In vitro neuronal and glial production and differentiation of human central neurocytoma cells

Human central neurocytoma cells were cultured and characterized immunophenotypically and electrophysiologically to clarify their developmental potential. We conducted systematic in vitro studies utilizing fresh tissues from three patients. Initially small homogenous cell clusters settled down onto the bottom of the culture flasks, and, after 2 weeks from plating, mature neuron-like cells developed from these clusters and expressed neurofilament proteins (NF: specific neuronal markers). On the other hand, approximately 80% of small round cell clusters and flat glial-like cells from which these clusters developed were positively stained for glial fibrillary acidic protein (GFAP: a specific glial marker). Furthermore these neuronal and glial cells showed distinct morphology, and dual-label, indirect immunohistochemistry for GFAP and NF-200 kD disclosed that the two antigens were not found co-localized in the same cells. In single-cell clonal analysis, neuronal, glial, and mixed neuronal and glial clones were generated. Electrophysiologically, the cells of neuronal morphology possessed sodium channels, and also L-type calcium channels in whole-cell voltage clamp. The sodium channels were of a characteristic neuronal phenotype which appears in neurons. These findings suggest that small round human central neurocytoma cells exhibit both neuronal and glial differentiations and have the properties reminiscent of precursor cells derived from subventricular matrix; thus, these cultured cells may be a potential source for investigations of human CNS neuronal and glial development and differentiation. J. Neurosci. Res. 51:526–535, 1998. © 1998 Wiley-Liss, Inc.     

Glial activation modulates glutamate neurotoxicity in cerebellar granule cell cultures

We studied the influence of glial cells on the neuronal response to glutamate toxicity in cerebellar granule cell cultures. We compared the effect of glutamate on neuronal viability in neuronal vs. neuronal-glial cultures and determined this effect after pretreating the cultures with the lipopolysaccharide (LPS) of Escherichia coli, agent widely used to induce glial activation. Morphological changes in glial cells and nitric oxide (NO) production were evaluated as indicators of glial activation. We observed that glutamate neurotoxicity in neuronal-glial cultures was attenuated in a certain range of glutamate concentration when compared to neuronal cultures, but it was enhanced at higher glutamate concentrations. This enhanced neurotoxicity was associated with morphological changes in astrocytes and microglial cells in the absence of NO production. LPS treatment induced morphological changes in glial cells in neuronal-glial cultures as well as NO production. These effects occurred in the absence of significant neuronal death. However, when LPS-pretreated cultures were treated with glutamate, the sensitivity of neuronal-glial cultures to glutamate neurotoxicity was increased. This was accompanied by additional morphological changes in glial cells in the absence of a further increase in NO production. These results suggest that quiescent glial cells protect neuronal cells from glutamate neurotoxicity, but reactive glial cells increase glutamate neurotoxicity. Therefore, glial cells play a key role in the neuronal response to a negative stimulus, suggesting that this response can be modified through an action on glial cells.     

Aluminum-induced degeneration of astrocytes occurs via apoptosis and results in neuronal death

The mechanisms by which aluminum interacts with the nervous system are only partly understood. In this study, we used cultured astrocytes and neurons to investigate the effects of long exposures to aluminum (1 mM). We found that aluminum accumulated both in neurons and astrocytes. After 8-12 days exposure, aluminum caused strong changes in the morphology of astrocytes including shrinkage of cell bodies and retraction of processes. Exposures over 15-18 days reduced astrocytes viability by 50%. Aluminum-induced degeneration of astrocytes involved the DNA fragmentation characteristic of apoptosis, and staining of aluminum-treated astrocytes with the DNA-binding fluorochrome Hoeschst 33258 revealed the typical apoptotic condensation and fragmentation of chromatin. Aluminum was also found to be neurotoxic, causing first (4-6 days) abnormal clustering and aggregation, and later (8-12 days) neuronal death. Interestingly, aluminum neurotoxicity occurred in neuroglial cultures containing approximately 10% astrocytes but not in near-pure neuronal cultures containing only 1% astrocytes. Staining of co-cultured cells with Hoeschst 33258 showed apoptotic condensation and fragmentation of chromatin in aluminum-treated astrocytes but not in co-cultured neurons. Our study demonstrates that aluminum can induce the apoptotic degeneration of astrocytes, and that this toxicity is critical in determining neuronal degeneration and death. Aluminum-mediated apoptosis of cultured astrocytes may be also a valuable model system to study the mechanisms underlying apoptosis in glial cells.     

