Skeletal muscle is highly adaptable in response to increases and decreases in contractile activity. The purpose of this study was to determine whether the preconditioning of skeletal muscle has a protective effect against subsequent denervation-induced apoptotic protein expression. To investigate this, we chronically stimulated the tibialis anterior and extensor digitorum longus muscles for 7 days (10 Hz, 3 h/day) before 7 days of denervation. Denervation reduced total cytochrome-c oxidase activity by 39%, which was likely a consequence of a decrease in subsarcolemmal (SS) mitochondria. This decrease in the SS subfraction was prevented by prior chronic stimulation and, as a result, maintained total mitochondrial content at control levels. The expression of Bax was elevated 2.2-fold by denervation, and prior chronic stimulation did not attenuate this increase. This produced a increase in the Bax-to-Bcl-2 ratio, indicating greater muscle apoptotic susceptibility. Denervation also decreased state 3 respiration in SS and intermyofibrillar mitochondria and elevated state 4 reactive oxygen species production within both mitochondrial subfractions. These changes were not prevented by prior chronic stimulation. Furthermore, the antioxidant protein MnSOD was also reduced by denervation, whereas Beclin-1 was markedly elevated. This suggests that autophagic cell death could also play a significant part in denervation-induced muscle atrophy. Thus, despite prior chronic stimulation, denervation increases the apoptotic susceptibility of skeletal muscle by altering the Bax-to-Bcl-2 ratio, by increasing reactive oxygen species production, and by reducing the expression of MnSOD. Whether a more extensive stimulation paradigm would be more effective in attenuating apoptosis before muscle disuse remains to be determined.
- mitochondrial biogenesis
- muscle atrophy
- reactive oxygen species
- protein degradation
skeletal muscle is an adaptable tissue that exhibits a remarkable range of plasticity in response to different levels of contractile activity (15). In particular, chronic contractile activity is known to alter the expression of a variety of proteins from the nuclear and mitochondrial genomes (15). These changes in protein expression result in an increase in mitochondrial volume and density (6, 16). In skeletal muscle, two distinct subfractions of mitochondria exist that are characterized by their anatomic localization. Subsarcolemmal (SS) mitochondria are located just beneath the sarcolemmal membrane, whereas intermyofibrillar (IMF) mitochondria are intertwined between skeletal muscle myofibrils (8). SS and IMF mitochondrial subfractions are significantly increased in response to chronic contractile activity. However, SS mitochondria are more sensitive to endurance training and, as a result, demonstrate larger increases in mitochondrial volume than the IMF subfraction (17). In addition to changes in mitochondrial volume, chronic contractile activity also alters organelle protein composition by elevating the expression of antiapoptotic factors and by reducing the release of proapoptotic proteins (1). Thus chronic contractile activity induces a number of adaptations in skeletal muscle, which result in an increase in muscular oxidative capacity and in a reduction in the factors that stimulate the induction of mitochondrially mediated apoptosis.
In contrast to chronic contractile activity, chronic muscle disuse results in a rapid loss of skeletal muscle mass (27) and an impairment in oxidative metabolism (12). These changes in muscle composition are facilitated by rapid increases in muscle protein degradation via the preferential expression of proteins that promote muscle atrophy (23). Prolonged periods of muscle disuse also result in a reduction in metabolic enzymes (27) and in a decrease in the proteins needed for mitochondrial biogenesis (3). As a result, disused skeletal muscle demonstrates a lower resistance to fatigue (27) and an increase in mitochondrial dysfunction (3). Furthermore, recent work has shown that chronic muscle disuse is also a potent activator of mitochondrially mediated apoptosis. Denervation-induced muscle disuse led to an increase in the expression of proapoptotic proteins and the fragmentation of nuclear DNA (3). Hence, chronic muscle disuse is a powerful regulator of skeletal muscle metabolism and apoptosis.
