ELECTROMECHANICAL MODULATION OF CATABOLIC AND ANABOLIC PATHWAYS IN CHRONICALLY INACTIVE, BUT NEURALLY INTACT, MUSCLES

Abstract

Electromechanical modulation of catabolic and anabolic pathways in chronically inactive, but neurally intact, muscles

INTRODUCTION

The extent and mechanisms by which neural input regulates skeletal muscle mass remain largely unknown.

METHODS

Adult spinal cord isolated (SI) rats were implanted unilaterally with a microstimulator while the contralateral limb served as an SI control (SI-C). A 100-Hz, 1-sec stimulus was delivered every 30 sec for 5 min followed by 5-min rest. This was repeated six times consecutively (SI-Stim1) or with a 9-hr interval after the third bout (SI-Stim2) for 30 days (1 min daily activity).

RESULTS

SI-Stim1 and SI-Stim2 paradigms attenuated plantaris atrophy by 20% and 38%, whereas only SI-Stim2 blunted soleus atrophy (24%) relative to SI-C. Muscle mass changes occurred independent of the IGF-1/PI3K/Akt pathway. No relationships between SI or electromechanical stimulation and expression of several atrophy markers were observed.

CONCLUSIONS

These data suggest that regulatory mechanisms for maintaining muscle mass previously shown in acute states of atrophy differ substantially from those observed in chronic states.

INTRODUCTION

The amplification level for control of gene expression from a single motoneuron to thousands of myonuclei among the hundreds of muscle fibers within a single motor unit is remarkable. Both activity-dependent and activity-independent neurally induced signals mediate this amplification. The activity-dependent events that modulate muscle gene expression can be mediated either neurally or non-neurally. For example, denervation and spinal cord isolation (SI) both result in near “inactivity” of muscle, however the motoneuron-muscle connectivity remains intact with SI. The maintenance of skeletal muscle mechanical and phenotypic properties closer to control after SI compared to denervation demonstrates that there are activity-independent neural factors in the SI model ¹.

Despite the amplified motoneuron influence on skeletal muscle protein homeostasis, the amount of neurally mediated signaling needed to exert a coordinated expression of genes among the myonuclei within and across fibers remains largely unknown. The multiple systems that modulate protein synthesis and degradation must be controlled in a highly coordinated fashion to maintain both muscle protein homeostasis and specific muscle fiber phenotypes. Significant progress has been made in identifying molecular and cellular pathways that regulate muscle mass, e.g., insulin-like growth factor-1 (IGF-1) and ubiquitin-proteasome signaling systems, but these data are based largely on muscles in a relatively acute, dynamic state ^2–4. IGF-1 stimulates muscle protein synthesis and hypertrophy via the phosphatidylinositol 3-kinase (PI3K)/Akt pathway ^5–8. Activation of this pathway suppresses protein degradation induced by atrophy-related ubiquitin ligases, e.g., atrogin-1 and muscle-specific ring finger 1 (MuRF-1) ⁹, whereas down-regulation of the pathway leads to up-regulation of these atrogenes and subsequent atrophy ^6,10,11. Combined, these data suggest a well-coordinated balance between catabolic and anabolic pathways to maintain muscle mass. These data, however, provide little insight into how the genes respond with prolonged absence of neurally mediated muscle activation. Furthermore, the effectiveness of imposing daily short periods of activation in modulating the regulatory pathways that can preserve muscle mass in a chronic inactivity model is unknown.

The goal of this study was to examine these issues from several novel perspectives. First, near inactivity of motoneurons in the isolated spinal cord region and associated musculature induced by SI allowed us to administer known levels of electromechanical stimuli to muscles that were otherwise devoid of neuromuscular activity. As such, we examined the effectiveness of a stimulation paradigm delivered in one vs. two daily sessions in preserving muscle mass during 30 days of SI. Second, the responsiveness of different signaling pathways was examined in a chronic in vivo preparation, whereas other studies have focused on acute or in vitro preparations.

The principal findings are that some, but not all, components of the IGF-1 pathway remain highly responsive after 30 days of SI and after brief periods of daily stimulation. Additionally, no relationships among SI, electromechanical stimulation, and expression of catabolic markers were evident, suggesting that these pathways play a minimal role in regulating the mass of chronically inactive muscles.

MATERIALS AND METHODS

Experimental groups

Adult female Sprague-Dawley rats (229 ± 2 g body weight, n = 6/group) were assigned randomly to a sedentary control (Con-P1 or Con-P2) or SI (SI-P1 or SI-P2) group. All rats in the SI groups underwent SI surgery. In addition to the SI surgery, SI-P1 and SI-P2 rats were implanted unilaterally with a functional microstimulator in the thigh region. Rats in the SI-P1 (protocol 1) and SI-P2 (protocol 2) groups received unilateral electromechanical hindlimb stimulation training (SI-Stim1 and SI-Stim2, respectively), while the contralateral limb served as an internal control (SI-C1 and SI-C2, respectively). The same amount of stimulation was delivered in either one session/day (SI-Stim1) or two sessions/day (SI-Stim2).

