Oxidative stress is commonly defined as the imbalance between the generation of pro-oxidants and antioxidant defence resulting in an accumulation of oxidized or damaged molecules. In skeletal muscle, increased oxidative stress is closely associated with a number of negative phenotypes including loss of muscle mass and contractile strength, disruption of neuromuscular junction (NMJ) integrity and function, damage to mitochondrial functions including a reduced oxygen consumption rate (OCR), and a higher production rate of reactive oxygen species (ROS).1 Elevated oxidative stress in skeletal muscle has been implicated in a number of pathological conditions, such as muscle disuse, age-associated muscle loss in sarcopenia, denervation, muscular dystrophies, and cancer cachexia.2 A major source of cellular oxidative stress is mitochondrial electron transport chain generation of superoxide anions and hydrogen peroxide.3
To study the impact of increased oxidative stress in sarcopenia, our group developed the Sod1KO mouse as an accelerated sarcopenia model that faithfully reproduces muscle and motor neuron phenotypes normally associated with age-related muscle atrophy and weakness but the onset is accelerated appearing in younger adult mice. Sod1KO mice lack CuZnSOD, a primarily cytosolic superoxide dismutase that converts superoxide anion to hydrogen peroxide, and exhibit a number of oxidative stress-induced phenotypes, including accelerated loss of muscle mass and weakness, altered NMJ function and morphology, and mitochondrial dysfunction (reduced oxygen consumption, ATP production calcium buffering capacity, and increased production of H2O2 and lipid hydroperoxides.1,4,5 Importantly, in human subjects, the antioxidant enzyme CuZnSOD has been found to be closely associated with sarcopenia, in that the content and activity of CuZnSOD are dramatically reduced in sarcopenic patients.6 Our studies in this model and mouse models designed to target deletion of CuZnSOD to either motor neurons or skeletal muscle have led us to hypothesize that loss of innervation induces muscle mitochondrial generation of both H2O2 and lipid peroxides, contributing to a significant loss of muscle function and atrophy.5 In this study, we have targeted removal of muscle mitochondrial electron transport chain-derived generation of H2O2 in the Sod1KO mouse model using muscle-specific expression of a mitochondrial-targeted H2O2 scavenger, human catalase (mMCAT).
Catalase is a peroxisomal antioxidant enzyme that catalyzes the decomposition of hydrogen peroxide generated by oxidation of a number of substrates including long-chain fatty acids. A whole-body mitochondrial catalase transgenic mouse (MCAT mice) was generated a decade and a half ago7 using a mitochondrial targeting sequence to direct the expression of the transgene specifically to the mitochondria where it could scavenge hydrogen peroxide generated by the electron transport chain. The MCAT mice displayed a significant enhancement of mitochondrial function, that is, reduced ROS production and energy imbalance.7 Global expression of MCAT has been shown to improve muscle function in mice, including enhanced exercise ability and improved maximum specific force, and also results in reduced pathology in other disease models.8 The tissue-specific (MCAT) transgenic mouse model utilized here is a unique tool that allowed us to generate skeletal muscle-specific expression through removal of a stop codon in the transgene under direction of a tissue-specific Cre recombinase to generate the mMCAT model.9 The mMCAT mice were crossed to the Sod1KO mice to determine whether scavenging muscle mitochondrial H2O2 would alter the sarcopenia phenotypes present in the Sod1KO mice.
Our results clearly show that muscle-specific mitochondrial expression of catalase is sufficient to prevent NMJ disruption and the associated muscle atrophy and weakness phenotypes in the Sod1KO mice, supporting an important role for muscle–neuron crosstalk and for maintaining muscle mitochondrial H2O2 in preventing deleterious muscle phenotypes associated with sarcopenia.
Materials and methods Generation of the whole-body CuZnSOD knockout (The Sod1KO mice used in this study were maintained on a C57BL/6J background. The mice were originally generated by Dr Charles Epstein's laboratory at the University of California San Francisco, and the details have previously been reported.4 Mitochondrial-targeted catalase (MCAT) flox mice were developed by inserting a human MCAT construct with a stop signal flanked by LoxP sites preceding the catalase gene between exons 2 and 8 of a human GAPDH bacterial artificial chromosome.9 Acta Cre transgenic mice10 (from JAX Stock No. 006149) were used to generate skeletal muscle-specific expression of the MCAT transgene through the removal of LoxP sites and the stop codon, allowing the expression of MCAT specifically in skeletal muscle (mMCAT mice). The expression of the ACTA Cre transgene is induced early in embryonic development10; thus, catalase expression was induced from early development as well. The mMCAT mice were bred to Sod1KO mice, to generate four groups of mice with different genotypes, wild-type (WT), WT/mMCAT, Sod1KO, and Sod1KO/mMCAT. Specifically, MCATflox/flox Acta Cre+/o Sod1−/− or MCATflox/flox Acta Cre+/o Sod1+/− male mice were bred to MCATflox/flox Acta Creo/o Sod1+/− female mice to generate MCATflox/flox Acta Cre+/o Sod1−/− (MCATSod1KO), and MCATflox/flox Acta Cre+/o Sod1+/+ (mMCAT Tg mice) and MCATflox/flox Acta Creo/o Sod1+/+ (WT) female control mice. All mice were caged in a pathogen-free environment with free access to standard chow and water and maintained on a 12 h light/dark cycle, and measurements were performed in female mice at around 12 months of age, unless otherwise stated. The Institutional Animal Care and Use Committee at Oklahoma Medical Research Foundation (Oklahoma City, OK, USA) approved all procedures.
