Glutamate and Neurotoxicity

1. HISTORICAL PERSPECTIVE

The toxic effects of glutamate exposure on neurons were first recognized nearly half a century ago, when Lucas and Newhouse observed that subcutaneous administration of glutamate caused loss of neurons in the inner nuclear layer of the retina in both adult and neonatal mice (1). Olney extended these findings to other regions of brain, including neurons in the roof of the third ventricle, the hypothalamus, and the dentate gyrus (2). Changes evolved rapidly, over minutes in adult mice to several hours in neonates, and were characterized by intracellular edema and pyknotic nuclei, consistent with necrosis. In the next few years the role of glutamate as the major excitatory neurotransmitter in the mammalian central nervous system (CNS) became clear (3-6) and the existence of specific glutamate receptors was demonstrated. Excitotoxicity, the effect of glutamate receptor activation to trigger neuronal cell death, was proposed to play a role in many pathological conditions, in large part based on the observations that injection of glutamate agonists, notably kainate, could result in neuronal death and biochemical abnormalities resembling the pathology seen in disorders such as Huntington's disease (7,8) and epilepsy (9,10). A role for endogenous glutamate release and subsequent glutamate receptor activation in triggering neuronal death under pathological conditions was further suggested by demonstrations that blockade of presynaptic glutamate release could attenuate neuronal injury in oxygen-deprived cultured hippocampal neurons (11) and that a blockade of the N-methyl-d-aspartate (NMDA) subtype of glutamate receptors attenuated neuronal injury in rodent models of global ischemia and hypoglycemic brain damage (12,13). Cell culture models were useful in exploring the ionic changes responsible for glutamate-mediated cell death (see Section 2). More recent observations suggest that receptor-mediated glutamate toxicity may not be limited to neurons, but may also affect oligodendrocytes (14-16). A non-receptor-mediated form of glutamate cytotoxicity due to cystine deprivation and lowering of intra-cellular glutathione has also been described (17,18), although the levels of sustained exposure required to induce this death are higher than expected in most in vivo situations.

The original description of glutamate-mediated neuronal death in retina and in brain fits with a morphological picture of necrosis, but glutamate receptor activation can also trigger cell death with features of apoptosis. More recent findings suggest that, in some circumstances, glutamate receptor-stimulated Ca2+ entry could promote neuronal survival rather than neuronal death and might specifically attenuate some forms of neuronal apoptosis (see Section 2).

2. GLUTAMATE NEUROTOXICITY IN VITRO

In vitro systems have made it possible to dissect out some of the ionic alterations and signaling pathways involved in glutamate-induced neuronal death. Three subtypes of ionotropic glutamate receptors have been characterized and the subunits comprising these receptors have been cloned

From: Contemporary Clinical Neuroscience: Glutamate and Addiction Edited by: Barbara H. Herman et al. © Humana Press Inc., Totowa, NJ

(19-21; see also Chapters 1 and 2). N-Methy1-d-aspartate receptors and a subset of a-amino-3-hydroxy-5-methy1-4-isoxazole propionic acid (AMPA) receptors are permeable to Ca2+ as well as Na+, whereas the majority of AMPA receptors and kainate receptors are permeable to Na+ but not Ca2+. Metabotropic glutamate receptors (mGluRs) are linked to G-proteins rather than ion channels (reviewed in refs.22 and 23); activation of these receptors does not appear to mediate excitotoxicity, but it can modulate it in complex ways (24-27).

Activation of ionotropic glutamate receptors causes an initial Na+ influx (with accompanying Cl-and water influx) and induces cell body swelling during glutamate overexposure (reviewed in ref. 28; see also ref. 29). Ca2+ influx through NMDA receptors and through the Ca2+-permeable subset of AMPA receptors (30), likely augmented by secondary Ca2+ influx through voltage-gated Ca2+ channels and reverse operation of neuronal Na+-Ca2+ exchangers (resulting from membrane depolarization and elevated intracellular Na+) (31), is the predominant factor in the neurodegeneration that occurs over subsequent hours. NMDA-induced neuronal death correlates well with the amount of calcium influx, as measured by the uptake of radiolabeled calcium (32), or with intracellular free levels ([Ca2+]i) measured with low-affinity indicator dyes (33). A similar correlation between [Ca2+]i and glutamate agonist-induced cell death has been observed in non-neuronal cells transfected with NMDA receptors (34-35).

The precise downstream mechanisms linking intracellular Ca2+ overload to cell death are still not entirely clear, but mitochondria probably play an important role. Mitochondrial calcium uptake following glutamate exposure may result in the uncoupling of electron transport from ATP synthesis, with resultant increased production of mitochondrial reactive oxygen species and derangements of energy metabolism (36-40). Indeed, free-radical scavengers attenuate glutamate neurotoxicity in vitro (41,42).

