Expressions of nitrotyrosine and TUNEL immunoreactivities in cultured rat spinal cord neurons after exposure to glutamate, nitric oxide, or peroxynitrite

Although excitotoxic and oxidative stress play important roles in spinal neuron death, the exact mechanism is not fully understood. We examined cell damage of primary culture of 11-day-old rat spinal cord by addition of glutamate, nitric oxide (NO) or peroxynitrite (PN) with detection of nitrotyrosine (NT) or terminal deoxynucleotidyl transferase-mediated dUTP-biotin in situ nick end labeling (TUNEL). With addition of glutamate, NOC18 (a slow NO releaser) or PN, immunoreactivity for NT became stronger in the cytoplasm of large motor neurons in the ventral horn at 6 to 48 hr and positive in the axons of the ventral horn at 24 to 48 hr. TUNEL positive nuclei were found in spinal large motor neurons from 24 hr, and the positive cell number greatly increased at 48 hr in contrast to the vehicle. Pretreatment of cultures with alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)/kainate receptor antagonist, NO-suppressing agent, and antioxidant protected the immunoreactivity for NT or TUNEL. The present results suggest that both excitotoxic and oxidative stress play an important role in the upregulation of NT nitration and the apoptotic pathway in cultured rat spinal neurons.     

Gene therapy and neurotrophic factor treatment for amyotrophic lateral sclerosis]

Although excitotoxic and oxidative stress play important roles in spinal neuron death, the exact mechanisms are not fully understood. We examined cell damage of primary culture of 11-day-old rat spinal cord by addition of glutamate, nitric oxide (NO) or peroxynitrite (PN) with detection of terminal deoxynucleotidyl transferase-mediated dUTP-biotin in situ nick end labeling (TUNEL). With addition of glutamate, NOC18 (a slow NO releaser) or PN, TUNEL positive nuclei were found in spinal large motor neurons from 24 h, and the positive cell proportion greatly increased at 48 h in contrast to the vehicle. The present results suggest that both excitotoxic and oxidative stress play important role in the apoptotic pathway in cultured rat spinal neurons. To examine a possible protective effect of exogenous glial cell line-derived neurotrophic factor (GDNF) gene expression in transgenic (Tg) mice carrying a Gly 93Ala (G93A) mutant SOD1 gene found in human familial ALS, a replication defective adenoviral vector containing GDNF gene was directly injected unilaterally into leg muscles. There were significantly more large motoneurons in GDNF-treated Tg mice than in untreated and Ad-Laz-treated group. The number of large motoneurons in GDNF-treated side of Tg mice were significantly more than that in untreated side. These observations demonstrate that GDNF gene therapy in a mouse model of FALS promotes the survival of motoneurons, suggesting that a similar approach might delay the progression of neurodegeneration of ALS.     

Glutamate enhances caspase-3 immunoreactivity in cultured spinal cord neurons of newborn rats

The role of glutamate in the mechanism of spinal neuron death is not fully understood. With addition of glutamate to primary culture of 11-day-old rat spinal cord, the number of caspase-3 positive small neurons of the dorsal horn greatly increased at 6-24 h in contrast to the case with vehicle. The addition of glutamate made caspase-3 immunoreactivity stronger in the cytoplasm of large motor neurons in the ventral horn. The present results show that excessive amount of glutamate enhances apoptotic pathway through caspase-3 in cultured spinal neurons of newborn rat.     

Glutamate enhances caspase-3 immunoreactivity in cultured spinal cord neurons of newborn rats

Abstract

The role of glutamate in the mechanism of spinal neuron death is not fully understood. With addition of glutamate to primary culture of 11-day-old rat spinal cord, the number of caspase-3 positive small neurons of the dorsal horn greatly increased at 6-24 h in contrast to the case with vehicle. The addition of glutamate made caspase-3 immunoreactivity stronger in the cytoplasm of large motor neurons in the ventral horn. The present results show that excessive amount of glutamate enhances apoptotic pathway through caspase-3 in cultured spinal neurons of newborn rat.     

