C82 THE POTENTIAL ROLE OF SOD IN THE PATHOGENESIS OF SPORADIC ALS
LUI J
California Pacific Medical Center, San Francisco, CA, USA
Email address for correspondence: [email protected]
More than 90% of ALS is of sporadic nature. Molecular mechanisms of sporadic ALS (sALS) are more difficult to study compared to those of known inherited forms of ALS. Studies of the SOD1-linked familial form of ALS over the last 15 years have developed increasing amounts of information and knowledge, together with a realization that the immense complexity of sALS limits the application of this model. Different approaches to identify genes and proteins that are involved in sALS will be reviewed. Recent work on a known blood enzyme that may be potentially important in sALS will be reported. Finally, thoughts on how one might approach studying sALS in the context of other neurodegenerative diseases that are largely of a sporadic nature will be discussed.
C83 MODULATION OF ENDOPLASMIC RETICULUM STRESS AS A POTENTIAL TREATMENT FOR ALS
WALKER A, FARG M, TURNER B, HORNE M, ATKIN J
Howard Florey Institute, University of Melbourne, Victoria, Australia
E-mail address for correspondence: [email protected]
Keywords: SOD1, aggregation, ER stress
Background: The causes of motor neuron death in ALS remain unknown, although protein misfolding and accumulation of disulfide bonded mutant SOD1 aggregates is linked to neurotoxicity in some familial forms of disease. The unfolded protein response (UPR), a homeostatic mechanism activated by accumulation of misfolded proteins within the endoplasmic reticulum (ER), is activated in mutant SOD1 expressing transgenic rodents as well as in spinal cords of human patients with non-SOD1 linked sporadic ALS. The UPR involves upregulation of chaperones and down-regulation of general protein synthesis, however if the process is prolonged ER stress occurs and apoptosis is triggered via CHOP, JNK and ER stress-specific caspases. Protein disulfide isomerase (PDI), a key ER stress response chaperone with disulfide bond modulating activity, is also upregulated as part of the ER stress response and is protective against mutant SOD1 aggregation in vitro.
Objectives: The aim of this study was to examine the effects of a small molecular mimic of PDI and ER stress modulating drugs on mutant SOD1 aggregation and ER stress in ALS models.
Methods: Motor neuron-like NSC34 cell lines expressing either wildtype or mutant SOD1-EGFP fusion proteins were treated with a small molecular mimic of PDI or pharmacological modulators of ER stress. Immunocytochemistry, confocal microscopy and immunoblotting were performed to detect protein inclusions and high-molecular weight insoluble protein aggregates, as well as markers of ER stress.
Results: Treatment with the PDI mimic or ER stress inhibitor decreased: the formation of large mutant SOD1 inclusions, levels of insoluble SOD1 protein and toxicity. Conversely, pharmacological induction of the ER stress response had the opposite effect. Mutant SOD1 induced expression of ER stress response proteins, including the apoptotic transcription factor CHOP, this induction was decreased in cells treated with either the PDI mimic or ER stress inhibitor.
Discussion and Conclusions: We have shown that a small molecule mimic of PDI retains PDI-like activity in decreasing mutant SOD1 aggregation and preventing induction of ER stress, and that inhibition of ER stress also decreases protein aggregation in vitro. These data suggest that PDI and the ER stress pathways are possible therapeutic targets for treatment of ALS, and further in vivo experiments will clarify whether or not these compounds could be beneficial in altering ALS disease course.
C84 TRAFFICKING FROM ER TO GOLGI IS DISRUPTED IN MUTANT SOD1 EXPRESSING CELLS
ATKIN J, FARG M, WALKER A, TURNER B, TINSLEY R, HORNE M
Howard Florey Institute, University of Melbourne, Australia
E-mail address for correspondence: [email protected]
Keywords: ER stress, cellular trafficking, SOD1 aggregation
Background: We and others showed recently that (a) SOD1 is secreted by an ER to Golgi-dependent route and dysfunction of the secretory pathway is linked to neurotoxicity in ALS and (b) that Endoplasmic reticulum (ER) stress is present in lumbar spinal cords of transgenic SOD1G93A mice and human patients with sporadic disease. The up-regulation of ER stress specific apoptotic markers (caspase-12 and CHOP) in these tissues suggests that this is an important pathway to cell death in both forms of ALS.
Objective: To determine the relationship between secretory pathway dysfunction, SOD1 aggregation and ER stress in ALS
Methods: NSC34 cells expressing mutant and wildtype SOD1 were examined by Western blotting and immunohistochemistry for the expression and distribution of ER and Golgi trafficking proteins.
