult rats with bilateral thermal lesions of the caudate nuclei (CN) show severe learning and memory deficits. The present study was designed to test the effects of an astroglial stimulating growth factor in this behavioral model. Immediately after receiving lesions of the CN, experimental subjects received an injection of one of three doses of glial maturation factor-beta (GMF-beta) directly in the lesion site. All subjects were then tested for twenty days on an active avoidance spatial alternation task. The behavioral recovery of the three groups of experimental animals was compared to that of animals having received the same brain damage and administration of a control substance (lesion controls), and to that of animals receiving a sham operation and no treatment (shams). The beneficial effects of administration were evident in the group of experimental animals receiving the lowest dose of GMF-beta. The performance of animals in this group was indistinguishable from that of the shams, and was significantly better than that of the lesion controls. The results suggest a behavioral role of GMF-beta which, in an in vitro system, is known to be a growth regulator of astroglial cells.

Felbamate neuroprotection against CA1 traumatic neuronal injury

Traumatic brain injury is a leading cause of disability and death. Since the anticonvulsant felbamate provides hypoxic neuroprotection, we investigated whether felbamate would provide protection against traumatic neuronal injury as well. Traumatic injury to CA1 neurons in hippocampal slices was induced by fluid percussion, and CA1 evoked response was monitored. Pre-treatment with felbamate was strongly protective against neuronal injury, and permitted CA1 antidromic population spike recovery to a mean 94 +/- 1% (S.E.M.) of initial amplitude, compared to unmedicated slices which regained only 15 +/- 6%. The felbamate EC50 for this protection was 136 mg/1, and significant protection was found at felbamate concentrations similar to those reported in felbamate monotherapy for seizures. Significant protection was also detected when felbamate was initiated 15 min after trauma. Slices given brief post-trauma felbamate treatment could demonstrate long-term potentiation when assessed 8 h after trauma. These studies indicate that felbamate is neuroprotective against CA1 traumatic neuronal injury.

Excitatory amino acid-induced currents in rat septal cholinergic neurons in culture

Whole-cell voltage-clamp recordings were used to study excitatory amino acid-induced currents in neurons isolated from the septum of fetal rat brains. The neurons were cultured for more than four weeks on a feeder layer composed of glial cells obtained from the septal region. Septal neurons were either fusiform, triangular or multipolar and 83% of cells showed acetylcholinesterase activity. L-Glutamate, kainate, quisqualate and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) applied by local perfusion produced inward currents (Iglu, Ikai, Iquis and IAMPA, respectively) at -44mV which increased in amplitude with increasing concentration of agonist; they desensitized when induced at higher concentrations except for the Ikai. The EC50s for the peak Ikai and sustained Iglu, Iquis and IAMPA were 55, 13, 0.39 and 3.5 microM, respectively. 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX) depressed Ikai and IAMPA evoked at a concentration of 10 microM (IC50s: 0.58 and 0.84 microM, respectively). Schild analysis for the CNQX action on Ikai gave a dissociation constant of 0.27 microM for CNQX. n-Methyl-D-aspartate (NMDA) (with glycine, 3 microM) produced an inward current (INMDA) at -44 mV whose peak amplitude enhanced with increased concentrations (EC50 = 32 microM). INMDA was potentiated by glycine (EC50 = 0.15 microM) and inhibited by D-2-amino-5-phosphovalerate (IC50 = 9.9 microM for INMDA evoked at a concentration of 50 microM). MK-801 (0.1-10 microM) inhibited INMDA in a dose- and use-dependent manner. INMDA was (0.1-10 microM) inhibited INMDA in a dose- and use-dependent manner. INMDA was potentiated by spermine (EC50 = 247 microM; 91% increase at 1mM) in a manner independent of holding potential (VH). INMDA was inhibited by Mg2+ and Zn2+ (IC50 = 673 and 39 microM, respectively, at -44 mV) in a manner dependent on VH; the magnitudes of a depolarization required for an e-fold increase in their IC50s in a range of -64 to -24 mV were 16 and 22 mV, respectively. The action of Zn2+ was independent of VH > -24 mV. Current-voltage relations for Ikai, Iquis and IAMPA exhibited outward rectification, while that of INMDA showed a region of negative conductance at VH < -30 mV, which disappeared in a Mg(2+)-free solution. Reversal potentials for Ikai, Iquis, IAMPA and INMDA were close to 0 mV, indicating the involvement of non-specific cation channels. Increasing extracellular Ca2+ concentration from 2.4 to 30 mM did not affect the Ikai and Iquis, reversal potential showing negligible Ca2+ component, but shifted INMDA reversal potential to a more positive potential, yielding a ratio of Ca2+ permeability to that of monovalent cation to be 13. Cholinergic septal neurons in culture express non-NMDA-(AMPA/kainate-) and NMDA-type of glutamate receptor channels. Their properties were quantitatively similar to those of glutamate receptor channels on other types of neurons in the brain except for the actions of endogenous neuromodulators (Mg2+, Zn2+ and spermine) on NMDA receptor channels. It is suggested that NMDA receptor channels on different types of neurons may play a distinct role depending on a difference in the actions of these neuromodulators.