Glutamate enhances DNA fragmentation in cultured spinal motor neurons of rat

The role of glutamate in the mechanism of spinal motor neuron death is not fully understood. With addition of glutamate to primary culture of 11-day-old rat spinal cord, terminal deoxynucleotidyl transferase-mediated dUTP-biotin in situ nick end labeling (TUNEL) positive nuclei were found in spinal large motor neurons from 24 h, and the number of TUNEL positive large motor neurons greatly increased at 48 h. In contrast, only a small number of large motor neurons became TUNEL positive at 48 h with addition of vehicle to the primary spinal cord culture. The present results show that excessive amount of glutamate enhances DNA fragmentation in developing large motor neuron of cultured spinal cord by involving in apoptotic process of the neurons.     

Protein modification by the lipid peroxidation product 4-hydroxynonenal in the spinal cords of amyotrophic lateral sclerosis patients

We report increased modification of proteins by 4-hydroxynoneal (HNE), a product of membrane lipid peroxidation, in the lumbar spinal cord of sporadic amyotrophic lateral sclerosis (ALS) patients versus that of neurologically normal controls. By immunohistochemistry, HNE-protein modification was detected in ventral horn motor neurons, and immunoprecipitation analysis revealed that one of the proteins modified by HNE was the astrocytic glutamate transporter EAAT2. Given that the function of proteins modified by HNE can be severely compromised as previously demonstrated for glutamate transporters in cortical synaptosome preparatations, our findings suggest a scenario in which oxidative stress leads to the production of HNE, impairment of gluatmate transport, and excitotoxic motor neuron degeneration in ALS.     

Characterization of spinal HSP72 induction and development of ischemic tolerance after spinal ischemia in rats

Induction of heat shock protein (HSP72) has been implicated in the development of ischemic tolerance in several tissue organs including brain and spinal cord. In the present study, using an aortic balloon occlusion model in rats, we characterized the effect of transient noninjurious (3 or 6 min) or injurious intervals (10 min) of spinal ischemia followed by 4-72 h of reflow on spinal expression of HSP72 and GFAP protein. In a separate group of animals, the effect of ischemic preconditioning (3 or 6 min) on the recovery of function after injurious interval of spinal ischemia (10 min) was studied. After 3 min of ischemia, there was a modest increase in HSP72 protein immunoreactivity in the dorsal horn neurons at 12 h after reperfusion. After 6 min of ischemia, a more robust and wide spread HSP72 protein expression in both dorsal and ventral horn neurons was detected. The peak of the expression was seen at 24 h after ischemia. At the same time point, a significant increase in spinal tissue GFAP expression was measured with Western blots and corresponded morphologically with the presence of activated astrocytes in spinal segments that had been treated similarly. After 10 min of ischemia and 24 h of reflow, a significant increase in spinal neuronal HSP72 expression in perinecrotic regions was seen. Behaviorally, 3 min preconditioning ischemia led to the development of a biphasic ischemic tolerance (the first at 30 min and the second at 24 h after preconditioning) and was expressed as a significantly better recovery of motor function after exposure to a second 10-min interval of spinal ischemia. After 6 min ischemic preconditioning, a more robust ischemic tolerance at 24 h after preconditioning then seen after 3-min preconditioning was detected. These data indicate that 3 min of spinal ischemia represents a threshold for spinal neuronal HSP72 induction, however, a longer sublethal interval (6 min) of preconditioning ischemia is required for a potent neuronal HSP72 induction. More robust neurological protection, seen after 6 min of preconditioning ischemia, also indicates that HSP72 expression in spinal interneurons seen at 24 h after preconditioning may represent an important variable in modulating ischemic tolerance observed during this time frame.     

Temporal and spatial distribution of activated caspase-3 after subdural kainic acid infusions in rat spinal cord

The molecular events initiating apoptosis following traumatic spinal cord injury (SCI) remain poorly understood. Soon after injury, the spinal cord is exposed to numerous secondary insults, including elevated levels of glutamate, that contribute to cell dysfunction and death. In the present study, we attempted to mimic the actions of glutamate by subdural infusion of the selective glutamate receptor agonist, kainic acid, into the uninjured rat spinal cord. Immunohistochemical colocalization studies revealed that activated caspase-3 was present in ventral horn motor neurons at 24 hours, but not 4 hours or 96 hours, following kainic acid treatment. However, at no time point examined was there evidence of significant neuronal loss. Kainic acid resulted in caspase-3 activation in several glial cell populations at all time points examined, with the most pronounced effect occurring at 24 hours following infusion. In particular, caspase-3 activation was observed in a significant number of oligodendroglia in the dorsal and ventral funiculi, and there was a pronounced loss of oligodendroglia at 96 hours following treatment. The results of these experiments indicate a role for glutamate as a mediator of oligodendroglial apoptosis in traumatic SCI. In addition, understanding the apoptotic signaling events activated by glutamate will be important for developing therapies targeting this cell death process.     