Cadmium modulates proliferation and differentiation of human neuroblasts

Cadmium is an environmental pollutant inducing numerous pathological effects, including neurological disorders and brain diseases. However, little is known about the molecular mechanisms of cadmium in affecting neurons and in inducing neurotoxicity in the development of the human brain. We have recently established, cloned, and propagated in vitro a primary long-term cell culture (FNC-B4) obtained from the human fetal olfactory neuroepithelium. In the present study, we show that different concentrations of cadmium chloride (CdCl(2)) induced dose-dependent biological effects in FNC-B4 cells. A low concentration (10 microM) of CdCl(2) stimulated neuroblast growth, whereas a high concentration (100 microM) inhibited the growth and the viability of neuroblasts inducing morphological and cytoskeletal alterations as well as apoptotic cell death. We also observed that CdCl(2) affected, in a dose-dependent manner, the differentiation of FNC-B4 neuroblasts, with increased mRNA and protein levels of differentiation markers and decreased expression levels of neuronal stem markers. Furthermore, differentiated cells co-expressed glial and neuronal markers. We suggest that CdCl(2) in FNC-B4 neuroblasts might represent a selective cue by which, in a heterogeneous primary culture, the more differentiated mature cells die, whereas the undifferentiated cells, at the same time glial and neuronal progenitors, are forced to access a state of differentiation.     

Evaluation of the Importance of Astrocytes When Screening for Acute Toxicity in Neuronal Cell Systems

Reliable, high throughput, in vitro preliminary screening batteries have the potential to greatly accelerate the rate at which regulatory neurotoxicity data is generated. This study evaluated the importance of astrocytes when predicting acute toxic potential using a neuronal screening battery of pure neuronal (NT2.N) and astrocytic (NT2.A) and integrated neuronal/astrocytic (NT2.N/A) cell systems derived from the human NT2.D1 cell line, using biochemical endpoints (mitochondrial membrane potential (MMP) depolarisation and ATP and GSH depletion). Following exposure for 72 h, the known acute human neurotoxicants trimethyltin-chloride, chloroquine and 6-hydroxydopamine were frequently capable of disrupting biochemical processes in all of the cell systems at non-cytotoxic concentrations. Astrocytes provide key metabolic and protective support to neurons during toxic challenge in vivo and generally the astrocyte containing cell systems showed increased tolerance to toxicant insult compared with the NT2.N mono-culture in vitro. Whilst there was no consistent relationship between MMP, ATP and GSH log IC₅₀ values for the NT2.N/A and NT2.A cell systems, these data did provide preliminary evidence of modulation of the acute neuronal toxic response by astrocytes. In conclusion, the suitability of NT2 neurons and astrocytes as cell systems for acute toxicity screening deserves further investigation.     

Neonatal mouse cortical but not isogenic human astrocyte feeder layers enhance the functional maturation of induced pluripotent stem cell‐derived neurons in culture