Glutamate as a hippocampal neuron survival factor: an inherited defect in the trisomy 16 mouse

The survival of cultured mouse hippocampal neurons was found to be greatly enhanced by micromolar concentrations of the excitatory neurotransmitter glutamate. Blockade of kainate/AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) glutamate receptors increased the rate of neuron death, suggesting that endogenous glutamate in the cultures promotes survival. Addition of glutamate (0.5-1 microM) further increased neuron survival, whereas glutamate in excess of 20 microM resulted in increased death. Thus, the survival vs. glutamate dose-response relation is bell-shaped with an optimal glutamate concentration near 1 microM. We found that hippocampal neurons from mice with the genetic defect trisomy 16 (Ts16) died 2-3 times faster than normal (euploid) neurons. Moreover, glutamate, at all concentrations tested, failed to increase survival of Ts16 neurons. In contrast, the neurotrophic polypeptide basic fibroblast growth factor did increase the survival of Ts16 and euploid neurons. Ts16 is a naturally occurring mouse genetic abnormality, the human analog of which (Down syndrome) leads to altered brain development and Alzheimer disease. These results demonstrate that the Ts16 genotype confers a defect in the glutamate-mediated survival response of hippocampal neurons and that this defect can contribute to their accelerated death.

Pretreatment with antisense oligodeoxynucleotides directed against the NMDA-R1 receptor enhances survival and behavioral recovery following traumatic brain injury in rats

Treatment with N-methyl-D-aspartate (NMDA) receptor antagonists limits tissue damage following CNS ischemia or trauma, supporting the hypothesis that NMDA receptors participate in the pathophysiology of such injuries. An alternative approach for evaluating this hypothesis is to examine the effects of selective inhibition of NMDA receptor synthesis, using antisense oligodeoxynucleotides. In the present studies, the effects of antisense oligodeoxynucleotides directed at NMDA-R1 receptor subunit, administered intracerebroventricularly (i.c.v.) prior to injury, were evaluated in a well-defined traumatic brain injury model in rats. Outcome measures included survival, motor recovery, and histological changes. Administration of antisense oligodeoxynucleotides (15 nmol/ml twice daily x 2 days) did not alter physiological variables or motor function prior to trauma. However, such treatment significantly decreased mortality and improved behavioral recovery at 2 weeks after trauma as compared to animals treated with the corresponding sense oligodeoxynucleotides. Although cell counts in hippocampus did not differ between treatment groups, astrocyte activation as reflected by glial fibrillary astrocytic protein (GFAP) immunocytochemistry was significantly reduced in antisense treated animals. These findings provide additional evidence that NMDA receptors contribute to secondary injury after brain trauma and may suggest an alternative treatment approach.