Selective vulnerability of spinal cord motor neurons to non-NMDA toxicity

We previously reported that alpha-motor neurons in organotypic cultures of rat spinal cord (OTC-SC) are resistant to excitotoxicity induced through NMDA receptors. Here we describe the effects of non-NMDA glutamate receptor agonists kainic acid (KA) and quisqualic acid (QUIS) on motor neurons in OTC-SC. Large ventral horn acetylcholinesterase-positive neurons (VHANs), most of which are motor neurons, were quite sensitive to QUIS and KA toxicity and displayed losses of 95% and 94%, respectively. Small VHANs were reduced by 41% and 61% only. Identical results were obtained in cultures stained for non-phosphorylated neurofilaments. These observations demonstrate that alpha-motor neurons are considerably more sensitive to KA and QUIS than to NMDA toxicity. The proposed excitotoxic mechanism of ALS, therefore, is most likely mediated through non-NMDA glutamate receptors.     

Beta-amyloid 42 accumulation in the lumbar spinal cord motor neurons of amyotrophic lateral sclerosis patients

Amyotrophic lateral sclerosis (ALS) is characterized by a progressive loss of large motor neurons in the brain and spinal cord. Amyloid precursor protein (APP), the transmembrane precursor of beta-amyloid (A beta), accumulates in the anterior horn motor neurons of ALS patients with mild lesions. APP undergoes an alternative proteolysis mediated by caspase-3, which is activated in motor neurons in a mouse model of ALS. The ALS spinal cord motor neurons also show evidence of increased oxidative damage, which is thought to alter APP processing. We sought to determine whether A beta42, the more pathogenic A beta species, accumulates in the postmortem lumbar spinal cord of ALS patients. While there was little or no A beta42 labeling in control spinal cord tissues, elevated A beta42 immunoreactivity occurred in ALS motor neuronal perikarya and axonal swellings in the anterior horn. A few A beta42-positive neurons exhibited thioflavine S staining. No extracellular A beta42 deposits were found. A beta42 coexisted with the oxidative damage markers malondialdehyde, 8-hydroxydeoxyguanosine, heme oxygenase-1, and nitrotyrosine in abnormal neurons. The neurons with intracellular A beta42 accumulation also displayed robust cleaved caspase-3 immunoreactivity. Very little A beta40 immunoreactivity occurred in motor neurons of both control and ALS. These results suggest that aberrant accumulation of A beta42 in ALS spinal cord motor neurons is associated with oxidative stress, and may play a role in the pathogenesis of neurodegeneration in ALS.     

Motor neurons rapidly accumulate DNA single-strand breaks after in vitro exposure to nitric oxide and peroxynitrite and in vivo axotomy

The mechanisms of neuronal degeneration in motor neuron disease are not fully understood. We tested the hypothesis that oxidative stress in vitro and axotomy in vivo induce single-strand breaks (SSB) in DNA, a form of early DNA damage, in adult motor neurons early during their degeneration. We developed and characterized a novel cell suspension system enriched in motor neurons from adult rat spinal cord ventral horn. This cell system is approximately 84% neurons, with approximately 86% of these neurons being motor neurons; approximately 72% of these motor neurons are alpha-motor neurons. Motor neuron viability in suspension is approximately 100% immediately after isolation and approximately 61% after 12 hours of incubation. During incubation, isolated motor neurons generate high levels of superoxide. We used single-cell gel electrophoresis (comet assay) to detect DNA-SSB in motor neurons. Exposure of motor neurons to nitric oxide (NO) donors (sodium nitroprusside or NONOate), H2O2, or NO donor plus H2O2 rapidly induces DNA-SSB and causes motor neuron degeneration, the occurrence of which is dose and time related, as represented by comet formation and cell loss. Motor neuron toxicity is potentiated by cotreatment with NO donor and H2O2 (at nontoxic concentrations alone). Peroxynitrite causes DNA-SSB in motor neurons. The DNA damage profiles (shown by the comet morphology and moment) of NO donors, NO donor plus H2O2, and peroxynitrite are similar. In an in vivo model of motor neuron apoptosis, DNA-SSB accumulate slowly in avulsed motor neurons before apoptotic nuclear features emerge, and the comet fingerprint is similar to NO toxicity. We conclude that motor neurons challenged by oxidative stress and axotomy accumulate DNA-SSB early in their degeneration and that the formation of peroxynitrite is involved in the mechanisms.     