Surgical procedures and animal care

SI surgery was performed under aseptic conditions as described previously ¹². Briefly, the rats were anesthetized with an intraperitoneal injection of a mixture of ketamine hydrochloride (100 mg/kg body weight) and xylazine (5 mg/kg body weight). The spinal cord was transected completely at a mid-thoracic and a high sacral level, and an intradural dorsal rhizotomy was performed bilaterally between the two transection sites. On the same day of the SI surgery, the rats assigned to the SI-Stim groups were implanted unilaterally with a microstimulator as described previously ¹³. Briefly, a skin incision (~10 mm) was made in the upper lateral thigh, a small opening was made in the fascia near the greater trochanter, and the microstimulator was inserted between the vastus lateralis and biceps femoris muscles in parallel to the sciatic nerve. The surgical areas and the skin incision were closed using 4-0 Dexon and 4-0 Ethilon, respectively.

Post-surgical care involved manual bladder expression three times daily for the first three days and twice daily thereafter. Both hindlimbs were manipulated passively once per day through a full range of motion to prevent joint fixation. Motor tests were performed routinely to verify that the muscles in both hindlimbs were inactive, i.e., toe-spreading, withdrawal reflex, and limb extension to toe pinching. None of the SI rats responded to these motor reflex tests throughout the study. All rats used in the study were fed rat chow and water ad libitum for the entire duration of the study. All surgical and post-surgical procedures have been detailed elsewhere ¹⁴. All experimental and animal care procedures used in this study were approved by the UCLA Chancellor’s Animal Research Committee and followed the American Physiological Society Animal Care Guidelines.

Electromechanical stimulation

The microstimulator is a wireless implantable radio frequency powered device (2.4 mm diameter × 16.7 mm length) designed for functional electrical stimulation. It is composed of a hermetically sealed ceramic capsule containing an electronic subassembly¹⁵. The microstimulator receives power and stimulation commands (pulse width, pulse amplitude, frequency stimulation duration, cycle time) via a 2 MHz magnetic field generated by an external radio frequency coil that is connected to a microstimulator control unit.

The hindlimb of each SI-Stim rat implanted with a microstimulator was secured in a removable cast (ankle at ~90° and knee at ~120°) during the stimulation sessions to optimize isometric contractions of the muscles of interest as described previously¹³. The stimulation paradigm for the SI-Stim1 rats consisted of one session of 6 bouts of a 1-sec tetanic isometric contraction (100 Hz) delivered once every 30 sec for 5 min followed by 5 min of rest. The parameters for the SI-Stim2 paradigm were the same, except that there were 2 sessions of 3 bouts of stimulation and the sessions were separated by a 9-hr interval. For both paradigms, the stimulation pulse width was maintained at 50 µs, and the threshold pulse amplitude (µA) was determined at the beginning of each stimulation bout by doubling the minimum pulse amplitude necessary to produce a maximum twitch response. Pulse amplitude was adjusted as necessary to maintain maximum torque at the ankle throughout a session.

Daily stimulation training sessions began 2 days after surgery and continued for 30 consecutive days. For both groups, the total duration of the training sessions was ~1 h/day, while the total stimulation duration was 1 min or 0.069% of a 24-h period. To assure that the contralateral, non-stimulated limb could serve as an internal SI control, an additional leg cast with a force transducer as described above was secured on the non-implanted leg of two rats during a number of stimulation bouts. No measurable force was observed in the contralateral non-stimulated limbs. In addition, we routinely palpated the non-stimulated limb of each rat to verify that the lower limb muscles were inactive during the stimulation bouts. The day after the final training session animals were overdosed with sodium pentobarbital (100 mg/kg), and several hindlimb muscles, including the plantaris and soleus, were dissected rapidly bilaterally from each rat, cleaned of excess fat and connective tissue, wet weighed, quick frozen on dry ice, and then stored at −80°C until used for cellular and molecular analyses.

Biochemical and molecular analyses

As previously described by Haddad et al.¹⁶, a portion from the mid-belly of the plantaris muscle was cut, weighed, and homogenized in 20 vol of ice-cold homogenization buffer (250 nM sucrose, 100 mM KCl, 5 mM EDTA, and 10 mM Tris·HCl, pH 6.8). A known volume of the total homogenate was used for myofibrillar protein extraction based on the modified protocol of Solaro et al.¹⁷. The muscle total protein and myofibrillar protein concentrations were calculated from the protein concentration of the total homogenate and myofibril suspension obtained with the Bio-Rad protein assay using gamma globulin as a standard. Total DNA concentration was determined using a volume of the total homogenate and a fluorometric assay^16,18. Total protein, myofibril, and DNA content were determined based on their concentration values multiplied by the plantaris whole muscle wet weight. The soleus was used only for RNA and RT-PCR experiments, because there was insufficient tissue to perform the other analyses.

