Glutamate AMPA receptors change in motor neurons of SOD1G93A transgenic mice and their inhibition by a noncompetitive antagonist ameliorates the progression of amytrophic lateral sclerosis-like disease

Amyotrophic lateral sclerosis (ALS) is a fatal neurological disorder involving the selective degeneration of motor neurons. In a small proportion of patients, ALS is caused by mutations in copper/zinc superoxide dismutase (SOD1), and mice overexpressing SOD1(G93A) mutant develop a syndrome that closely resembles the human disease. Excitotoxicity mediated by glutamate AMPA receptors has been suggested to be implicated in the selective susceptibility of motor neurons occurring in ALS. In SOD1(G93A) mice, we found that levels of GluR2 AMPA subunit, which plays a pivotal role in the maintenance of calcium impermeability of AMPA receptors, are decreased in spinal motor neurons before symptom onset in concomitance with a modest increase of GluR3 expression, a calcium-permeable AMPA subunit. This effect can result in a higher number of calcium-permeable AMPA receptors on motor neurons of SOD1(G93A) mice, predisposing these cells to be injured by AMPA-mediated glutamate firing. In support of this, we showed that treatment with a new noncompetitive AMPA antagonist, ZK 187638, partially protected motor neurons, improved motor function, and prolonged the survival of SOD1(G93A) mice.     

GluR2 deficiency accelerates motor neuron degeneration in a mouse model of amyotrophic lateral sclerosis

AMPA receptor-mediated excitotoxicity has been implicated in the selective degeneration of motor neurons in amyotrophic lateral sclerosis (ALS). Motor neurons in vitro are particularly vulnerable to excessive AMPA receptor stimulation and one of the factors underlying this selective vulnerability is the presence of a large proportion of Ca2+ -permeable (i.e. GluR2-lacking) AMPA receptors. However, the precise role of GluR2-lacking AMPA receptors in motor neuron degeneration remains to be defined. We therefore studied the impact of GluR2 deficiency on motor neuron death in vitro and in vivo. Cultured motor neurons from GluR2-deficient embryos displayed an increased Ca2+ influx through AMPA receptors and an increased vulnerability to AMPA receptor-mediated excitotoxicity. We deleted the GluR2 gene in mutant SOD1G93A mice by crossbreeding them with GluR2 knockout mice. GluR2 deficiency clearly accelerated the motor neuron degeneration and shortened the life span of mutant SOD1G93A mice. These findings indicate that GluR2 plays a pivotal role in the vulnerability of motor neurons in vitro and in vivo, and that therapies that limit Ca2+ entry through AMPA receptors might be beneficial in ALS patients.     

Cultured motor neurons possess calcium-permeable AMPA/kainate receptors

We examined the biology of AMPA/kainate-induced motor neuron degeneration using dissociated spinal cord cultures and motor neuron-specific antibodies which enable characterization of individual motor neurons in culture. Cobalt, which is thought to pass through Ca2+-permeable AMPA/kainate receptors following kainate exposure, labeled motor neurons in spinal cord cultures. The analysis of AMPA subunit distribution in dissociated motor neurons revealed a unique pattern of glutamate receptor (GluR) subunits in those cells; the GluR1 subunit was found in all spinal cord neurons, but the GluR2 subunit was not found in identified dissociated motor neurons. These data suggest that selective sensitivity of motor neurons to non-NMDA receptor activation is due, at least in part, to the presence of Ca2+-permeable AMPA/kainate receptors.     

The cellular mRNA expression of GABA and glutamate receptors in spinal motor neurons of SOD1 mice

ALS is a fatal neurodegenerative disorder characterized by a selective loss of upper motor neurons in the motor cortex and lower motor neurons in the brain stem and spinal cord. About 10% of ALS cases are familial, in 10-20% of these, mutations in the gene coding for superoxide dismutase 1 (SOD1) can be detected. Overexpression of mutated SOD1 in mice created animal models which clinically resemble ALS. Abnormalities in glutamatergic and GABAergic neurotransmission presumably contribute to the selective motor neuron damage in ALS. By in situ hybridization histochemistry (ISH), we investigated the spinal mRNA expression of the GABAA and AMPA type glutamate receptor subunits at different disease stages on spinal cord sections of mutant SOD1 mice and control animals overexpressing wild-type SOD1 aged 40, 80, 120 days and at disease end-stage, i.e. around 140 days) (n=5, respectively). We detected a slight but statistically significant decrease of the AMPA receptor subunits GluR3 and GluR4 only in end stage disease animals.     

