Trends in Neurosciences
The Yin and Yang of NMDA receptor signalling
Section snippets
Importance of physiological levels of synaptic NMDA receptor activity in neuronal survival
Neuronal survival is often dependent on physiological levels of electrical activity – blocking electrical activity in vivo or in vitro causes cell death (10, 11 and Refs therein). But what aspect of this activity is neuroprotective? Synaptic communication, both that exacted on target neurons and received from afferents, can lead to exposure to neurotrophins 10, 12, 13, 14. However, synaptic signals (from glutamatergic neurons) can also cause Ca2+ influx via NMDA receptors, and there is evidence
Mechanism of NMDA-receptor-dependent survival
Detailed knowledge of the mechanisms behind synaptically activated NMDA-receptor-induced neuroprotection is currently lacking. NMDA-receptor-dependent neuroprotective pathways have been traditionally studied in vitro using the bath application of sub-toxic levels of glutamate or NMDA (which chronically activate both synaptic and extrasynaptic NMDA receptors), which could provide clues as to which synaptic routes lead to neuronal survival (although there are likely to be differences). These
CREB-dependence of neuronal survival?
Mice lacking CREB (and that also lack the cAMP-response-element modulator CREM, another CREB family member) in the CNS during development show extensive neuronal apoptosis [38]. Postnatal disruption of the CREB gene (again in a CREM−/− background) leads to progressive neurodegeneration in the hippocampus and dorsolateral striatum [38]. In mice deficient in CREB only, sensory neurons show excessive apoptosis [39], although neurons of the CNS appear normal, possibly owing to compensatory actions
NFκB-dependence of neuronal survival?
There is considerable evidence supporting a role for NFκB in mediating several neuroprotective pathways in the CNS and PNS [37]. Mice lacking the p50 subunit of NFκB show increased neuronal damage following hippocampal seizure [45] and cultures from these mice are more vulnerable to excitotoxic cell death. Use of a transgenic mouse that reports NFκB activity in vivo showed that activity is constitutively high in neurons from many areas of the developing and mature CNS and that this activity is
Downstream genes involved in synaptic NMDA-receptor-induced protection?
Figure 1 shows some CREB- and NFκB-regulated genes that might be involved in NMDA-receptor-dependent survival. The harmful effects of chronic NMDA receptor blockade could be explained, in part, by inhibition of the expression of some of these genes. Conversely, enhanced synaptic NMDA receptor activation and upregulation of these genes might underlie the neuroprotection afforded by environmental enrichment in rats [51] where, interestingly, certain trauma-resistant neurons in the hippocampus
Negative control of CREB by extrasynaptic NMDA receptor activation
The ability of synaptic NMDA receptor activity to induce CREB in vivo and in vitro is at odds with the fact that activation of NMDA receptors by the bath application of glutamate is often a very poor activator of CRE-dependent gene expression 52, 53. In neonatal hippocampal neurons cultured for over one week, bath-applied glutamate cannot reproduce the level of induction achieved by synaptic NMDA receptor activation, despite generating similar NMDA-receptor-dependent intracellular Ca2+ levels
Pro-death signalling from the NMDA receptor
Pathological activation of NMDA receptors, with consequent disruption to intracellular Ca2+ regulation, is the primary cause of neuronal death following acute excitotoxic trauma (e.g. brain ischaemia, hypoxia and mechanical insults) and is also associated with chronic neurodegenerative diseases 3, 4, 5, 6, 7, 8. Cultures of neurons taken from many areas of the brain respond to high levels of bath-applied glutamate by undergoing delayed Ca2+ deregulation, which precedes and predicts subsequent
‘Source specificity’ of Ca2+ toxicity?
For many years, glutamate toxicity has been believed to be due to excessive cellular Ca2+ loading imposed on the cell by the chronic activation of NMDA receptors 60, 73, 74. The work of Tymianski and colleagues supports a variation to this hypothesis: although Ca2+ influx specifically through the strong activation of NMDA receptors causes cell death, equivalent initial Ca2+ loads generated by influx through L-type voltage gated channels are tolerated 75, 76. This could be because Ca2+ influx
Molecular basis for the source-specificity model
NMDA receptors do not exist in isolation but are linked via the C termini of their subunits to large complexes of cytoplasmic proteins. These include scaffolding, adaptor, cell adhesion and cytoskeletal proteins, as well as components of signal transduction pathways (some modifiable by Ca2+) 79, 80. Thus, Ca2+ entry through the NMDA receptor will occur in an immediate molecular environment distinct from that in other parts of the cell.
The source-specificity hypothesis predicts a functional or
Do synaptic and extrasynaptic NMDA receptors signal differently to mitochondria?
Similar to the dramatic differences in signaling from synaptic and extrasynaptic NMDA receptors with regard to CREB activation, these receptors also have different effects on the mitochondrial membrane potential [23]. Ca2+ influx that depends on intense synaptic NMDA receptor activation is well tolerated by cells, whereas activation of extrasynaptic NMDA receptors, either on their own or in the presence of synaptic NMDA receptor activation, causes a loss of mitochondrial membrane potential and
A model for neuronal fate as a function of NMDA receptor activity
Figure 3a shows how neuronal survival and NMDA receptor activity have been proposed to be linked [89]. The ‘inverted U-shape’ [89] incorporates the evidence that physiological levels of NMDA receptor activity generally promote survival, and that a complete absence of NMDA receptor activity is, thus, deleterious to the cell. However, in neuropathological conditions, extracellular levels of glutamate build up, for example by reversed uptake (in stroke) or glutamate release from dying cells
Concluding remarks
Clinical trials for many of the disorders mentioned above involving NMDA receptor antagonists have been plagued by problems of poor efficacy and tolerance (24, 91, 92 and Refs therein). This is understandable given the important role of physiological NMDA receptor signalling in the brain, as well as in crucial adaptive processes such as synaptic plasticity and neuronal survival and resistance to trauma. It is no coincidence that perhaps the most successful NMDA receptor antagonist to date,
References (92)
Targeted disruption of NMDA receptor 1 gene abolishes NMDA response and results in neonatal death
Neuron
(1994)The changing landscape of ischaemic brain injury mechanisms
Nature
(1999)Excitotoxic mechanisms of epileptic brain damage
Adv. Neurol.
(1986)Glutamate signaling and the fetal alcohol syndrome
Mental Retardation And Developmental Disabilities Research Reviews
(2001)Activity-dependent transfer of brain-derived neurotrophic factor to postsynaptic neurons
Science
(2001)The survival of developing neurons: a review of afferent control
Neuroscience
(1994)Blockade of NMDA receptors increases cell death and birth in the developing rat dentate gyrus
J. Comp. Neurol.
(1994)Chronic pre-explant blockade of the NMDA receptor affects survival of cerebellar granule cells explanted in vitro
Brain Res. Dev. Brain Res.
(1997)- et al.
Why did NMDA receptor antagonists fail clinical trials for stroke and traumatic brain injury?
Lancet
(2002) Direct evidence for biphasic cAMP responsive element-binding protein phosphorylation during long-term potentiation in the rat dentate gyrus in vivo
J. Neurosci.
(1999)