Fluorescence-Based Approaches for Quantitative Assessment of Protein Carbonylation, Protein Disulfides, and Protein Conformation in Biological Tissues.

Protein oxidation and misfolding have been considered as key players for progression of aging and etiology of various pathological conditions. However, few attempts have been made to develop sensitive and reproducible assays to quantify the changes in protein oxidation and alteration in structure. Here we describe three distinct fluorescence-based assays to quantify changes in protein oxidation, namely carbonylation and disulfides and alteration in protein surface hydrophobicity as a reporter for protein conformation. These techniques will provide investigators the opportunity to address important biological questions in their experimental models.

Neuronal cells but not muscle cells are resistant to oxidative stress mediated protein misfolding and cell death: role of molecular chaperones.

Our recent study in a mouse model of familial-Amyotrophic Lateral Sclerosis (f-ALS) revealed that muscle proteins are equally sensitive to misfolding as spinal cord proteins despite the presence of low mutant CuZn-superoxide dismutase, which is considered to be the key toxic element for initiation and progression of f-ALS. More importantly, we observed differential level of heat shock proteins (Hsp's) between skeletal muscle and spinal cord tissues prior to the onset and during disease progression; spinal cord maintains significantly higher level of Hsp's compared to skeletal muscle. In this study, we report two important observations; (i) muscle cells (but not neuronal cells) are extremely vulnerable to protein misfolding and cell death during challenge with oxidative stress and (ii) muscle cells fail to mount Hsp's during challenge unlike neuronal cells. These two findings can possibly explain why muscle atrophy precedes the death of motor neurons in f-ALS mice.

Genetic ablation of 12/15-lipoxygenase but not 5-lipoxygenase protects against denervation-induced muscle atrophy.

Skeletal muscle atrophy is a debilitating outcome of a number of chronic diseases and conditions associated with loss of muscle innervation by motor neurons, such as aging and neurodegenerative diseases. We previously reported that denervation-induced loss of muscle mass is associated with activation of cytosolic phospholipase A2 (cPLA2), the rate-limiting step for the release of arachidonic acid from membrane phospholipids, which then acts as a substrate for metabolic pathways that generate bioactive lipid mediators. In this study, we asked whether 5- and 12/15-lipoxygenase (LO) lipid metabolic pathways downstream of cPLA2 mediate denervation-induced muscle atrophy in mice. Both 5- and 12/15-LO were activated in response to surgical denervation; however, 12/15-LO activity was increased ~2.5-fold versus an ~1.5-fold increase in activity of 5-LO. Genetic and pharmacological inhibition of 12/15-LO (but not 5-LO) significantly protected against denervation-induced muscle atrophy, suggesting a selective role for the 12/15-LO pathway in neurogenic muscle atrophy. The activation of the 12/15-LO pathway (but not 5-LO) during muscle atrophy increased NADPH oxidase activity, protein ubiquitination, and ubiquitin-proteasome-mediated proteolytic degradation. In conclusion, this study reveals a novel pathway for neurogenic muscle atrophy and suggests that 12/15-LO may be a potential therapeutic target in diseases associated with loss of innervation and muscle atrophy.

Differential effects of mutant SOD1 on protein structure of skeletal muscle and spinal cord of familial amyotrophic lateral sclerosis: role of chaperone network.

Protein misfolding is considered to be a potential contributing factor for motor neuron and muscle loss in diseases like Amyotrophic lateral sclerosis (ALS). Several independent studies have demonstrated using over-expressed mutated Cu/Zn-superoxide dismutase (mSOD1) transgenic mouse models which mimic familial ALS (f-ALS), that both muscle and motor neurons undergo degeneration during disease progression. However, it is unknown whether protein conformation of skeletal muscle and spinal cord is equally or differentially affected by mSOD1-induced toxicity. It is also unclear whether heat shock proteins (Hsp's) differentially modulate skeletal muscle and spinal cord protein structure during ALS disease progression. We report three intriguing observations utilizing the f-ALS mouse model and cell-free in vitro system; (i) muscle proteins are equally sensitive to misfolding as spinal cord proteins despite the presence of low level of soluble and absence of insoluble G93A protein aggregate, unlike in spinal cord, (ii) Hsp's levels are lower in muscle compared to spinal cord at any stage of the disease, and (iii) G93ASOD1 enzyme-induced toxicity selectively affects muscle protein conformation over spinal cord proteins. Together, these findings strongly suggest that differential chaperone levels between skeletal muscle and spinal cord may be a critical determinant for G93A-induced protein misfolding in ALS.

Elevated protein carbonylation, and misfolding in sciatic nerve from db/db and Sod1(-/-) mice: plausible link between oxidative stress and demyelination.