Assessment of muscle contractile properties, functional denervation, and neuromuscular junction (NMJ) functionIsometric contractile force generation was measured in situ in gastrocnemius (GTN) muscle based on the methods described previously.11 Briefly, mice were anaesthetized with isoflurane, and the whole GTN muscle was isolated and cleaned from surrounding muscles and connective tissues. After dissection, the isolated muscle was tied with silk suture on the distal tendon, and the tendon was severed and mounted onto the force transducer (Model 305B, Aurora Scientific). The mouse was placed on a temperature-controlled platform at 37°C and provided with continuous anaesthesia. The electrode was placed on the surface of GTN muscle directly, and the muscle optimal length was adjusted with single 0.2 ms stimulation pulses until a maximum twitch reached. At the muscle optimal length, a series of 300 ms stimulus pulses were applied to achieve the maximum isometric tetanic force. After muscle stimulation, the electrode was moved from muscle to the sciatic nerve, and the nerve filament hooked firmly by the electrode. The same pulses of tetanic stimulus were applied to the nerve to achieve the nerve-induced maximum isometric tetanic force. All of the earlier tetanic twitch protocols were repeated several times to confirm the reproducibility and reliability of the data. Comparing muscle and nerve stimulated contractile force generation allows us to measure the extent of function denervation or loss of contractile force generation due to loss of intact NMJ innervation.
After all force measurements were completed, muscles were carefully removed and weighed, and the maximum tetanic force was normalized to the muscle cross-sectional area (CSA) calculated by the length and weight of the GTN muscle (dividing the muscle mass, mg, by the optimal length, mm, and the density of mammalian skeletal muscle, 1.06 g/cm3) to give the specific force (N/cm2). The NMJ function was presented by normalizing the nerve-induced force to the muscle-induced force as a percentage.
Measurement of mitochondrial respiration and reactive oxygen species (ROS) productionMitochondrial function was measured in permeabilized muscle fibres. The permeabilization protocol has been previously described by our laboratory.12 In general, a small piece of red GTN muscle was excised from the body and finely dissected to separate the muscle fibres along their striations in cold buffer X containing (in mM): 7.23 K2EGTA, 2.77 CaK2EGTA, 20 imidazole, 0.5 DTT, 20 taurine, 5.7 ATP, 14.3 PCr, 6.56 MgCl2·6H2O, and 50 K-MES (pH 7.1); 30 μg/mL saponin was added to the fibres to induce permeabilization for 30 min, followed by 5 min washes for three times in washing buffer containing (in mM): 105 K-MES, 30 KCl, 10 K2HPO4, 5 MgCl2·6H2O, 0.5 mg/mL BSA, and 0.1 EGTA (pH 7.1). After washing, the permeabilized fibres were placed into the Oxygraphy-2K (O2k, Oroboros Instruments, Austria) following the protocols described before.13 The OCR was determined using the oxygen probe, while the ROS production rate was measured by the O2K-Fluo LED2-Module Fluorescence-Sensor Green with Amplex UltraRed Reagent (Invitrogen, A36006). In a few random samples, 10 μM cytochrome c was added into the system to confirm the integrity of the mitochondrial outer membrane14 (Supporting Information, Figure S4). An increase >15% is considered to be evidence of a compromised preparation. An H2O2 standard curve was measured each time before the actual experiments to calibrate the level of H2O2. All data generated from O2K were normalized to the muscle wet weights and analysed with the official O2K software, DataLab Version 7.0.
Sarco(endo)plasmic reticulum ATPase (SERCA) activity assayThe sarco(endo)plasmic reticulum ATPase (SERCA) enzyme activity was measured in muscle homogenates at 37°C using a spectrophotometric assay as previously described.15 In brief, all muscle samples were homogenized following the ratio 1:10 with the SERCA homogenizing buffer, containing (in mM): 250 sucrose, 5 HEPES, 0.2 PMSF, and 0.2% NaN3. After centrifugation of the homogenates, the supernatant was taken with the protein amount of 100 μg and mixed with the SERCA assay buffer containing (in mM) 200 KCl, 20 HEPES, 10 NaN3, 1 EGTA, 15 MgCl2, 5 ATP, and 10 phosphoenolpyruvate, to generate a 3 mL mixture. At this point, 18 U/mL of lactate dehydrogenase and pyruvate kinase (PK/LDH) and 1 mM Ca2+ ionophore A-23187 (C-7522; Sigma) were added into the mixture. This reaction mixture was aliquoted and mixed with CaCl2 to form eight different calcium concentrations with pCa points from 7.6 to 4.2 and a blank and then loaded into a pre-warmed 37°C quartz plate. The reaction was initiated by adding 1 mM NADH into the mixture, and the kinetic assay was performed by the following settings: temp = 37°C, time = 30 min, λ = 340 nm, and shaking between readings. The SERCA activity was calculated using the formula
Skeletal muscle myosin heavy chain (MHC) composition was determined by distinguishing all four MHC isoforms, MHCIIa, MHCIIx, MHCIIb, and MHCI. In this study, GTN muscle samples were homogenized and denatured as described previously.16,17 Then the denatured muscle sample was separated with a special SDS-PAGE gel, the separating gel consisted of 32% v/v glycerol, 8% w/v acrylamide with a 50:1 ratio of acrylamide to N′-ethylenebisacrylamide or Bis, 0.2 M Tris–HCl (pH 8.8), 0.1 M glycine, 0.4% w/v SDS, 0.1% w/v ammonium persulfate, and 0.05% v/v N,N,N′,N′-tetramethylethylenediamine (TEMED). The stacking gel comprised 32% v/v glycerol, 4% w/v acrylamide, and Bis at the same ratio (50:1) to acrylamide, 70 mM Tris–HCl (pH 6.7), 4 mM EDTA, 0.4% w/v SDS, 0.1% w/v ammonium persulfate, and 0.05% v/v TEMED. The gel was then run using two different buffers, the lower running buffer consisting of 0.05 M Tris (base), 75 mM glycine, and 0.05% w/v SDS, and the upper running buffer, which was at 6× the concentration of the lower running buffer and had β-ME added (final concentration: 0.12% v/v). The gel underwent electrophoresis at 150 V at 4°C for 24 h, and immediately after running, Coomassie Brilliant Blue G250 was used to stain the gel and visualize the MHC bands. The images were collected by G:BOX Chemi (Syngene, USA), and the densitometry analysis was conducted using ImageJ software (ImageJ, Fiji).
