**4. Possible patterns of motor neurone loss**

In normal healthy individuals, it has been shown that there is little loss of functioning motor neurones before the age of 60. The normal aging process then accounts for loss of approximately 3.9% of the original motor neurone pool per annum after the age of 60. (Brown, 1972). In this situation, the number of motor neurones remain fairly constant up to the age of 60, after which there is a gradual steady decline with age.

MND may be due to a slow attrition of motor neurones over time (Pattern 1 in Figure 2). If this were the case, pre-symptomatic motor neurone loss may be identifiable in SOD1 mutation carriers, as eventually there may be a gradual decline over time (Figure 2).

Another possible course of MND is that normal numbers of motor neurones are maintained until sudden, rapid multi-focal cell death of motor neurones occurs, corresponding with the development of symptoms (Pattern 2 in Figure 2). If this situation, it would be expected that SOD1 mutation carriers have a normal number of motor neurones during the presymptomatic phase. In this case, cell death occurs as neurones gradually accumulate damage, secondary to the mutation, which ultimately overwhelms cellular homeostasis. This is the cumulative damage hypothesis. (Clarke et al., 2000).

630 Amyotrophic Lateral Sclerosis

39 on exon 4 and 29 on exon 5 (Figure 1). There have also been 2 non-exon mutations identified on intron 4 and intron 1 and 14 'apparently' sporadic cases described with 6

Most are autosomal dominant in inheritance, but there is one confirmed autosomal recessive mutation, the D90A mutation in exon 4. This is unique in that it exists in dominant families in a heterozygous state, but in a number of pedigrees, specifically those of Scandinavian

Mutations in the heavy polypeptide 200kDa subunit of neurofilaments (NEFH) have been identified in sporadic MND cases, (Figlewicz et al., 1994) and in one FALS case. (Al-Chalabi et al., 1999). Accumulation of neurofilaments in cell bodies and axons of motor neurons is a pathological hallmark of early stages of many neurodegenerative diseases. These mutations lie in the region of the protein involved in cross-linking and thus may disrupt normal aggregation of filaments. Thus far, 1 insertion and 5 deletion mutations have been identified on exon 4. Analysis of the NEFH locus on chromosome 22 however has failed to detect linkage in MND families. (Vechio et al., 1996). Genome search on a large pedigree with autosomal dominant juvenile onset MND found strong evidence for linkage to chromosome 9q34 (ALS4). The average age of onset is 17 years, with slow progression of disease. (Chance et al., 1996). There is also an autosomal recessive, juvenile onset MND, with linkage to a

The other 90% of all MND patients have the sporadic form. There is no recognisable phenotypic difference between FALS and sporadic MND. The male: female ratio is 1:1 in FALS and 1.7:1 in sporadic MND. (De Belleroche et al., 1995). This decreases with increasing age of onset and approaches 1:1 after the age of 70. (Haverkamp et al., 1995). The site of onset is variable. Survival does not seem to be affected by age or gender, but rather the site of symptom onset. Generally, bulbar onset disease has a worse prognosis, and upper limb

ancestry, homozygous mutations are required for disease. (Anderson et al., 1997).

different SOD1 mutations. (Shaw et al., 1998).

Fig. 1. Number of SOD1 mutations identified for each exon

locus on chromosome 15 (ALS5). (Hentati et al., 1998).

onset is more favourable. (Mulder et al., 1986).

Fig. 2. Diagrammatic representation of possible patterns for motor neurone loss in an individual.

Protection of Motor Neurons in Pre-Symptomatic Individuals Carrying

homeostasis of copper and/or protein aggregation. (Clevland, 1999).

SOD 1 Mutations: Results of Motor Unit Number Estimation (MUNE) Electrophysiology 633

pathogenesis of FALS (Figure 3). (Hand & Rouleau, 2002). At present the favoured hypotheses is that the mutation causes disease as a result of a toxic gain of function by the mutant SOD1 provoking selective neurotoxicity, probably disrupting the intracellular

Fig. 3. Pathways that have been implicated in motor neurone cell death in amyotrophic lateral sclerosis (Reproduced from Hand CK. Familial Amyotrophic Lateral Sclerosis.