Human induced pluripotent stem (iPS) cell‐derived neurons and astrocytes are attractive cellular tools for nervous system disease modeling and drug screening. Optimal utilization of these tools requires differentiation protocols that efficiently generate functional cell phenotypes in vitro. As nervous system function is dependent on networked neuronal activity involving both neuronal and astrocytic synaptic functions, we examined astrocyte effects on the functional maturation of neurons from human iPS cell‐derived neural stem cells (NSCs). We first demonstrate human iPS cell‐derived NSCs can be rapidly differentiated in culture to either neurons or astrocytes with characteristic cellular, molecular and physiological features. Although differentiated neurons were capable of firing multiple action potentials (APs), few cells developed spontaneous electrical activity in culture. We show spontaneous electrical activity was significantly increased by neuronal differentiation of human NSCs on feeder layers of neonatal mouse cortical astrocytes. In contrast, co‐culture on feeder layers of isogenic human iPS cell‐derived astrocytes had no positive effect on spontaneous neuronal activity. Spontaneous electrical activity was dependent on glutamate receptor‐channel function and occurred without changes in I_Na, I_K, V_m, and AP properties of iPS cell‐derived neurons. These data demonstrate co‐culture with neonatal mouse cortical astrocytes but not human isogenic iPS cell‐derived astrocytes stimulates glutamatergic synaptic transmission between iPS cell‐derived neurons in culture. We present RNA‐sequencing data for an immature, fetal‐like status of our human iPS cell‐derived astrocytes as one possible explanation for their failure to enhance synaptic activity in our co‐culture system.     

Humoral and contact interactions in astroglia/stem cell co-cultures in the course of glia-induced neurogenesis

Astroglial cells support or restrict the migration and differentiation of neural stem cells depending on the developmental stage of the progenitors and the physiological state of the astrocytes. In the present study, we show that astroglial cells instruct noncommitted, immortalized neuroectodermal stem cells to adopt a neuronal fate, while they fail to induce neuronal differentiation of embryonic stem cells under similar culture conditions. Astrocytes induce neuron formation by neuroectodermal progenitors both through direct cell-to-cell contacts and via short-range acting humoral factors. Neuron formation takes place inside compact stem cell assemblies formed 30- 60 h after the onset of glial induction. Statistical analyses of time-lapse microscopic recordings show that direct contacts with astrocytes hinder the migration of neuroectodermal progenitors, while astroglia-derived humoral factors increase their motility. In non-contact co-cultures with astrocytes, altered adhesiveness prevents the separation of frequently colliding neural stem cells. By contrast, in contact co-cultures with astrocytes, the restricted migration on glial surfaces keeps the cell progenies together, resulting in the formation of clonally proliferating stem cell aggregates. The data indicate that in vitro maintained parenchymal astrocytes (1) secrete factors, which initiate neuronal differentiation of neuroectodermal stem cells; and (2) provide a cellular microenvironment where stem cell/stem cell interactions can develop and the sorting out of the future neurons can proceed. In contrast to noncommitted progenitors, postmitotic neuronal precursors leave the stem cell clusters, indicating that astroglial cells selectively support the migration of maturing neurons as well as the elongation of neurites.     

The neurotoxicity of 1-methyl-4-phenylpyridinium in cultured cerebellar granule cells

Cerebellar granule cells in enriched primary culture are susceptible to the neurotoxic effects of 1-methyl-4-phenylpyridinium (MPP+). Relatively high MPP+ concentrations are required to elicit neurotoxic effects at early culture times, but lower concentrations of MPP+ produce comparable neurotoxic effects at later culture times. Under identical culture conditions 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is not neurotoxic. Preincubation with the glutamate uptake blockers, DL-threo-3-hydroxyaspartic acid or dihydrokainate, or the dopaminergic uptake blocker mazindol, protects the granule cells from the cytotoxic effects of MPP+. Although MPTP is not neurotoxic in an enriched granule cell culture, in coculture with cerebellar astrocytes MPTP is toxic to granule cells, presumably because it is converted in astrocytes to MPP+. Cerebellar astrocytes remain confluent and viable. The addition of pargyline to the coculture abolishes the neurotoxicity consistent with a role of MAO B in bioactivation of MPTP. The concentration of MPP+ in the coculture medium (13 microM) was less than that required for the toxic effect in enriched neuronal cultures at earlier culture times, suggesting that an astroglial-neuronal interaction, perhaps by proximity, enhances the neurotoxicity of MPP+. These results might explain reported effects of MPTP on some cerebellar cells in mice.     