Excitotoxicity and ALS: what is unique about the AMPA receptors expressed on spinal motor neurons?

It has been repeatedly reported that spinal motor neurons are selectively vulnerable to AMPA receptor-mediated excitotoxicity. Therefore, identifying the uniqueness of AMPA receptors that are expressed on motor neurons, especially in individuals affected with sporadic amyotrophic lateral sclerosis (ALS) is essential for elucidating the etiology of this disorder. The mechanism that initiates motor neuronal death appears to be an exaggerated influx of Ca(2+) through AMPA receptors. The determinants that affect this Ca(2+) influx are Ca(2+) permeability, which is regulated by the presence of the GluR2 subunit and by RNA editing at the Q/R site of GluR2; channel desensitization, which is regulated by alternative splicing at the flip/flop site and by RNA editing at the R/G site of GluR subunits; and receptor density on the cell surface, which is controlled by many factors including regulatory proteins, direct phosphorylation and RNA editing at the Q/R site. This review focuses on recent progress on the molecular dynamics of AMPA receptors and discusses the pathophysiology of selective motor neuron death mediated by AMPA receptors in individuals affected with sporadic ALS.     

Quantitative assessment of AMPA receptor mRNA in human spinal motor neurons isolated by laser capture microdissection

Disturbance of glutamate neurotransmission may contribute to the motor neuron injury seen in amyotrophic lateral sclerosis. Previous studies have suggested that human spinal motor neurons express a specific profile of the AMPA subtype of glutamate receptor with low mRNA expression for the GluR2 AMPA receptor subunit but other studies have contested this finding. The present study uses laser capture microdissection to isolate specifically identified neurons coupled with quantitative RT-PCR to demonstrate that the level of expression of the GluR2 subunit is lower in spinal motor neurons than in dorsal horn neurons from the same spinal cord region. Thus, it is likely that human spinal motor neurons express a proportion of Ca2+-permeable AMPA receptors which may contribute to the selective vulnerability of these cells in amyotrophic lateral sclerosis.     

Subcellular localization of calcium-permeable AMPA receptors in spinal motoneurons

Activation of Ca(2+)-permeable alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors has been linked to potent effects on survival and dendritic outgrowth of spinal motoneurons. Ca(2+) permeability of AMPA receptors is controlled by the GluR2 subunit. Whole-cell electrophysiological studies have suggested that GluR2-containing and GluR2-lacking AMPA receptors may coexist in individual motoneurons. However, there has not been a direct demonstration of heterogeneity in AMPA receptor subunit composition in single motoneurons, nor of distinct subcellular distributions of GluR2-containing and GluR2-lacking receptors. In the present study, we have used confocal microscopy, immunocytochemistry and Ca(2+) imaging to characterize the subcellular localization of AMPA receptors in cultured rat spinal motoneurons. Immunoreactivity for GluR2 and GluR4 was concentrated in clusters, the vast majority of which were found in dendrites at synapses. Double-labelling for GluR2 and GluR4 revealed variability in relative expression of GluR2 and GluR4 between clusters within individual motoneurons; most AMPA receptor clusters were immunoreactive for both GluR2 and GluR4, but a significant minority of clusters were immunoreactive for GluR2 only or for GluR4 only. The majority of GluR2-immunonegative AMPA receptor clusters was present in dendrites, but the relative proportion of GluR2-immunonegative and GluR2-immunopositive clusters was similar in dendrites and soma. Imaging of [Ca(2+)](i) rises triggered by AMPA receptor activation confirmed Ca(2+) influx in motoneuron dendrites. These findings strongly support a model in which GluR2-containing and GluR2-lacking AMPA receptors coexist in motoneurons, clustered at synapses, and mixed in a relative proportion that varies considerably between cell membrane microdomains.     

