Glutamate Neurotoxicity In Animal Models Of Neuronal Injury

Extracellular glutamate levels are elevated in brain following ischemia (82), seizures (83), and head trauma (84). Although Simon's original observation that NMDA antagonists attenuate hip-pocampal neuronal death following global ischemia has not been consistently confirmed [(12); but see ref. 85], AMPA receptor antagonists have reduced hippocampal injury following global ischemia in many studies (85-88) and also reduce infarct volume following focal ischemia (89,90). NMDA antagonists, especially if administered prior to the onset of ischemia, reduce infarct size in rodent and feline models of both transient and permanent focal ischemia (91-93). Administration of glutamate antagonists improves neurological outcome in rodent models of traumatic brain injury (94,95) and spinal cord injury (96).

Other factors present in the injured nervous system could cause neurons to become vulnerable to glutamate neurotoxicity even when the synaptic release and extracellular concentration of glutamate are not especially elevated [e.g., when neuronal homeostatic mechanisms are compromised by energy depletion (97) or mitochondrial dysfunction (98-100)]. Glutamate-mediated excitotoxicity could thus contribute, at least in a secondary fashion, to the neuronal loss associated with chronic neurodegenera-tive diseases such as Huntington's disease (7,8,101), Alzheimer's disease (e.g., see refs. 102 and 103 for reviews), or Parkinson's disease (104,105). In particular, loss of transporter-mediated glutamate uptake has been postulated to promote the excitotoxic death of motor neurons in amyotrophic lateral sclerosis (106).

Glutamate-mediated neuronal death in cultured cells can have mixed features of both necrosis and apoptosis (see Section 2); similar observations have been made in models of excitotoxic cell death in vivo. Neurons in adult rat brains typically die a morphologically necrotic death after intrastriatal injection of glutamate receptor agonists (107,108), but these same neurons may exhibit some features of apoptosis, including TUNEL positivity and transient DNA laddering (109); injection of non-NMDA agonists also induces chromatin clumping (110). Selective neuronal death following global ischemia can evolve over 2-3 d (111,112) and exhibits many features of apoptosis, although features of necrosis can also be present (113; reviewed in ref. 114). This death is sensitive to inhibition of caspases or overexpression of the anti-apoptotic gene bcl-2 (115,116). In contrast, neuronal death following focal ischemic insults was thought to evolve rapidly via necrosis, but even in this more fulminant injury, recent experiments have suggested that neuronal death can evolve over several days after the onset of injury (117,118) and may exhibit morphological and biochemical features of apoptosis, including sensitivity to antiapoptotic strategies, even when treatment is delayed for up to 6 h after the onset of ischemia (117-120). Apoptosis has also been implicated in neuronal death occurring after brain or spinal cord trauma (121-124) or in association with several chronic neurodegenerative disease states, including Alzheimer's disease (125-127) and Huntington's disease (108).


