Loss of synaptophysin-positive boutons on lumbar motor neurons innervating the medial gastrocnemius muscle of the SOD1G93A G1H transgenic mouse model of ALS.

Amyotrophic lateral sclerosis (ALS) is a common form of motor neuron disease (MND) that involves both upper and lower nervous systems. In the SOD1G93A G1H transgenic mouse, a widely used animal model of human ALS, a significant pathology is linked to the degeneration of lower motor neurons in the lumbar spinal cord and brainstem. In the current study, the number of presynaptic boutons immunoreactive for synaptophysin was estimated on retrogradely labeled soma and proximal dendrites of alpha and gamma motor neurons innervating the medial gastrocnemius muscle. No changes were detected on both soma and proximal dendrites at postnatal day 60 (P60) of alpha and gamma motor neurons. By P90 and P120, however, alpha motor neuron soma had a reduction of 14 and 33% and a dendritic reduction of 19 and 36%, respectively. By P90 and P120, gamma motor neuron soma had a reduction of 17 and 41% and a dendritic reduction of 19 and 35%, respectively. This study shows that levels of afferent innervation significantly decreased on surviving alpha and gamma motor neurons that innervate the medial gastrocnemius muscle. This finding suggests that the loss of motor neurons and the decrease of synaptophysin in the remaining motor neurons could lead to functional motor deficits, which may contribute significantly to the progression of ALS/MND.

Impaired extracellular secretion of mutant superoxide dismutase 1 associates with neurotoxicity in familial amyotrophic lateral sclerosis.

Mutations in the intracellular metalloenzyme superoxide dismutase 1 (SOD1) are linked to neurotoxicity in familial amyotrophic lateral sclerosis (ALS) by an unclear mechanism. Golgi fragmentation and endoplasmic reticulum stress are early hallmarks of spinal motor neuron pathology in transgenic mice overexpressing mutant SOD1, suggesting that dysfunction of the neuronal secretory pathway may contribute to ALS pathogenesis. We therefore proposed that mutant SOD1 directly engages and modulates the secretory pathway based on recent evidence of SOD1 secretion in diverse human cell lines. Here, we demonstrate that a fraction of active endogenous SOD1 is secreted by NSC-34 motor neuron-like cells via a brefeldin-A (BFA)-sensitive pathway. Expression of enhanced green fluorescent protein-tagged mutant human SOD1 (hSOD1-EGFP) in NSC-34 cells induced frequent cytoplasmic inclusions and protein insolubility that correlated with toxicity. In contrast, transfection of non-neuronal COS-7 cells resulted in mutant hSOD1-EGFP cytoplasmic inclusions, oligomerization, and fragmentation without detectable toxicity. Importantly, impaired secretion of hSOD1-EGFP was common to all 10 SOD1 mutants tested relative to wild-type protein in NSC-34 cells. Treatment with BFA inhibited hSOD1-EGFP secretion with pronounced BFA-induced toxicity in mutant cells. Extracellular targeting of mutant hSOD1-EGFP via SOD3 signal peptide fusion attenuated cytoplasmic inclusion formation and toxicity. The effect of elevated extracellular SOD1 was then evaluated in a transgenic rat model of ALS. Chronic intraspinal infusion of exogenous wild-type hSOD1 significantly delayed disease progression and endpoint in transgenic SOD1(G93A) rats. Collectively, these results suggest novel extracellular roles for SOD1 in ALS and support a causal relationship between mutant SOD1 secretion and intraneuronal toxicity.

Magnetic resonance imaging reveals neuronal degeneration in the brainstem of the superoxide dismutase 1 transgenic mouse model of amyotrophic lateral sclerosis.

Magnetic resonance imaging (MRI) is becoming the preferred neuroimaging modality for the diagnosis of human amyotrophic lateral sclerosis (ALS). A useful animal model of ALS is the superoxide dismutase 1G93A G1H transgenic mouse, which shows many of the clinico-pathological features of the human condition. We have employed a 4.7-Tesla MRI instrument to determine whether a noninvasive imaging approach can reveal pathological changes in the nervous system of this animal model. Our T2-weighted MRI revealed consistent changes in brain and brainstem of these mice. Hyperintensities, indicative of neuropathology, were observed in several areas including the nucleus ambiguus, facial nucleus, trigeminal motor nucleus, rostroventrolateral reticular nucleus, lateral paragigantocellular nucleus and the substantia nigra. Histology analysis including neuronal counts of the imaged brains confirmed the T2-weighted MRI findings. Enlarged ventricles and hypointense striations, indicative of global atrophy, were also observed in the brain and cerebellum. This atrophy was confirmed by fresh brain weight data. The extensive global degeneration involving multiple structures suggests a multisystem disease that is similar to human ALS.