RNA extraction

Total RNA from the plantaris and soleus muscles from each rat was extracted from pre-weighed frozen samples (~25 to 40 mg) using the TRI reagent (Molecular Research Center, Cincinnati, OH) per the manufacturer’s instructions. RNA concentration was determined by optical density at 260 nm (using an optical density 260-unit equivalent to 40 µg/ml). Total RNA concentration for each sample was calculated based on the total RNA yield and the weight of the analyzed sample. The RNA samples were stored at −80° C and used for subsequent reverse transcription-polymerase chain reaction (RT-PCR) analyses.

RT-PCR analyses

One µg of total RNA was reverse transcribed for each muscle sample with SuperScript II RT (Invitrogen, Carlsbad, CA) per the manufacturer’s specifications. One µl of each RT reaction (0- to 20-fold dilution depending on target mRNA abundance) was used for the PCR amplification. The PCR reactions were carried out in the presence of optimized concentration of MgCl₂ using standard PCR buffer (Bioline), 0.2 mM dNTP, 1µM specific primer set, and 0.75 unit of Biolase DNA polymerase (Bioline, Genesee, San Diego, CA) in 25 µl total volume as described previously^5,19,20. Amplifications were carried out in a Stratagene Robocycler with an initial denaturing step of 3 min at 96°C, followed by 25 cycles of 1 min at 96°C, 45 sec at 55°C (55–60°C depending on primers), 45 sec at 72°C, and a final step of 3 min at 72°C. The PCR data for a specific mRNA target were normalized to muscle weight and thus are reported as arbitrary units/mg muscle weight as reported previously^21,22. The sequences for the various primers used for these experiments are shown in Table 1 (Supplemental Material). All primers were purchased from Operon Biotechnologies (Huntsville, AL), and for each primer set, PCR conditions (cDNA dilutions and number of PCR cycles) were optimized so that the target mRNA product yields were in the linear range of the semi-log plot when the yield is expressed as a function of the number of cycles. All samples were run in duplicate.

Western blot analysis

The total and phosphorylated states of p70S6k at Thr 389, Akt at Thr 308, mammalian target of rapamycin (mTOR) at Ser 2448 (Cell Signaling Technology, Beverly, MA), and 4E-BP (Santa Cruz Biotechnology, Santa Cruz, CA) were examined by immunoblotting using total and phospho-specific antibodies. Total protein from pre-weighed frozen plantaris muscle (~30–35 mg) was extracted as previously described ¹⁹. Briefly, muscles were homogenized in 7 vol of ice cold homogenization buffer, centrifuged at 12,000 × g for 30 min at 4° C. The supernatant was transferred and saved in aliquots at −80° C. The supernatant concentration was determined by using the Bio-Rad protein assay with BSA as the standard. Approximately 50 µg of protein was run on a 6% (mTOR), 10% (p70S6k) or 12.5% (Akt, 4E-BP) SDS-PAGE gel and transferred to a polyvinylidene difluoride membrane (Immobilon-P, Millipore, Billerica, MA). The enhanced chemiluminescence method was used for detection (Pierce, Rockford, IL). Signal intensity was determined by laser scanning densitometry (Molecular Dynamics/Image Quant). For each antibody, all of the samples were run under identical (previously optimized) conditions. A positive control provided by the antibody company was run on each gel to allow for normalization. For each set of Western blotting and detection conditions, the detected signal was directly proportional to the amount of protein loaded on the gel over a range of 20–150 µg (data not shown).

Statistical analysis

All data are presented as means ± SEM. Group differences for all analyses were determined by one-way ANOVA followed by the Neuman-Keuls post hoc test using the Prism software package (GraphPad). The significance level for all analyses was set at P ≤ 0.05.

RESULTS

Body and muscle weights

The mean body weights for the control groups from both studies were combined, as they were not statistically different. The mean body weights for the SI-P1 and SI-P2 groups were 21% and 32% lower than in the Con group after 30 days of SI (Fig. 1A). In addition, the mean body weight of the SI-P2 group was 11% lower than the SI-P1 group. The significant decrease in body weights after 30 days of SI are consistent with previous observations at the same time point¹².

Figure 1.

Body weight (A) for control (Con), spinal cord isolated protocol 1 (SI-P1) and SI protocol 2 (SI-P2) groups. Absolute (B, D) and relative (to body weight; C, E) plantaris and soleus weights, respectively, for Con, SI non-stimulated control for P1 (SI-C1), SI stimulated group for P1 (SI-Stim1), SI non-stimulated control for P2 (SI-C2), and SI stimulated group for P2 (SI-Stim2) groups. Values are presented as means ± SEM. *, †: significantly different from Con, SI-C, or SI-Stim1, respectively, at P ≤ 0.05.

The plantaris and soleus weights from the control groups from both studies were combined as they were not statistically different. After 30 days, the mean absolute plantaris weights were 46% and 52% smaller in the SI-C1 and SI-C2 rats compared to the Con group, respectively (Fig. 1B). The absolute plantaris weights were 20% and 39% larger in the SI-Stim1 and SI-Stim2 than in the SI-C1 and SI-C2 groups, respectively, but remained smaller than in the Con group. Relative (to body weight) plantaris weights were 32% and 28% smaller in the SI-C1 and SI-C2 groups than in the Con1 and Con2 groups, respectively (Fig. 1C). The relative weights of the SI-Stim1 and SI-Stim2 groups were 20% and 38% larger than their respective SI-C1 and SI-C2 groups, with the relative weight of the SI-Stim2 group being not different from Con.