Light and electron microscopic distribution of the AMPA receptor subunit,GluR2, in the spinal cord of control and G86R mutant superoxide dismutase transgenic mice

Excitotoxicity has been hypothesized to contribute to amyotrophic lateral sclerosis (ALS) neurodegeneration. The similar pattern of vulnerability in the spinal cord of mutant superoxide dismutase (SOD-1) transgenic mice and mice treated with excitotoxins supports a role for excitotoxicity in the mechanism of degeneration. The distribution of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) class of glutamate receptors (GluRs) with different calcium permeabilities has been proposed as an explanation for this differential vulnerability. GluR2 appears to be the dominant determinant of calcium permeability for AMPA receptors; thus, it is critical for their contribution to excitotoxic mechanisms. In this study, we investigate the distribution of GluR2 immunoreactivity in the spinal cord of control and SOD-1 transgenic mice. GluR2 immunoreactivity is present equally within vulnerable neurons (i.e., motor neurons and calretinin-immunoreactive neurons) as well as nonvulnerable neurons (i.e., calbindin-immunoreactive neurons and dorsal horn neurons). In addition, postembedding immunoelectron microscopy reveals that GluR2 is present in synapses of dorsal and ventral horn neurons and that the percentage of labeled synapses and numbers of immunogold particles per synapse do not vary between these spinal cord regions. Comparing control mice with SOD-1 transgenic mice, at both the light and the electron microscopic levels, the distribution and intensity of GluR2-immunoreactivity do not appear to be altered. These results suggest that the cellular and synaptic distribution of GluR2 is not a determinant of the selective vulnerability observed in SOD-1 transgenic mice or in ALS patients. J. Comp. Neurol. 395:523–534, 1998. © 1998 Wiley-Liss, Inc.     

Ontogeny of descending serotonergic innervation and evidence for intraspinal 5-HT neurons in the mouse spinal cord

Neuronal networks in the mouse spinal cord express serotonin (5-HT)-induced rhythmic motor activity at early developmental stages (embryonic day (E) 12.5). Later in development, by post-natal day (P) 10, the 5-HT-evoked rhythmic motor activity matures and acquires an adult locomotor-like pattern. With the view to establishing a relationship between the ontogeny of locomotor networks and the maturation of spinal 5-HT systems, we have traced 5-HT immunoreactivity in the mouse spinal cord from E12.5 to PN10. By E12.5, descending 5-HT immunoreactive (5-HT-ir) fibers that likely originate from raphe nuclei were detected in the ventral and lateral funiculi, at anterior cervical spinal levels, but not at more caudal levels. Descending 5-HT-ir axons reached thoracic levels at E14.5 and lumbar levels at E16.5. Some 5-HT-ir fibers could be detected in the ventral and intermediate gray matter by E16.5, whereas the dorsal gray matter was not invaded before PN0. At PN10, a dense serotonergic innervation was restricted to the gray matter with a high concentration of 5-HT-ir fibers in three areas: dorsal horn, ventral horn (where motoneurons are located) and intermediate area. Surprisingly, from E16.5 to PN10, 5-HT-ir intraspinal neurons were found, exclusively at sacral levels. Their somata lay in the gray matter around the central canal and preferentially in the ventro-median part of the ventral horn. The functional significance of these sacral 5-HT-ir neurons is discussed.     

Expression of metabotropic glutamate receptor mRNAs in the human spinal cord: implications for selective vulnerability of spinal motor neurons in amyotrophic lateral sclerosis

To elucidate the relevance of metabotropic glutamate receptors (mGluRs) to the selective vulnerability of motor neurons in the spinal cord in patients with amyotrophic lateral sclerosis (ALS), we investigated the distribution of mRNAs coding mGluR1-5 in the normal human spinal cord. The mRNAs for mGluR1, 4 and 5 were observed in the spinal gray matter, whereas mGluR2 mRNA was absent in the spinal cord and mGluR3 mRNA was displayed only on glial cells in the white matter. Signals for mGluR1 and mGluR5 were enriched in the dorsal horn, while mGluR4 mRNA was abundant in the ventral horn. Since agonists to group I mGluRs (mGluR 1 and 5) have been demonstrated to have neuroprotective effects on spinal motor neurons, less expression of mRNAs coding mGluR1 and mGluR5 in the ventral horn than in the dorsal horn may be implicated in the selective susceptibility of spinal motor neurons in ALS.     