To date, little research has focused on elucidating the possible protective effects that chronic contractile activity may have on the reductions in oxidative capacity and the incidence of mitochondrially mediated apoptosis induced by muscle disuse. Encouraging work performed by Allen et al. (4) has shown that a training period before hindlimb suspension can reduce the incidence of apoptosis in skeletal muscle. The purposes of the present study were to determine whether prior chronic contractile activity could attenuate the denervation-induced effects on 1) mitochondrial function and 2) apoptotic protein expression. We hypothesized that the reductions in mitochondrial function and the increases in mitochondrially mediated apoptosis, which have been shown to be caused by denervation, would be attenuated by prior chronic contractile activity.
MATERIALS AND METHODS
Experimental design 1: chronic stimulation.
Male Sprague-Dawley rats (n = 30, 400–450 g; Charles River, St. Constant, Quebec, Canada) were given food and water ad libitum. The implantation of electrodes and the chronic stimulation procedures were performed as previously described (26) and were approved in accordance with the regulations of the York University Animal Care Committee. Briefly, rats were anesthetized with an intraperitoneal injection of xylazine-ketamine (0.2 ml/g body wt). Under sterile conditions, a 1-cm incision was made in the left hindlimb of the animal. Two platinum electrodes were sutured down on either side of the common peroneal nerve. Stimulating wires were then passed subcutaneously and exteriorized at the base of the neck. The electrodes were then fastened to a portable stimulation pack, which was attached to the animal. Incisions made in the muscle were closed with sutures, and the skin was closed with staples. Animals were given 1 wk to recuperate from the surgery, and then the chronic stimulation protocol (10 Hz, 0.1-ms duration, 3 h/day) was initiated for the next 7 days. The contralateral hindlimb was sham operated. Thus, in experimental design 1, every animal was subjected to a chronic stimulation surgery on their left hindlimb and a sham-operation on their right hindlimb.
Experimental design 2: denervation.
After the chronic stimulation protocol was completed, animals were either 1) denervated or 2) sham operated. Denervation surgeries were performed as previously described (27). Briefly, rats were anesthetized with an intraperitoneal injection of xylazine-ketamine (0.2 ml/g body wt), and a 1-cm incision was made in both hindlimbs. A 1-cm portion of the common peroneal nerve was removed from both hindlimbs. Thus the tibialis anterior (TA) and the extensor digitorum longus (EDL) muscles in the left and the right hindlimbs were denervated. The muscle incisions were closed with sutures, and the skin was closed with staples. The sham operation followed all of the experimental procedures of the denervation protocol except for the excision of the common peroneal nerve. Both the sham-operated and the denervated animals remained in the animal care facility for the next 7 days and were then euthanized. The TA muscles were extracted and used for the mitochondrial isolation procedure, whereas the EDL muscles were extracted and immediately freeze clamped for protein analysis via immunoblotting. Therefore, after the completion of the two experimental protocols, there were four experimental groups: 1) a sham group that received two sham operations, 2) a stimulated group that received a chronic stimulation surgery followed by a sham operation, 3) a denervated group that received a sham operation followed by a denervation surgery, and 4) a prior chronic stimulation group that received a stimulation surgery followed by a denervation surgery.
The TA muscles were extracted, weighed, blotted to remove excess liquid, quickly minced, and homogenized. Differential centrifugation was used to separate the SS and the IMF mitochondrial subfractions as previously described (19). Isolated SS and IMF mitochondrial subfractions were resuspended in medium (100 mM KCl, 10 mM MOPS, 0.2% BSA) and used for the analysis of mitochondrial respiration, reactive oxygen species (ROS) production, and protein composition.
SS and IMF mitochondrial respirations were evaluated with a Clarke oxygen electrode (Yellow Springs Instruments, Yellow Springs, OH). Mitochondrial oxygen consumption was recorded at 30°C with constant stirring. A volume of 300 μl of mitochondrial sample was added to 2 ml of VO2 buffer (250 mM sucrose, 50 mM KCl, 25 mM Tris·HCl, 10 mM K2HPO4, 0.2% BSA, pH 7.4). Respiration was measured in the presence of 1) 11 mM glutamate (state 4 respiration) and 2) 0.4 mM ADP (state 3 respiration) as previously done (2, 8) (n = 20).