Histological staining Muscle H&E stainingAfter the muscle was excised, it was placed in O.C.T mounting medium (Tissue-Tek, VWR, Radnor, PA, USA) and snap-frozen in isopentane pre-cooled in liquid nitrogen, so the myofibrils remained aligned properly. The muscle was cryosectioned into 8–10 μm cross-sections from the middle area of the whole muscle, and the muscle sections were then mounted onto slides. Slides were stained with the H&E staining kit from Thermo Fisher Scientific (Rapid Chrome H & E frozen section staining kit, #9990001), and then the slides were sealed with cover slips. Images were collected with Nikon Microphot-FXA microscope at ×20.
Immunohistochemistry staining for MHC isoformsAfter collecting the muscle sections as described earlier, the sections were left at room temperature for ~10 min, and then the fixative (4% paraformaldehyde in PBS; PFA) was added onto the sections for 30 min. Afterwards, the fixative was washed off for 3 × 10 min by PBS, which was then followed by adding the immunobuffer (50 mM glycine, 0.033% saponin, 0.25% BSA, 5% corresponding serum same to secondary antibody host, and 0.05% sodium azide) onto the sections with incubating for 2 h at room temperature as described previously.18 Then the immunobuffer-diluted primary antibody (MHCIIb (DSHB, BF-F3) with 1:20 dilution) was added onto the sections, and the sections were incubated for overnight at room temperature in an airtight, saturated container. After the primary antibody incubation, all slides were washed with immunobuffer followed by adding the secondary antibody (goat anti-mouse Alexa 647, Life Technology, #A21238) and incubating for 2 h in darkness. After the incubation, all slides were washed with PBS for 2 × 10 min and then sealed with coverslip under the presence of mounting medium (EMS, #17989). Immunofluorescent images were taken at the same magnificent (×20) with Nikon confocal microscope (Nikon Eclipse).
Confocal microscopy for NMJ morphologyAfter the muscle was excised from the body, small muscle pieces were taken under the cold PBS in a Petri dish along the fibre directions with a careful removal of fat and connective tissues. Muscle samples were transferred into a 24-well plate with 10% STUmol (Poly Scientific R&D, #2832) for 1 h to fix the tissue with gently shaking. After the fixation, washed tissues for 3 × 5 min in PBS at room temperature, and then permeabilized the tissue under the 2% Triton in PBS for 30 min on a shaker. After permeabilization, the tissues are placed into the blocking buffer containing, 4% BSA, 1% Triton, 5% serum matches the host of secondary antibody diluted in PBS, blocking for overnight in the cold room at 4°C. After blocking, primary antibodies, 1:50 SV2 (DSHB) for nerve terminals and 1:50 2H3 (DSHB) for neurofilaments, were added. The tissues were incubated overnight at 4°C, washed for 6 × 30 min in PBS at room temperature, and subsequently incubated with the secondary antibodies, 1:100 BTX-Alexa 488 (Invitrogen, #B13422), and 1:250 goat anti-mouse Cy3. After the incubation for secondary antibodies for overnight at 4°C, all tissues were then washed 6 × 30 min with PBS. Afterwards, tissues were transferred onto slides and mounted with mounting medium and sealed with nail polish. Images of NMJ were taken by the Nikon confocal microscope under the magnification of ×20, and Z-stacks were taken to show the 3D structure of intact NMJ. The total thickness of optical sections is around 20 to 60 μm, and the stack interval was set at every 1 to 2 μm, so there are around 20 to 30 images per stack.
The NMJ area was analysed as the area occupied by each individual-labelled AChR, and only the NMJs facing forward were analysed. The fragmentation level was quantified by counting the fragmented pieces of each NMJ, and if there are five or more pieces per junction, the NMJ is considered as a fragmented NMJ. The denervation score was quantified as follows: Score 0: there is no denervation, the NMJ is fully overlapped with neural filament; Score 1: partial denervation, the NMJ is partially overlapped with neural filament; and Score 2: complete denervation, the NMJ has no overlap with neural filament. The statistical analysis for all images taken in this section was performed by ImageJ software (ImageJ).