Impact of Plant-Derived Flavonoids on Neurodegenerative Diseases

Neurodegenerative disorders have a common characteristic that is the involvement of different cell types, typically the reactivity of astrocytes and microglia, characterizing gliosis, which in turn contributes to the neuronal dysfunction and or death. Flavonoids are secondary metabolites of plant origin widely investigated at present and represent one of the most important and diversified among natural products phenolic groups. Several biological activities are attributed to this class of polyphenols, such as antitumor activity, antioxidant, antiviral, and anti-inflammatory, among others, which give significant pharmacological importance. Our group have observed that flavonoids derived from Brazilian plants Dimorphandra mollis Bent., Croton betulaster Müll. Arg., e Poincianella pyramidalis Tul., botanical synonymous Caesalpinia pyramidalis Tul. also elicit a broad spectrum of responses in astrocytes and neurons in culture as activation of astrocytes and microglia, astrocyte associated protection of neuronal progenitor cells, neuronal differentiation and neuritogenesis. It was observed the flavonoids also induced neuronal differentiation of mouse embryonic stem cells and human pluripotent stem cells. Moreover, with the objective of seeking preclinical pharmacological evidence of these molecules, in order to assess its future use in the treatment of neurodegenerative disorders, we have evaluated the effects of flavonoids in preclinical in vitro models of neuroinflammation associated with Parkinson’s disease and glutamate toxicity associated with ischemia. In particular, our efforts have been directed to identify mechanisms involved in the changes in viability, morphology, and glial cell function induced by flavonoids in cultures of glial cells and neuronal cells alone or in interactions and clarify the relation with their neuroprotective and morphogetic effects.     

The use of astrocytes in culture as model systems for evaluating neurotoxic-induced-injury.

The prevailing thought that astrocytes function predominantly as passive metabolic or even physical support for neurons has faded over the last 20 years. Today these stellar shaped cells are credited with an expanded role, playing key functions in CNS development, homeostasis, and pathology. In probing their expanded roles, primary astrocyte culture systems have proven to be an indispensable tool. Astrocytes have been implicated in both a defensive and facilitatory capacity for many toxic injuries. Evidence for a protective role of astrocytes in modulating CNS toxicity is afforded by observations that the toxicity of glutamate to cortical neurons is diminished upon astrocytic enrichment of the cell culture (Rosenberg and Aizenman, 1989). In cultures of rat cerebral cortex in which astrocyte proliferation is stringently suppressed, glutamate neurotoxicity occurs at low glutamate concentrations similar to those which are normally found in the extracellular space in the hippocampus. In the presence of excess astrocytes, concentrations of glutamate one-hundred fold higher are required to produce equivalent neurotoxicity (Rosenberg and Aizenman, 1989). Astrocytes can facilitate the action of neurotoxins via a modulating process which takes place within the astrocyte or by a direct cytotoxic effect. Whereas primary astrocyte cultures remain unaffected by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP; Marini et al., 1989), they function prominently in the selective destruction of dopaminergic neurons of the nigrostriatal pathway in humans, other primates and rodents (Davis et al. 1979; Langston et al., 1983; Burns et al., 1983; Langston et al., 1984; Heikkila et al., 1984; Jarvis and Wagner, 1985). Thus, while MPTP by itself is not toxic to cerebellar cells in co-culture with cerebellar astrocytes, MPTP is toxic to the granule cells (Marini et al, 1989). This is thought to be due to an astrocyte-mediated conversion of MPTP to its highly polar and toxic metabolite, 1-methyl-4-phenylpyridinium ion (MPP+; Chiba et al. 1984). There is compelling evidence that astrocytes respond directly or indirectly to a number of other neurotoxins. Direct cytotoxic effects on astrocytes constitute the major morphologic feature in hyperammonemia (Norenberg, 1981), a condition implicated as an etiologic factor in several CNS disorders. In addition, a predisposition of astrocytes for methylmercury uptake (Aschner et al., 1990 a,b) offers a possible explanation for the observed neurotoxicity of this heavy metal, since a direct toxic effect on astrocytes would result in failure of astrocyte homeostatic functions, indirectly resulting in neuronal impairment, injury and death.     