ROS production was measured in SS and IMF mitochondria (n = 10). This assay was designed to evaluate total ROS production, which includes various forms of ROS such as hydrogen peroxide and peroxynitrite (9). Briefly, mitochondria (50 μg) were incubated with VO2 buffer and 50 μM dichlorodihydrofluorescein diacetate in a black polystyrene 96-well plate. ROS production was assessed during state 4 respiration by adding 10 mM glutamate immediately before the addition of dichlorodihydrofluorescein diacetate. ROS production was measured with a Synergy HT microplate reader, and the data were compiled with KC4 (v3.0) software.
Whole muscle protein extracts or isolated mitochondria were separated by 10–12% SDS-PAGE and transferred to nitrocellulose membranes with a wet electrotransfer apparatus (Mini Trans-Blot electrophoretic transfer cell, Bio-Rad, Mississauga, Canada). Nitrocellulose membranes were blocked in a 5% skim milk solution containing 1× TBST [Tris-buffered saline-Tween 20, 25 mM Tris·HCl (pH 7.5), 1 mM NaCl, 0.1% Tween 20]. Membranes were then incubated overnight at 4°C with the appropriate concentration of primary antibody (Bax, 1:500; Bcl-2, 1:200; p45, 1:200; Beclin-1, 1:500; MnSOD, 1:2,500; cyclophilin D, 1:400; Tim23, 1:500). All antibodies were obtained from Santa Cruz, with the exception of Beclin-1 (Cell Signaling) and p45 (a kind gift of Dr. G. DeMartino, University of Texas Southwestern Medical School). Membranes were subsequently washed in TBST to remove any excess primary antibody and then incubated with the appropriate secondary antibody at room temperature for 60 min. Membranes were then washed in TBST and developed with the enhanced chemiluminescence method. Films were scanned and quantified using SigmaScan Pro (version 5) software (Jandel Scientific, San Rafael, CA). To control for loading, protein quantifications were corrected with GAPDH immunoblotting (1:20,000). Our data indicate that there was no effect of denervation or chronic stimulation on the expression of GAPDH over the time course of the experimental design used in this study.
Cytochrome-c oxidase activity.
Whole muscle extracts were diluted 20-fold in extraction buffer (0.1 M KH2PO4, 2 mM EDTA, pH 7.2) and sonicated to disrupt cellular membranes. After centrifugation, the supernatant fractions were recovered as previously described (8). Enzyme activity was determined by the maximal rate of cytochrome c oxidation (COX) at 550 nm. COX activity was measured with a clear 96-well plate in the Synergy HT microplate reader, and the data were compiled by KC4 software.
Data are expressed as means ± SE. Two-way repeated-measures ANOVA was used to determine whether there was a main effect of denervation and/or an effect of stimulation. A Bonferroni post hoc test was applied to detect differences within groups. Data were considered statistically different if P < 0.05.
Body weight and muscle mass.
Final body weights for the sham-operated and denervated animals were not different from each other (Table 1). TA muscle mass was not altered by 7 days of chronic stimulation, but muscle mass was decreased by ∼35% after 7 days of denervation (Table 1). Importantly, prior chronic stimulation did not prevent the denervation-induced reduction in TA muscle mass.
Mitochondrial yield and COX activity.
SS and IMF mitochondrial yields were measured as an estimate of how each subfraction responded to altered levels of contractile activity. A main effect of denervation was observed on SS mitochondrial yield, which was reduced by 37% (P < 0.05; Fig. 1A). There was no overall effect of chronic stimulation on SS mitochondrial yield; however, the denervation-induced decrease was prevented by prior chronic stimulation (P < 0.05; Fig. 1A). Interestingly, there was no main effect of denervation on IMF mitochondria yield (Fig. 1B), but the yield was increased by 1.8-fold as a result of chronic stimulation (P < 0.05; Fig. 1B). Furthermore, prior chronic stimulation maintained the elevation in IMF mitochondrial content despite the 7 days of chronic muscle disuse.