Western blotting and antibody informationGastrocnemius muscles were homogenized in RIPA buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, and protease inhibitors. Then the total protein was quantified with the Bio-Rad protein assay kit (Sigma-Aldrich, Poole, UK), and the same amount of protein was loaded and separated with SDS-PAGE gels at certain percentages, that is, 10% or 12%. The gel was then run at 200 V for 1 h and wet transferred onto 0.45 μm nitrocellulose membranes (Bio-Rad) with the conditions of 100 V, 30 min at 4°C, same as described before.19 After the transfer of the gel, total proteins in each lane were quantified using Ponceau staining (Sigma, #P3504), and then the membrane was washed with ddH2O to remove the Ponceau staining and blocked with 1% BSA solution in TBST for at least 1 h at room temperature. Soon after blocking, primary antibodies were added onto the membrane and incubated for overnight at 4°C. After the incubation of primary antibody, the membrane was washed with blocking buffer and then exposed to the secondary antibody for 30–60 min. After the secondary antibody, membrane was washed with TBST for the last time to clean the background. Protein bands were visualized and quantified using GeneTool system (SynGene, Frederick, MD, USA). The relative content of each protein measured using western blot analysis was normalized to sample total protein content measured using Ponceau stain and densitometry of total Ponceau in that sample lane. Primary antibody information was as follows: rabbit anti-Sod1 (Enzo, #ADI-SOD-101-E), rabbit anti-human catalase (Athens Research & Technology, #0105030000), rabbit anti-GAPDH (Sigma, #9545), SERCA1 (DSHB, #CAF2-5D2), SERCA2 (Cell Signaling, 4388S), CSQ1/2 (Abcam, #ab3516), and parvalbumin (Abcam, #ab11427).
Statistical analysisAll results are presented as mean values ± standard deviation, and comparisons among different groups were performed with one-way ANOVA and Tukey's multiple-comparison test. The statistical analysis was undertaken by GraphPad Prism 8, and the statistical significance was set at P values <0.05.
Results Validation ofThe protein levels of CuZnSOD (~17 kDa) and mMCAT human catalase (~63 kDa) were determined by western blotting for all groups using whole GTN muscle homogenates. Lack of CuZnSOD expression was confirmed in Sod1KO groups (with and without mMCAT) (Figure 1A and 1B). Muscles from WT/mMCAT and Sod1KO/mMCAT mice clearly show a similar level of elevated expression of the human catalase transgene, and there is no detectable expression of human catalase in WT and Sod1KO groups (Figure 1A and 1C). It is important to note that we confirmed that no human MCAT expression is detected in tissues from other organs (Figure S3). Moreover, the western blot image with isolated mitochondria from GTN muscle shows that the expression of the human catalase is limited to mitochondria and is not present in the cytosol (Figure 1D). Thus, the Sod1KO/mMCAT mice have increased expression of catalase specifically in the muscle mitochondria.
We have previously reported functional defects of muscle mitochondria from Sod1KO mice.1 In order to detect whether the expression of mMCAT in Sod1KO mice improves mitochondrial function, we measured the mitochondrial OCR and the ROS production rate as peroxide emission using permeabilized muscle fibres from red GTN muscle. Respiratory substrates were added to stimulate electron flow through different complexes (I, I + II, II, and IV), and fibres from the Sod1KO mice exhibited a lower OCR in all complexes compared with fibres from WT mice (Figure 2A). In contrast, the expression of mMCAT returned the OCR of all complexes in Sod1KO mice back to about the same level as seen in WT mice (Figure 2A). Moreover, in accordance with our previous findings, we found that the ROS production rate in State 1 respiration (mitochondria respiring without addition of external substrate) was significantly elevated in fibres from Sod1KO mice compared with fibres from WT mice (Figure 2B). mMCAT overexpression reduced ROS production rate in Sod1KO mice to the same level to WT mice (Figure 2B).
The measurement of body composition using quantitative magnetic resonance in the 12- to 14-month-old female Sod1KO mice revealed a significantly lower fat mass (~17%) compared with WT mice (~31%) (Figure 3A). The expression of mMCAT in the Sod1KO mice is associated with a level of fat mass in the normal control range (~32%) (Figure 3A). In agreement with our previous findings, we also found a significant decline in the absolute muscle mass of the GTN muscle (~46%, Figure 3B). When normalized to the body mass, the normalized muscle mass was also found to be significantly decreased (~30%) in Sod1KO mice as compared with WT mice (Figure 3C). However, in Sod1KO mice with mMCAT expression, both absolute and relative muscle mass in the Sod1KO mice was restored to the same level as in the WT mice (Figure 3B and 3C). Performance on a treadmill revealed that the running ability (distance to exhaustion) is significantly lower in Sod1KO compared with WT mice (305 vs. 709 m) but restored by ~35% to 463 m with the expression of mMCAT (Figure 3D).