The mean absolute soleus weight was 58% and 65% smaller in the SI-C1 and SI-C2 groups compared to the Con group, respectively (Fig. 1D). SI-Stim1 had no effect on absolute muscle weight, whereas SI-Stim2 resulted in a 24% larger absolute soleus weight compared to SI-C2. The soleus in the SI-Stim2 group remained significantly smaller (56%) than Con. Relative (to body weight) soleus weights showed similar trends as observed for absolute weights, with the SI-Stim2 group showing a larger (24%) relative weight than the contralateral SI-C2 (Fig. 1E).

Cellular properties

The cellular data from the plantaris muscle from the control groups from both studies were combined, as they were not statistically different. At the 30-day time point, the total protein concentration was not different from Con (239 ± 8 mg/g) in the SI-C1, SI-Stim1, and SI-Stim2 groups (204 ± 5, 239 ± 6, and 231 ± 19 mg/g, respectively) (Fig. 2A). It was lower than Con in the SI-C2 group (181 ± 9 mg/g), resulting in a 28% higher total protein concentration in the SI-Stim2 than SI-C2 group. The total protein content was lower in all SI-C and SI-Stim groups (33 ± 4, 26 ± 2, 46 ± 3, and 47 ± 3 mg/muscle for the SI-C1, SI-C2, SI-Stim1, and SI-Stim2 groups, respectively) than in Con (73 ± 3 mg/muscle) (Fig. 2B). Total protein content was 45% and 84% higher in the SI-Stim1 and SI-Stim2 compared to the SI-C1 and SI-C2 groups, respectively.

Figure 2.

Cellular properties of the plantaris after 30 days of SI. (A) Protein concentration (mg/g), (B) protein content (mg/plantaris), (C) myofibril concentration (mg/g), and (D) myofibril content (mg/plantaris), (E) total DNA concentration (mg/g), (F) DNA content (µg/plantaris), (G) total protein-to-DNA ratio (mg/µg), (H) RNA concentration (mg/g), and (I) total RNA content (µg/plantaris). Values are presented as means ± SEM. *, †, ‡: significantly different from Con or SI-C, respectively, at P ≤ 0.05. Abbreviations same as in Fig. 1.

The total myofibril protein concentration was lower than Con (85 ± 2 mg/g) in the SI-C1, SI-C2, and SI-Stim1 (54 ± 3, 62 ± 3, and 67 ± 7 mg/g, respectively) groups, but it was not different from Con in the SI-Stim2 (91 ± 4 mg/g) group (Fig. 2C). Myofibril concentration was 48% higher in the SI-Stim2 than SI-C2 group, whereas there was no difference between the SI-Stim1 and SI-C1 groups. The total myofibril content was lower in all SI-C and SI-Stim groups (9 ± 1, 8 ± 0, 13 ± 2, and 17 ± 1 mg/muscle for the SI-C1, SI-C2, SI-Stim1, and SI-Stim2 groups, respectively) compared to the Con group (26 ± 1 mg/muscle) (Fig. 2D). Myofibril content was 51% and 108% higher in the SI-Stim1 and SI-Stim2 than SI-C1 and SI-C2 groups, respectively. The effect of electromechanical stimulation on total myofibril concentration and content was greater in the SI-Stim2 than the SI-Stim1 group.

Total mean DNA concentration was higher than Con (1.16 ± 0.03 mg/g) in the SI-C1, SI-C2, and SI-Stim1 (2.03 ± 0.15, 1.98 ± 0.14, and 1.71± 0.07, respectively) groups, but not different from Con in the SI-Stim2 (1.42 ± 0.12 mg/g) group (Fig. 2E). Mean DNA concentration was 27% lower in the SI-Stim2 than the SI-C2 group, whereas there was no difference between the SI-Stim1 and SI-C1 groups. The effect of electromechanical stimulation on total DNA concentration was lower in the SI-Stim2 than in the SI-Stim1 group. The mean DNA content was not different from Con (339 ± 12 mg/muscle) across all SI-C and SI-Stim groups (~325 for the SI-C1 and SI-Stim1 and ~271 mg/g, for the SI-C2, SI-Stim2 groups) and SI-C and SI-Stim groups (Fig. 2F). The protein-to-DNA ratio was lower in all SI-C and SI-Stim groups (0.10 ± 0.01, 0.09 ± 0.01, 0.14 ± 0.01, and 0.16 ± 0.01 mg/µg for the SI-C1, SI-C2, SI-Stim1, and SI-Stim2 groups, respectively) than in the Con group (0.21 ± 0.01 mg/µg) (Fig. 2G). This ratio was 38 and 81% higher in the SI-Stim1 and SI-Stim2 groups, respectively, compared to their respective SI-C groups.