COX activity was measured to provide a more accurate assessment of whole muscle mitochondrial content. Denervation depressed COX activity by ∼40% (P < 0.05; Fig. 1C). In contrast, a main effect (P < 0.05) of chronic stimulation was observed in both the sham and denervated groups. Compared with that shown in the sham-operated control muscle, after chronic stimulation, COX activity was induced by 38% (P < 0.05; Fig. 1C). Chronic stimulation effectively attenuated the decrease in COX activity to a level that was not significantly different from that shown in sham-operated control muscle.
SS and IMF mitochondrial respiration.
Mitochondrial respiration was measured in SS and IMF subfractions to assess organelle function. Chronic stimulation did not have an effect on either state 3 or state 4 SS mitochondrial respiration values (Fig. 2, A and C). In contrast, a main effect of denervation was observed during both state 3 and state 4 respiration within the SS subfraction (P < 0.05; Fig. 2, A and C). This decline was not prevented by prior chronic stimulation (Fig. 2, A and C). Similarly, state 3 respiration in IMF mitochondria was reduced after both denervation and prior chronic stimulation (P < 0.05; Fig. 2B). However, chronic stimulation also resulted in a reduced state 3 respiration in IMF mitochondria (P < 0.05; Fig. 2B). State 4 respiration within IMF mitochondria remained unchanged in response to all four experimental conditions (Fig. 2D).
ROS production during mitochondrial respiration.
ROS production in SS and IMF mitochondria was not affected by chronic stimulation during state 3 respiration (data not shown). However, in response to denervation, both mitochondrial subfractions displayed increases in ROS production during state 4 respiration (P < 0.05; Fig. 3). This denervation-induced increase was not prevented by prior chronic stimulation in either SS or IMF mitochondria (Fig. 3).
Chronic stimulation did not affect the protein expression of MnSOD, cyclophilin D (Cyp D), or Tim23 (Fig. 4). However, denervation decreased the expression of all three proteins (P < 0.05; Fig. 4). Prior chronic stimulation attenuated the denervation-induced decrease in both Cyp D (P < 0.05; Fig. 4C) and Tim23 (Fig. 4A). MnSOD protein expression was not significantly affected by prior chronic contractile activity (Fig. 4B), but this treatment did appear to maintain MnSOD expression closer to sham-operated control levels.
Expression of the anti-apoptotic protein Bcl-2 and expression of the ubiquitin-protein ligase p45 were not affected by chronic stimulation or denervation (Fig. 5, A and B). Bax protein expression was also not affected by chronic stimulation (Fig. 5C). However, denervation resulted in a 2.2-fold increase (P < 0.05) in the expression of Bax, which was not reduced by prior chronic stimulation (Fig. 5C). Thus the Bax-to-Bcl-2 ratio was increased in response to denervation and was not reduced by prior chronic contractile activity. A similar pattern of expression was observed for the autophagy-related protein Beclin-1. To assess whether autophagy could be activated in response to muscle disuse, we measured Beclin-1 protein expression. In response to denervation, Beclin-1 expression was increased by 2.7-fold (P < 0.05), a change that was not attenuated by prior chronic stimulation (Fig. 5D).
Chronic contractile activity leads to an improvement in muscular endurance performance (13, 16). Increases in muscular activity, induced by either electrical stimulation or treadmill running, have been shown to increase the expression of oxidative enzymes and to result in an elevation in mitochondrial content (14, 16). Importantly, chronic electrical stimulation of skeletal muscle has also been shown to have a protective effect against mitochondrially mediated apoptosis (1). In contrast, skeletal muscle as a result of denervation has been shown to dramatically reduce skeletal muscle mass and to decrease muscle endurance performance (3, 27). These adaptations have been attributed, at least in part, to increases in the rate of muscle protein degradation and to the activation of mitochondrially mediated apoptosis (3, 20, 21). Thus the primary purpose of this study was to determine whether chronic contractile activity, induced by electrical stimulation before denervation, could have a protective effect against the expression of apoptotic proteins in skeletal muscle.