Muscle fibre type is categorized by the MHC isoforms present. There are four different MHC isoforms present in adult mouse skeletal muscle, MHCI, MHCIIa, MHCIIx, and MHCIIb. The predominant isoforms in GTN muscle are the fast isoforms, MHCIIa, MHCIIx, and MHCIIb, and a small amount of the slow isoform MHCI. Because they have different molecular weights, all MHC isoforms can be separated by SDS-PAGE gel. In comparison with WT mice, there is a dramatic alteration in MHC composition in Sod1KO mice, in that the abundance of MHCIIb fibres is decreased approximately 50%. The level of MHCIIb fibres is restored by the presence of mMCAT to a level similar to WT mice, while the amount of MHCIIa and IIx remained the same (Figure 4A and 4B).
We also measured the CSA by H&E staining (Figure 4C). The statistical analysis shows that fibre diameter is ~20% smaller in muscle from Sod1KO mice compared with WT mice, and the reduced fibre size in Sod1KO mice is partially reversed by mMCAT expression, resulting in only a 10% smaller fibre size compared with WT mice (Figure 4D). Confocal images with MHCIIb staining indicate a significant reduction of CSA in MHCIIb fibres from Sod1KO mice (17%), which approximates the reduction of the total fibre CSA (Figure 4E and 4F). The presence of mMCAT fully restores the CSA of MHCIIb fibres in Sod1KO mice to the level measured in WT mice (Figure 4F).
Force generation and the structure and function of neuromuscular junction (NMJ) are improved inThe muscle maximum specific force and the function of NMJ were measured using in situ electrical stimulation. In Sod1KO mice, in agreement with our previous work, we found that the maximum specific force (16 N/cm2) is dramatically decreased compared with WT mice (25 N/cm2). Strikingly, the expression of mMCAT in Sod1KO mice is associated with the restoration of force to a comparable level measured in WT mice (24 N/cm2) (Figure 5A). The impact on NMJ function was determined by comparing the force generated using nerve stimulation through the NMJ to force generated by direct muscle stimulation, bypassing the NMJ. We found that the NMJ function in the Sod1KO mice is significantly impaired, resulting in a ~30% functional denervation (i.e. nerve force/muscle force = 70%) (Figure 5B). In contrast, WT mice show a nerve/muscle directed force ratio close to 100%, indicative of full NMJ function (Figure 5B). With the expression of mMCAT, the nerve-induced force in Sod1KO mice is fully returned to 100% of the muscle-induced force, equivalent to WT mice (Figure 5B). The confocal microscopy images support these findings showing that the structure of the NMJ in Sod1KO mice is damaged with fragmentation and denervation, and mMCAT expression restores the NMJ in Sod1KO mice to a normal structure, which is similar to WT mice (Figure 5C). The quantification and statistical analysis indicate that in the Sod1KO mice, the NMJ area is smaller (722 μm2) compared with WT mice (849 μm2), the proportion of fragmented NMJ is remarkably higher (74%) than WT mice (11%), and the denervated NMJ proportion is extremely high (96%) compared with WT mice (0.7%) (Figure 5D–5F). In the presence of mMCAT in the Sod1KO mice, all morphological NMJ parameters are similar to the WT mice; that is, the NMJ area is 909 μm2, the fragmented NMJ is 18%, and the denervated NMJ is 1.2% (Figure 5D–5F).
The activity of the SERCA pump, which is responsible for returning calcium to the SR following contraction and maintaining cytoplasmic calcium homeostasis, is sensitive to inactivation by oxidative stress,20 potentially contributing to sarcopenia. Figure 6A shows SERCA activity in GTN muscle homogenates. In comparison with WT mice, both the maximum and the overall (area under activity curve) SERCA activity is dramatically reduced (~35%) in Sod1KO mice (Figure 6B and 6C). The reduction of SERCA activity in Sod1KO mice is completely reversed with the presence of mMCAT, showing no difference to WT mice (Figure 6B and 6C).
Our previous work has shown that in addition to reduced SERCA activity, proteins involved in calcium homeostasis show altered expression in ageing and in muscle from Sod1KO mice.21 These changes can alter muscle metabolic pathways as well as EC coupling and contribute to sarcopenia. To determine whether mMCAT expression can modulate expression of calcium regulatory proteins, we measured protein abundance of calcium-regulating proteins in skeletal muscle, including the calcium storage protein inside SR, calsequestrin (CSQ1 and CSQ2), the SR transmembrane transport protein, SERCA (SERCA1 and SERCA2a), and the cytosolic calcium-binding and calcium-transporting protein, parvalbumin, using western blot analysis in whole muscle homogenates. The abundance of both CSQ1 and CSQ2, as well as the fast isoform SERCA1, was stable across all four groups with no change (Figure 7A, 7C, 7E, and 7F). However, the amount of slow SERCA isoform, SERCA2a, was decreased for ~50% in Sod1KO mice compared with WT mice (Figure 7B and 7D), and this reduction was restored by the presence of mMCAT, showing no difference compared with WT mice (Figure 7D). Moreover, the deletion of Sod1 also altered the expression of parvalbumin, and it was found that the amount of parvalbumin in Sod1KO mice increased significantly with 60% higher comparing with the WT mice (Figure 7B and 7G). The expression of mMCAT in Sod1KO mice reduced parvalbumin amount to that observed in WT mice (Figure 7G).