Altered calcium homeostasis in motor neurons following AMPA receptor but not voltage-dependent calcium channels' activation in a genetic model of amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis (ALS) is a late-onset progressive neurodegenerative disease characterized by a substantial loss of motor neurons in the spinal cord, brain stem and motor cortex. By combining electrophysiological recordings with imaging techniques, clearance/buffering capacity of cultured spinal cord motor neurons after a calcium accumulation has been analyzed in response to AMPA receptors' (AMPARs') activation and to depolarizing stimuli in a genetic mouse model of ALS (G93A). Our studies demonstrate that the amplitude of the calcium signal in response to AMPARs' or voltage-dependent calcium channels' activation is not significantly different in controls and G93A motor neurons. On the contrary, in G93A motor neurons, the [Ca(2+)](i) recovery to basal level is significantly slower compared to control neurons following AMPARs but not voltage-dependent calcium channels' activation. This difference was not observed in G93A cultured cortical neurons. This observation is the first to indicate a specific alteration of the calcium clearance linked to AMPA receptors' activation in G93A motor neurons and the involvement of AMPA receptor regulatory proteins controlling both AMPA receptor functionality and the sequence of events connected to them.     

Molecular and synaptic changes in the hippocampus underlying superior spatial abilities in pre-symptomatic G93A+/+ mice overexpressing the human Cu/Zn superoxide dismutase (Gly93 --> ALA) mutation

Although amyotrophic lateral sclerosis (ALS) is mainly considered as a motor disease, extramotor neural and cognitive alterations have also been reported in ALS patients. There is evidence that mutations in the Cu/Zn superoxide dismutase (SOD1) gene are implicated in about 20% of familiar ALS and transgenic mice overexpressing the human Cu/Zn superoxide dismutase (GLY(93) --> ALA) mutation show an ALS-like phenotype. However, while motor behavior has been extensively analyzed in these mutants, little is known on their cognitive abilities. To characterize the pre-symptomatic cognitive profile of G93A+/+ mice, we estimated their capability to detect spatial novelty and examined several indexes of their hippocampal function. We found an enhancement of spatial abilities in mutant mice associated with (1) a higher expression of hippocampal AMPA subunit GluR1 mRNA and of GluR1 protein levels, and (2) an increased induction and maintenance of long-term potentiation (LTP) at Schaffer collateral-CA1 synapses. Thus, before leading to extensive neuronal excitotoxicity, the high endogenous levels of glutamate present in the brain of pre-symptomatic G93A+/+ mice could mediate site-specific molecular and synaptic changes providing favorable conditions to spatial information processing. These findings suggest that identification of pre-symptomatic behavioral changes in murine models of ALS may point to early neural abnormalities selectively associated with mutations in the Cu/Zn superoxide dismutase (SOD1) gene.     

Novel toxicity of the unedited GluR2 AMPA receptor subunit dependent on surface trafficking and increased Ca2+-permeability

RNA editing modifies the GluR2 AMPA receptor subunit pore loop at the Q/R site and limits receptor Ca(2+) permeability. Editing failure is implicated in neurodegenerative diseases, including amyotrophic lateral sclerosis. We show that channels with unedited GluR2 are highly toxic in cultured hippocampal neurons. Toxicity exceeds that of other Ca(2+)-permeable AMPA receptor types and is influenced by agonist binding site mutations, ability to desensitize, and extracellular Ca(2+) levels. Significantly, toxicity also depends on GluR2's constitutive surface trafficking, a function dependent on GluR2 C-terminal domain interaction with NSF, a specialized chaperone. We have exploited the interaction between unedited GluR2 and NSF to reduce GluR2 surface levels. We show that a peptide that blocks the GluR2-NSF interaction reduces unedited GluR2 toxicity by reducing receptor surface expression. Peptide block of trafficking provides a model for design of drugs to reduce unedited GluR2 excitotoxicity in neurodegenerative diseases that result from editing failure.     