The total mean RNA concentration was higher than Con (0.80 ± 0.04 mg/g) in the SI-C1 and SI-C2 groups (1.12 ± 0.09 mg/g) and not different from Con in the SI-Stim1 and SI-Stim2 groups (0.8 ± 0.08 and 0.90 ± 0.05, respectively) (Fig. 2H). Mean RNA concentration was 26% and 21% lower in the SI-Stim1 and SI-Stim2 groups compared to their respective SI-C groups. The total mean RNA content was lower in all SI-C and SI-Stim groups (174 ± 14, 151 ± 16, 153 ± 13, and 162 ± 12 µg/muscle for the SI-C1, SI-C2, SI-Stim1, SI-Stim2 groups, respectively) than in the Con group (230 ± 14 µg/muscle) (Fig. 2I), and not different between the respective SI-Stim and SI-C groups.

The only cellular data obtained for the soleus muscle was RNA concentration and content, because there was insufficient tissue to perform both the cellular and RT-PCR experiments. Total mean RNA concentration in the soleus muscle was similar to that observed in the plantaris (data not shown). RNA concentration was higher in the SI-C1 and SI-C2 groups (1.32 ± 0.07 and 1.35 ± 0.04 mg/g, respectively) than in the Con group (1.10 ± 0.03 mg/g) and not different from Con in the SI-Stim1 and SI-Stim2 groups (1.11 ± 0.07 and 0.95 ± 0.03 mg/g, respectively). RNA concentration in the SI-Stim1 and SI-Stim2 groups was 16 and 30% lower than in the SI-C1 and SI-C2 groups, respectively. Total mean RNA content was lower in all SI-C and SI-Stim groups (71 ± 15, 61± 3, 57 ± 8, 53 ± 3 µg/muscle for the SI-C1, SI-C2, SI-Stim1, and SI-Stim2 groups, respectively) than in the Con group (140 ± 6 µg/muscle). There were no differences in the RNA content between the respective SI-Stim and SI-C groups.

Anabolic markers of muscle mass

These data show that administering the same amount of electromechanical stimulation in two bouts per day was more effective than one bout per day in preserving muscle mass and myofibril concentration and content. Therefore, we focused our molecular analyses of the pathways known to regulate muscle mass, i.e. the IGF-1 and the ubiquitin-proteasome signaling systems, on the muscles from the Con2, SI-C2, and SI-Stim2 groups. Only RT-PCR data are reported for the soleus because of the small amount of tissue available.

IGF-1 mRNA levels (normalized to muscle weight) were 3-fold higher in the SI-C2 than Con2 group and similar in the SI-Stim2 and Con2 group (Fig. 3A) despite the significant atrophy observed after 30 days of inactivity (Fig. 1B). Mechano growth factor (MGF), a spliced variant of IGF-1, and IGF-1 binding proteins BP-4 and BP-5 showed similar trends as for IGF-1. All mRNA levels were higher in the SI-C2 than the Con2 group, and there were no differences between the Con2 and SI-Stim2 group (Figs. 3B–D). IGF-1, MGF, BP-4, and BP-5 mRNA expression in the soleus muscle showed similar trends as for the plantaris. The mRNA levels were higher in the SI-C2 than Con2 group, and there were no differences between the SI-Stim2 and Con2 groups, with the exception of BP-5 mRNA which was 2-fold higher than in Con2 (Fig. 3D).

Figure 3.

mRNA expression levels of insulin-like growth factor-1 (IGF-1, A), mechano growth factor (MGF, B), and IGF-1 binding protein-4 (BP-4, C) and -5 (BP-5, D) in plantaris and soleus based on RT-PCR analyses. Each panel includes a representative gel of the PCR products. Data are presented as arbitrary units (AU) per weight of the muscle (mg/muscle). Values are presented as means ± SEM. *, †: significantly different from Con2 or SI-C2, respectively, at P ≤ 0.05. Abbreviations same as in Fig. 1.

In the plantaris muscle, phosphorylated Akt (Fig. 4A) protein expression was 89% and 97% lower in the SI-C2 and SI-Stim2 groups, respectively, than in the Con2 group (Fig. 4A), whereas total Akt protein expression was not different across all groups (Fig. 4B). In contrast, the phosphorylated state of Akt was 91 and 95% lower in the SI-C2 and SI-Stim2 groups, respectively, than in the Con2 group (Fig. 4C). Phosphorylated mTOR (Fig. 5A) protein expression was 50% higher in the SI-C2 compared to SI-Stim2 group and not different between the Con2 and SI-C2 groups. Total mTOR protein expression and the phosphorylyated state of mTOR were not different across all groups (Figs. 5B and 5C).

Figure 4.

Analysis of phosphorylated (^T308p) (A), total (B), and phosphorylated state (ratio of phosphorylated to total) (C) of Akt protein in the plantaris based on Western blot analyses. A representative blot is included for the phosphorylated and total measures. Values are presented as means ± SEM. *: significantly different from Con2 at P ≤ 0.05. Abbreviations same as in Fig. 1.

Figure 5.