The mutant SOD1 enzyme has altered reactivity with certain substrates, (Noor et al., 2003), in addition to the major superoxide dismutase activity. The SOD1 enzyme catalyses the reduction of hydrogen peroxide (H2O2), therefore acting as a peroxidase. This leads to the formation of hydoxyl radicals that can also alter the neurofilament network. Motor neurones have highenergy requirements and thus contain many mitochondria that generate superoxide radials (O2-) through normal metabolism. SOD1 is an anti-oxidant defence which catalyses conversion of superoxide free radical anion (O2-) to hydrogen peroxide (H2O2), which is reduced to H2O and O2 by catalse. Mutations at SOD1 binding sites, alter the redox behaviour of the enzyme and destabilise the SOD1 ligand, leading to increased oxidative damage as hydrogen peroxide

This supports the hypothesis that the pathogenesis of SOD1 related FALS may be due to increased peroxidase activity of mutant SOD1 resulting in oxidative damage mainly to

Muscle Nerve 2002; 25:137).

lipids of the cell membrane.

and its derivatives are toxic to the cell. (Yim et al., 1990).

One of the mechanisms most frequently proposed to underlie cumulative damage is oxidative stress, in which an imbalance between the production of reactive oxygen species and cellular antioxidant mechanisms results in chemical modifications of macromolecules, thereby disrupting cellular structure and function. (Robberecht, 2000). A key prediction of the cumulative damage hypothesis is that the probability that any individual neurone will become committed to apoptosis increases as damage accrues within it. A mutant neurone in an older patient will have accumulated a greater amount of damage and is therefore be more likely to die than in a younger patient. Consequently, early in the course of disease, the chance of a cell containing a sufficient amount of damage to initiate apoptosis is small, and the rate of cell loss is correspondingly low. However, as the amount of intracellular damage increases, the chance that a cell will die also increases

It has been shown that the kinetics of neuronal death in a number of inherited neurodegenerative diseases was best explained by models in which the risk of cell death remains constant throughout life of the neurone and that cell death occurred randomly in time and was independent of any other neurone. This implies a "one-hit" biochemical phenomenon in which the mutant imposes an abnormal mutant steady state on the neurone and a single catastrophic event randomly initiates cell death and apoptosis. The principal features of the mutant steady state are that the living mutant neurones function very well for years or even decades and that the predominant feature of the mutant neurones is that they are all at a risk of death. This argues against the multiple environmental factors hypothesis as a cause of MND, as a random process is probably responsible for the initiation of disease. (Clarke et al., 2001).

### **5. Cu/Zn superoxide dismutase (SOD1) mutations**

Linkage studies for familial MND (FALS) on chromosome 21q22.1 led to the identification of point mutations in the gene for Cu/Zn superoxide dismutase (SOD1) as a cause of MND. (Siddique 1991). Superoxide (O2-) is an unstable and highly active molecule, which causes oxidation of cell constituents either directly or through toxic and stable derivatives. The major superoxide dismutase activity in cytoplasm is from SOD1, which consists of 5 small exons that encode 153 highly conserved amino acids with a molecular weight of 16Kda. SOD1 is a homodimer. Within each monomer, there is an active site containing one atom each of copper and zinc. (Radunovic & Leigh, 1996).

The most common SOD1 gene mutation seen in FALS is an alanine to valine shift at codon 4 (Ala4Val). This accounts for 50% of all mutations in the USA. (Rosen, 1993). Of all the clinical variables, only bulbar onset and three specific mutations seem to influence age of onset of MND. Bulbar patients are older when their illness begins, whereas the Gly37Arg and Leu38Val mutations predict an earlier age of onset.). Leu38Val is associated with the earliest onset (mean 35.5 years) and Ile113Thr with the latest onset (mean 58.9 years).

In terms of survival, Ala4Val correlated with the shortest survival of 1.5 years. Whereas, Gly37Arg, Gly41Asp, and Gly93Ala mutation predicted longer survival. The mutations that predict earlier onset are not the same as those that correlate with shortest duration of disease. (Cudkowicz et al., 1997). This suggests that the factors that influence onset of disease differ from those that influence the rate of progression of the disease.