The novel findings presented in this study show that the muscle-specific mitochondrial expression of catalase is sufficient to prevent the majority of muscle atrophy and weakness phenotypes present in the CuZnSOD null (Sod1KO) mouse model of oxidative stress-induced accelerated sarcopenia. Importantly, the expression of catalase localized specifically in skeletal muscle mitochondria (mMCAT) in the Sod1KO mice not only reduced hydrogen peroxide release by muscle mitochondria but also prevented the disruption of NMJ morphology and function that normally occurs in response to lack of CuZnSOD and elevated oxidative stress in the Sod1KO mouse model. These findings are important because they suggest a link between skeletal muscle mitochondria and the NMJ and a crosstalk between the muscle and motor neurons that can modulate skeletal muscle atrophy and weakness.
Mitochondria are a predominant site for the generation of superoxide anion, producing superoxide at complexes I and III, that is released to the mitochondrial matrix and intermembrane space.22 The mitochondrial intermembrane space contains localized CuZnSOD (Sod1) that can scavenge superoxide generated from complex III, preventing superoxide release to the cytoplasm while contributing to increased potential H2O2 release to the cytosol, as well as potentially back into the mitochondrial matrix by diffusion. In the absence of CuZnSOD in the Sod1KO mice, superoxide generated at complex III may be released to the cytoplasm or potentially act to damage the electron transport chain causing mitochondrial dysfunction and further mitochondrial ROS generation. Conversely, superoxide released to the matrix from complexes I and III will be converted by MnSOD to H2O2 that is freely diffusible out of the mitochondrial matrix through the intermembrane space and out to the cytosol. The mitochondrial catalase, presumably in concert with the other mechanisms present to detoxify H2O2 such as peroxiredoxins and glutathione peroxidase, can scavenge this H2O2, preventing damage to the electron transport chain or mitochondrial membranes and prevent its release to the cytosol, thus protecting both the mitochondria and the cell from increased oxidative stress. Consistent with this model, the whole-body MCAT mice have been reported to have a number of positive effects as a result of enhancing antioxidant activity of mitochondria.8 In this study, we measured a number of indicators of mitochondrial function, including increased peroxide generation and reduced oxygen consumption, using permeabilized muscle fibres from WT, Sod1KO, and mMCATSod1KO mice. Mitochondrial proteins, in particular iron–sulfur centres of the respiratory enzymes (e.g. succinate dehydrogenase and NADH dehydrogenase), are specific targets of ROS and prone to oxidative modifications, potentially impacting the function of the electron transport chain and mitochondrial function.22 Therefore, our results showing that scavenging the level of H2O2 in mitochondria by expression of catalase completely restored the reduction in OCR and generation of mitochondrial peroxide in muscle from Sod1KO mice to WT levels support the hypothesis that controlling H2O2 levels can maintain mitochondrial function.
The Sod1KO mice exhibit a number of age-dependent muscle atrophy phenotypes, such as lower muscle mass, and an altered body index with less fat mass that are likely caused in response to high oxidative stress.4 In agreement with our previous findings, the results presented here show a similar effect on these parameters, as well as a reduction in running ability. Our results in this study also clearly confirm that the absence of Sod1 leads to a reduction in maximum specific force and a disruption of NMJ structures, and a failure of neuromuscular transmission and NMJ function. The whole-body Sod1 deletion likely increases mitochondrial ROS in both motor neurons and muscles, and the impairment of muscle and NMJ function is facilitated by compromised redox homeostasis.23 Previously, we showed that replacing CuZnSOD in neurons in the Sod1KO mice prevents NMJ disruption and the deleterious muscle phenotypes in the Sod1KO mice.24 The results we present here further demonstrate that expression of catalase restricted to muscle mitochondria can reduce muscle-specific mitochondrial ROS in the Sod1KO mice, preserving NMJ function and preventing muscle phenotypes as well. The NMJ fragmentation and increase in the percentage of denervated synapses in the Sod1KO mice are dramatically reversed by the expression of mitochondrial catalase in the Sod1KOxmMCAT mice. We saw a similar preservation of the NMJ and downstream muscle phenotypes in both ageing WT and Sod1KO mice maintained on dietary restriction.25 The NMJ preservation was associated with reduced mitochondrial ROS generation and reduced oxidative damage, consistent with and antioxidant mechanism of protection. It is possible that H2O2 release in the Sod1KO mice results in alterations in NMJ-associated proteins, the local environment of the synapse, or even in the motor neurons themselves through a muscle neuron retrograde pathway that can be reversed by increased expression of catalase and increased antioxidant protection. Fischer et al. found a similar preservation of the NMJ in a mouse model expressing human CuZnSOD targeted to the mitochondrial intermembrane space (mitoSod1) and crossed to the Sod1KO mice. Although the authors did not report changes in muscle mass, the expression of CuZnSOD in the mitochondrial intermembrane space was sufficient to prevent NMJ denervation in the Sod1KO mice up to 12 months of age and also prevented loss of grip strength.26 The mitoSod1 overexpression in this model was in several tissues including both neuronal and muscle tissues. These findings are consistent with our results indicating that reduced muscle mitochondrial oxidative stress may protect NMJ structure and function. Both muscle and nerve are highly metabolic and likely sensitive to changes in mitochondrial function that can alter metabolism. One potential candidate for retrograde communication between muscle and neurons is peroxisome proliferator-activated receptor gamma coactivator-1α (PGC-1α), which has been shown to mediate muscle nerve communication during activity and to control the expression of a number of NMJ-related proteins including AChRs, muscle-specific kinase, and utrophin.27
Muscle dysfunction can be associated with a number of potential factors, among them, an impairment in calcium regulation is considered to play a significant role.28 One of the key enzymes in the calcium-regulating system is the SERCA, which executes the reuptake of cytoplasmic calcium into SR lumen maintaining the low Ca2+ concentration and protecting the contractile proteins after muscle contraction.29 However, the SERCA pump is impaired in many chronic pathologies, including ageing,30 muscular dystrophy,31 and denervation,32 and oxidative modification of SERCA has been reported as a potential underlying cause in the inactivation of SERCA pumps.33 We have previously shown that Sod1KO mice have reduced SERCA activity in hindlimb skeletal muscle.20 In the current study, we found that the expression of mMCAT fully restored SERCA activity in muscle from Sod1KO mice, suggesting that the reduction of hydrogen peroxide and oxidative stress by mMCAT efficiently protects SERCA function. It is probable that the restored SERCA activity contributes to the overall reduced atrophy and function phenotypes in the mMCAT × Sod1KO mice.