Effect of p75 neurotrophin receptor antagonist on disease progression in transgenic amyotrophic lateral sclerosis mice.

Neurotrophin level imbalances and altered p75 neurotrophin receptor (p75(NTR)) expression are implicated in spinal motor neuron degeneration in human and mouse models of amyotrophic lateral sclerosis (ALS). Recently, elevated reactive astrocyte-derived nerve growth factor (NGF) was linked to p75(NTR)-expressing motor neuron death in adult transgenic ALS mice. To test the role of NGF-dependent p75(NTR)-mediated signalling in ALS, we examined the effects of a cyclic decapeptide antagonist of p75(NTR) ligand binding by using neurotrophin-stimulated cell death assays and transgenic ALS mice. Murine motor neuron-like (NSC-34) cell cultures expressed full-length and truncated p75(NTR), tyrosine receptor kinase B (TrkB), and the novel neurotrophin receptor homolog-2 (NHR2) but were TrkA deficient. Accordingly, treatment of cells with NGF induced dose-dependent cell death, which was significantly blocked by the cyclic decapeptide p75(NTR) antagonist. Application of brain-derived neurotrophic factor, neurotrophin-3, or neurotrophin-4 to cultures increased cell proliferation, and such trophic effects were abolished by pretreatment with the tyrosine kinase inhibitor K-252a. Systemic administration of a modified cyclic decapeptide p75(NTR) antagonist conjugated to the TAT4 cell permeabilization sequence to presymptomatic transgenic SOD1(G93A) mice affected neither disease onset nor disease progression, as determined by hindlimb locomotor, grip strength, and survival analyses. These studies suggest that disrupting NGF-p75(NTR) interactions by using this approach is insufficient to alter the disease course in transgenic ALS mice. Thus, alternate ligand-independent pathways of p75(NTR) activation or additional NGF receptor targets may contribute to motor neuron degeneration in ALS mice.

Antisense peptide nucleic acid targeting GluR3 delays disease onset and progression in the SOD1 G93A mouse model of familial ALS.

Glutamate excitotoxicity is strongly implicated as a major contributing factor in motor neuron degeneration in amyotrophic lateral sclerosis (ALS). Excitotoxicity results from elevated intracellular calcium ion (Ca(2+)) levels, which in turn recruit cell death signaling pathways. Recent evidence suggests that alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor subunit (GluR) stoichiometry is a dominant factor leading to excess Ca(2+) loading in neurodegeneration. In particular, the Ca(2+) permeable glutamate receptor subunit 3 (GluR3) has been implicated in several neurologic conditions such as bipolar disorder and epilepsy. Recent proteomic analysis within our group on the copper zinc superoxide dismutase (SOD1)(G93A) transgenic mouse model of familial ALS (FALS) reveals a potentially deleterious upregulation of GluR3 in spinal cord compared to that in wild-type littermates. Based on this finding we designed a 12mer antisense peptide nucleic acid (PNA) directed against GluR3. This sequence significantly reduced levels of GluR3 protein and protected neuroblastoma x spinal cord (NSC-34) cells against death induced by the AMPA receptor-specific agonist (S)-5-fluorowillardiine. We subsequently treated SOD1(G93A) mice thrice weekly with intraperitoneal injections of the antisense PNA (2.5 mg/kg) commencing at postnatal day 50. Mice treated with the antisense sequence had significantly extended survival compared to mice injected with a nonsense sequence. Western blot analysis, however, did not reveal a significant reduction in GluR3 protein levels in whole extracts of the lumbar spinal cord. These results suggest that interference with the GluR3 component of the AMPA receptor assembly may be a novel strategy for controlling excitotoxic destruction of motor neurons and may lead to new therapeutic opportunities for the treatment of human ALS.

Inducible superoxide dismutase 1 aggregation in transgenic amyotrophic lateral sclerosis mouse fibroblasts.