As discussed earlier, glutamate-mediated elevations in [Ca2+]i play a central role in excitotoxic neuronal necrosis. However, developmental neuronal apoptosis has been linked to an opposite change in calcium levels, i.e., a drop below an optimal "set point" (128). This idea has been most extensively studied in sympathetic neurons deprived of nerve growth factor and in cerebellar granule cells switched from high to low potassium media. In cultured sympathetic neurons, [Ca2+]i at early timepoints correlates with survival: lowering [Ca2+]i (reduced extracellular [Ca2+], voltage-gated calcium channel blockers, or intracellular Ca2+ chelators) enhances apoptosis (128-130), whereas raising [Ca2+]i (increased extracellular [Ca2+], increased extracellular K+, BayK 8644, or nicotinic receptor agonists) blocks apoptosis. Similarly, lowering extracellular K+ lowers [Ca2+]i and enhances cerebellar granule apoptosis (131-133), whereas raising [Ca2+]i by increasing Ca2+ release from intracellular stores or increasing Ca2+ entry by exposure to elevated K+ or glutamate blocks this apoptosis (134-136). Reduced levels of [Ca2+]i are found in cultured neocortical neurons undergoing apoptosis following oxygen-glucose deprivation in the presence of glutamate antagonists (137); NMDA antagonists and agonists, respectively, enhance or block neocortical neuronal apoptosis in culture (138). The association of reduced [Ca2+]i with apoptosis is not limited to neurons. For example, calcium chelators induce apoptosis in astrocytes and lymphoid cells (139-143). Some authors point to a reduction in intracellular calcium stores rather than a reduction in overall [Ca2+]i as a mediator of apoptosis [e.g., in lymphoid cells undergoing apoptosis after glucocorticoid exposure, intracellular calcium stores were reduced (144,145)]. Elevations in [Ca2+]i have been associated with apoptosis as well, particularly with activation of some of the mediators of the programmed cell (e.g., see refs. (146-149). It is possible that deviation of [Ca2+]i from a calcium set point, either up or down, could trigger apoptosis, with elevations in calcium occurring more universally later in the apoptotic process (150).

If lowering [Ca2+]i can promote neuronal death via apoptosis, it is possible that glutamate antagonists might be neurotoxic under some conditions. This possibility was supported by the observation of pathological changes, most notably vacuolization in the cingulate gyrus, in rats treated with the NMDA antagonists phenylcyclidine (PCP), MK-801, and ketamine (151), although this may be the result of released circuit overexcitation rather than Ca2+ deprivation-induced apoptosis (152,153). More specific support for glutamate antagonist-induced, Ca2+ deprivation-induced neuronal apoptosis was provided by observations that NMDA antagonists could induce neuronal apoptosis in culture (154) and widespread apoptotic neuronal death in the developing nervous system in vivo (155). These observations not only raise caution against the use of compounds with NMDA antagonist activity in the young brain, but may also, in part, underlie the neuronal death seen in fetal alcohol syndrome (156). Studies of head trauma in infant rats demonstrate the two-edged potential of glutamate antagonists: early excitotoxic neuronal death is reduced, but later apoptotic neuronal death is enhanced when NMDA antagonists are administered (157). As discussed above, apoptosis may not be limited to the developing nervous system and has been described following ischemia or trauma in the adult brain (see above), as well as in neurodegenerative diseases. Recent studies in our laboratory have suggested that ischemic apoptosis and apoptosis associated with proteasome inhibition might be associated with reductions in neuronal [Ca2+]i and that raising [Ca2+]i might attenuate these forms of neuronal death (137, 161).

Although there have been as of yet no reports of glutamate antagonists enhancing apoptosis when administered following injury in the adult nervous system, the apoptosis-promoting effects of glutamate antagonists could underlie the disappointing outcome of trials of glutamate antagonists in human disease (e.g., see ref. 158; reviewed in ref 60). The timing of the administration of NMDA antagonists may be crucial to the efficacy of this therapeutic approach, with administration in the immediate peri-ischemic period able to reduce intracellular calcium levels and attenuate acute excitotoxicity, whereas later administration might tend to exacerbate the apoptotic component of ischemic cell death. It is plausible that the more complex watersheds in human gyrencephalic brain compared to those of the lissencephalic rodent brain, together with the not uncommonly stuttering onset of human stroke, would favor an increased apoptotic component in human versus rodent ischemic brain injury. These concerns may not be limited to ischemic injury and raise a note of caution regarding the proposed use of NMDA antagonists in diseases such as Huntington's disease (159,160) for which proteasome inhibition and resultant reductions in neuronal [Ca2+]i might occur.

Unraveling Alzheimers Disease

Unraveling Alzheimers Disease

I leave absolutely nothing out! Everything that I learned about Alzheimer’s I share with you. This is the most comprehensive report on Alzheimer’s you will ever read. No stone is left unturned in this comprehensive report.

Get My Free Ebook

Post a comment