As a correlate to SERCA activity, we also measured the levels of calcium-regulating proteins such as SERCA, CSQ, and parvalbumin. SERCA and CSQ, two SR-related calcium-regulating proteins, both have two main different isoforms, with SERCA1 and CSQ1 being the fast isoform, and SERCA2a and CSQ2 being the slow isoform.17 Our analysis showed that the amount of SERCA1 protein remained the same across all groups, supporting concept that the change in SERCA activity is not due to a change in protein abundance. In contrast, the amount of SERCA2a protein was decreased dramatically in Sod1KO mice. The muscle used for this analysis was GTN muscle, which is almost entirely composed of fast-twitch fibres, and <5% of fibres are slow-twitch fibres.34 SERCA2a is a slow isoform, and therefore, the change in the level of SERCA2a is not likely to contribute significantly to the activity change in GTN muscle. In addition, the MHC analysis in this study suggests that the reduction of SERCA2a is most likely due to the decreased abundance of the MHCI fibre type in Sod1KO mice. The amounts of both CSQ1 and CSQ2 proteins are stable across the different groups, indicating that the SR calcium storing ability is likely not affected by the elevated oxidative stress in Sod1KO mice. Interestingly, the abundance of parvalbumin was increased significantly in Sod1KO mice, with levels more than 50% higher than in WT mice, and the increase was reversed by expression of mMCAT. Parvalbumin is a cytosolic calcium/magnesium binding and buffering protein that facilitates the SR Ca2+ sequestration from cytosol side. In skeletal muscles, it is predominantly expressed in fast-twitch fibres where it functions to bind and transport calcium ions off from troponin C to SERCA during the relaxation. The expression of parvalbumin is therefore highly sensitive to the cytosolic Ca2+ concentration. It is possible that the elevated amount of parvalbumin in Sod1KO mice is attributable to the defect in SERCA function, resulting in increased cytosolic Ca2+ concentration or a prolonged clearance time for cytosolic Ca2+, therefore increasing parvalbumin concentration. In addition, our previous findings also noted that the half-relaxation time is longer in Sod1KO mice,20 suggesting a prolonged period of high of Ca2+ levels in the cytosol, which might also be responsible for the increased expression of parvalbumin. Taken together, the elevated expression of parvalbumin may be reversed by the restoration of SERCA activity, consistent with our findings in the Sod1KO mice with the presence of mMCAT.
In accordance to our previous findings, we described a phenotype in Sod1KO mice, in which the relative abundance of MHCIIb fibres is reduced compared with WT mice,1 while other MHC isoforms are not affected. For this measurement, we used GTN muscle, a predominantly fast-twitch muscle consisting of predominantly fast-twitch fibres, MHCIIa, IIb, and IIx.34 Previous studies have reported that type II fibres are more prone to ageing-induced atrophy,1 and there is a tendency for fibre type switching from type II to type I fibres.35 Here, we observed that only type IIb fibres are affected in the Sod1KO mice, not all type II fibres. In addition, we did not find conversion from type IIb fibres to other fibre types, rather we find a decrease in type IIb fibre diameter, suggesting that this is a fibre atrophy rather than a compositional change. The atrophy in type IIb fibres perfectly matches the decrease of the total CSA in whole muscle, supporting the fact that the total muscle atrophy is due primarily to the atrophy occurring in type IIb fibres. The potential underlying mechanism of this atrophy may be due to the regulating effects from PGC-1α.36 It has been well documented that PGC-1α is sensitive and up-regulated under the high oxidative stress environments, such as after highly intensive exercise,37 and activated likely by phosphorylation of the PGC-1α protein by p38 MAPK together with NF-κB, with both of which are known to be activated by ROS.38 PGC-1α plays a pivotal role on regulating muscle fibre compositions, by inhibiting the generation of glycolytic fibres (IIb/IIx) and promoting the synthesis of oxidative fibres (IIa/I).39 Another potential mechanism could be the compromised muscle function in Sod1KO mice, where increased sedentary behaviour and reduced demand for activity may contribute to loss of the IIb fibres, the predominant active fibre type in rodent fast-twitch muscles.17 The expression of mMCAT was highly effective in restoring the altered fibre composition in Sod1KO mice. This could be due to reduced impact of oxidative stress on PGC-1α and maintenance of the type IIb fibre size in Sod1KO mice. Although we did not measure spontaneous cage activity in the Sod1KO or Sod1KOxmMCAT mice, it is possible that MCAT expression may contribute indirectly to increased activity that along with maintained muscle contractile function may preserve the healthy fibre composition in Sod1KO mice. Notably, mMCAT expression did significantly improve the running ability in Sod1KO mice, although it remained lower than in WT mice. However, this is not necessarily surprising as treadmill running exercise is a complex whole-body network of actions that requires many other organs that are critical in running exercise, such as heart, artery, and lung functions. Finally, attributed to the restoration of the normal fibre contribution, the reduced muscle maximum specific force in Sod1KO mice is fully restored.