It has been well established that skeletal muscle mitochondrial content is particularly sensitive to changes in contractile activity (1, 3, 7, 14, 26, 27). In the present study, we show that chronic stimulation results in an increase in whole muscle mitochondrial content (Fig. 1C) and that this increase is primarily attributable to an elevation in IMF mitochondria (Fig. 1B). Conversely, whole muscle mitochondrial content was reduced in response to denervation (Fig. 1C), likely as a result of a decline in SS mitochondria (Fig. 1A). This denervation-induced decrease in the SS subfraction was prevented by prior chronic stimulation, which maintained whole muscle mitochondrial content at sham-operated control levels. To our knowledge, this is the first study to show that decrements in skeletal muscle mitochondria, caused by denervation, can be attenuated by prior chronic stimulation. Furthermore, we were not overly surprised by the preferential reduction in SS mitochondria in response to denervation, as this subfraction has been shown to have a heightened sensitivity to alterations in muscular activity (11, 16).
To determine whether altered levels of contractile activity have an influence on whole muscle apoptotic susceptibility, we measured the expression of Bax and Bcl-2. The Bax-to-Bcl-2 ratio is used as an index to evaluate whether permeability transition pores such as the mitochondrial permeability transition pore (mtPTP) (22) and the mitochondrial apoptotic-inducing channel (10) are being formed in the mitochondrial membrane. In agreement with previous denervation studies (3, 25), our data show a robust increase in Bax protein expression with denervation (Fig. 5C), which was not prevented by prior chronic stimulation. This change in Bax protein expression shifts the Bax-to-Bcl-2 ratio in favor of apoptosis. Other studies have shown that the denervation-induced change in the Bax-to-Bcl-2 ratio contributes to a faster rate of mtPTP opening (3) and to higher concentrations of apoptosis-inducing factor (AIF) and cytochrome c within the cytosol (25). In addition, we also measured a decrease in the expression of Cyp D in response to denervation. Consequently, the formation of the mtPTP could be inhibited during denervation-induced muscle disuse, whereas other mitochondrial transition pores, such as the mitochondrial apoptotic-inducing channel, may be mediating more protein release than originally hypothesized. Together, these data suggest that denervation increases the apoptotic susceptibility of skeletal muscle by elevating the expression of proapoptotic proteins and that prior chronic stimulation of this length and duration does not prevent these changes. If the pro- and antiapoptotic proteins that were assessed in this study responded to increases in skeletal muscle contractile activity, we would have anticipated a greater anti-apoptotic effect. Nonetheless, it remains to be determined whether a longer duration of chronic stimulation before denervation would have exerted a greater effect on the expression of proapoptotic proteins. However, our results, along with our previous work (3), suggest that a chronic contractile activity model that is properly matched with an equal disuse paradigm has the potential to exert a protective effect against disuse-induced apoptosis by increasing the expression of anti-apoptotic factors (1), which could subsequently disrupt the activation of apoptotic signaling.