Diabetic peripheral polyneuropathy is associated with decrements in motor/sensory neuron myelination, nerve conduction and muscle function; however, the mechanisms of reduced myelination in diabetes are poorly understood. Chronic elevation of oxidative stress may be one of the potential determinants for demyelination as lipids and proteins are important structural constituents of myelin and highly susceptible to oxidation. The goal of the current study was to determine whether there is a link between protein oxidation/misfolding and demyelination. We chose two distinct models to test our hypothesis: 1) the leptin receptor deficient mouse (dbdb) model of diabetic polyneuropathy and 2) superoxide dismutase 1 knockout (Sod1(-/-) ) mouse model of in vivo oxidative stress. Both experimental models displayed a significant decrement in nerve conduction, increase in tail distal motor latency as well as reduced myelin thickness and fiber/axon diameter. Further biochemical studies demonstrated that oxidative stress is likely to be a potential key player in the demyelination process as both models exhibited significant elevation in protein carbonylation and alterations in protein conformation. Since peripheral myelin protein 22 (PMP22) is a key component of myelin sheath and has been found mutated and aggregated in several peripheral neuropathies, we predicted that an increase in carbonylation and aggregation of PMP22 may be associated with demyelination in dbdb mice. Indeed, PMP22 was found to be carbonylated and aggregated in sciatic nerves of dbdb mice. Sequence-driven hydropathy plot analysis and in vitro oxidation-induced aggregation of purified PMP22 protein supported the premise for oxidation-dependent aggregation of PMP22 in dbdb mice. Collectively, these data strongly suggest for the first time that oxidation-mediated protein misfolding and aggregation of key myelin proteins may be linked to demyelination and reduced nerve conduction in peripheral neuropathies.

Protein misfolding, mitochondrial dysfunction and muscle loss are not directly dependent on soluble and aggregation state of mSOD1 protein in skeletal muscle of ALS.

Mutant superoxide dismutase 1 (mSOD1) is often found as aggregates at the outer-membrane of mitochondria in motor neurons of various mouse models and familial amyotrophic lateral sclerosis (f-ALS) patients. It has been postulated that disruption of mitochondrial function by physical association of misfolded mSOD1 aggregates may actually be the trigger for initiation of degeneration of motor neurons in ALS. However, it was not clear if the same mechanism is involved in muscle degeneration and mitochondrial dysfunction in skeletal muscles of ALS. Recent study from our laboratory show that two skeletal muscle proteins, namely creatine kinase (CK) and glyceraldehydes-3-phosphate dehydrogenase (GAPDH) undergo major conformational and functional changes in the f-ALS mouse model of ALS (G93A). In this paper, we report two intriguing observations which are as follows:(i) G93A protein does not form aggregates in skeletal muscle at any stages of disease process probably due to high chymotrypsin-like activity of proteasome and thus G93A protein aggregates have no direct effects on progressive loss of muscle mass and global changes in protein conformation in ALS, and (ii) the soluble G93A protein does not have direct effects on mitochondrial dysfunction as determined by quantifying the release of reactive oxygen species (ROS) in skeletal muscle mitochondria; instead, the proteins affected by G93A possibly affect mitochondrial ROS release. These data strongly suggest for the first time that unlike in motor neurons, the soluble and aggregation states of the G93A protein do not have direct effects on protein misfolding and mitochondrial dysfunction in skeletal muscle during ALS.

A Novel mouse model of enhanced proteostasis: Full-length human heat shock factor 1 transgenic mice.

The heat shock response (HSR) is controlled by the master transcriptional regulator heat shock factor 1 (HSF1). HSF1 maintains proteostasis and resistance to stress through production of heat shock proteins (HSPs). No transgenic model exists that overexpresses HSF1 in tissues of the central nervous system (CNS). We generated a transgenic mouse overexpressing full-length non-mutant HSF1 and observed a 2-4-fold increase in HSF1 mRNA and protein expression in all tissues studied of HSF1 transgenic (HSF1(+/0)) mice compared to wild type (WT) littermates, including several regions of the CNS. Basal expression of HSP70 and 90 showed only mild tissue-specific changes; however, in response to forced exercise, the skeletal muscle HSR was more elevated in HSF1(+/0) mice compared to WT littermates and in fibroblasts following heat shock, as indicated by levels of inducible HSP70 mRNA and protein. HSF1(+/0) cells elicited a significantly more robust HSR in response to expression of the 82 repeat polyglutamine-YFP fusion construct (Q82YFP) and maintained proteasome-dependent processing of Q82YFP compared to WT fibroblasts. Overexpression of HSF1 was associated with fewer, but larger Q82YFP aggregates resembling aggresomes in HSF1(+/0) cells, and increased viability. Therefore, our data demonstrate that tissues and cells from mice overexpressing full-length non-mutant HSF1 exhibit enhanced proteostasis.