Roles of metabotropic glutamate receptors in brain plasticity and pathology

In summary, the mGluRs are a large family of receptor subtypes with diverse properties in terms of transduction coupling, pharmacology, and anatomical distribution. Many divergent studies have demonstrated that activation of these receptors can result in either neuroprotection or neuropathology. We hypothesized that the mGluRs of astrocytes may have a role in determining the response following administration of mGluR agonists in vivo, and we have defined a suitable in vitro model for the study of these receptors. The experimental plasticity demonstrated in the astrocyte culture model may represent a more general principle that conditions in the microenvironment may differentially alter mGluR subtype expression as part of development, functional specialization, or pathology. This astrocyte model of receptor regulation provides a system suitable for studying the effects of specific growth factors, neurotrophins, cytokines, and other substances released by neurons and glia that may act in both autocrine and paracrine fashions. Alteration in the ratios of receptors by such variables could then modify future signaling properties and neuroglial interactions, a form of conditioning of the astrocytic response that would alter the physiological output following glutamate release. One measure of the value of this model will be its usefulness in stimulating the generation of hypotheses that can be tested in vivo. For example, the morphology of the astrocytes when cultured in the defined medium has similarities to the morphology of astrocytes undergoing reactive gliosis in pathological states. It is also interesting to note that treatments that have been reported to increase excitatory amino acid-stimulated PI hydrolysis in ex vivo brain slices (lesions, ischemia, and kindling) are accompanied by reactive gliosis. Those findings combined with the present in vitro results lead us to speculate that mGluR5 expression may also be altered in vivo during reactive gliosis. If so, it will be important to examine the functional consequences of such a change with regard to the astrocytic response to injury and maintaining the balance between excitatory transmission and excitotoxicity.

Quantitative measurement of neuronal degeneration in organotypic hippocampal cultures after combined oxygen/glucose deprivation

Organotypic hippocampal cultures were used to study cell degeneration during the recovery period after defined periods (30 and 60 min) of combined oxygen/glucose deprivation mimicking transient ischemic conditions. Staining with the fluorescent dye propidium iodide allowed detection of damaged cells. Fluorescence intensity was measured by an image analysis system and used to quantify cell damage at different time points during the recovery period (up to 22 h). At 30 min of oxygen/glucose deprivation cells in the CA1 area were relatively more sensitive compared to CA3 and dentate gyrus cells, with respect to the time course of degeneration and the percentage of affected cells. Expanding the oxygen/glucose deprivation period from 30 to 60 min drastically increased the percentage of cells dying in all hippocampal areas. Still, however, cells in CA1 degenerated faster compared to those in the CA3 area and dentate gyrus. A histological analysis of toluidine blue as well as MAP2-immunostained sections revealed that almost all neurons degenerated in all hippocampal areas following the 60-min deprivation period, whereas GFAP-stained astrocytes appeared to be unaffected. Therefore, neuronal degeneration could be quantified by taking the fluorescence intensity values 22 h after 60 min of oxygen/glucose deprivation as 100% neuronal damage. The possibility to quantify neuronal damage in organotypic cultures offers a useful tool for detailed studies on mechanisms of neuronal cell death in a cell culture system which is closer to in situ conditions than monolayer cell cultures.

Delayed neuronal injury induced by sub-lethal NMDA exposure in the hippocampal slice

Stroke produces neuronal death by two general processes which differ in their temporal course. Acute neuronal death occurs within minutes, while delayed neuronal death evolves within 24 h. To better examine mechanisms of delayed death, we developed a new in vitro model of delayed neuronal injury using extended electrophysiological recordings in paired hippocampal slices. We exposed one hippocampal slice of each pair to 10 microM N-methyl-D-aspartate (NMDA) until the orthodromic CA1 PS disappeared. Thereafter, NMDA-treated slices regained near full recovery of PS amplitude within one hour. However, 10 h later, NMDA-treated slices demonstrated a rapid decline in PS amplitude of 82% +/- 15. CA1 orthodromic evoked PS was lost completely at an average 12.4 +/- 1.6 h after NMDA exposure. This sudden loss of response contrasted with paired, untreated slices, where CA1 PS could be elicited for 22.6 +/- 4.0 h (P < 0.05). Treatment with 10 mM MgCl2 begun after NMDA exposure and continued for 35 min, prevented delayed loss of CA1 orthodromic PS, which then could be elicited for 20.3 +/- 2.1 h. These results indicate that delayed injury can be evaluated using the hippocampal slice. They also suggest that activation of NMDA receptors can induce delayed neuronal injury in CA1 neurons, and that magnesium treatment after NMDA can prevent this injury.