New role for spinal Stargazin in alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor-mediated pain sensitization after inflammation

Considerable evidence has demonstrated that alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor blockade has an antinociceptive effect on inflammatory pain. Stargazin (STG) is the first transmembrane protein known to associate with AMPA receptors and regulate their synaptic targeting. However, it is not known whether STG is involved in inflammatory pain processing by regulating AMPA receptor function. In the present study, we investigated the effect of knockdown of spinal STG on AMPA receptor-mediated pain sensitization after inflammation. Antisense technology was employed to knock down STG expression in the spinal cord. We show that STG was expressed and interacted with AMPA receptor subunit GluR2 in the spinal cord. Intrathecally injected STG antisense oligodeoxyribonucleotide (ODN) specifically decreased STG expression in the lumbar spinal cord and dose dependently inhibited formalin-induced inflammatory pain in the second phase. More important was our finding for the first time that this specific STG antisense ODN diminished AMPA (0.1 mug)-enhanced formalin pain and lost its effect if pretreated with AMPA receptor antagonist CNQX. Our results demonstrate a new role for STG in central sensitization of inflammatory pain by interacting with AMPA receptors in the spinal cord.     

Glutamate receptor plasticity and activity-regulated cytoskeletal associated protein regulation in the phrenic motor nucleus may mediate spontaneous recovery of the hemidiaphragm following chronic cervical spinal cord injury

High cervical spinal cord hemisection results in paralysis of the ipsilateral hemidiaphragm; however, functional recovery of the paralyzed hemidiaphragm can occur spontaneously. The mechanisms mediating this recovery are unknown. In chronic, experimental contusive spinal cord injury, an upregulation of the NMDA receptor 2A subunit and a downregulation of the AMPA receptor GluR2 subunit have been correlated with improved hind limb motor recovery. Therefore, we hypothesized that NR2A is upregulated, whereas GluR2 is down-regulated following chronic C2 hemisection to initiate synaptic strengthening in respiratory motor pathways. Since NMDA receptor activation can lead to the delivery of AMPA receptor subunits to the post-synaptic membrane, we also hypothesized that there would be an upregulation of the GluR1 AMPA receptor subunit and that activity-regulated cytoskeletal associated protein may mediate the post-synaptic membrane delivery. Female rats were hemisected at C2 and allowed to recover for different time points following hemisection. At these time points, protein levels of NR2A, GluR1, and GluR2 subunits were assessed via Western blot analysis. Western blot analysis revealed that there were increases in NR2A subunit at six and twelve weeks post C2 hemisection. At six, twelve, and sixteen weeks post hemisection, the GluR1 subunit was increased over controls, whereas the GluR2 subunit decreased sixteen weeks post hemisection. Immunocytochemical data qualitatively supported these findings. Results also indicated that activity-regulated cytoskeletal associated protein may be associated with the above changes. These findings suggest a role of NR2A, GluR1, and GluR2 in mediating chronic spontaneous functional recovery of the paralyzed hemidiaphragm following cervical spinal cord hemisection.     

A prolonged pharmacological blockade of type-5 metabotropic glutamate receptors protects cultured spinal cord motor neurons against excitotoxic death

The causes of amyotrophic lateral sclerosis (ALS) are mostly undefined; however, excitotoxic injury and astrogliosis may contribute to motor neuron (MN) degeneration. Group I metabotropic glutamate (mGlu) receptors are over-expressed in reactive astrocytes in ALS, but the functional significance of this over-expression is presently unknown. We examined the role of group I mGlu receptors on excitotoxic death of spinal cord MNs grown in cultures enriched of astrocytes bearing a reactive phenotype. A prolonged exposure to the selective non-competitive mGlu5 receptor antagonist MPEP reduced AMPA-mediated toxicity and cobalt uptake in MNs. Expression levels of the GluR1 (but not GluR2) AMPA receptor subunit and levels of brain-derived neurotrophic factor (BDNF) were reduced in mixed spinal cord cultures pretreated with MPEP. In addition, neuroprotection by MPEP was less than additive with that produced by a neutralizing anti-BDNF antibody and a treatment with exogenous BDNF masked the protective effect of MPEP, suggesting that mGlu5 receptors and BDNF converge in facilitating excitotoxic MN death. The protective effect of MPEP was absent in cultures with a reduced number of astrocytes. We suggest that blocking astrocytic mGlu5 receptors is a potential therapeutic strategy in ALS.     