High molecular weight detergent-insoluble complexes of superoxide dismutase 1 (SOD1) enzyme are a biochemical abnormality associated with mutant SOD1-linked familial amyotrophic lateral sclerosis (FALS). In the present study, SOD1 protein from spinal cords of transgenic FALS mice was fractionated according to solubility in saline, zwitterionic, non-ionic or anionic detergents. Both endogenous mouse SOD1 and mutant human SOD1 were least soluble in SDS, followed by NP-40 and CHAPS, with an eight-fold greater detergent resistance of mutant protein overall. Importantly, high molecular weight mutant SOD1 complexes were isolated with SDS-extraction only. To reproduce SOD1 aggregate pathology in vitro, primary fibroblasts were isolated and cultured from neonatal transgenic FALS mice. Fibroblasts expressed abundant mutant SOD1 without spontaneous aggregation over time with passage. Proteasomal inhibition of cultures using lactacystin induced dose-dependent aggregation and increased the SDS-insoluble fraction of mutant SOD1, but not endogenous SOD1. In contrast, paraquat-mediated superoxide stress in fibroblasts promoted aggregation of endogenous SOD1, but not mutant SOD1. Treatment of cultures with peroxynitrite or the copper chelator diethyldithiocarbamate (DDC) alone did not modulate aggregation. However, DDC inhibited lactacystin-induced mutant SOD1 aggregation in transgenic fibroblasts, while exogenous copper slightly augmented aggregation. These data suggest that SOD1 aggregates may derive from proteasomal or oxidation-mediated oligomerisation pathways from mutant and endogenous subunits respectively. Furthermore, these pathways may be affected by copper availability. We propose that non-neural cultures such as these transgenic fibroblasts with inducible SOD1 aggregation may be useful for rapid screening of compounds with anti-aggregation potential in FALS.

Brain beta-amyloid accumulation in transgenic mice expressing mutant superoxide dismutase 1.

Oxidative stress is implicated in both the deposition and pathogenesis of beta-amyloid (Abeta) protein in Alzheimer's disease (AD). Accordingly, overexpression of the antioxidant enzyme superoxide dismutase 1 (SOD1) in neuronal cells and transgenic AD mice reduces Abeta toxicity and accumulation. In contrast, mutations in SOD1 associated with amyotrophic lateral sclerosis (ALS) confer enhanced pro-oxidative enzyme activities. We therefore examined whether ALS-linked mutant SOD1 overexpression in motor neuronal cells or transgenic ALS mice modulates Abeta toxicity or its accumulation in the brain. Aggregated, but not freshly solubilised, substrate-bound Abeta peptides induced degenerative morphology and cytotoxicity in motor neuron-like NSC-34 cells. Transfection of NSC-34 cells with human wild-type SOD1 attenuated Abeta-induced toxicity, however this neuroprotective effect was also observed for ALS-linked mutant SOD1. Analysis of the cerebral cortex, brainstem, cerebellum and olfactory bulb from transgenic SOD1G93A mice using enzyme-linked immunosorbent assay of acid-guanidine extracts revealed age-dependent elevations in Abeta levels, although not significantly different from wild-type mouse brain. In addition, brain amyloid protein precursor (APP) levels remained unaltered as a consequence of mutant SOD1 expression. We therefore conclude that mutant SOD1 overexpression promotes neither Abeta toxicity nor brain accumulation in these ALS models.

Antisense peptide nucleic acid-mediated knockdown of the p75 neurotrophin receptor delays motor neuron disease in mutant SOD1 transgenic mice.

Re-expression of the death-signalling p75 neurotrophin receptor (p75NTR) is associated with injury and neurodegeneration in the adult nervous system. The induction of p75NTR expression in mature degenerating spinal motor neurons of humans and transgenic mice with amyotrophic lateral sclerosis (ALS) suggests a role of p75NTR in the progression of motor neuron disease (MND). In this study, we designed, synthesized and evaluated novel antisense peptide nucleic acid (PNA) constructs targeting p75NTR as a potential gene knockdown therapeutic strategy for ALS. An 11-mer antisense PNA directed at the initiation codon, but not downstream gene sequences, dose-dependently inhibited p75NTR expression and death-signalling by nerve growth factor (NGF) in Schwann cell cultures. Antisense phosphorothioate oligonucleotide (PS-ODN) sequences used for comparison failed to confer such inhibitory activity. Systemic intraperitoneal administration of this antisense PNA to mutant superoxide dismutase 1 (SOD1G93A) transgenic mice significantly delayed locomotor impairment and mortality compared with mice injected with nonsense or scrambled PNA sequences. Reductions in p75NTR expression and subsequent caspase-3 activation in spinal cords were consistent with increased survival in antisense PNA-treated mice. The uptake of fluorescent-labelled antisense PNA in the nervous system of transgenic mice was also confirmed. This study suggests that p75NTR may be a promising antisense target in the treatment of ALS.