The Ca2+ influx triggered by glutamate receptor activation can also directly activate catabolic enzymes: calcium-dependent proteases, phospholipases, and endonucleases. For example, calpain is activated following glutamate receptor activation, and inhibition of calpain attenuates glutamate agonist-induced death in vitro and also reduces neuronal death in transient global ischemia (43-45).

The cellular swelling and calcium overload triggered by the glutamate receptor over activation typically results in neuronal death with features consistent with necrosis, including early cell body swelling and loss of plasma membrane integrity, organelle disruption, and insensitivity to inhibitors of protein synthesis or caspase activity (e.g., refs. 46-49). Some features of apoptosis, such as positive TUNEL (terminal transferase-mediated dUTP-digoxigenin nick end labeling) staining, internucleoso-mal DNA fragmentation (DNA laddering), and nuclear and chromatin condensation have been reported in cultured cerebellar granule cells and neocortical neurons exposed to glutamate agonists (50-52). TUNEL staining and DNA ladders have also been observed in neurons dying by excitotoxic necrosis (49,53), and in isolation, they do not support an important role for apoptosis in excitotoxic neuronal death. Under certain circumstances, pharmacological or genetic inhibitors of apoptosis reduce glutamate-mediated neuronal death, but many such studies have used relatively immature neurons. For example, protein and RNA synthesis inhibitors attenuate NMDA-induced neuronal death in cultured retinal ganglion cells and immature neocortical neurons (maintained in culture for 3-5 d) (54,55). The caspase inhibitor Z-VAD.FMK attenuates NMDA-induced death in more mature (15 d in culture) rat neocortical cultures exposed to NMDA in the absence of Mg2+ (56), but does not attenuate NMDA-induced neuronal death in murine neocortical cultures (48). Deletion of the bax gene attenuates glutamate and kainate-induced death in neocortical neurons grown in culture for 4 d (57). At least some of the glutamate-induced neuronal death in such immature neurons could be secondary to cystine depletion rather than receptor-mediated excitotoxicity, an idea supported by the observation that bax gene deletion did not alter the vulnerability of more mature (14 d in culture) murine neocortical neurons to NMDA-mediated excitotoxicity (Gottron and Choi, unpublished observation). Taken together, available observations support a model in which excitotoxic glutamate receptor overactivation favors necrosis, but can also lead to apoptosis under certain circumstances. In particular, factors such as milder insult intensity (see ref. 54), cell immaturity (associated with fewer glutamate receptors and intrinsically higher propensity to undergo apoptosis) (58), low intracellular Ca2+, and low trophic factor availability may favor apoptosis after any insult, including excitotoxic insults. Mixed forms of death may be particularly prominent in vivo, where an initial excitotoxic insult may be followed by loss of trophic factor or surface factor support resulting from damage to inputs or surrounding cells.

Ca2+ may not be the only divalent cation important to excitotoxicity, as toxic neuronal uptake of Zn2+ released from glutamatergic presynaptic nerve terminals has been suggested to interact importantly with excitotoxicity (for review, see refs. 59-61). A chelatable pool of zinc is located in synaptic vesicles within glutamatergic nerve terminals throughout the telencephalon (see ref 59). Zn2+ directly inhibits the NMDA receptor (62-64) and modulates multiple other receptors and channels, the former including GABA, glycine, and purine receptors (reviewed in ref. 65). It may also induce a delayed upregulation of NMDA receptor function and NMDA receptor-mediated excitotoxicity (66). In addition to its neuromodulatory role, Zn2+ can enter postsynaptic neurons in toxic quantities via routes facilitated by AMPA/kainate and NMDA receptor activation (67), and thereby contribute to neuronal death after transient global ischemia (68,69) and seizures (70,71).

A central role for K+ efflux in promoting apoptosis has been increasingly suspected, and in neurons, this efflux may be mediated both by the delayed rectifier Ik as well as glutamate receptors (72-74). The blockade of potassium channels reduces neuronal death in focal and global ischemia (75,76), contrary to the conventional expectation that this maneuver would enhance excitotoxicity and thus increase ischemic neuronal death.

Although many studies in cultured neurons, like the early in vivo studies of glutamate toxicity, have focused on neuronal death induced by exogenously added glutamate agonists, cell culture models have also provided insights into the role of endogenously released glutamate in neuronal death after other insults. For example, the blockade of NMDA receptors attenuates oxygen-glucose deprivation-induced death in cultured neocortical neurons (77,78) and death induced by exposure to the mitochondrial toxin 3-nitropropionic acid in organotypic corticostriatal slice cultures (79). Oxygen-glucose deprivation-induced neuronal death is associated with enhancement in neuronal [Ca2+]i and with uptake of radiolabeled Ca2+ (80,81).

Was this article helpful?

0 0
Natural Detox

Natural Detox

Are you looking for a full total body detox? If so, then you might want to try a 10 day detox or some refer to it as the 2-week detox. A 10-day detox is a full body detox that usually means taking several different steps to reach your total body transformation. It might involve a change in diet, exercise and more.

Get My Free Ebook


Post a comment