ROS are potent activators of mitochondrially mediated apoptosis, and their generation has been shown to be inversely related to the rates of mitochondrial respiration (2, 24). As a result, we assessed ROS production and the rate of mitochondrial respiration in SS and IMF mitochondria during state 3 (active) and state 4 (passive) respiration. Previous work has demonstrated that denervation decreases SS mitochondrial state 3 and state 4 respiration (3, 18), with a concomitant increase in ROS production (3). Our results extend these findings to show that denervation suppresses mitochondrial state 3 and state 4 respiration but that prior chronic stimulation does not attenuate this reduction in either SS or IMF mitochondria (Fig. 2). Moreover, these reductions in respiration were accompanied by increases in both SS and IMF mitochondrial ROS production (Fig. 3A). Thus we believe that denervation induces a change in mitochondrial composition, resulting in an elevation in ROS production, and a consequent reduction in organelle function. Furthermore, we speculate that these decrements in mitochondrial function could be a consequence of 1) a decrease in the expression of electron transport chain proteins (3), 2) a mitochondrial protein import dysfunction, or 3) ROS-mediated damage of the electron transport chain, which accelerates organelle change in a positive feedback manner (3). Consistent with the second possibility is the fact that the expression of Tim23, a component of the protein import machinery (Fig. 4A), is markedly disrupted by denervation. The enhanced ROS production observed during denervation could also be partly attributable to a reduction in the antioxidant defense enzyme MnSOD (Fig. 4A). Interestingly, prior chronic stimulation appears to attenuate this decline in MnSOD (Fig. 4A) and may have contributed to a slightly lower level of ROS production observed in mitochondria isolated from these muscles (Fig. 3B). However, further investigation is needed to determine whether the expressions of other antioxidant enzymes (i.e., glutathione peroxidase, catalase) are also affected by denervation and whether prior chronic stimulation prevents their decrease. Furthermore, the elevations in chronic contractile activity-induced SS ROS production were not unexpected, as we have shown this adaptation previously (1) and determined that modest increases in SS ROS production function to activate redox sensitive transcription factors rather than promote apoptosis.
In addition to apoptotic cell death, denervation-induced muscle atrophy is also attributable to an increase in the rate of protein degradation by the ubiquitin proteasome pathway and may also involve the activation of autophagic cell death. In this study, we provide a preliminary assessment of how these two pathways respond to altered levels of contractile activity (Fig. 5). Autophagy was evaluated by measuring the expression of Beclin-1, an essential protein for this type of cell death. In response to denervation, Beclin-1 expression was dramatically increased, and this increase was not prevented by prior chronic stimulation. These data provide an initial evaluation of the role that autophagy might play during denervation-induced muscle disuse, warranting further investigation. Conversely, the expression of p45, a component of the 26S proteasome, was not altered by any of the experimental treatments. Thus, on the basis of this initial information, we believe that autophagic cell death is invoked by denervation despite prior chronic stimulation and that it may contribute to denervation-induced muscle atrophy. The lack of change in p45 expression does not necessarily imply an absence of proteasome-mediated protein degradation because the proteasome is composed of many different proteins and p45 is only one component within the 19S complex (5). Thus other proteins within this oligomeric complex may more accurately represent possible changes in proteasome-mediated protein degradation in response to different levels of contractile activity, as previously reported (20, 21, 28).
In conclusion, based on our previous work (1, 3), we hypothesized that prior chronic stimulation would attenuate the increases in apoptosis observed during denervation. Our results show that prior chronic stimulation does not inhibit the expression of the proapoptotic protein Bax or prevent changes in organelle function, which lead to an increase ROS production. However, there is mounting evidence to suggest that prior chronic stimulation could potentially ameliorate the deleterious effects of muscle disuse by increasing the expression of antiapoptotic factors, such as MnSOD, which could impede the activation of apoptotic signaling cascades. Our previous work (1, 3), which supports this hypothesis, has shown that mitochondrially mediated apoptosis is increased in muscle tissue that has reduced mitochondrial content and is decreased in those muscles that possess higher levels of mitochondria. In addition to these observations, both of our previous studies also noted that, along with the changes in mitochondrial content, there was a subsequent alteration in the apoptotic environment of the cell. In our present study, we utilized an extreme model of muscle disuse that completely prevents contractile activity of skeletal muscle, and, as a consequence, it is possible that chronic denervation is overwhelming the antiapoptotic effects of chronic contractile activity that have been previously observed. Therefore, it is reasonable that a milder model of muscle disuse, such as hindlimb suspension, would provide more insight into the protective effects of chronic contractile activity. Further research in the area is warranted because the elucidatation of whether chronic contractile activity has a protective effect against myonuclear cell death could have a profound influence on how therapeutic treatments are designed for those with muscular diseases, such as sarcopenia.
This work was supported by a National Sciences and Engineering Research Council of Canada grant to D. A. Hood.
We are grateful to Sophia Kapchinsky for technical assistance during this study.
D. A. Hood holds a Canada Research Chair in Cell Physiology.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society