Results: In cells expressing mutant SOD1, CHOP was up-regulated prior to the formation of the aggresome, indicating that the smaller aggregates or oligomers are responsible for ER stress and toxicity in ALS. Trafficking proteins of the ER Golgi Intermediate Compartment (ERGIC) were down-regulated in comparison to cells expressing wildtype SOD1. Vesicle associated protein COPII also had an altered distribution and co-localised with SOD1 aggregates. However, markers of the endosome pathway were not altered in these cells indicating that the post-Golgi secretory pathway was not affected by mutant SOD1 expression.
Conclusion: Small ‘microaggregates’ or oligomers of mutant SOD1 lead to ER stress, dysfunction of the ERGIC compartment, and neurotoxicity.
C85 ZINC-DEFICIENT MONOMERS ARE WELL-POPULATED SOD1 UNFOLDING INTERMEDIATES: IMPLICATIONS FOR AMYOTROPHIC LATERAL SCLEROSIS PATHOGENESIS
MULLIGAN V1, KERMAN A2, HO S2, CHAKRABARTTY A2
1Department of Biochemistry, University of Toronto, 2Department of Biochemistry and Medical Biophyics, Ontario Cancer Instutute, University of Toronto, Ontario, Canada
E-mail address for correspondence: [email protected]
Keywords: SOD1, protein folding, kinetics
Background: Mutations in the Cu,Zn superoxide dismutase (SOD1) are known to cause a subset of cases of familial amyotrophic lateral sclerosis (ALS). This protein is a homodimer in which each monomer binds one catalytic copper atom and one structural zinc atom. Mutant SOD1 cytoxicity is believed to be due to increased conformational flexibility, which either allows aberrant substrates to access the SOD1 active site, giving rise to novel toxic catalytic activity, or increases the propensity of SOD1 to misfold and to aggregate. SOD1 protein stability is greatly affected by its metal ligands, but little is known about the role these ligands play in the folding, unfolding, and misfolding processes.
Objectives: This study aims to generate a more detailed model of the SOD1 unfolding process, focussing particularly on the release of bound copper and zinc as the protein denatures. Such a model would identify potentially disease-relevant folding intermediates and describe their metallation states. These intermediates may be useful targets for therapeutic agents designed to slow or halt ALS disease progression.
Method: SOD1 denaturation in molar concentrations of guanidine HCl was monitored in several ways. A variation of the 4-(2-pyridylazo)resorcinol (PAR) assay used previously to quantify SOD1 metal content was employed to measure rates of copper and zinc release. Rates of local conformational changes in the SOD1 beta-barrel were measured by changes in tryptophan fluorescence intensity. A time-resolved glutaraldehyde cross-linking assay was employed to quantify the rate of dimer dissociation.
Results: Distinct rates and mechanisms of copper and zinc release were observed. Where zinc is released rapidly, by a simple two-state mechanism, copper release was a slower process showing a distinct lag phase that could only be fit by a three-state sequential release model. Dissociation of the SOD1 dimer and conformational changes in the SOD1 beta-barrel monitored by tryptophan fluorescence occurred by two-state mechanisms.
Discussion and Conclusions: SOD1 unfolds by a complex mechanism involving a number of distinct processes, including release of bound copper and zinc, dissociation of the SOD1 homodimer, and conformational changes in the SOD1 beta-barrel. Correlations in the rates of these processes across a range of guanidine concentrations reveal linkages in them. Dimer dissociation and zinc release appear to happen rapidly and simultaneously. Copper release occurs over a much longer period of time, and is necessarily preceded by a conformational change in the SOD1 beta-barrel. The majority of SOD1 molecules denature by a four-state mechanism, in which zinc release and dimer dissociation occur first followed by the slower beta-barrel conformational shift, which in turn is followed by rapid copper release. This establishes a zinc-deficient, copper-loaded SOD1 monomer as a well-populated SOD1 unfolding intermediate, and a species likely to be populated under conditions of denaturational stress. Since this species has previously been reported to possess enhanced neurotoxicity in cell culture and enhanced aggregation propensity, this is a likely candidate to be the disease-causing neurotoxic species in SOD1-associated ALS.
C86 ISOLATION AND PROTEOMIC CHARACTERIZATION OF MUTANT SOD1-CONTAINING INCLUSION BODIES
BERGEMALM D, NILSSON K, OLOFSSON A, GRAFFMO KS, BRANNSTROM T, MARKLUND S
Umea University, Sweden
E-mail address for correspondence: [email protected]
Keywords: SOD1, proteomics, inclusions
Introduction: A characteristic finding in ALS patients and transgenic murine models carrying SOD1 mutations are inclusion bodies displaying SOD1-immunoreactivity. A possible toxic property of mutant SOD1 is interaction with proteins that become inactivated, depleted or erroneously activated. These proteins could co-aggregate with SOD1 and become part of cellular inclusion bodies.