In summary, the results in this study clearly demonstrate that mitochondrial expression of catalase can prevent sarcopenia-like phenotypes induced by genetic deletion of Sod1, including muscle atrophy and weakness, NMJ disruption, inactivation of SERCA, mitochondrial defects, and a significant exercise intolerance. Our results clearly show that enhanced hydrogen peroxide scavenging potential in muscle mitochondria can maintain the NMJ structure and function in the Sod1KO mice and prevent the sequelae of events that occur downstream of loss of innervation to induce phenotypes associated with muscle atrophy and weakness in the Sod1KO mice. These finding support the importance of NMJ maintenance in sarcopenia.
FundingThis work was supported by a P01 grant (NIA-AG051442) and R01 grant (NIA-AG050676) from the National Institute on Aging (NIA) and also the VA Senior Research Career Scientist Award of the U.S. Department of Veterans Affairs (IK6 BX005234) to H.V.R. A.R. is also supported by a Senior Research Career Scientist Award (1IK6BX005238) from the Department of Veterans Affairs.
Author contributionsH.X., A.R., and H.V.R. contributed in the conception and design of the research; H.X. and R.R. performed the experiments and analysed the data; H.X., A.R., and H.V.R. interpreted the results of the experiments; H.X. prepared the figures and drafted the manuscript; H.X., A.R., and H.V.R. edited and revised the manuscript; and H.X., R.R., A.R., and H.V.R. approved the final version of the manuscript.
AcknowledgementsWe gratefully acknowledge the mCAT conditional mouse model that was provided by Dr Peter Rabinovitch from the University of Washington Seattle. The original generation and characterization of this mouse model was supported by the NIH P01 grant (NIH P01 AG001751) to Dr Rabinovitch. The authors certify that they comply with the ethical guidelines for authorship and publishing of the Journal of Cachexia, Sarcopenia and Muscle.40
Conflict of interestNone declared.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2021. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Background
Oxidative stress and damage are associated with a number of ageing phenotypes, including age‐related loss of muscle mass and reduced contractile function (sarcopenia). Our group and others have reported loss of neuromuscular junction (NMJ) integrity and increased denervation as initiating factors in sarcopenia, leading to mitochondrial dysfunction, generation of reactive oxygen species and peroxides, and loss of muscle mass and weakness. Previous studies from our laboratory show that denervation‐induced skeletal muscle mitochondrial peroxide generation is highly correlated to muscle atrophy. Here, we directly test the impact of scavenging muscle mitochondrial hydrogen peroxide on the structure and function of the NMJ and muscle mass and function in a mouse model of denervation‐induced muscle atrophy CuZnSOD (Sod1−/− mice, Sod1KO).
Methods
Whole‐body Sod1KO mice were crossed to mice with increased expression of human catalase (MCAT) targeted specifically to mitochondria in skeletal muscle (mMCAT mice) to determine the impact of reduced hydrogen peroxide levels on key targets of sarcopenia, including mitochondrial function, NMJ structure and function, and indices of muscle mass and function.
Results
Female adult (~12‐month‐old) Sod1KO mice show a number of sarcopenia‐related phenotypes in skeletal muscle including reduced mitochondrial oxygen consumption and elevated reactive oxygen species generation, fragmentation, and loss of innervated NMJs (P < 0.05), a 30% reduction in muscle mass (P < 0.05), a 36% loss of force generation (P < 0.05), and a loss of exercise capacity (305 vs. 709 m in wild‐type mice, P < 0.05). Muscle from Sod1KO mice also shows a 35% reduction in sarco(endo)plasmic reticulum ATPase activity (P < 0.05), changes in the amount of calcium‐regulating proteins, and altered fibre‐type composition. In contrast, increased catalase expression in the mMCAT × Sod1KO mice completely prevents the mitochondrial and NMJ‐related phenotypes and maintains muscle mass and force generation. The reduction in exercise capacity is also partially inhibited (~35%, P < 0.05), and the loss of fibre cross‐sectional area is inhibited by ~50% (P < 0.05).
Conclusions
Together, these striking findings suggest that scavenging of mitochondrial peroxide generation by mMCAT expression efficiently prevents mitochondrial dysfunction and NMJ disruption associated with denervation‐induced atrophy and weakness, supporting mitochondrial H2O2 as an important effector of NMJ alterations that lead to phenotypes associated with sarcopenia.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
1 Aging & Metabolism Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA
2 Department of Biochemistry & Molecular Biology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA; Oklahoma City VA Medical Center, Oklahoma City, OK, USA
3 Aging & Metabolism Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA; Oklahoma City VA Medical Center, Oklahoma City, OK, USA