Ligand specific effects on aluminum incorporation and toxicity in neurons and astrocytes

Aluminum is present in many manufactured foods and medicines and is added to drinking water for purification purposes. It has been proposed that aluminum is a contributing factors to several neurodegenerative disorders such as Alzheimer's disease. However, this remains controversial primarily due to the unusual properties of aluminum and a lack of information on its cellular sites of action. To resolve some of these questions, we have examined aluminum uptake in both neuronal and astroglial cells as well as the role of metal speciation. The relative accumulation of four aluminum salts, aluminum maltolate, aluminum lactate, aluminum chloride and aluminum fluoride, was investigated and correlated with cell viability and intracellular distribution as determined by morin staining. Significant differences in aluminum incorporation and toxicity were observed in both neuronal and glia cells with the largest effects exhibited by the maltol species. This was accompanied by a nuclear accumulation in the neuronal cell line that was contrasted by the perinuclear, vesicular distribution in astrocytes that partially co-localized with cathepsin D, a lysosomal marker. These findings demonstrate differences in aluminum species and highlights the importance of these factors in modulating the toxic effect of aluminum.     

Compartmentalized microfluidic culture platform to study mechanism of paclitaxel-induced axonal degeneration

Chemotherapy induced peripheral neuropathy is a common and dose-limiting side effect of anticancer drugs. Studies aimed at understanding the underlying mechanism of neurotoxicity of chemotherapeutic drugs have been hampered by lack of suitable culture systems that can differentiate between neuronal cell body, axon or associated glial cells. Here, we have developed an in vitro compartmentalized microfluidic culture system to examine the site of toxicity of chemotherapeutic drugs. To test the culture platform, we used paclitaxel, a widely used anticancer drug for breast cancer, because it causes sensory polyneuropathy in a large proportion of patients and there is no effective treatment. In previous in vitro studies, paclitaxel induced distal axonal degeneration but it was unclear if this was due to direct toxicity on the axon or a consequence of toxicity on the neuronal cell body. Using microfluidic channels that allow compartmentalized culturing of neurons and axons, we demonstrate that the axons are much more susceptible to toxic effects of paclitaxel. When paclitaxel was applied to the axonal side, there was clear degeneration of axons; but when paclitaxel was applied to the soma side, there was no change in axon length. Furthermore, we show that recombinant human erythropoietin, which had been shown to be neuroprotective against paclitaxel neurotoxicity, provides neuroprotection whether it is applied to the cell body or the axons directly. This observation has implications for development of neuroprotective drugs for chemotherapy induced peripheral neuropathies as dorsal root ganglia do not possess blood-nerve-barrier, eliminating one of the cardinal requirements of drug development for the nervous system. This compartmentalized microfluidic culture system can be used for studies aimed at understanding axon degeneration, neuroprotection and development of the nervous system.     

Cholinergic neurotoxicity induced by ethylcholine aziridinium (AF64A) in neuron-enriched cultures