Objectives: To identify proteins in SOD1 inclusion bodies.
Methods: Spinal cord homogenates from terminal G127X and G85R mice were subjected to ultracentrifugation in density gradients. From these separations fractions containing SOD1-inclusions, free from other organelles, could be isolated. There were no inclusions in corresponding fractions of presymptomatic or control animals. Relevant fractions were further evaluated by atomic force microscopy, 2-dimensional gel analysis/MALDI-TOF and by LC-MS/MS.
Results: In density gradient centrifugations SOD1 inclusions entered the gradient but showed a large heterogeneity in density. A portion displayed a density higher than that of any organelle (mitochondria, lysosomes or peroxisomes) or markers for plasma membrane. Prolongation of centrifugation time from 1.5 to 4 hr did not change the distribution pattern. By addition of 0.5% of the detergent NP-40 to the homogenates, virtually all SOD-1 material in the gradient appeared at the bottom of the gradient, as expected for aggregated proteins. This suggests that the inclusions are not simple protein aggregates but more complex in structure. By atomic force microscopy, the inclusions were mainly round shaped with a large heterogeneity in size. In 2D gels from G85R and G127X mice about 20–30 protein spots could be seen, most overlapping between the models. In control animals only a few or no protein spots could be found in the corresponding fractions. SOD1 accounted for about 50% of the total amount of protein in the inclusions. About ten proteins have so far been identified, some previously known from immunohistochemistry to be constituents of SOD1-inclusion bodies.
Discussion: By this method we are able to find previously unknown SOD1-aggregation partners as well as other proteins that might be trapped in these inclusions. Knowledge of the identity of such proteins might be valuable for the development of understanding the toxic mechanism of mutant SOD1s and the pathogenesis of ALS.
C87 ALS2/ALSIN-DEFICIENT SOD1H46R TRANSGENIC MICE EXHIBIT INCREASED ACCUMULATION OF INSOLUBLE PROTEINS IN THE SPINAL CORD
HADANO S1, OTOMO A1, KUNITA R1, SUZUKI-UTSUNOMIYA K1, ONOE K1, OSUGA H1, AOKI M2, ITOYAMA Y2, IKEDA JE1
1Department of Molecular Life Sciences, Tokai University School of Medicine, Isehara, Kanagawa, Japan, 2 Department of Neurology, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan
e-mail address for correspondence: [email protected]
Keywords: ALS2/alsin, SOD1, protein degradation
Background: ALS2 is a causative gene for a juvenile autosomal recessive form of motor neuron diseases (MNDs), including ALS2, PLSJ, and IAHSP. The ALS2-coded protein, ALS2/alsin, activates Rab5 small GTPase and involves in macropinocytosis-associated endosome trafficking and fusion, and neurite outgrowth in the cells. Loss of these functions accounts for motor dysfunction and axonal degeneration in the ALS2-linked ALS/MNDs. Recently, we have shown that loss of ALS2 exacerbates motor dysfunction in SOD1H46R mice, indicative of a close relationship between the ALS2-associated cellular function and mutant SOD1-mediated pathogenesis in vivo. However, molecular mechanisms for the acceleration of disease progression in ALS2-deficient/mutant SOD1-expressing mice are still unknown.
Objectives: To clarify the biochemical basis of the ALS2-mediated functions that are associated with disease progression in SOD1H46R transgenic mice.
Methods: We generated mice with 6 different genotypes; Als2+/ + (wild-type), Als2+/-, Als2-/-, SOD1H46R, Als2+/-;SOD1H46R, and Als2-/-;SOD1H46R, by crossing male Als2+/-;SOD1H46R and female Als2+/- mice. Brain and spinal cord tissues were obtained from these mice at 8, 12, 16 and 20 weeks of age. The Triton X-soluble and insoluble fractions were prepared and aliquots of samples were separated by SDS-PAGE. The various protein bands with an increased intensity as the disease progressed were excised and analyzed by liquid chromatography-tandem mass spectrometry with computational peptide mass fingerprinting (LC-MS/MS-PMF). Further, tissue samples were subjected to Western blot analyses. A panel of antibodies, which included anti-ALS2, anti-SOD1, anti-ubiquitin, anti-p62/SQSTM1, anti-GFAP, anti-vimentin, anti-neurofilament HC, anti-peripherin, anti-beta-tubulin, and anti-GAPDH antibodies, were used in this study.