Analysis of phosphorylated (^S2448p) (A), total (B), and phosphorylated state (ratio of phosphorylated to total) (C) of mTOR protein in the plantaris based on Western blot analyses. A representative blot is included for the phosphorylated and total measures. Values are presented as means ± SEM. †: significantly different from SI-C2 at P ≤ 0.05. Abbreviations same as in Fig. 1.

The two downstream targets of Akt, i.e., p70S6k and 4E-BP, showed similar responses. Phosphorylated (Fig. 6A) and total p70S6k (Fig. 6B) protein levels were 61 and 34% higher, respectively, in the SI-C2 compared to SI-Stim2 group and not different between the Con2 and SI-C2 groups. The phosphorylated state of p70S6k was ~12- and 15-fold higher in the SI-C2 than in the Con2 and SI-Stim2 groups, respectively (Fig. 6C). Phosphorylated 4E-BP (Fig. 7A) and total 4E-BP (Fig. 7B) protein levels were 92 and 35% higher, respectively, in the SI-C2 compared to Con2 group and not different between the SI-C2 and SI-Stim2 groups. The phosphorylated state of 4E-BP was ~7-fold higher in the SI-C2 compared to Con2 group and not different between the SI-C2 and SI-Stim2 groups (Fig. 7C).

Figure 6.

Analysis of phosphorylated (^T389p) (A), total (B), and phosphorylated state (ratio of phosphorylated to total) (C) of p70S6k protein in the plantaris based on Western blot analyses. A representative blot is included for the phosphorylated and total measures. Values are presented as means ± SEM. *, †: significantly different from Con2 or SI-C2, respectively, at P ≤ 0.05. Abbreviations same as in Fig. 1.

Figure 7.

Analysis of phosphorylated (p) (A), total (B), and phosphorylated state (ratio of phosphorylated to total) (C) of 4E binding protein (4EBP) in the plantaris based on Western blot analyses. A representative blot is included for the phosphorylated and total measures. The utilized antibody reacts with 4EBP, with the phosphorylated form migrating slower on the gel, i.e., top band. The total 4EBP expression was calculated as the sum of the density of all bands. The phosphorylated state of 4EBP was calculated as the ratio between the density of the top band and the sum of the top and bottom bands. Values are presented as means ± SEM. *: significantly different from Con2 at P ≤ 0.05. Abbreviations same as in Fig. 1.

Catabolic markers of muscle mass

The mRNA levels of key markers involved in skeletal muscle protein degradation, i.e., atrogin-1 (Fig. 8A) and MuRF-1 (Fig. 8B), were not different across groups in both the plantaris and soleus muscles. In the plantaris muscle, myostatin mRNA levels also were not different across groups. These values in the soleus, however, were 87 and 68% lower in the SI-C2 and SI-Stim2 groups, respectively, compared to the Con2 group and 54% higher in the SI-Stim2 than SI-C2 group (Fig. 8C).

Figure 8.

mRNA expression levels of skeletal muscle atrophy genes atrogin-1 (A), muscle ring finger-1 (MuRF-1, B), and myostatin (C) in the plantaris and soleus based on RT-PCR analyses. Data are presented as arbitrary units (AU) per weight of muscle (mg/muscle). Each panel includes a representative gel of the PCR products. Values are presented as means ± SEM. *, †: significantly different from Con2 or SI-C2, respectively, at P ≤ 0.05. Abbreviations same as in Fig. 1.

DISCUSSION

Amount and pattern of electromechanical stimulation needed to preserve the mass of inactive, but neurally intact, fast and slow muscle

Our results indicate that the twice per day stimulation protocol was more effective for both the plantaris and soleus muscles and, in fact, the soleus was responsive only when it was stimulated twice per day. The greater effectiveness of the twice per day stimulation protocol in the plantaris was evident in measures of total myofibrillar protein concentration and content, and DNA concentration. We previously have reported that the same amount of electrical stimulation administered for 30 consecutive days in two sessions per day resulted in maintenance of the mass and mechanical properties of the rat medial gastrocnemius muscle closer to control levels than if administered in one session per day¹³. Moreover, preliminary data from the medial gastrocnemius muscle of rats after 7 days of SI clearly show that the frequency at which muscles are stimulated and the stimulation duration are key parameters, whereas the number of bouts delivered per session and the total activity delivered per day have a lesser influence on preserving muscle mass²³. Consistent with this, a recent acute (7 days) SI study showed that electromechanical stimulation delivered at 100 Hz for 4 sec delivered in two sessions (3 bouts per session) per day for a total period of 4 min (0.28% of a 24 hr period) maintained the relative muscle mass near control in both the medial gastrocnemius and soleus muscles²⁰.

Other studies of chronic models of decreased activity including denervation²⁴, hindlimb unloading²⁵, and SI (cat)¹ also clearly demonstrate the efficacy of administering a small amount of stimulation or training to maintain muscle mass. Combined, these data indicate that, under conditions of acute and chronic decreased use, muscle mass can be maintained by a relatively small amount of total daily activity and that distributing the activity in multiple sessions per day is more effective than administering the same amount of activity in a single session.