Determining the mechanism by which mutations in the Cu/Zn superoxide dismutase (SOD1) gene triggers the destruction of motor neurones causing MND remains a challenging and complex problem. Five primary hypotheses have been postulated for the

One of the mechanisms most frequently proposed to underlie cumulative damage is oxidative stress, in which an imbalance between the production of reactive oxygen species and cellular antioxidant mechanisms results in chemical modifications of macromolecules, thereby disrupting cellular structure and function. (Robberecht, 2000). A key prediction of the cumulative damage hypothesis is that the probability that any individual neurone will become committed to apoptosis increases as damage accrues within it. A mutant neurone in an older patient will have accumulated a greater amount of damage and is therefore be more likely to die than in a younger patient. Consequently, early in the course of disease, the chance of a cell containing a sufficient amount of damage to initiate apoptosis is small, and the rate of cell loss is correspondingly low. However, as the amount of intracellular damage

It has been shown that the kinetics of neuronal death in a number of inherited neurodegenerative diseases was best explained by models in which the risk of cell death remains constant throughout life of the neurone and that cell death occurred randomly in time and was independent of any other neurone. This implies a "one-hit" biochemical phenomenon in which the mutant imposes an abnormal mutant steady state on the neurone and a single catastrophic event randomly initiates cell death and apoptosis. The principal features of the mutant steady state are that the living mutant neurones function very well for years or even decades and that the predominant feature of the mutant neurones is that they are all at a risk of death. This argues against the multiple environmental factors hypothesis as a cause of MND, as a random process is probably responsible for the initiation

Linkage studies for familial MND (FALS) on chromosome 21q22.1 led to the identification of point mutations in the gene for Cu/Zn superoxide dismutase (SOD1) as a cause of MND. (Siddique 1991). Superoxide (O2-) is an unstable and highly active molecule, which causes oxidation of cell constituents either directly or through toxic and stable derivatives. The major superoxide dismutase activity in cytoplasm is from SOD1, which consists of 5 small exons that encode 153 highly conserved amino acids with a molecular weight of 16Kda. SOD1 is a homodimer. Within each monomer, there is an active site containing one atom

The most common SOD1 gene mutation seen in FALS is an alanine to valine shift at codon 4 (Ala4Val). This accounts for 50% of all mutations in the USA. (Rosen, 1993). Of all the clinical variables, only bulbar onset and three specific mutations seem to influence age of onset of MND. Bulbar patients are older when their illness begins, whereas the Gly37Arg and Leu38Val mutations predict an earlier age of onset.). Leu38Val is associated with the earliest

In terms of survival, Ala4Val correlated with the shortest survival of 1.5 years. Whereas, Gly37Arg, Gly41Asp, and Gly93Ala mutation predicted longer survival. The mutations that predict earlier onset are not the same as those that correlate with shortest duration of disease. (Cudkowicz et al., 1997). This suggests that the factors that influence onset of

Determining the mechanism by which mutations in the Cu/Zn superoxide dismutase (SOD1) gene triggers the destruction of motor neurones causing MND remains a challenging and complex problem. Five primary hypotheses have been postulated for the

onset (mean 35.5 years) and Ile113Thr with the latest onset (mean 58.9 years).

disease differ from those that influence the rate of progression of the disease.

increases, the chance that a cell will die also increases

**5. Cu/Zn superoxide dismutase (SOD1) mutations** 

each of copper and zinc. (Radunovic & Leigh, 1996).

of disease. (Clarke et al., 2001).

pathogenesis of FALS (Figure 3). (Hand & Rouleau, 2002). At present the favoured hypotheses is that the mutation causes disease as a result of a toxic gain of function by the mutant SOD1 provoking selective neurotoxicity, probably disrupting the intracellular homeostasis of copper and/or protein aggregation. (Clevland, 1999).

Fig. 3. Pathways that have been implicated in motor neurone cell death in amyotrophic lateral sclerosis (Reproduced from Hand CK. Familial Amyotrophic Lateral Sclerosis. Muscle Nerve 2002; 25:137).

The mutant SOD1 enzyme has altered reactivity with certain substrates, (Noor et al., 2003), in addition to the major superoxide dismutase activity. The SOD1 enzyme catalyses the reduction of hydrogen peroxide (H2O2), therefore acting as a peroxidase. This leads to the formation of hydoxyl radicals that can also alter the neurofilament network. Motor neurones have highenergy requirements and thus contain many mitochondria that generate superoxide radials (O2-) through normal metabolism. SOD1 is an anti-oxidant defence which catalyses conversion of superoxide free radical anion (O2-) to hydrogen peroxide (H2O2), which is reduced to H2O and O2 by catalse. Mutations at SOD1 binding sites, alter the redox behaviour of the enzyme and destabilise the SOD1 ligand, leading to increased oxidative damage as hydrogen peroxide and its derivatives are toxic to the cell. (Yim et al., 1990).