Results: SOD1H46R, Als2+/-;SOD1H46R, and Als2-/-;SOD1H46R mice all exhibited progressive motor dysfunction and paralysis. Notably, a mean survival of Als2-/-;SOD1H46R mice (152.4±5.0 days; n = 67) was significantly shorter than those in SOD1H46R (164.5±11.1 days; n = 146) or Als2+/-;SOD1H46R mice (164.8±10.3; n = 190) (p < 0.001). PAGE analysis of insoluble spinal cord proteins demonstrated two bands (p55 and p50) with an increased intensity as the disease progressed. LC-MS/MS-PMF analysis identified that p55 and p50 represent vimentin and GFAP, respectively. Although levels of neuronal intermediate filament proteins; neurofilament HC and peripherin were unchanged, other insoluble proteins such as vimentin, p62/SQSTM1, and mutant SOD1 were more progressively accumulated in the spinal cord of Als2-/-;SOD1H46R mice, from an approximately 8 weeks prior to disease-onset, than in those of SOD1H46R or Als2+/-;SOD1H46R mice.
Discussion and Conclusions: Our results suggest that loss of ALS2 hinders protein degradation and/or accelerates the accumulation of a number of insoluble proteins such as astrocyte-associated intermediate filaments and autophagy-associated protein (p62/SQSTM1) in the spinal cord, thereby exacerbating motor dysfunction in SOD1H46R mice. Further characterization of these mice will clarify the implication of the ALS2-mediated functions in mutant SOD1-linked ALS/MND in vivo.
C88 ANALYSIS OF THE ROLE OF DYNEIN MUTATIONS IN ATTENUATING THE PHENOTYPE OF SOD1G93A TRANSGENIC MICE
MORSI EL-KADI A1, DENG W1, MOORE A1, BANKS G2, GREENSMITH L2, FISHER E2, HAFEZPARAST M1
1University of Sussex, Brighton, East Sussex, United Kingdom, 2Institute of Neurology, University College London, United Kingdom
E-mail address for correspondence: [email protected]
Keywords: dynein, Loa, SOD1
Background: Cu/Zn SOD1G93A transgenic mice develop an ALS-like phenotype, which is characterized by motor neuron degeneration and muscle paralysis. Previously, we demonstrated that a missense point mutation in the gene encoding the heavy chain subunit of cytoplasmic dynein causes degeneration of motor neurons in the Legs at odd angles (Loa) mouse Citation[1]. We also showed that double mutant (Loa/SOD1G93A) transgenic mice have a significant improvement in SOD1G93A -mediated disease, including a reduction in motor neuron degeneration Citation[2].
Here we present data indicating a subcellular redistribution of mutant SOD1 in the Loa/SOD1G93A, which could provide a mechanistic explanation for the amelioration of the disease in these mice.
Objectives: To elucidate at the molecular level the role of the dynein mutations in attenuating the phenotype of SOD1G93A transgenic mice.
Methods: We have used density and buoyant gradient sedimentation assays to fractionate homogenates from the mouse spinal cord and brain tissues into subcellular fractions. Fractions were analysed on SDS-PAGE and Western blots.
Results: Our density gradient sedimentations of SOD1G93A protein in tissues from SOD1G93A and Loa/SOD1G93A littermates show a clear difference in the sedimentation patterns of this protein in the brain and spinal cord tissues from these mice. There is significantly more SOD1G93A protein in the denser fractions of the 90 and 121-day SOD1G93A mice compared with those of Loa/SOD1G93A. Importantly, these denser fractions contain mitochondria, as determined by using an antibody to the mitochondrial marker COX4.
Discussion: Our previously reported amelioration of the disease phenotype in SOD1G93A by defective dynein highlighted the dynein mediated retrograde axonal transport as a potential target for therapy. The dramatic depletion of SOD1G93A from denser cellular fractions, containing mitochondria, of Loa/SOD1G93A in density gradient sedimentation assays suggests that mutant dynein alters the sub-cellular distribution of toxic SOD1G93A protein, which may reflect a decreased level of association of mutant SOD1 with mitochondria. In support of this hypothesis we have evidence that the mitochondrial membrane potential is depolarized in SOD1G93A spinal cord, but in Loa/SOD1G93A it returns to normal levels (abstract submitted separately by V. Bros). We are currently investigating the mitochondrial function in SOD1G93A versus Loa/SOD1G93A.
References
- Hafezparast M, Klocke R, Ruhrberg C, et al. Science 2003; 300: 808–812
- Kieran D, Hafezparast M, Bohnert S, et al. J. Cell Biol. 2005; 169: 561–567