**7. Effects of metabolic phenotype on ROS production by tgBM and tgSCM**

Figure 7A shows generation of ROS by tgBM and tgSCM isolated from tgSOD1 rats in 2007, and figure 7B presents corresponding results for tgBM and tgSCM isolated from tgSOD1 rats in 2010. The figures clearly show that in spite of higher basic ROS production with

Role of Neuronal Mitochondrial Metabolic Phenotype in Pathogenesis of ALS 241

protein damage and promote mitochondrial dysfunction. Loss of metals by mutant SOD1 leads to formation of amyloid-like aggregates (Durer et al., 2009). Mutations in SOD1 amplify reactions with H2O2, increase the lifetime of incorrectly folded states, and if exposed to even mild oxidative stress, incorrect disulfide links form and stabilize larger aggregates that may be resistant to the degradation by the quality control machinery of the cell, and thus increase association with mitochondria (Field et al., 2003; Furukawa, O'Halloran, 2008).

There is strong evidence that tgSOD1 selectively binds to the outer (Vande Velde et al., 2008) and inner mitochondrial membranes (Liu et al., 2004; Ahtoniemi et al., 2008). Mutant SOD1 protein associated with mitochondria forms cross-linked oligomers and causes a shift in the redox state of respiratory components (Ferri et al., 2006). We propose a mechanism by which tgSOD1 might increase generation of ROS associated with reverse electron transport. The hypothesis is based on our recently published data on the effects of cholesterol β-Dglucoside and cycad phytosterol glucosides on ROS generation by BM (Panov et al., 2010b). These compounds are neurotoxic and suspected as the cause of the cluster of neurodegenerative disorders in the western Pacific termed amyotrophic lateral sclerosis– parkinsonism dementia complex (ALS-PDC) (Wilson et al., 2002). When added *in vitro* to mitochondria, cholesterol β-D-glucoside increased ROS production, which was specifically dependent on activity of SDH. Cholesterol β-D-glucoside is known to diminish the surface area of the membranes and thus affect the activity of the membrane's enzymes (Deliconstantinos et al., 1989). Here we suggest, that tgSOD1 upon its interaction with the mitochondria may change physical-chemical properties of the membranes and increase the rate of reverse electron transport in a way similar to that of cholesterol β-D-glucoside.

**10. Oxidative stress increases sensitivity of mitochondria to Ca2+-dependent** 

Increased oxidative stress makes mitochondria more sensitive to the Ca2+-induced permeability transition, which may initiate apoptotic or necrotic cell death (Halestrap et al., 2000; Nicholls, 2008b, 2009). Mitochondria from tgSOD1 animals have high sensitivity to the deleterious effects of calcium (Martin et al., 2009). SCM are particularly sensitive to calcium overload (Sullivan et al., 2004, Panov et al., 2011a). This scenario becomes very likely to occur in tgSOD1 animals in view of the fact that spinal cord tissue contains 8 times more calcium than brain tissue (Panov et al., 2011a). We estimated that when normalized for 1 gram of tissue, BM could sequester several times more Ca2+ than was available in the whole tissue. SCM, on the other hand, could sequester only approximately 20% of the total spinal cord tissue Ca2+ (Panov et al., 2011a). We suggest that the high content of Ca2+ is necessary to hold together the sheets of myelin, which protects the neurons. We have shown recently that increased oxidative stress promotes demyelination in brains of OXYS rats with genetically accelerated aging, which was ameliorated by feeding of animals with malate (Kolosova et al. 2011). Underwood et al. (2010), using magnetic resonance imaging, have shown signs of demyelination in the lumbar spinal cord of ALS-affected SOD1 mice, which were limited to white matter tracts arising from the motor neurons, whereas sensory white matter fibers were preserved. Damages to myelin sheets of motor neurons may evidently

**9. Proposed mechanisms of increased ROS production in tgSOD1** 

**mitochondria** 

**excitotoxicity** 

glutamate and pyruvate by BM and SCM in 2010 (Fig. 5A and 5B), the overall rates of ROS generation with physiologically relevant substrate mixture - glutamate + pyruvate + malate were 5 times lower for tgBM and 10 times for tgSCM , when compared with transgenic mitochondria in 2007 (Fig. 7A and 7B).

**A.** 2007; **B.** 2010. Dark grey – brain mitochondria, light gray – spinal cord mitochondria. Incubation conditions as in Fig. 5. Statistics: \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001. The data for tgSCM were compared with the corresponding results for tgBM.

### Fig. 7. **Generation of ROS by brain and spinal cord mitochondria from tgSOD1 rats, isolated in 2007 and 2010, oxidizing various substrates.**

Similar multiple-fold differences in ROS production existed also with succinate and succinate containing substrate mixtures (Figs. 7A, 7B). In tgBM and tgSCM, the increases in ROS generation were substrate specific and depended on activity of SDH. These results led us to conclusion that changes in the metabolic phenotype of neuronal mitochondria, which occurred in 2008, resulted in a dramatic decrease in production of ROS in tgSOD1 rats.