This supports the hypothesis that the pathogenesis of SOD1 related FALS may be due to increased peroxidase activity of mutant SOD1 resulting in oxidative damage mainly to lipids of the cell membrane.

Protection of Motor Neurons in Pre-Symptomatic Individuals Carrying

SOD 1 Mutations: Results of Motor Unit Number Estimation (MUNE) Electrophysiology 635

To date, the only effective approved treatment for amyotrophic lateral sclerosis is Riluzole, (Cheah et al., 2010), which has a neuroprotective role, possibly due to pre-synaptic inhibition of glutamate release. (Doble, 1996). Treatment of human ALS patients or transgenic Cu, Zn superoxide dimutase 1 (SOD 1) mice, most commonly produce a modest but significant increase in survival. (Bensimon et al., 1994). It has also been shown to have a small beneficial effect on bulbar function, but not muscle strength. (Miller et al., 2007). Apoptosis is characterised by a series of cellular changes leading to non-inflammatory cell death. Mitochondrial involvement in the apoptotic pathway also leads to the release of cytochrome c, an activator of the initiator caspase-9, which in turn activates caspase-3, which are executioners in the breakdown of essential cellular proteins. There is evidence that the mutant SOD1 transgene causes motor neurone death in mice through caspase-mediated programmed cell death. (Li et al., 2000). This may then be a target for inhibiting the apoptotic cascade, as it has been shown in a SOD1 transgenic mouse model that a small peptide caspase inhibitor (*zVAD-fmk*), prolonged survival after onset of disease by nearly 70%. (Kosti et al., 1997). It has also been reported that there are elevated levels of *bax* protein

in MND spinal motor neurones, which promotes apoptosis. (Mu et al., 1996).

telephone by the department's genetic counsellor and informed about the study.

The Department of Molecular Medicine at Concord hospital had a large database of family members with a known family history of MND, who had blood samples collected for DNA, as part of a previous linkage study. From this database, family members were contacted by

The regional committees for Ethics in Medical Research from Central Sydney Area Health Service, Royal North Shore Hospital and Prince Charles Hospital, approved this study. All individuals participated without knowledge of their mutation status and on the understanding that this would not be revealed to them. Subjects were also aware that the results obtained from the study would not be available to them and that the information would only be used for research purposes. New consents were obtained from all individuals who participated in the study. The neurologist performing the MUNE studies also had no knowledge of their mutation status. The mutation status was only used in the final analysis of results. Subsequently, they were divided into "SOD1 negative family controls" and

In addition, studies were also carried out on normal individuals, such as department technicians, spouses of SOD1 family members and individuals from the general population who attended MND support meeting and had an interest in helping to advance research into MND. This group was used as "population controls", to test the validity and

Sporadic MND subjects were also initially studied once the MUNE technique had been validated to demonstrate that the MUNE technique used was able to detect a loss of motor

Motor unit number estimation (MUNE) estimates the number of functioning lower motor neurones innervating a muscle or a group of muscles and is a measure of the primary pathologic process of motor neurone loss. The concept of motor unit number estimation

**6. Methods** 

"asymptomatic SOD1 mutation carriers".

reproducibility of the MUNE technique used.

**6.1 Motor unit number estimation** 

neurones, when present. These were used as "positive controls".

Mapping of the mutation sites predicted that these mutations destabilise the protein structure, leading to a less active enzyme i.e. "loss of function". This is however not supported by the fact that transgenic mice over expressing SOD1 gene developed disease similar to MND in humans, while those over-expressing normal SOD1 remained unaffected. This suggests that the mutant mice develop the disease independent of the level of SOD1 activity and suggests that the mutant protein itself is selectively toxic to motor neurones and that there is a "gain of toxic function" rather than a "loss of function". (Gurney et al., 1993). Also, although most mutations in SOD1 gene cause decrease in steady state of cytosolic SOD1 activity, Gly37Arg and Asp90Ala, have no significant decrease in SOD1 activity. (Shaw et al., 1998).