The sequence of events in neuronal changes induced by the cholinotoxin ethylcholine aziridinium (AF64A) was studied. Neuron-enriched cultures derived from 8-day-embryonic chick cerebra were treated with AF64A at concentrations of 10(-5), 10(-4) and 10(-3) M. Choline acetyltransferase (ChAT) was used as an index of cholinergic neurons. Changes in cell morphology, the immunocytochemical and biochemical presence of ChAT, and DNA and protein content were assessed. Neuron-enriched cultures exposed to AF64A showed a dose-dependent response; after 24 h of exposure to 10(-3) M toxin all cells were dead, whereas a concentration of 10(-5) M did not alter culture morphology or DNA and protein contents. Despite the lack of cytological changes and the presence of ChAT immunoreactivity, biochemically assessed ChAT activity was reduced 36% in 10(-5) M treated cultures. Thus, the implicated decrease in acetylcholine synthesis in these cells cannot entirely account for the neuronal degeneration. Simultaneous exposure of cultures to both AF64A and 10 times higher concentrations of choline chloride delayed or diminished the neurotoxic changes. The protective effect of high choline concentrations was interpreted as evidence of competition between choline and AF64A for the high affinity choline transport system and as constituents in the cell membrane. Examination of the temporal sequence of cytotoxic changes in 10(-4) M exposed cultures revealed that disruption of neuronal aggregates and fragmentation of neurites occurred between 4 and 8 hours of exposure. After 24 h, some neurons survived but with attenuated arbors; in contrast, astrocytes appeared intact, suggesting that glial cells are more resistant than neurons to the toxic effects of AF64A. These findings suggest this culture model may be useful to further elucidate the mechanisms of AF64A drug action and study differentiation of cultured neuronal populations in the absence of cholinergic cells.     

Structure-activity comparison of organotin species: dibutyltin is a developmental neurotoxicant in vitro and in vivo

Human exposure to the organotins can occur due to their use as polyvinyl chloride heat stabilizers and as marine biocides. The consequences of this exposure for human health are unknown. We initially compared the toxicity of monomethyltin, dimethyltin, and dibutyltin to the known neurotoxicant trimethyltin using an in vitro model of neuronal development in PC12 cells. Dibutyltin, a compound traditionally thought to target the immune system, was the most potent neurotoxicant. Dibutyltin significantly inhibited neurite outgrowth and caused cell death at concentrations approximately 40-fold lower than the lowest toxic concentrations of trimethyltin. Dimethyltin was less potent than trimethyltin and monomethyltin was not toxic at any concentration examined. These results suggested the importance of prioritizing in vivo neurotoxicity testing with dibutyltin. To accomplish this, pregnant rats were dosed orally with low levels of dibutyltin from gestational day 6 through weaning. In response to developmental dibutyltin exposure, the incidence of apoptotic cell death, measured by DNA fragmentation and TUNEL staining, was increased in the neocortex and hippocampus of postnatal day 38 offspring. No effect was observed at other ages examined.     

Morphological differentiation of neuropeptide Y neurons in aggregate cultures of dissociated fetal cortical cells: a model system for glia-neuron paracrine interactions

The temporal changes in the morphological profiles of neuropeptide Y (NPY) neurons and their topographical relationship with glial cells (astrocytes) were characterized in aggregate cultures derived from fetal cortical tissue using immunocytochemical procedures. On day 6 of culture, structures labelled with NPY antibodies were small and uneven in size but many resembled neuronal cell bodies. On day 14, neuronal perikarya were well defined and several morphological types of NPY neurons could be distinguished most of which gave rise to beaded processes: unipolar or multipolar bitufted neurons whose processes branch in close proximity to the cell body; bipolar neurons; and multipolar neurons. On day 23, heavily punctate and asymmetrically labelled cell bodies were dispersed throughout the aggregate; neuronal processes were less conspicuous. At 14 and 23 days, cells expressing glial fibrillary acidic protein (GFAP) and neuronal specific enolase (NSE) were abundantly distributed throughout the aggregate. Using a double immunoreaction on 14-day-old aggregates revealed that GFAP + cells and their processes were in close apposition to and engulfing the NPY neurons. Thus, dissociated fetal NPY neurons undergo morphological differentiation in culture along with astrocytes (GFAP +) and other neuronal cell types (NSE +). Based on the topographical association of astrocytes and neurons, particularly NPY neurons, we propose that the aggregate culture system can serve as a model to study the role of paracrine interactions in the regulation of the expression of NPY.     