Glutamate uptake is attenuated in spinal deep dorsal and ventral horn in the rat spinal nerve ligation model

Alteration of glutamatergic (GLU) neurotransmission within the spinal cord contributes to hyperalgesic and allodynic responses following nerve injury. In particular, changes in expression and efficacy of glutamate transporters have been reported. Excitatory, pain transmitting primary afferent neurons utilizing glutamate as an excitatory neurotransmitter project to both superficial (I-II) and deep (III-V) laminae of the dorsal horn. These experiments were designed to examine changes in glutamate uptake occurring concomitantly within the spinal deep dorsal and ventral horn in situ after experimentally induced neuropathic pain. In vivo voltammetry, using microelectrode arrays configured for enzyme-based detection of GLU were employed. Sprague-Dawley rats had either sham surgery or tight ligation of L5 and L6 spinal nerves (SNL). Four to six weeks later, the L4-L6 spinal cord of chloral hydrate-anesthetized animals was exposed, and ceramic-based glutamate microelectrodes equipped with glass micropipettes 50 microm from the recording surfaces were placed stereotaxically at sites within the spinal cord. Pressure ejection of GLU into the ipsilateral L5-L6 spinal cord resulted in a 72% reduction of GLU uptake in SNL rats compared to sham controls in the ipsilateral L5-L6 deep dorsal horn and a 96% reduction in the ventral horn. In contrast, in the same animals, the contralateral L5-L6 or the ipsilateral L4 spinal cord showed no change in glutamate uptake. The data suggest that spinal nerve ligation produced attenuated glutamate uptake activity extending into the deep dorsal and ventral horn. The study suggests that plasticity related to spinal nerve injury produces widespread alteration in glutamate transporter function that may contribute to the pathophysiology of neuropathic pain.     

Interneuronal synapses formed by motor neurons appear to be glutamatergic

Acetylcholine release at motor neuron synapses has been long established; however, recent discoveries indicate that synaptic transmission by motor neurons is more complex than previously thought. Using whole-cell patch clamp, we show that spontaneous excitatory postsynaptic currents of rat motor neurons in primary ventral horn cultures are entirely glutamatergic, although the cells respond to exogenous acetylcholine. Motor neurons in cultures express the vesicular glutamate transporter VGlut2, and culturing motor neurons for weeks with glutamate receptors blocked upregulates glutamate signaling without increasing cholinergic signaling. In spinal cord slices, motor neurons showed no decrease in spontaneous excitatory synaptic potentials after blocking acetylcholine receptors. Our results suggest that motor neuron synapses formed on other neurons are largely glutamatergic in culture and the spinal cord.     

Amyotrophic lateral sclerosis: glutamate dehydrogenase and transmitter amino acids in the spinal cord.

Measurements were taken of the activity of glutamate dehydrogenase (GDH) and the levels of transmitter amino acids in anatomically dissected regions of cervical and lumbar spinal cord in eight patients dying with amyotrophic lateral sclerosis (ALS) and in 11 neurologically normal controls. GDH activity was considerably increased in lateral and ventral white matter and in the dorsal horn of the ALS cervical spinal cord, but normal in the ventral horn and the dorsal columns. Similar, although less pronounced, GDH changes were found in the lumbar enlargement. The mean concentrations of aspartate and glutamate were reduced in all regions of ALS spinal cord investigated. Taurine concentrations were significantly increased in several subdivisions of cervical spinal cord, but normal in lumbar regions. Glycine levels were significantly reduced in lumbar ventral and dorsal horns. There was no striking change in spinal cord GABA levels in our ALS patients. It is suggested that the reduced levels of glutamate and aspartate as well as the elevated GDH activity in the spinal cord of ALS patients may reflect an overactivity of the neurons releasing these potentially excitotoxic amino acids and thus may be causally related to the spinal neuro-degenerative changes characteristic of ALS.     

Enhanced Fos expression in rat lumbar spinal cord cultured with cerebrospinal fluid from patients with amyotrophic lateral sclerosis.