As most SOD1 mutations destabilise SOD1 protein (except Asp90Ala), it is possible that the mutant protein, with altered conformation may become unstable and precipitate to form aggregates or inclusions in motor neurons. These aggregates may then disturb normal cell function and lead to cell death. They are easily formed when SOD1 protein stability is decreased because this protein exists in large amounts accounting for 0.5-1% of total cytosolic protein in neurons. Alternations in the length of the coding sequence, folding, solubility or degradation results in the formation of aggregates. (Yim et al., 1990). Structural changes of mutant SOD1 may distort the rim of the electrostatic guidance channel and allow the catalytic site to become exposed and shallow. Molecules that are normally excluded may gain access to the catalytic reactive site. This results in less buffering of copper and zinc, which then become neurotoxic. (Radunovic & Leigh, 1996).

The nitric oxide (NO) produced by nitric oxide synthase (NOS) reacts spontaneously with O2- to generate peroxynitrite (ONOO-), which nitrosylates proteins leading to damage. Excess NO may also cause an increase in O2- production by inhibition of mitochondrial electron flow, resulting in further generation of peroxynitrite. This facilitates nitrosylation of tyrosine residues of critical cytosolic proteins thus injuring cells. This reaction is copper dependent. The source of free copper may be mutant SOD1, which cannot accept the ion from the copper chaperone (CCS) protein. Mutant SOD1 possibly exhibit metal mediated cytotoxicities by disrupting the intracellular homeostasis of Cu and Zn, which are potential neurotoxins. (Gurney & Tomasselli, 2000).

The target proteins for nitrosylation include the neurofilament (NF) subunits, which may result in abnormal NF accumulation and subsequent disruption of the NF network and axonal transport, as there is a high neurofilament content in motor neurones. It has also been demonstrated that transgenes encoding mutant NF subunits can directly cause selective degeneration and death of motor neuones. (Cleveland, 1999). Conformational changes have been described in the mutations, Ala4Val, Gly37Arg and His6Arg that may affect the rim of the electrostatic guidance channel coded by exon 3.(Sjalander et al., 1995).

Glutamate is released from the presynaptic terminal activates the glutamate receptor on the postsynaptic cell membrane. It is then cleared from the synaptic cleft by specific glutamate transporters such as EAAT2. (Trotti et al., 1999). Astrocyte (glial cell) dysfunction may result in selective loss of EAAT2, interfering with the normal clearance of glutamate and allowing it to accumulate in the cell membrane and continue to activate the receptor. (Bruijin et al., 1997). Once activated, the glutamate receptor causes a calcium influx and a cascade of toxicity. The neurone does not have the capacity to buffer this efficiently due to a deficiency in calcium binding proteins (CBP's). This results in disturbances in mitochondrial metabolism and as a consequence, motor neurone cell death. (Beal, 1996).

To date, the only effective approved treatment for amyotrophic lateral sclerosis is Riluzole, (Cheah et al., 2010), which has a neuroprotective role, possibly due to pre-synaptic inhibition of glutamate release. (Doble, 1996). Treatment of human ALS patients or transgenic Cu, Zn superoxide dimutase 1 (SOD 1) mice, most commonly produce a modest but significant increase in survival. (Bensimon et al., 1994). It has also been shown to have a small beneficial effect on bulbar function, but not muscle strength. (Miller et al., 2007).

Apoptosis is characterised by a series of cellular changes leading to non-inflammatory cell death. Mitochondrial involvement in the apoptotic pathway also leads to the release of cytochrome c, an activator of the initiator caspase-9, which in turn activates caspase-3, which are executioners in the breakdown of essential cellular proteins. There is evidence that the mutant SOD1 transgene causes motor neurone death in mice through caspase-mediated programmed cell death. (Li et al., 2000). This may then be a target for inhibiting the apoptotic cascade, as it has been shown in a SOD1 transgenic mouse model that a small peptide caspase inhibitor (*zVAD-fmk*), prolonged survival after onset of disease by nearly 70%. (Kosti et al., 1997). It has also been reported that there are elevated levels of *bax* protein in MND spinal motor neurones, which promotes apoptosis. (Mu et al., 1996).