Primary cultures from fetal bovine brain

This study describes a method of obtaining primary cultures from the cerebral cortex and the hypothalamus of bovine fetuses. We describe here the influence of tissue origin, developmental stage and culture medium conditions on cell differentiation and prevalence of neurons vs glial cells. To identify optimal conditions for obtaining and growing viable neurons and astrocytes in culture, we tested early, middle and late stages of development. Explants from cortex, early stages (week 10 of pregnancy out of 36) and low fetal calf serum concentration (1%) yielded maximum amounts of neurons. Fresh and thawed tissues gave comparable results.     

Cultured adult porcine astrocytes and microglia express functional interferon-γ receptors and exhibit toxicity towards SH-SY5Y cells

In vitro cultures of various glial cell types are common systems used to model neuroinflammatory processes associated with age-dependent human neurodegenerative diseases. Even though most researchers choose to use neonatal rodent brain tissues as the source of glial cells, there are significant variations in glial cell functions that are species and age dependent. It has been established that human and swine immune systems have a number of similarities, which suggests that cultured porcine microglia and astrocytes may be good surrogates for human glial cell types. Here we describe a method that could be used to prepare more than 90% pure microglia and astrocyte cultures derived from adult porcine tissues. We demonstrate that both microglia and astrocytes derived from adult porcine brains express functional interferon-γ receptors (IFN-γ-R) and CD14. They become toxic towards SH-SY5Y neuroblastoma cells when exposed to proinflammatory mediators. Upon such stimulation with lipopolysaccharide (LPS) and interferon-γ (IFN-γ), adult porcine microglia, but not astrocytes, secrete tumor necrosis factor-α (TNF-α) while both cell types do not secrete detectable levels of nitric oxide (NO). Comparison of our experimental data with previously published studies indicates that adult porcine glial cultures have certain functional characteristics that make them similar to human glial cells. Therefore adult porcine glial cells may be a useful model system for studies of human diseases associated with adulthood and advanced age. Adult porcine tissues are relatively easy to obtain in most countries and could be used as a reliable and inexpensive source of cultured cells.     

Gliotoxicity in hippocampal cultures is induced by transportable, but not by nontransportable, glutamate uptake inhibitors

Extracellular glutamate is kept below a toxic level by glial and neuronal glutamate transporters. Here we show that the transportable glutamate uptake inhibitor L-trans-pyrrolidine-2,4-dicarboxylate (t-PDC) induced cell death in mature, but not in immature, hippocampal neuron-enriched cultures. The cell death produced by a 24-hr treatment with t-PDC was dose-dependent and reached 85% of the cell population at a 250 microM concentration at 23 days in vitro (DIV). Immunocytochemistry experiments showed that, under these experimental conditions, t-PDC killed not only neurons as expected but also glial cells. The N-methyl-D-aspartate (NMDA) antagonist D-2-aminophosphonovalerate (D-APV; 250 microM) only partially reversed this toxicity, completely protecting the neuronal cell population but not the glial population. The antioxidant compounds alpha-tocopherol or Trolox, used at concentrations that reverse the oxidative stress-induced toxicity, did not block the gliotoxicity specifically produced by t-PDC in the presence of D-APV. The nontransportable glutamate uptake inhibitor DL-threo-beta-benzyloxyaspartate (TBOA) elicited cell death only in mature, but not in immature, hippocampal cultures. The TBOA toxic effect was dose dependent and reached a plateau at 100 microM in 23-DIV cultures. About 50% of the cell population died. TBOA affected essentially the neuronal population. D-APV (250 microM) completely reversed this toxicity. It is concluded that nontransportable glutamate uptake inhibitors are neurotoxic via overactivation of NMDA receptors, whereas transportable glutamate uptake inhibitors induce both an NMDA-dependent neurotoxicity and an NMDA- and oxidative stress-independent gliotoxicity, but only in mature hippocampal cultures.