Expression of protein translation and synthesis markers mediated by IGF-1 are affected differentially after chronic inactivity and electromechanical stimulation

IGF-1 induces the phosphorylation of Akt which then activates mTOR and its downstream targets p70S6k and 4E-BP to initiate protein translation and synthesis^26,27. IGF-1 also can prevent atrophy by suppressing atrogin-1 and MuRF-1 transcription via Akt⁹. Presently, IGF-1 mRNA levels were significantly higher in the soleus and plantaris after 30 days of SI, consistent with that observed in the soleus after 8 and 15 days of SI, which the authors attributed to a counter-regulatory response to conserve muscle mass at a new homeostatic state¹⁹. Interestingly, phosphorylated Akt protein levels in the plantaris were markedly lower compared to Con and did not reflect the large increase in IGF-1 mRNA levels, whereas phosphorylated mTOR, p70S6k and 4E-BP proteins were higher than Con after 30 days of inactivity. Although our data seem to contrast with observations made using in vitro and acute and chronic models of decreased use, they are consistent with an increase in phosphorylated p70S6k protein observed in transgenic mice that overexpress IGF-1 that occurred via an Akt-independent mechanism²⁸. When combined, these data strongly suggest an alternative mechanism. Evidence from in vivo studies suggests that mTOR can be activated by a variety of stimuli including amino acids, contractile activity, insulin, IGF-1, and other growth factors and that this activation occurs independent of Akt to activate its downstream targets p70S6k and 4E-BP^29–31. Consistent with this, mice that overexpress IGF-1 showed increased phosphorylated mTOR independent of Akt activity and further demonstrated the anabolic effect of IGF-1 on the activation of the mTOR/p70S6k pathway²⁸. The increased expression of phosphorylated mTOR, p70S6k, and 4E-BP independent of Akt activity in our study also strongly supports the possibility of an mTOR-dependent mechanism initiated by IGF-1.

These data also are consistent with previous reports of variable expression of p70S6k (phosphorylated or total – no significant changes) and 4E-BP (phosphorylated or total – no significant changes) proteins in the soleus muscle after 2-, 4-, 8-, and 15-days of SI¹⁹. Furthermore, adult rats showed an overall down-regulation of phosphorylated and total mTOR and p70S6k and total AKT protein levels in the soleus 10 weeks after complete spinal cord transection³², which is in contrast to an up-regulation of phosphorylated mTOR and p70S6k and an overall down-regulation of phosphorylated AKT and up-regulation of total AKT observed in the plantaris in this study. Biopsies taken from the vastus lateralis muscle from chronic spinal cord injured patients (3 mos to 30 yrs) also demonstrated reduced expression of phosphorylated and total p70S6k and 4E-BP proteins, and this occurred independent of Akt and mTOR activity, with no change observed in the expression of either protein³³. Combined, these results indicate that there is no clear understanding of the role these protein synthesis markers are playing in regulating muscle mass in muscles subjected to chronic levels of decreased neuromuscular activity, as their expression is highly variable and appears to be specific to the muscle analyzed.

Electromechanical stimulation returned IGF-1 mRNA levels to near Con values, and there were no concomitant changes observed in phosphorylated Akt expression. Similarly, phosphorylated mTOR, p70S6k and 4E-BP protein levels returned to near Con values, and relative plantaris weight was not different from Con. Intermittent mechanical stimulation (passive stretch) of the extensor digitorum longus ex vivo activated mTOR-dependent signaling events independent of changes in Akt activity, and this activation was necessary for increasing the translational efficiency as assessed by total RNA content and p70S6k phosphorylation³⁴. The same stimulation paradigm administered to the extensor digitorum longus muscles in Akt^−/− mice also showed a mechanically induced increase in phosphorylated p70S6k. Although the direction of changes in p70S6k phosphorylation differs between studies, the mechanism may be similar – stimulation was effective in reversing the changes induced by SI, and Hornberger et al.³⁴ showed increases in markers of protein synthesis via mechanical stimulation in an intact system. Combined, these data demonstrate that mechanical stimulation can activate protein synthesis markers in either situation via an mTOR-mediated mechanism and that Akt is not required. The changes observed in these factors after SI, however, are not well understood with respect to their role in muscle protein homeostasis, and they appear paradoxical.

Protein degradation markers are differentially regulated in chronically inactive muscles

Downstream targets of the IGF-1/PI3K/Akt pathway include the atrophy-related ubiquitin ligases atrogin-1 and MuRF-1^6,10,11. Atrogin-1 and MuRF-1 mRNA expression levels in the SI groups were not different from Con for both plantaris and soleus, consistent with no changes observed in these transcripts after 10 weeks of spinal cord transection in the soleus ³⁵. However, these data are in contrast to the significant up-regulation of these atrogenes observed in soleus and medial gastrocnemius in acute SI and denervation studies^19,20,36. Previous time-course studies showed that after the initial rapid rate of atrophy induced by SI (4 days), atrogin-1 mRNA expression increased significantly and then returned to near baseline at ~14 days^19,36. A similar trend was observed for MuRF-1 mRNA expression. Interestingly, Leger and colleagues³³ observed a significant down-regulation of atrogin-1 and MuRF-1 mRNA expression in the vastus lateralis muscle relative to control in chronic complete spinal cord injured patients that the authors suggest may be due to an internal mechanism aimed at reducing further loss in muscle proteins. Combined, these data suggest that the expression levels of these atrogenes appear to have a clear role in modulating muscle mass during the early rapid phase of muscle atrophy, but their role in regulating muscle mass homeostasis in models of chronic disuse remain largely undefined.