Neuromuscular accumulation of mutant superoxide dismutase 1 aggregates in a transgenic mouse model of familial amyotrophic lateral sclerosis.

Superoxide dismutase 1 (SOD1) aggregates are a histological and biochemical correlate of disease progression in neural tissues from mutant SOD1-linked forms of familial amyotrophic lateral sclerosis (FALS). In the present study, we assayed the monomeric and high molecular weight mutant SOD1 content of nervous, muscle and visceral tissues from transgenic SOD1(G93A) mice using immunoblotting and zymograms. A progressive age-dependent increase in mutant SOD1 level, aggregation and stabilisation by cross-species heterodimers was determined in lumbar spinal cord, sciatic nerve and gastrocnemius muscle. Such biochemical abnormalities were not present in cervical spinal cord, brainstem and diaphragm muscle, nor common to endogenous mouse SOD1. Mutant dismutase activity in general did not increase correspondingly with accumulating protein at later ages. These results suggest that peripheral targets such as hindlimb skeletal muscle and nerve accumulate mutant SOD1 aggregates and may therefore be susceptible to mutant SOD1-mediated toxicity, in addition to lower and upper motor neurons of the central nervous system in transgenic FALS mice.

Behavioural and anatomical effects of systemically administered leukemia inhibitory factor in the SOD1(G93A G1H) mouse model of familial amyotrophic lateral sclerosis.

We investigated the anatomical and behavioural effects of daily intraperitoneal injection of 25 microg/kg of LIF in the SOD1(G93A G1H) mouse model of familial ALS. We found some subtle beneficial behavioural changes in LIF treated mice. These included later onset of clinical disease in females as determined by clinical scoring; better grip strength in males; and delayed development of motor impairment in males as determined by the rotarod test. However, we found no significant rescue of motoneurons or prolongation of survival as a result of this systemic dose of LIF in these mice.

Design and application of a peptide nucleic acid sequence targeting the p75 neurotrophin receptor.

Novel antisense peptide nucleic acid (PNA) constructs targeting p75NTR as a potential therapeutic strategy for amyotrophic lateral sclerosis (ALS) were designed, synthesised and evaluated against phosphorothioate oligonucleotide sequences (PS-ODN). An 11-mer antisense PNA directed at the initiation codon dose-dependently inhibited p75NTR expression and death signalling by nerve growth factor in Schwann cell cultures. Inhibition of p75NTR production was not detected in cultures treated with the nonsense PNA or antisense PNA directed at the 3'-terminus sequence. The 19-mer PS-ODN sequences also failed to confer any activity against p75NTR but, unlike the PNA sequences, were toxic in vitro at comparable doses.

Low-affinity neurotrophin receptor with targeted mutation of exon 3 is capable of mediating the death of axotomized neurons.

1. In vivo studies have shown that the low-affinity 75 kDa neurotrophin receptor (p75NTR) is involved in axotomy-induced cell death of sensory and motor neurons. To further examine the importance of p75NTR in mediating neuronal death in vivo, we examined the effect of axotomy in the p75NTR-knockout mouse, which has a disrupted ligand-binding domain. 2. The extent of sensory and motor neuron loss in the p75NTR-knockout mouse following axotomy was not significantly different to that in wild-type mice. This suggests that disruption of the ligand-binding domain is insufficient to block the cell death process in axotomized neurons. 3. Immunohistochemical studies showed that axotomized neurons continue to express this mutant receptor with its intracellular death-signalling moiety intact. 4. Treatment with antisense oligonucleotides targeted against p75NTR resulted in significant reduction in the loss of axotomized neurons in the knockout mouse. 5. These data suggest that the intracellular domain of p75NTR is essential for death-signalling and that p75NTR can signal apoptosis, despite a disrupted ligand-binding domain.