Cerebrospinal fluid from amyotrophic lateral sclerosis patients preferentially elevates intracellular calcium and toxicity in motor neurons via AMPA/kainate receptor

Several lines of evidence in the literature purport the contribution of glutamate mediated excitotoxicity in the etiology of amyotrophic lateral sclerosis (ALS) but the cellular mechanisms responsible for selective loss of motor neurons are still obscure. Elevation of intracellular Ca(2+) is considered as the early event in glutamate mediated cell injury. We have studied the changes in [Ca(2+)](i) and cytotoxicity in motor neurons and other spinal neurons in culture upon exposure to cerebrospinal fluid (CSF) from ALS patients. CSFs from 20 ALS patients and 20 disease control patients were examined. Eighteen out of twenty (90%) ALS-CSF samples induced a transient but pronounced elevation of [Ca(2+)](i) in neurons, whereas only 1/20 (5%) sample from disease control patients induced a marginal elevation of [Ca(2+)](i). Strikingly the [Ca(2+)](i) rise was 2-3-fold higher and longer lasting in motor neurons in comparison to the other spinal neurons. Exposure of cells to ALS-CSF drastically decreased the survival rate of motor neurons to 32.26+/-2.06% whereas a moderate decrease was observed in case of other spinal neurons (67.90+/-2.04%). In cultures treated with disease control CSF, a small decrease was observed in the survival rate with 80.14+/-2.00% and 90.07+/-1.37% survival of motor neuron and other spinal neurons respectively. The AMPA/kainate receptor antagonist NBQX rendered significant protection against the ALS-CSF induced Ca(2+) influx and neurotoxicity while the NMDA receptor antagonist APV showed a mild effect. Our data demonstrate that the exposure of spinal cord neurons to ALS-CSF differentially elevates [Ca(2+)](i) and neurotoxicity in motor neurons by activation of glutamate receptors, the AMPA/kainate receptor playing the major role.     

AMPA receptor‐mediated neuronal death in sporadic ALS

α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionate (AMPA) receptor‐mediated excitotoxicity has been proposed to play a role in death of motor neurons in amyotrophic lateral sclerosis (ALS). We demonstrated that RNA editing of GluR2 mRNA at the glutamine/arginine (Q/R) site was decreased in autopsy‐obtained spinal motor neurons, but not in cerebellar Purkinje cells, of patients with sporadic ALS. This molecular change occurs in motor neurons of sporadic ALS cases with various phenotypes, but not in degenerating neurons of patients with other neurodegenerative diseases, including SOD1‐associated familial ALS. Because GluR2 Q/R site‐editing is specifically catalyzed by adenosine deaminase acting on RNA 2 (ADAR2), it is likely that regulatory mechanism of ADAR2 activity does not work well in the motor neurons of sporadic ALS. Indeed, ADAR2 expression level was significantly decreased in the spinal ventral gray matter of sporadic ALS as compared to normal control subjects. It is likely that ADAR2 underactivity selective in motor neurons induced deficient GluR2 Q/R site‐editing, which results in the neuronal death of sporadic ALS. Thus, among multiple different molecular mechanisms underlying death of motor neurons, it is likely that an increase of the proportion of Q/R site‐unedited GluR2‐containing Ca²⁺‐permeable AMPA receptors initiates the death of motor neurons in sporadic ALS. To this end, normalization of ADAR2 activity in motor neurons may become a therapeutic strategy for sporadic ALS.     

Glutamate receptor subunit expression after spinal cord injury in young rats

To investigate the possibility that glutamate receptor levels in the spinal cord are altered following injury to young rats, we used a previously characterized model of spinal cord contusion that produces a reliable injury in rats at postnatal day 14-15. Quantitative Western blot analysis was used to measure relative amounts of protein for several glutamate receptor subunits acutely (24 h) and chronically (28 days) after spinal cord injury (SCI). Acutely after injury significant decreases were observed in the GluR1, GluR2, and GluR4 subunits of the alpha-amino-3-hydroxy-5-methyl-4-isoxazole proprionate (AMPA) receptor, and the NR2A and NR2B subunits, but not the NR1 subunit, of the N-methyl-d-aspartate (NMDA) receptor. However, 28 days after injury only one subunit (GluR4) was shown to be altered. These widespread changes that occur acutely in receptor subunit expression may be an attempt to protect cells from glutamate-induced death. The injured spinal cord in these young animals, however, appears to have the capacity to regulate receptor subunit levels to normal within a month of injury.