Myostatin, a negative regulator of skeletal muscle mass³⁷, differentially affected the soleus and plantaris during 30 days of SI. Myostatin mRNA expression was significantly lower in soleus despite a 65% loss in mass, whereas it was unchanged in plantaris despite a 52% loss in mass. Interestingly, there was no change in myostatin mRNA expression in the soleus after 10 weeks of spinal cord transection³⁵. In contrast, no change was observed in myostatin mRNA expression after 7 days of SI in medial gastrocnemius and soleus despite a 41% and 39% loss in mass, respectively ²⁰. There also was no change in myostatin mRNA expression in the soleus muscle hindlimb unloaded for 7 days despite undergoing 42% atrophy³⁸. Myostatin mRNA levels are higher in predominantly slow than predominantly fast muscles in control rats^{20,39 and present study}. SI and denervation induce a type I to type II shift in myosin heavy chain composition in soleus^12,40, and the observed decrease in myostatin mRNA expression in soleus, but not plantaris, supports this fiber-type specific expression. Similarly, myostatin protein expression decreased markedly in soleus, but not medial gastrocnemius and plantaris, 28 days after denervation³⁹. Furthermore, 7 days of SI¹² and hindlimb unloading⁴¹ is insufficient to induce a shift in the MHC composition from type I to type II in the soleus, and no significant changes occurred at ~14 days. This supports findings in acute studies that show no change in myostatin mRNA expression despite significant muscle atrophy^20,38. Together, these results indicate that modulation of myostatin expression is dependent on the type of muscle (slow vs. fast), and type (SI, denervation, spinal cord transection) and duration (acute vs. chronic) of the models of disuse.

Perspective

It is clear that the IGF-1/PI3K/Akt pathway modulates genes associated with protein synthesis and degradation, and plays an essential role in defining the relative atrophy-hypertrophy state of skeletal muscles. Our data, however, demonstrate how incomplete the dynamics of the regulation of these factors and pathways in vivo are understood. Figure 9 summarizes the changes observed in the expression of select genes examined in the present study after SI and electromechanical stimulation in the plantaris muscle. It is apparent that the short- and long-term dynamics and the physiological triggers that modulate these pathway dynamics must be determined more clearly to further elucidate these regulatory mechanisms.

Figure 9.

Summary schematic illustrating the adaptations of select genes of the IGF-1/PI3K/Akt pathway involved in protein synthesis and degradation analyzed in the plantaris muscle. Phosphorylated (p) proteins analyzed are noted in bold and dark gray and mRNA analyses are noted in bold and shaded in light gray. ◇ indicates up-regulation after SI and down-regulation after electromechanical stimulation; ⇔ indicates no change after either SI or electromechanical stimulation; ▽ indicates down-regulation after SI and electromechanical stimulation.

Other key mechanistic questions are raised by the differential responses of slow vs. fast muscles to the once and twice per day stimulation paradigms. The absent, or small, effect of stimulation on soleus mass suggests that neither of these growth-related factors of the IGF-1/PI3K/Akt pathway played a critical role in defining the different responses of a slow vs. fast muscle to brief periods of simulation administered over 30 days. Similarly, the role that atrogin-1, MuRF-1, and myostatin may play in defining the ability of the muscle to maintain its mass when receiving brief periods of stimulation was not evident in either muscle. Combined, these results indicate that there must be other important mechanisms that control the mass of a muscle when it is in a relatively constant state of atrophy that are not reflected in the assays used in this study, e.g., FoxO, locally acting factors, mechanoreceptors. These results indicate that a detailed analysis of the dynamics of regulation of neurally mediated activity-dependent and activity-independent growth- and differentiation-associated genes is needed.

LIST OF ABBREVIATIONS

4E-BP

4E-binding protein

BP-4

IGF-1 binding protein 4

BP-5

IGF-1 binding protein 5

Con

control

Con-P1

control stimulation protocol 1

Con-P2

control stimulation protocol 2

FOXO

forkhead transcription factor

IGF-1

insulin-like growth factor-1

MGF

mechano growth factor

mTOR

mammalian target of rapamycin

MuRF-1

muscle-specific ring finger 1

PI3K

phosphatidylinositol-3-kinase

RT-PCR

reverse-transcription-polymerase chain reaction

SI

spinal cord isolation

SI-C

spinal cord isolated control

SI-C1

spinal cord isolated control 1

SI-C2

spinal cord isolated control 2

SI-Stim1

spinal cord isolated and stimulation paradigm 1

SI-Stim2

spinal cord isolated and stimulation paradigm 2