The etiology of amyotrophic lateral sclerosis (ALS) remains unknown although an existence of neurotoxic substances in cerebrospinal fluid (CSF) from ALS patients have been postulated. In order to investigate a possible effect of CSF from ALS patients on cellular signaling in spinal neurons, we compared Fos-like immunoreactivity (Fos-LI) in organotypic cultures of rat lumbar spinal cord after addition of CSF from ALS patients or another neurologic disease. Fos-LI was normally present predominantly in dorsal horn neurons, whereas only a few ventral horn neurons were positive for Fos-LI. The number of Fos-LI positive neurons significantly increased in dorsal horn with addition of CSF from ALS patients as well as glutamate at 100 microM. However, the increase was not observed with addition of CSF from other neurologic diseases. The increase in Fos-LI positive neurons in dorsal horn was reversed by a further supplement of MK801, an N-methyl-D-aspartate (NMDA) receptor antagonist, but not of CNQX, an alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)/kainate antagonist. These results indicate that there may be substances in CSF from ALS patients that stimulate Fos expression in certain populations of spinal neurons via the NMDA receptors.     

Spinal cord injury induced heat shock protein expression is reduced by an antioxidant compound H-290/51. An experimental study using light and electron microscopy in the rat

Summary. The possibility that oxidative stress participates in heat shock protein 72 kD (HSP 72) expression following a focal trauma to the spinal cord was examined using a potent antioxidant compound H-290/51 in a rat model. A focal spinal cord injury (SCI) inflicted by making a longitudinal incision on the right dorsal horn of the T10–T11 segment under equithesin anaesthesia resulted in profound upregulation of HSP 72 expression in the adjacent spinal cord segments T9 and T12. This expression of HSP was most marked in the ipsilateral cord at 5 h after SCI. Pretreatment with H-290/51 (50 mg/kg, p.o.) 30 min before SCI markedly attenuated HSP expression in the spinal cord seen at 5 h. The motor functions of traumatized rats were also improved in the drug treated group. At this time, structural changes in the spinal cord and edema formation were considerable reduced compared to the untreated traumatized rats. Taken together, these observations suggest that (i) oxidative stress participates in HSP response following trauma, and (ii) the antioxidant compound H-290/51 attenuates cellularstress, improves motor functions and induces considerable neuroprotection in the early phase of SCI. Further studies using post-injury treatment with H-290/51 is needed to explore its therapeutic potentials in clinical settings.     

Motor neuron impairment mediated by a sumoylated fragment of the glial glutamate transporter EAAT2

Dysregulation of glutamate handling ensuing downregulation of expression and activity levels of the astroglial glutamate transporter EAAT2 is implicated in excitotoxic degeneration of motor neurons in amyotrophic lateral sclerosis (ALS). We previously reported that EAAT2 (a.k.a. GLT-1) is cleaved by caspase-3 at its cytosolic carboxy-terminus domain. This cleavage results in impaired glutamate transport activity and generates a proteolytic fragment (CTE) that we found to be post-translationally conjugated by SUMO1. We show here that this sumoylated CTE fragment accumulates in the nucleus of spinal cord astrocytes of the SOD1-G93A mouse model of ALS at symptomatic stages of disease. Astrocytic expression of CTE, artificially tagged with SUMO1 (CTE-SUMO1) to mimic the native sumoylated fragment, recapitulates the nuclear accumulation pattern of the endogenous EAAT2-derived proteolytic fragment. Moreover, in a co-culture binary system, expression of CTE-SUMO1 in spinal cord astrocytes initiates extrinsic toxicity by inducing caspase-3 activation in motor neuron-derived NSC-34 cells or axonal growth impairment in primary motor neurons. Interestingly, prolonged nuclear accumulation of CTE-SUMO1 is intrinsically toxic to spinal cord astrocytes, although this gliotoxic effect of CTE-SUMO1 occurs later than the indirect, noncell autonomous toxic effect on motor neurons. As more evidence on the implication of SUMO substrates in neurodegenerative diseases emerges, our observations strongly suggest that the nuclear accumulation in spinal cord astrocytes of a sumoylated proteolytic fragment of the astroglial glutamate transporter EAAT2 could participate to the pathogenesis of ALS and suggest a novel, unconventional role for EAAT2 in motor neuron degeneration.